__group__,ticket,summary,component,version,type,severity,owner,status,created,_changetime,_description,_reporter 1.0 Release,9619,"Side-by-side editing fails if Preview significantly longer than source; needs smart, vertical locking!",wiki system,0.12-stable,defect,major,Christian Boos,reopened,2010-09-15T12:10:03+02:00,2023-01-10T14:50:43+01:00,"The side-by-side editing is rendered useless, if the Preview is significantly longer (due to large text, tables and inline images) than the source code. You might end up in a situation where the related source code to what you currently see in the Preview is 1 or 2 pages up the screen and it is currently impossible to see source code and related Preview together at once. This is a serious issue in my opinion. You can make a simple test by positioning a few large images using ![[Image(XYZ.jpg)]] underneath each other: The source code only takes a few lines, whereas the last image might by two screens down the window scroll bar. A solution should look like this: The preview get dynamically positioned in such a way, that it (optionally) aligns vertically with the cursor position in the source code window, so that one really sees what one's editing.",klein@… 1.6.1 Release,12681,Restore the t.e.o specific customization after the switch to Jinja2,project,1.3dev,defect,normal,Christian Boos,assigned,2017-02-08T10:55:04+01:00,2023-09-23T22:47:00+02:00,cf. TracDev/Proposals/Jinja/TracInterfaceCustomization,Christian Boos 1.6.1 Release,13613,Remove code for creating Windows installers,general,,enhancement,normal,Ryan J Ollos,assigned,2023-09-23T22:59:55+02:00,2023-09-23T22:59:55+02:00,Remove code in Makefile and elsewhere for creating the Windows installers.,Ryan J Ollos 1.6.1 Release,13607,sqlite3.version_info is deprecated since Python 3.12 and will be removed in Python 3.14,database backend,,defect,normal,Jun Omae,assigned,2023-05-31T10:34:48+02:00,2023-05-31T19:23:26+02:00,"https://docs.python.org/3.12/library/sqlite3.html#sqlite3.version_info After the following patch, it will retrieve `pysqlite3.version_info` only when `pysqlite3` is used. {{{#!diff diff --git a/contrib/make_status.py b/contrib/make_status.py index 65278624a..ab4962a8d 100755 --- a/contrib/make_status.py +++ b/contrib/make_status.py @@ -51,8 +51,7 @@ PACKAGES = [ (""Jinja2"", 'jinja2.__version__'), (""multipart"", 'multipart.__version__'), (""Babel"", 'babel.__version__'), - (""sqlite3"", ('sqlite3.version', - 'sqlite3.sqlite_version')), + (""sqlite3"", 'sqlite3.sqlite_version'), (""PySqlite3"", ('__main__._pysqlite3_version()', 'pysqlite3.dbapi2.sqlite_version')), (""PyMySQL"", 'pymysql.__version__'), diff --git a/trac/db/sqlite_backend.py b/trac/db/sqlite_backend.py index dc0bf4ca9..8f38af387 100644 --- a/trac/db/sqlite_backend.py +++ b/trac/db/sqlite_backend.py @@ -37,13 +37,13 @@ try: import pysqlite3.dbapi2 as sqlite except ImportError: import sqlite3 as sqlite + pysqlite3_version = None +else: + pysqlite3_version = get_pkginfo(sqlite).get('version', '%d.%d.%s' % + sqlite.version_info) sqlite_version = sqlite.sqlite_version_info sqlite_version_string = sqlite.sqlite_version -pysqlite_version = sqlite.version_info -pysqlite_version_string = get_pkginfo(sqlite).get('version', - '%d.%d.%s' - % pysqlite_version) min_sqlite_version = (3, 0, 0) @@ -273,7 +273,8 @@ class SQLiteConnector(Component): def get_system_info(self): yield 'SQLite', sqlite_version_string - yield 'pysqlite', pysqlite_version_string + if pysqlite3_version: + yield 'pysqlite3', pysqlite3_version @lazy def _extensions(self): diff --git a/trac/tests/env.py b/trac/tests/env.py index e0f949949..80a700208 100644 --- a/trac/tests/env.py +++ b/trac/tests/env.py @@ -520,8 +520,6 @@ class SystemInfoTestCase(unittest.TestCase): self.assertEqual(info_before, info_after) self.assertRegex(get_info(info_before, 'SQLite'), r'^\d+(\.\d+)+$') - self.assertRegex(get_info(info_before, 'pysqlite'), - r'^\d+(\.\d+)+$') else: self.fail(""Unknown value for dburi %s"" % self.env.dburi) }}}",Jun Omae 1.6.1 Release,13606,DeprecationWarning: pkg_resources is deprecated as an API,general,,defect,normal,,new,2023-05-31T05:27:03+02:00,2023-05-31T11:02:14+02:00,"`setuptools` warns `pkg_resources is deprecated as an API` since 67.5.0. - https://github.com/pypa/setuptools/commit/a1aeda391a0c462ea53627bcdf50dd4c0daadaed - https://github.com/pypa/setuptools/blob/v67.8.0/pkg_resources/__init__.py#L16-L20 ==== Migration guide - https://importlib-resources.readthedocs.io/en/latest/migration.html - https://importlib-metadata.readthedocs.io/en/latest/migration.html ---- {{{ $ git grep -w pkg_resources | wc -l 86 $ git grep -h '^from *pkg_resources\>\|\ | sed -e 's/.*\(\\).*/\1/' \ > | sort -u from pkg_resources import DistributionNotFound from pkg_resources import DistributionNotFound, get_distribution from pkg_resources import parse_version from pkg_resources import parse_version as pv from pkg_resources import resource_exists, resource_filename from pkg_resources import resource_filename from pkg_resources import resource_listdir, resource_string from pkg_resources import working_set, DistributionNotFound, \ pkg_resources.Distribution pkg_resources.DistributionNotFound pkg_resources.Environment pkg_resources.find_distributions pkg_resources.get_distribution pkg_resources.parse_version pkg_resources.require pkg_resources.resource_exists pkg_resources.resource_filename pkg_resources.resource_listdir pkg_resources.set_extraction_path pkg_resources.working_set }}}",Jun Omae 1.6.1 Release,13599,The first parameter in the API docs for the interface's methods is incorrectly stripped,general,,defect,normal,,new,2023-05-09T04:27:17+02:00,2023-05-13T19:35:14+02:00," - https://www.edgewall.org/docs/trac-trunk/html/api/trac_web_api.html#trac.web.api.IRequestHandler.match_request - Missing `req` parameter - https://trac.readthedocs.io/en/latest/api/trac_web_api.html#trac.web.api.IRequestHandler.match_request - The API docs outdated but `req` parameter is not stripped",Jun Omae 1.6.1 Release,13567,Send summary of workflow to trac-builds mailing list on workflow_run event,project,,enhancement,normal,,new,2023-02-12T12:30:09+01:00,2023-02-24T08:33:31+01:00,"The [wiki:TracDev/AutomaticBuilds automatic builds] has been restored, however trac-builds mailing list is not working yet. I created a new workflow to send the summary to trac-builds on `workflow_run` event (smtp server and the user to send are required). Proposed changes in [678ea4140/jomae.git] ([log:jomae.git@t13567]). ==== Configuration: - `SMTP_URL`: set `smtps://:465` or `smtp://:587` - `SMTP_USER`: set `:` to authenticate for the smtp server. - `SMTP_MAIL_FROM`: set sender address - `SMTP_MAIL_RCPT`: set `trac-builds@googlegroups.com` build-completed.yml workflow in the changes works once committing to trunk (default branch).",Jun Omae 1.6.1 Release,13564,TicketQuery with format=progress doesn't show progress when no matched ticket (part of #13474),query system,1.4.3,defect,normal,,new,2023-02-11T03:21:35+01:00,2023-09-23T22:47:00+02:00,"Copied from [ticket:13474#comment:4]: > Now, with the implemented fix, a progress bar for a query that yields no tickets looks as follows: > [[Image(ticket:13474:fix_1.png,border=1)]] > > It is definitely better than a python error message but it does not look especially pretty. One could try to display an empty progress bar here instead? > > I would not opt for a ""No Results"" message because when using `format=table` we get an empty table and not ""No Results"". An empty progress bar would be more consistent. > > Open to suggestions! What is the expected behavior? Or would handling None values in `templates/progress_bar.html` be more preferable? ",Jun Omae 1.6.1 Release,13563,ResourceWarning raised when tracd is killed by CTRL-C (SIGINT),web frontend/tracd,1.5.3,defect,normal,,new,2023-02-11T02:46:30+01:00,2023-02-23T22:38:38+01:00,"`ResourceWarning` raised when `tracd` is killed by CTRL-C, however not raised when killed by SIGTERM. {{{ $ python -Wdefault trac/web/standalone.py -p 3000 /var/trac/1.5-sqlite Server starting in PID 184511. Serving on 0.0.0.0:3000 view at http://127.0.0.1:3000/ Using HTTP/1.1 protocol version ^Ctrac/web/standalone.py:385: ResourceWarning: unclosed pass ResourceWarning: Enable tracemalloc to get the object allocation traceback }}}",Jun Omae 1.6.1 Release,13550,OverflowError: Python int too large to convert to SQLite INTEGER,database backend,1.0.20,defect,minor,,new,2023-02-02T01:38:54+01:00,2023-09-23T22:47:00+02:00,"Passing a large integer to `version` parameter of wiki handler, `OverflowError` is raised. {{{ 09:20:38 AM Trac[main] ERROR: [192.168.11.15] Internal Server Error: , referrer 'http://192.168.11.122:3000/1.5-sqlite/wiki/WikiStart?version=999999999999999999999999999999999' Traceback (most recent call last): File ""trac/web/main.py"", line 610, in dispatch_request dispatcher.dispatch(req) File ""trac/web/main.py"", line 302, in dispatch raise e File ""trac/web/main.py"", line 248, in dispatch resp = chosen_handler.process_request(req) File ""trac/wiki/web_ui.py"", line 126, in process_request versioned_page = WikiPage(self.env, pagename, version) File ""trac/wiki/model.py"", line 68, in __init__ self._fetch(name, version) File ""trac/wiki/model.py"", line 87, in _fetch self.env.db_query(sql, args): File ""trac/db/api.py"", line 50, in execute return db.execute(query, params) File ""trac/db/util.py"", line 129, in execute cursor.execute(query, params if params is not None else []) File ""trac/db/util.py"", line 62, in execute r = self.cursor.execute(sql_escape_percent(sql), args) File ""trac/db/sqlite_backend.py"", line 89, in execute result = PyFormatCursor.execute(self, *args) File ""trac/db/sqlite_backend.py"", line 64, in execute return self._rollback_on_error(sqlite.Cursor.execute, sql, File ""trac/db/sqlite_backend.py"", line 56, in _rollback_on_error return function(self, *args, **kwargs) OverflowError: Python int too large to convert to SQLite INTEGER }}}",Jun Omae 1.6.1 Release,13514,Support zoneinfo tzinfo objects since Python 3.9,general,,enhancement,normal,,new,2023-01-11T04:08:16+01:00,2023-02-26T05:51:42+01:00," - [https://docs.python.org/3/library/zoneinfo.html#module-zoneinfo zoneinfo — IANA time zone support] - [https://peps.python.org/pep-0615/ PEP 615 – Support for the IANA Time Zone Database in the Standard Library | peps.python.org] - [https://pytz-deprecation-shim.readthedocs.io/en/latest/migration.html Migration Guide — pytz_deprecation_shim 0.1.0.post0 documentation]",Jun Omae 1.6.1 Release,13482,6 failures in unit-test with Babel 2.10.1,i18n,,defect,normal,Jun Omae,assigned,2022-04-24T01:10:16+02:00,2023-02-23T22:38:38+01:00,"[https://github.com/python-babel/babel/releases/tag/v2.10.1 Babel 2.10.1] has been released recently, which upgraded [https://github.com/python-babel/babel/blob/v2.10.1/CHANGES.rst#version-2100 CLDR 41] from [https://github.com/python-babel/babel/blob/v2.10.1/CHANGES.rst#version-291 CLDR 37]. ---- {{{ ====================================================================== FAIL: test_i18n_datetime_hint (trac.util.tests.datefmt.I18nDateFormatTestCase) ---------------------------------------------------------------------- Traceback (most recent call last): File ""/home/jun66j5/src/tracdev/git/trac/util/tests/datefmt.py"", line 1250, in test_i18n_datetime_hint self.assertIn(datefmt.get_datetime_format_hint(fr), AssertionError: 'd MMM y, HH:mm:ss' not found in ('d MMM yyyy HH:mm:ss', 'd MMM y HH:mm:ss', ""d MMM y 'à' HH:mm:ss"") ====================================================================== FAIL: test_i18n_format_datetime (trac.util.tests.datefmt.I18nDateFormatTestCase) ---------------------------------------------------------------------- Traceback (most recent call last): File ""/home/jun66j5/src/tracdev/git/trac/util/tests/datefmt.py"", line 1175, in test_i18n_format_datetime self.assertIn(datefmt.format_datetime(t, tzinfo=tz, locale=fr), AssertionError: '28 août 2010, 13:45:56' not found in ('28 août 2010 13:45:56', '28 août 2010 à 13:45:56') ====================================================================== FAIL: test_i18n_format_time (trac.util.tests.datefmt.I18nDateFormatTestCase) ---------------------------------------------------------------------- Traceback (most recent call last): File ""/home/jun66j5/src/tracdev/git/trac/util/tests/datefmt.py"", line 1233, in test_i18n_format_time self.assertIn(datefmt.format_time(t, tzinfo=tz, locale=zh_CN), AssertionError: '13:45:56' not found in ('下午01:45:56', '下午1:45:56') ====================================================================== FAIL: test_i18n_parse_date_datetime (trac.util.tests.datefmt.I18nDateFormatTestCase) ---------------------------------------------------------------------- Traceback (most recent call last): File ""/home/jun66j5/src/tracdev/git/trac/util/tests/datefmt.py"", line 1367, in test_i18n_parse_date_datetime self.assertEqual(expected, AssertionError: datet[16 chars]0, 8, 28, 13, 45, 56, tzinfo=) != datet[16 chars]0, 8, 28, 1, 45, 56, tzinfo=) ====================================================================== FAIL: test_i18n_parse_date_datetime_meridiem (trac.util.tests.datefmt.I18nDateFormatTestCase) ---------------------------------------------------------------------- Traceback (most recent call last): File ""/home/jun66j5/src/tracdev/git/trac/util/tests/datefmt.py"", line 1398, in test_i18n_parse_date_datetime_meridiem self.assertEqual(expected_am, AssertionError: datet[16 chars]1, 2, 22, 0, 45, 56, tzinfo=) != datet[16 chars]1, 2, 22, 12, 45, 56, tzinfo=) ====================================================================== FAIL: test_i18n_parse_date_roundtrip (trac.util.tests.datefmt.I18nDateFormatTestCase) ---------------------------------------------------------------------- Traceback (most recent call last): File ""/home/jun66j5/src/tracdev/git/trac/util/tests/datefmt.py"", line 1454, in test_i18n_parse_date_roundtrip self.assertEqual(tz_t, actual, AssertionError: datet[16 chars]0, 8, 28, 13, 45, 56, tzinfo=) != datet[16 chars]0, 8, 28, 1, 45, 56, tzinfo=) : datetime.datetime(2010, 8, 28, 13, 45, 56, tzinfo=) != datetime.datetime(2010, 8, 28, 1, 45, 56, tzinfo=) ('2010年8月28日 B1:45:56' Locale('zh', territory='TW', script='Hant')) ---------------------------------------------------------------------- Ran 2762 tests in 41.802s FAILED (failures=6, skipped=5) }}}",Jun Omae 1.6.1 Release,13456,distutils is deprecated and will be removed in Python 3.12,general,1.5.3,defect,normal,Jun Omae,assigned,2022-01-04T01:58:05+01:00,2023-07-14T11:42:33+02:00,"https://docs.python.org/3/whatsnew/3.10.html#distutils-deprecated {{{ $ git grep -w distutils mirror/trunk mirror/trunk:.travis.yml: print_sitelib='from distutils.sysconfig import get_python_lib; print(get_python_lib())' mirror/trunk:trac/dist.py:`distutils.command.build` and `setuptools.command.install_lib` classes mirror/trunk:trac/dist.py:from distutils import log as distlog mirror/trunk:trac/dist.py:from distutils.cmd import Command mirror/trunk:trac/dist.py:from distutils.command.build import build as _build mirror/trunk:trac/dist.py:from distutils.errors import DistutilsOptionError }}}",Jun Omae 1.6.1 Release,13430,Translation of past time expressions,i18n,1.4,defect,normal,,new,2021-10-10T12:17:20+02:00,2023-02-23T22:38:38+01:00,"If someone comments on a ticket, the timestamp is stored and gives a human-readable output when was something made. It's a great feature, but it seems the time parts are translated separately. Example: `2 months ago` appears, and translated as `2 `. However, in Hungarian (possibly many more languages), the `months` and `ago` can't be translated separately, because it's an expression. I suggest merging the two translation units into one because the current implementation results in mixed translations. Possibly there are other places than comments when this appears, but I don't know Trac well enough to tell all of them :(",anonymous 1.6.1 Release,13406,Replace jQuery UI Timepicker,web frontend,,enhancement,normal,Ryan J Ollos,assigned,2021-05-27T22:38:42+02:00,2023-02-23T22:31:25+01:00,"The [https://trentrichardson.com/examples/timepicker/ Timepicker add-on] is deprecated (see JavaScript). Since we are doing a major release, we should replace it with an equivalent that is maintained.",Ryan J Ollos 1.6.1 Release,13349,Changesets are shown without repository name in search view,search system,,defect,normal,,new,2021-02-15T09:14:33+01:00,2023-09-23T22:47:00+02:00,"I think we should show changesets with repository name. Currently, all changesets are shown without repository name even if the repository is not default. Example:: `[662]: merge r661 from trunk into 0.9 maintenance branch` in demo-1.2:search:maintenance?noquickjump=1&changeset=on Should be:: `[662/babel.svn]: merge r661 from trunk into 0.9 maintenance branch` {{{#!diff diff --git a/trac/versioncontrol/web_ui/changeset.py b/trac/versioncontrol/web_ui/changeset.py index 7563f2351..271e3bbcb 100644 --- a/trac/versioncontrol/web_ui/changeset.py +++ b/trac/versioncontrol/web_ui/changeset.py @@ -1154,6 +1154,8 @@ class ChangesetModule(Component): cset = repos.resource.child(self.realm, rev) if 'CHANGESET_VIEW' in req.perm(cset): uids_seen.add(uid) + if repos.reponame: + drev = '%s/%s' % (drev, repos.reponame) yield (req.href.changeset(rev, repos.reponame or None), '[%s]: %s' % (drev, shorten_line(log)), from_utimestamp(ts), author, }}}",Jun Omae 1.6.1 Release,13332,Use pytidylib for HTML5 validation with jinjachecker,general,,enhancement,normal,Jun Omae,assigned,2020-12-19T04:06:40+01:00,2023-02-27T00:53:42+01:00,Discussed in comment:30:ticket:11988.,Ryan J Ollos 1.6.1 Release,13311,Toolbar button to quote text as a discussion citation,general,,enhancement,normal,Ryan J Ollos,assigned,2020-06-04T15:53:40+02:00,2023-02-23T22:31:25+01:00,"All Trac text input fields (like this ticket description field) have a little toolbar with buttons to make text bold or italic, insert a heading, a link, a code block, a horizontal rule, a new paragraph, a line break or an image. It would be useful to have an additional button to quote the currently selected text as [WikiFormatting#DiscussionCitations DiscussionCitations] by adding `>` before each selected line.",anonymous 1.6.1 Release,13287,MS Outlook improvements for 1.4.1 shows only empty fields,general,1.4.1,defect,normal,,new,2020-05-07T14:41:52+02:00,2021-05-28T21:54:30+02:00,"Dear Ryan, I updated the e-mail template by replacing the `${ticket_props}` as suggested here in TracNotification#CustomizingContent, with CookBook/Notification/Templates. {{{ -------------------------------------------------------------------------- # set sel = ('Reporter', 'Owner', 'Type', 'Status', 'Priority', 'Milestone', 'Component', 'Severity', 'Resolution', 'Keywords') # for a in ticket_props.replace('|', '\n').splitlines()[1:-1]|map('trim'): # set a = a.split(':', 1)|map('trim')|list # if len(a) > 1 and a[0] in sel: ${'%-*s %s'|format(12, a[0] + ':', a[1])} # endif # endfor -------------------------------------------------------------------------- }}} So, notification works - no error, but I have only empty fields in the e-mail as it is shown here below: {{{ #1703: NEST for coping with bunching of regulated traffic -------------------------------------------------------------------------- Reporter: Owner: Type: Status: Priority: Milestone: Component: Severity: Resolution: Keywords: -------------------------------------------------------------------------- Comment (by Laszlo Elbert): testx -- Ticket URL: NEST NEST project }}} Thanks for the advise in advance Laszlo",laszlo.elbert@… 1.6.1 Release,13275,Improve option documentation,general,,task,normal,Ryan J Ollos,assigned,2020-04-25T04:11:04+02:00,2023-02-23T22:31:25+01:00,"For 1.5.x there are a few tasks we could do on the option doc strings: * Review and improve * Ensure special values are documented. Example, what does the value `0` mean for [TracIni#changeset-max_diff_files-option changeset.max_diff_files]? * Remove //(since ...)// for versions < 1.1",Ryan J Ollos 1.6.1 Release,13270,Default value column doesn't wrap on TracIni page,general,,defect,normal,,new,2020-04-22T21:21:21+02:00,2021-05-28T21:54:30+02:00,See TracIni#trac-section. It appears to expand to the width of TracIni#trac-permission_policies-option.,Ryan J Ollos 1.6.1 Release,13231,Egg files without SOURCES.txt not displayed correctly on plugin admin page,general,,defect,normal,,new,2019-12-05T05:33:03+01:00,2021-05-28T21:54:30+02:00,"Discussed in gmessage:trac-users:AUBRFehR83c/ZRjKm-SnCAAJ Trac Debian package should have `*.egg-info/SOURCES.txt` file. Searching the `SOURCES.txt` file in debian packages, several packages have SOURCES.txt file. https://packages.debian.org/file:.egg-info/SOURCES.txt (If *.dist-info structure is used, RECORD file must be included. (2)) However, according to The Internal Structure of Python Eggs (1), it seems that it is possible to have no `SOURCES.txt` file in egg-info structure. Work around: {{{#!diff diff --git a/trac/util/__init__.py b/trac/util/__init__.py index 1d3485e09..8808ebf90 100644 --- a/trac/util/__init__.py +++ b/trac/util/__init__.py @@ -700,6 +700,7 @@ def get_sources(path): """"""Return a dictionary mapping Python module source paths to the distributions that contain them. """""" + path = os.path.normpath(os.path.abspath(path)) sources = {} for dist in find_distributions(path, only=True): if not dist.has_metadata('top_level.txt'): @@ -717,6 +718,18 @@ def get_sources(path): for row in reader if any(row[0].startswith(top) for top in toplevels)) continue + for top in toplevels: + dir_ = os.path.normpath(os.path.join(dist.location, + top.rstrip('/'))) + if not os.path.isdir(dir_): + continue + for root, dirs, files in os.walk(dir_): + for filename in files: + if not filename.endswith('.py'): + continue + filename = os.path.join(root, filename) + sources[filename[len(path):].strip(os.sep) + .replace(os.sep, '/')] = dist return sources }}} 1. https://svn.python.org/projects/sandbox/trunk/setuptools/doc/formats.txt 1. https://www.python.org/dev/peps/pep-0376/#one-dist-info-directory-per-installed-distribution",Ryan J Ollos 1.6.1 Release,9740,Impossible to CC a username with spaces,ticket system,0.12-stable,defect,normal,,new,2010-11-01T14:22:20+01:00,2021-05-28T21:54:30+02:00,"It is impossible to CC a username with spaces in the username such as ""Joe Bloggs"", attempting to enter that treats as two different users Joe and Bloggs. Suggest first attempt to establish that usernames separated by commas as known usernames before splitting by spaces for known usernames.",david+trac@… 1.7.1 Release,13600,Upgrade jQuery to 3.7.1,general,,enhancement,normal,,new,2023-05-11T21:49:00+02:00,2023-08-29T23:05:03+02:00,https://blog.jquery.com/2023/08/28/jquery-3-7-1-released-reliable-table-row-dimensions/,Ryan J Ollos 1.7.1 Release,13282,Ticket timeline events can't be filtered using a permission policy,ticket system,,defect,normal,,new,2020-05-02T05:05:15+02:00,2021-05-28T21:28:12+02:00,I'm trying to implement a solution for private ticket comments (th:#13776) and finding that the permission checking for timeline events is too coarse to implement a permissions policy that filters by resource realm `comment` and resource id `cnum`.,Ryan J Ollos 1.7.1 Release,13273,"Add a ""trac-admin session rename"" command",admin/console,,enhancement,normal,,new,2020-04-24T07:23:07+02:00,2022-07-08T20:37:06+02:00,The `session rename` command would be particularly useful if it updates table entries that store the `sid`.,Ryan J Ollos 1.7.1 Release,13215,Adding a repository does NOT trim trailing blanks,version control,1.2.2,defect,normal,,new,2019-10-27T15:57:50+01:00,2021-05-28T21:28:12+02:00,"Adding a new git repository fails to remove trailing blanks from the repository path. It took some time for me to figure out. This happened with trac 1.2.2, I didn't test newer versions",strk@… 1.7.1 Release,13012,Allow direct docutils configuration,wiki system,,enhancement,normal,,new,2018-04-20T01:15:22+02:00,2021-05-28T21:28:12+02:00,"Add an option for the path to a docutils configuration file. This isn't strictly needed because there are [http://docutils.sourceforge.net/docs/user/config.html#configuration-files other ways to specify a configuration], but it might be a convenience. Discussed in gmessage:trac-users:bxWbIZwZbh0/VHnIdD-4AAAJ.",Ryan J Ollos 1.7.1 Release,12925,Admin pages should add warnings rather than raising TracErrors,admin/web,,enhancement,normal,,new,2017-09-27T19:00:53+02:00,2021-05-28T21:28:12+02:00,"Noticed while working on #12923, we can make this more user-friendly by trapping `TracError`s and `add_warning`s.",Ryan J Ollos 1.7.1 Release,12406,Standardize nomenclature/options for cached repository,version control,,enhancement,normal,,new,2016-03-16T22:25:08+01:00,2022-03-22T03:19:05+01:00,"Cached SVN repository is specified using the type //svn//. Cached Git repository is specified using the option `[git]` `cached_repository`. I think the latter is more clear, however it can't be specified per-repository. In order to be more consistent, we could make `cached` a repository attribute, and use it for Git and Subversion repositories. We could allow the attribute to be set from the //Manage Repositories// admin page. An upgrade could be added to migrate users to the new configuration. Also, the admin panel shows two additional types: //direct-svnfs// and //svnfs//. Are these just aliases?",Ryan J Ollos 1.7.1 Release,11469,Add a custom field admin panel,ticket system,,enhancement,normal,,new,2014-02-03T05:27:06+01:00,2022-11-16T06:31:12+01:00,"The th:CustomFieldAdminPlugin seems to be a reasonable start for integrating a custom field admin panel to the Trac core. Here is an outline of the changes: * Rework the template for minor enhancements (e.g. #11027 and #11056), improved styling and to make the markup more consistent with the other ticket `admin_*` templates. * Add a `CustomField` class to `trac.ticket.model` that wraps the configuration. Later on, the storage can be changes from //trac.ini// to the database without changing the interface. * Modify the `CustomFieldAdminPanel` class to inherit from `TicketAdminPanel` and refactor. * Add functional test cases. Proposed changes will be posted within the next few days.",Ryan J Ollos 1.7.1 Release,10136,Way to configure by default displayed branch,version control/browser,0.12.2,enhancement,normal,,new,2011-04-18T02:19:33+02:00,2022-06-14T01:15:13+02:00,"For (at least Mercurial) it would be great to be able to configure which branch should be displayed as default in source code browser. Currently tip is displayed, but sometimes it is useful that default is some more ""stable"" branch.",Mitar 1.7.1 Release,1660,Patch to get email notification (with diff) of wiki changes,notification,devel,enhancement,major,,new,2005-06-10T19:43:58+02:00,2021-05-28T21:28:12+02:00,"This is a little patch to get notification, with diff included, of wiki changes. That's implemented against trunk/, revision 1777, as a component. I did'nt know where to put plugins, so I put it in Notify.py (I use NotifyEmail class), but I don't like that too much ... (patch follows ...)",pierre.palatin+trac@… next-dev-1.7.x Release,10267,Version controls diffs of large text files kill the system,version control/changeset view,0.12.2,defect,major,,new,2011-07-15T18:25:07+02:00,2020-12-29T09:45:39+01:00,"We have some large text files (they are some data files we deal with) and when you try to view ""last revision"" or a revision with them it bombs (i.e. the service's cpu usage goes to 100%, and things come to a screeching halt). Looks like the issue is in: versioncontrol/web_ui/changeset.py The `_content_changes` method tries to do a diff of anything that isn't a binary. This is fine for most text files, unless they are big (like even 5 to 10 megs seems a bit much for the diffing method used). Anyway, I simply added the following to `_content_changes`: {{{#!python if (len(old_content) > 100000 or len(new_content) > 100000): return None }}} (Sorry if it isn't proper pep8) It looks like you have a max_diff_bytes but that only effects the diff results. Ideally you would want a threshold before you have the diff results. ",anonymous next-dev-1.7.x Release,13304,Implement code formatting and style checking tools,general,,enhancement,normal,,new,2020-05-26T00:44:06+02:00,2022-04-07T12:57:20+02:00,"''(from comment:93:ticket:12130)'' In the release checklist, `reindent` is used to fix indentation. ​pypi:reindent doesn't support Python 3. It appears it's just packaged ​reindent.py from the Python Core. I've emailed the package owner to see if I can get maintainership to update it for Python 3.5+. Use formatting and style checking tools such as: * [pypi:autopep8] * [pypi:pycodestyle] Other candidates are: * [pypi:black] * [pypi:flake8] * [pypi:autoflake8] * [pypi:pylint] Implement ignore rules, as needed, in `setup.cfg`. Example: {{{#!ini [pycodestyle] max-line-length = 80 ignore = E126, E127, E128, E129, E131, E241, E731 }}} Format code to remove all warnings and errors. Add a Makefile command `make check-code`, which executes the style check tool and `contrib/jinja2checker.py`. The tool can be run before submitting a ""pull request"" and before releases. Revise as needed: * TracDev/CodingStyle * TracDev/DevelopmentWorkflow * TracDev/ReleaseChecklist * TracDev/SubmittingPatches * TracDev/UnitTests * TracDev/FunctionalTests `make status` echoes the version of pypi:coverage, but the tool isn't integrated into our development process, as far as I'm aware. It's use should be documented, or remove it from `make status`. {{{ $ make status Python: /Users/rjollos/.pyenv/shims/python Package Version ---------------------------------------------------------------------------- Python : 3.6.10 (default, Jan 27 2020, 16:14:13) : [GCC 4.2.1 Compatible Apple LLVM 11.0.0 (clang-1100.0.33.17)] Setuptools : 46.4.0 Pip : 20.1.1 Wheel : 0.34.2 Jinja2 : 2.11.2 Babel : 2.8.0 sqlite3 : 2.6.0 (3.28.0) PySqlite : not installed PyMySQL : 0.9.3 Psycopg2 : 2.8.5 (dt dec pq3 ext lo64) SVN bindings : 1.15.0 (under development) Mercurial : 5.4 Pygments : 2.6.1 Textile : 4.0.1 Pytz : 2020.1 Docutils : 0.16 Selenium : 3.141.0 LXML : 4.5.1 coverage : 5.1 Variables: ... }}} There is a `make test-coverage` command that runs unit tests with coverage. Before r17381, it also ran functional tests and figleaf was used for functional test coverage, but figleaf was removed for reasons described in comment:19:ticket:11988. Consider adding a `requirements-dev.txt` for installing all development requirements.",Ryan J Ollos next-dev-1.7.x Release,13302,Display plugin enable/disable message near anchor,admin/web,,enhancement,normal,Ryan J Ollos,assigned,2020-05-20T08:08:07+02:00,2020-12-29T09:45:39+01:00,"See comment:1:ticket:11269. On submit the page refreshes and scrolls to the anchor (e.g. `/admin/general/plugin#no21`). The message is displayed at the top of the page, but is not visible. [[Image(Screen Shot 2020-05-19 at 23.07.28.jpg,100%)]] [[Image(Screen Shot 2020-05-19 at 23.06.33.jpg,100%)]]",Ryan J Ollos next-dev-1.7.x Release,13301,"Remove ""Can't synchronize"" warnings or only show if user has TRAC_ADMIN",version control,,enhancement,normal,Ryan J Ollos,assigned,2020-05-20T08:02:15+02:00,2020-12-29T09:45:39+01:00,"First mentioned in comment:18:ticket:11269. This warning describes a configuration problem and we shouldn't display it to a non-admin user.",Ryan J Ollos next-dev-1.7.x Release,13283,"Per-realm always_{cc, bcc} options",notification,,enhancement,normal,,new,2020-05-02T20:37:33+02:00,2020-12-29T09:45:39+01:00,"Copied from [ticket:9148#comment:4]: > I think it might be nice that `AlwaysEmailSubscriber` respects `[notification] smtp_always_{cc|bcc}.` option and falls back to `smtp_always_{cc|bcc}` option (e.g. `smtp_always_cc.ticket`, `smtp_always_cc.wiki`, ...). ",Ryan J Ollos next-dev-1.7.x Release,13277,Add a macro for rendering code from the repository,version control,,enhancement,normal,,new,2020-04-29T19:59:03+02:00,2021-06-02T19:31:57+02:00,"The macro should render code from the repository, optionally with line numbers. * Specify revision id to render * Specify line range to render * Render changesets from repository * Path as an optional argument to only render changeset for specified path Examples: * [th:IncludeMacro] * [th:CiteCodeMacro] * [th:CodeExampleMacro]",Ryan J Ollos next-dev-1.7.x Release,13208,Add methods or property that formats default value for NotificationSystem.smtp_subject_prefix,notification,,enhancement,normal,,new,2019-09-17T21:15:32+02:00,2022-01-24T00:19:06+01:00,"Implementing notification systems for some plugins, I found it would be useful to have a method that handles the `__default__` value. I've named the method `get_smtp_subject_prefix`, but a suitably named property might be better.",Ryan J Ollos next-dev-1.7.x Release,13206,Add ICustomFieldTypeProvider interface,ticket system,,enhancement,normal,,new,2019-09-06T20:51:38+02:00,2023-04-07T05:38:06+02:00,"Copied from [ticket:11469#comment:4]: > Replying to [comment:3 rjollos]: > > * th:#7156 - API for defining new types of custom fields? > > Previously, I've worked it in https://github.com/jun66j5/trac-customfieldadminplugin/commit/8532c814 by reference to FieldRefactoring. In the branch, it added `ICustomFieldTypeProvider` interface which allows to define new types. I'd like to consider rebasing these changes and including in a 1.5.x release. Use cases: * th:comment:8:ticket:10284",Ryan J Ollos next-dev-1.7.x Release,13185,Keywords: remove duplicates and fixup separators,ticket system,,enhancement,normal,,new,2019-07-30T23:10:45+02:00,2022-07-12T05:47:17+02:00,"In comment:16:ticket:13179, the following changes to the keywords field were proposed: * Fixup separators. Example: `kw1, kw2; kw3 kw4` -> `kw1, kw2, kw3, kw4` (or use whitespace rather than comma separators). This would match the behavior for the CC field. * Remove duplicates.",Ryan J Ollos next-dev-1.7.x Release,13081,IWikiMacroProvider.is_inline should take macro name as argument,general,1.0,defect,normal,,new,2018-09-19T23:47:30+02:00,2020-12-29T09:45:39+01:00,"`IWikiMacroProvider.is_inline` was added in r10905. Implementers of `IWikiMacroProvider` can define multiple macros, so the other interface methods take `name` as an argument. It would seem that `is_inline` should also take `name` as an argument since the different macros might produce different types of content.",Ryan J Ollos next-dev-1.7.x Release,13080,Add [trac] mysql_storage_engine option instead of editting my.cnf,database backend,,enhancement,major,,new,2018-09-19T09:39:01+02:00,2020-12-29T09:45:39+01:00,"Cloned from ticket:10993#comment:8: > This issue might be revisited in the future, for now I'm just summarizing the discussion in gdiscussion:trac-users:MO_KSSQMdnU. > > > The requirement to set `default_storage_engine` and `default_tmp_storage_engine` in `my.cnf` could be removed if: > > * The storage engine was specified in a trac.ini variable such as `[trac]` `mysql_storage_engine` or `[mysql]` `storage_engine` > > * `MySQLConnector.to_sql` was modified to include the `ENGINE` table option in `TABLE CREATE` statements > > * The modules in the `upgrade` package were modified to use the `ENGINE` table option either directly or indirectly by use of a helper function (such as `MySQLConnector.to_sql`) We could set the `default_storage_engine` and `default_tmp_storage_engine` variables in session on connecting database, to remove the requirement. {{{ SET @@session.default_storage_engine = 'InnoDB'; }}} PoC: [f877c7e1d/jomae.git] ([log:jomae.git@t10993_storage_option]). If `[trac] mysql_storage_engine` option is empty (default), the session variables is not set and global variables is referred.",Jun Omae next-dev-1.7.x Release,13029,Attachment notifications despite Never notify: I update a ticket,attachment,1.2.2,defect,normal,,new,2018-05-16T17:04:17+02:00,2020-12-29T09:45:39+01:00,"I have configured: Subscription rules: Never notify: I update a ticket And this works well most of the time, but I still get a notification when I attach a file to a ticket.",anonymous next-dev-1.7.x Release,13014,Deprecate and remove IPermissionStore.get_users_with_permission,general,,enhancement,normal,,new,2018-04-22T23:13:14+02:00,2021-05-28T21:29:49+02:00,"This is a follow-on to comment:15:ticket:12915. The proposed changes are: * Change `DefaultPermissionStore.get_user_permissions` so that `IPermissionGroupProviders` are not searched. `PermissionSystem.get_user_permissions` will search `IPermissionGroupProvider`s. `DefaultPermissionStore` will no longer have any knowledge of `IPermissionGroupProvider`s that it does not implement. * Deprecate `IPermissionStore.get_users_with_permissions` and remove it in 1.5.1. * The method was added in #4245. * [th:LdapPlugin] is the only plugin on trac-hacks.org that implements `IPermissionStore`, and it has [th:browser:/ldapplugin/0.12/ldapplugin/api.py@16527:225-232#L222 copied the implementation] of [browser:tags/trac-1.2.2/trac/perm.py@:217-224#L209 DefaultPermissionStore.get_users_with_permissions]. * `PermissionSystem.get_users_with_permission` can do the equivalent work for all `IPermissionStore`s. * Change `PermissionSystem.get_users_with_permission` so that it does not call `IPermissionStore.get_users_with_permissions`. ",Ryan J Ollos next-dev-1.7.x Release,12928,Pass author to wiki_page_deleted and wiki_page_version_deleted,wiki system,,defect,normal,,new,2017-10-11T19:19:45+02:00,2020-12-29T09:45:39+01:00,"See [https://github.com/trac-hacks/trac-wikinotification/issues/11 trac-wikinotification issue #11]. One possibility would be to modify the `page` object before passing to the change listener: * Set the `author` to the user deleting the page * Set the `content` to an empty string * Set the `version` to `None`? Similar changes made be wanted elsewhere, such as the `Ticket` model.",Ryan J Ollos next-dev-1.7.x Release,12912,AuthzPolicy fine permissions for timeline and search,general,1.2.2,defect,normal,,new,2017-09-10T21:12:38+02:00,2020-12-29T09:45:39+01:00,"Hi, I'm using trac for years now, and during all this time I've been using AuthzPolicy to define permissions. Today I noticed that all my users never saw the SEARCH and TIMELINE buttons, so I went my `authzpolicy.conf` file and added this {{{ [*] @administrators = TRAC_ADMIN authenticated = SEARCH_VIEW, TIMELINE_VIEW }}} but it didn't work. Checking the logs I saw this for SEARCH (notice the None at the end) {{{ 2017-09-10 13:57:38,827 Trac[authz_policy] DEBUG: Checking SEARCH_VIEW on *:*@* 2017-09-10 13:57:38,828 Trac[authz_policy] DEBUG: *:*@* matched section *:*@* for user johanna 2017-09-10 13:57:38,831 Trac[authz_policy] DEBUG: Checking WIKI_VIEW on *:*@* 2017-09-10 13:57:38,831 Trac[authz_policy] DEBUG: *:*@* matched section *:*@* for user johanna 2017-09-10 13:57:38,832 Trac[perm] DEBUG: No policy allowed johanna performing WIKI_VIEW on None }}} and this for TIMELINE (note the None at the end): {{{ 2017-09-10 13:14:24,654 Trac[authz_policy] DEBUG: Checking TIMELINE_VIEW on timeline:*@* 2017-09-10 13:14:24,655 Trac[authz_policy] DEBUG: timeline:*@* matched section timeline:*@* for user johanna 2017-09-10 13:14:24,658 Trac[authz_policy] DEBUG: Checking WIKI_VIEW on *:*@* 2017-09-10 13:14:24,658 Trac[authz_policy] DEBUG: *:*@* matched section *@* for user johanna 2017-09-10 13:14:24,659 Trac[perm] DEBUG: No policy allowed johanna performing WIKI_VIEW on None }}} I tried with `[timeline:*]` and `[search:*]` as resources but it didn't worked. So, I went to check the code and I found that the `resource` is `None` at this file https://trac.edgewall.org/browser/trunk/trac/perm.py#L518 and it's passed as `None` to https://trac.edgewall.org/browser/trunk/tracopt/perm/authz_policy.py#L143. For some reason, the resource is marked as `None` so there is no way to catch it (I also tried with `[None]` :) I'm using this python package versions {{{ Genshi==0.7 Pygments==2.2.0 Trac==1.2.2 ## FIXME: could not find svn URL in dependency_links for this package: TracSectionEditPlugin==0.2.6-r11208 argparse==1.2.1 click==6.7 configobj==5.0.6 docutils==0.14 pur==4.0.2 regex==2017.07.28 six==1.10.0 textile==2.3.16 wsgiref==0.1.2 }}} Please, let me know if I can add more information. Thanks!",humitos@… next-dev-1.7.x Release,12898,Add interface for changes on cached repository,version control,,enhancement,normal,,new,2017-08-27T08:11:42+02:00,2020-12-29T09:45:39+01:00,"Plugins such as [th:TracTicketChangelogPlugin] implement an `IAdminCommandProvider` command to perform an initial synchronization. {{{#!sh $ trac-admin $env ticketlog sync }}} It would be nice if we could hook into the `repository sync` and `repository resync` commands rather than having a separate command for the plugin. I considered adding a `changeset_synchronized` method to `IRepositoryChangeListener`, but the pattern didn't seem quite right. All of the other `I*ChangeListener` interfaces send notifications when the an instance of the model has changed. `IRepositoryChangeListener` sends a notification when the repository that is external to Trac has changed, and Trac uses that notification to synchronize the cache with the repository on disk. That led to my thinking that a better solution would be an interface that sends notifications on changes to `CachedRepository`. {{{#!python class ICachedRepositoryChangeListener(Interface): def cache_deleted(repos): """"""Called when the cache is deleted."""""" def changeset_added(repos, changeset): """"""Called when a changeset is added."""""" def changeset_modified(repos, changeset, old_changeset): """"""Called when a changeset is modified."""""" }}}",Ryan J Ollos next-dev-1.7.x Release,12893,Empty time values should be stored as NULL,general,,defect,normal,,new,2017-08-16T18:10:33+02:00,2020-12-29T09:45:39+01:00,"Empty time values in the ticket system are stored as `0`. They should probably be stored as `NULL` instead. Example: [browser:tags/trac-1.2.2/trac/ticket/model.py@:1304#L1292 Version.update] {{{ sqlite> SELECT time FROM version; 0 1496041616000000 1472327798000000 0 0 }}} The issue is caused by behavior of [browser:tags/trac-1.2.2/trac/util/datefmt.py@:191-192#L189 to_utimestamp]. We should add a helper function: {{{#!python def _to_timestamp(time): return to_utimestamp(time) if time else None }}} Timestamps are stored in the following fields: * `attachment.time` * `wiki.time` * `revision.time` * `ticket.time` * `ticket.changetime` * `ticket_change.time` * `milestone.due` * `milestone.completed` * `version.time` * `notification_subscription.time` * `notification_subscription.changetime`",Ryan J Ollos next-dev-1.7.x Release,12646,Add group toggle functionality to timeline filters,timeline,,defect,normal,,new,2016-12-23T07:52:14+01:00,2020-12-29T09:45:39+01:00,Like was done in #12376 for the Search filters.,Ryan J Ollos next-dev-1.7.x Release,12567,Image macro should support protocol-relative URLs,wiki system,,enhancement,normal,,new,2016-08-25T00:18:48+02:00,2020-12-29T09:45:39+01:00,"For example, `//www.edgewall.org/gfx/trac_bullet.png` could be used on WikiStart. URLs starting with `//` should be interpreted as protocol-relative. Unfortunately, the Image macro interprets URLs starting with `//` as //server-relative//. If we were starting again, I might suggest we use something like `^//` for server-relative. The only idea I have is to use `://` to indicate protocol-relative URLs. Other ideas are welcome.",Ryan J Ollos next-dev-1.7.x Release,12161,Add a Revision Log macro,version control,,enhancement,normal,,new,2015-08-08T06:02:44+02:00,2020-12-29T09:45:39+01:00,This would be similar to the [th:ChangeLogMacro]. That changes could be based on that macro if the code was refactored and tests added.,Ryan J Ollos next-dev-1.7.x Release,11871,Notification subscriber for @name mentions,notification,,enhancement,normal,,new,2014-12-08T22:20:16+01:00,2022-10-01T15:18:08+02:00,"Many websites allow users to mention `@name` (e.g. `@psuter`) in comments to notify that user of the comment. We could add an optional `INotificationSubscriber` implementing this for ticket comments.",Peter Suter next-dev-1.7.x Release,11870,"Optional notification subscribers for watching components, users, groups, individual resources, ...",notification,,enhancement,normal,,new,2014-12-08T20:28:24+01:00,2020-12-29T09:45:39+01:00,"th:AnnouncerPlugin has `TicketComponentSubscriber` so users' can subscribe to all tickets of certain components. For the [wiki:TracDev/Proposals/AdvancedNotification#Advancedsubscriptions Advanced Notification proposal] we could add it (and more) to `tracopt.notification`. It would use a preference child panel (#11853) to allow users to check components for watching, and the `notify_watch` DB table (#11869) to store these preferences.",Peter Suter next-dev-1.7.x Release,11861,Add email fields property to TicketSystem,ticket system,,enhancement,normal,,new,2014-12-03T16:02:21+01:00,2020-12-29T09:45:39+01:00,In comment:3:ticket:11594 adding `email_fields` to the `TicketSystem` was discussed. A patch that needs more work was proposed in [a6ea05df/rjollos.git]. The proposed change will be further investigated in this ticket.,Ryan J Ollos next-dev-1.7.x Release,11856,Default values of ticket workflow fields should be configured in the [ticket-workflow] section,ticket system,,enhancement,normal,,new,2014-12-02T01:21:31+01:00,2023-07-18T15:15:19+02:00,"The fields //resolution// and //status// in the list of `Ticket.protected_fields` ([browser:/tags/trac-1.0.2/trac/ticket/model.py@:51-52#L49]) are only modified by the workflow. #2045 moves Trac one step closer to adding `owner` to `Ticket.protected_fields` and only allowing the workflow to change the owner field (see comment:60:ticket:2045). Having `default_resolution` and `default_owner` as attributes of a ticket workflow action and configured through the `[ticket-workflow]` section rather than the `[ticket]` section has a few advantages: * It should be clear that these fields are controlled by the ticket workflow. * Unique default values can be provided for each action in the workflow. I've also found the attribute names `set_owner` and `set_resolution` to be a bit confusing. We might consider renaming these to `resolutions` / `owners` or `allowed_resolutions` / `allowed_owners`. For completeness, we could also consider adding `restrict_owner` as a workflow attribute, replacing or overriding `[ticket] restrict_owner`. This was mentioned in comment:1:ticket:11976.",Ryan J Ollos next-dev-1.7.x Release,11706,"""Can't synchronize"" warning message persists longer than it should",version control,,defect,normal,,new,2014-08-04T02:45:20+02:00,2020-12-29T09:45:39+01:00,"To reproduce: 1. Add or modify an SVN repository (or any type) and set its `type` to be any other type. 1. Save the changes and find the following message when redirected back to the //Manage Repositories// page: {{{#!html
close Warning: Can't synchronize with repository ""(default)"" ('/home/user/Workspace/tracdev/repos/tracdev' does not appear to contain a repository (Mercurial 2.2.2 says RepoError: repository /home/user/Workspace/tracdev/repos/tracdev not found)). Look in the Trac log for more information.
}}} 1. Select the repository and set its type correctly. 1. //Submit changes// and on redirect back to the //Manage Repositories// page, notice that the warning message is still present. 1. Refresh the page to cause the warning message to disappear. ",Ryan J Ollos next-dev-1.7.x Release,11675,IAttachmentStorage interface for pluggable attachment storage,general,,enhancement,normal,,new,2014-07-14T00:29:46+02:00,2021-05-28T21:29:49+02:00,"The interface would allow attachments to be saved to external storage, such as [https://www.dropbox.com/developers/core/start/python DropBox]. This has been discussed for implementation in bloodhound:#486. Related to #11339.",Ryan J Ollos next-dev-1.7.x Release,11432,The href of customized navigation items doesn't work as documented,general,,defect,normal,,new,2014-01-08T17:32:57+01:00,2020-12-29T09:45:39+01:00,"TracNavigation#Notes suggests that the following mappings should occur: * `/projects/project1/wiki/WikiStart` -> `/projects/projects1/wiki/WikiStart` * `wiki/WikiStart` -> `/projects/project1/wiki/WikiStart` (when in the `project1` environment). This is not the case however. An `href` beginning with `/` is treated as project-relative. There does not seem to be a way to create an interproject link in a multiproject/multienvironment setup sharing the same base url link without using the base url in the `href` or using `..` in the `href` (e.g. `../project2/wiki/WikiStart`). Using an `href` such as `wiki/WikiStart` results in repeated clicks on the mainnav button mapping to `/projects/project1/wiki/WikiStart` -> `/projects/project1/wiki/WikiStart/wiki/WikiStart` -> `/projects/project1/wiki/WikiStart/wiki/WikiStart/wiki/WikiStart`. Unfortunately I think the proper fix would change the behavior for existing users: {{{#!diff diff --git a/trac/web/chrome.py b/trac/web/chrome.py index f627ae5..7b392f3 100644 --- a/trac/web/chrome.py +++ b/trac/web/chrome.py @@ -724,7 +724,8 @@ class Chrome(Component): label = category_section.get(name + '.label') href = category_section.get(name + '.href') if href: - if href.startswith('/'): + if not href.startswith(('http://', + 'https://', '/')): href = req.href + href if label: item = tag.a(label) # create new label }}}",Ryan J Ollos next-dev-1.7.x Release,11409,"Add a ""Versions released"" filter to the timeline",timeline,,enhancement,normal,,new,2013-12-19T17:59:38+01:00,2020-12-29T09:45:39+01:00,"//Milestones reached// is shown on the timeline (renamed to //Milestones completed// in #11051). It would probably make sense to also have versions entries in the timeline and a //Versions released// filter. Related to the proposed work in #1233.",Ryan J Ollos next-dev-1.7.x Release,11367,"Ticket change history not updated when deleting or renaming Components, Versions and Ticket System Enums",ticket system,,defect,normal,,new,2013-11-16T06:29:59+01:00,2020-12-29T09:45:39+01:00,"Similar to what we saw in #4582 for milestones, the ticket change history of associated tickets is not updated when deleting or renaming the following ticket system items: * Components * Enums (Priorities, Resolution, Severities, Ticket Types) * Versions The current behavior is that on rename the ticket field's value has the new name, and on delete the ticket field's value has the deleted name.",Ryan J Ollos next-dev-1.7.x Release,11343,"Retarget tickets when executing the ""trac-admin milestone remove/completed"" commands",admin/console,1.0-stable,enhancement,normal,,new,2013-10-20T20:59:53+02:00,2020-12-29T09:45:39+01:00,"Tickets are not retargeted when milestones are deleted or closed through trac-admin. To be consistent with the web interface, milestones should be retargeted to the //None// milestone by default. Additionally, we might consider adding a //retarget// parameter to specify which milestone tickets should be retargeted to.",Ryan J Ollos next-dev-1.7.x Release,11339,Swappable Configuration implementations,general,,enhancement,normal,,new,2013-10-18T16:00:42+02:00,2020-12-29T09:45:39+01:00,"On some systems, it may be useful to initialize the Trac Environment with a non-standard Configuration object. Use cases include: * Deploying Trac on Heroku with configuration from environment variables instead of `trac.ini`. Heroku provides [https://devcenter.heroku.com/articles/dynos#ephemeral-filesystem an ephemeral filesystem]; directly writing changes to files (vs. updating files locally in a git repository and pushing those changes to your Heroku instance) is risky, because those changes may disappear at any time (e.g. when redeploying code or when Heroku decides to reboot a system) and are not shared across application instances if you scale up your deployment. Meanwhile, Heroku *does* provide [https://devcenter.heroku.com/articles/config-vars a persistent environment variable layer] that can store up to 16kb of data. For a Heroku Trac deployment, a Configuration class that reads and writes its values via a JSON-structured environment variable would be more appropriate than a file-based configuration. * Preventing through-the-web changes to the configuration altogether to ensure that formatting and comments in a hand-edited `trac.ini` are not lost, until #7378 is closed. * Ensuring that `trac.ini` remains version-controlled by extending `Configuration.save()`to write out a diff from the previous file, or to commit changes directly to a repository in a subprocess. See [[https://groups.google.com/d/topic/trac-dev/z4-gWs269X4|[Trac-dev] Setting database string from environmental variable?""]] for motivation.",ethan.jucovy@… next-dev-1.7.x Release,11148,Generic ChangeListener events,general,1.1.1dev,enhancement,normal,,new,2013-04-11T11:47:24+02:00,2020-12-29T09:45:39+01:00,"The proposal is inspired by #8834. Initial proposition suggests a new IResourceChangeListener interface: {{{#!python class IResourceChangeListener(Interface): """"""Extension point interface for components that require notification when resources are created, modified, or deleted. 'resource' parameters is instance of the a resource e.g. ticket, milestone etc. 'context' is an action context, may contain author, comment etc. Context content depends on a resource type. """""" def get_subscribed_resources(): """""" Implementation should return iterator of resource types for which the listener has to be notified. None or empty list means all types of resources. """""" def resource_created(resource, context): """""" Called when a resource is created. """""" def resource_changed(resource, old_values, context): """"""Called when a resource is modified. `old_values` is a dictionary containing the previous values of the resource properties that changed. Properties are specific for resource type. """""" def resource_deleted(resource, context): """"""Called when a resource is deleted."""""" def resource_version_deleted(resource, context): """"""Called when a version of a resource has been deleted."""""" }}} Please consider discussion related to this ticket on [https://groups.google.com/forum/?fromgroups=#!topic/trac-dev/aoJEUYm1wYM Trac Development group]. There is discussion on Bloodhound project about addition another !ChangingListener interface for within transaction events and unifying I*Manipulator interfaces. So, the solution for the generic !ChangeListener events implementation can be used as basis for other generic events.",andrej@… next-dev-1.7.x Release,11024,Losslessly compress images,general,,enhancement,normal,,new,2013-01-23T09:26:30+01:00,2021-06-04T02:18:31+02:00,Hi. Compressing the png images losslessly saves quite a few bytes.,anonymous next-dev-1.7.x Release,10948,[PATCH] Clone functionality should also be available for users with TICKET_CREATE permission,ticket system,1.0,enhancement,normal,,new,2012-11-12T09:20:15+01:00,2020-12-29T09:45:39+01:00,"As described in ticket [ticket:4686#comment:14], it would be nice to enable clone functionality also for ""normal"" users, which have `TICKET_CREATE` permission. As you could copy / clone a ticket manually with `TICKET_CREATE` permission (by copy-paste) it should be sufficient to check `TICKET_CREATE` permission in `clone.py`. As described in [ticket:10856#comment:6] it would also make more sense to put the clone-button on top of the properties table, since when you clone a ticket you clone '''all''' properties and not only the description.",franz.mayer@… next-dev-1.7.x Release,10802,Cleanup Javascript code,general,1.0-stable,task,normal,,new,2012-08-11T13:47:48+02:00,2023-02-16T08:24:10+01:00,"Once we have Javascript minification (#10672) in place, we could: * Switch from two spaces indentation to four spaces. * Add copyright headers. * Add documentation and other comments. * Perhaps take the opportunity to cleanup code in general. TracDev/CodingStyle#JavaScript advises to follow Douglas Crockford's style (with minor modifications). Perhaps using his [http://www.jslint.com/ JSLint] might help doing this more consistently. ''(See #633, #5997 for CSS cleanup.)''",Peter Suter next-dev-1.7.x Release,10672,Automatically minify Javascript and CSS,general,,enhancement,normal,,new,2012-04-21T13:54:28+02:00,2020-12-29T09:45:39+01:00,"As [comment:5:ticket:10284 discussed] [comment:20:ticket:10245 previously] it might be a good idea to introduce a minification step that pre-processes the !JavaScript and CSS files stored in source control. This would allow developing with clean, well-commented source files while improving performance. We could then switch our [TracDev/CodingStyle#JavaScript JavaScript coding style] to 4 spaces indentation and add copyright and documentation comments to file headers. (See #10802) The step / a related step could do JSLint checking. How should this be implemented? * [comment:5:ticket:10284 One suggestion] was a `make minify` step. * Another idea would be to add it to `setup.py`. [http://pypi.python.org/pypi/minify minify] or [https://github.com/bmathieu33/distwebres distwebres] provide setuptools commands (both use YUI compressor, which requires Java) * Alternatively, we could hook it into `trac-admin deploy` somehow. * Or the web frontend could dynamically minify and cache and combine just the files required for a request into one file. (Including scripts by plugins.) An important requirement might be to keep this step optional.",Peter Suter next-dev-1.7.x Release,10606,GitPlugin: Improve performance,plugin/git,1.0dev,enhancement,critical,,new,2012-03-04T15:22:25+01:00,2022-11-11T15:58:38+01:00,"''Created as part of the [ticket:10594 move of GitPlugin]. Tickets originally reported for th:GitPlugin: #746'' Quoting [th:ticket746:comment:6 hvr]: > - try to create an additional `libgit-thin`-based `Storage` in `PyGIT.py` to evaluate whether the exec+fork+parse overhead is still signifikant > - most external `git` calls can be avoided now thanks to extensive caching of meta-data; one of the remaining speed killers is listing directories in the source browser (`cat-file -s` calls can be optimized, by using GIT 1.5.3+ `-l` option to `ls-tree`; but having to call `rev-list` for each folder element still remains an issue... maybe it'll be possible to get an enhancement to `ls-tree` merged upstream to provide that information as well...) ---- See also: TracDev/Performance/Git",Peter Suter next-dev-1.7.x Release,10532,Show word diff when summary changes (instead of to ),ticket system,0.13dev,enhancement,minor,,new,2012-01-19T09:47:24+01:00,2020-12-29T09:45:39+01:00,"In this [http://trac.edgewall.org/ticket/8135#comment:11 summary change] you do not see any changes at once - and I guess there is not a change at all (which might be a defect). It would be better readable if summary changes would be displayed in a diff view instead of "" to "" manner, since most summary changes are typos or small changes. If it is a bigger change (maybe using some smart algorithm to determine that it has changed more than 50%) Trac could display it in this way: to .",framay next-dev-1.7.x Release,10239,Enhancement: set default_owner from WebAdmin `Components`,admin/web,0.12.2,enhancement,minor,,new,2011-06-29T21:30:35+02:00,2020-12-29T09:45:39+01:00,"I have a few projects where they either want all new tickets to go to one person for filtering and/or they do not use the 'Components' option which would set the default owner (configurable through WebAdmin). I know I can set [ticket] `default_owner` in trac.ini but I think it would be nice to be able to modify this from the WebAdmin `Components` interface. See also: http://groups.google.com/group/trac-users/browse_thread/thread/fefcec45c2e5e364 which includes some useful info on how `default_owner` works... ",py.hieroglyph@… next-dev-1.7.x Release,10040,Support for integer/float custom fields,ticket system,,enhancement,normal,,new,2011-02-22T12:38:30+01:00,2020-12-29T09:45:39+01:00,"See http://article.gmane.org/gmane.comp.version-control.subversion.trac.devel/6561 for some discussion. We'd like to be able to mark some custom fields as ""contains integer"" or ""contains floating number"". This would then let us sort numerically, validate user input, add an aggregation/summing feature, and so on. We're not thinking of changing the ''storage'' type, nor, currently, the Python type that exists in the ticket.fields list. Instead, we just validate the user input and then later we can also attempt to convert the field value to the numeric type if we want to do sorting or summing. This can fix #3080, but walks in the wrong way for FieldRefactoring. ",pipern next-dev-1.7.x Release,9331,trac-admin should upgrade static files in environment,admin/console,0.11.7,enhancement,normal,,new,2010-05-11T18:30:26+02:00,2021-01-02T20:35:00+01:00,"Trac 0.11 doesn't upgrade static files in environment. Therefore bugfixes in JS code and CSS styles are not propagated into environment. `trac-admin upgrade` should try to update them (if it knows where these static files are - and ideally it should know this even if these files are fetched only through web-server directly). If Trac can not update these resources automatically - it should at least instruct users how to update manually, i.e. use `trac-admin deploy` and compare to detect what files are modified. If Trac has embedded differ, it may try to do this automatically. At the very least it should propose to read wiki TracUpgrade page for more information. P.S. Plugin installation/upgrade instructions should also include `trac-admin deploy` step to refresh resources.",anatoly techtonik next-dev-1.7.x Release,9222,"""full delete"" of wiki page could be a non-fatal operation",wiki system,,enhancement,normal,,new,2010-04-12T17:09:32+02:00,2020-12-29T09:45:39+01:00,"When I delete a file from SVN, it stays there in previous versions. Would be nice to have the same possibility in Trac (maybe configured from trac.ini): When I delete a page from Wiki, it goes off and I don't see it in Index and I '''can't''' reach it from web. But I can reach its previous versions! I just deleted the page in version 5. When I go to this page from web - Trac says that the page is absent and asks me whether I want to create it. But when I go to this page with `@5` suffix, I get directly to the deleted page, its latest version. Such a behavior looks more effective/consistent for me, than ""full delete"", as it is now.",team@… next-dev-1.7.x Release,8069,[PATCH] Support list of users as options list in ticket custom field,ticket system,0.11.3,enhancement,normal,,new,2009-02-18T19:57:54+01:00,2020-12-29T09:45:39+01:00,"In the process of migrating from a Borland StarTeam install to Trac+SVN, I needed the ability to assign multiple users to a ticket. Borland provided the ability to include custom Change Request fields which could be limited to the list of valid users within the StarTeam project. Attached is a patch which implements similar behavior in Trac. The patch introduces a simple variable replacement scheme to the options attribute of a custom ticket field. The INI section looks like this: {{{ [ticket-custom] devtest_assigned = select devtest_assigned.label = Dev Tester Assigned devtest_assigned.options = $userlist }}} Code in ticket/api.py checks for a select custom field which has an options attribute of literally ""$userlist"". If such a field is found, the field is passed to the existing function eventually_restrict_owner() to insert the listing of valid users. The attached trivial three-line patch appears to work well against Trac-0.11.3.",Zachary Bedell next-dev-1.7.x Release,7820,[PATCH] Syslog Logging Not Working on FreeBSD,general,none,defect,normal,,new,2008-11-14T16:39:07+01:00,2020-12-29T09:45:39+01:00,"FreeBSD's syslog doesn't listen on /dev/log. It uses /var/run/log. Attached is a patch that allows you to specify the path to the UNIX socket instead of hardcoding it. It basically uses the logfile parameter, if it's None use /dev/log otherwise use it. This bug affects all versions of Trac since about 0.7 it looks like.",david@… next-dev-1.7.x Release,7628,[PATCH] (suggested) autocomplete users on ticket forms,general,,enhancement,normal,,new,2008-09-12T23:43:16+02:00,2021-06-04T02:13:28+02:00,"I've written a plugin, th:AutocompleteUsersPlugin , that uses AJAX to complete users in the owner and CC fields for new and existing tickets (for the latter, at least for the default workflow). The attached patch is the plugin in patch form, fairly unmodified. Performance should degrade gracefully for no JS or when {{{restrict_owner}}} is true. The autocompleter completes from known users who have the TICKET_MODIFY permission, mirroring the performance of the select widget when {{{restrict_owner}}} is true and populates the fields with the user name, though (obfuscated) email and full names are shown in the drop down and are included in the completion algorithm. I have included {{{autocomplete.js}}} from http://www.dyve.net/jquery/?autocomplete which is an updated version of the jquery plugin that source:trunk/trac/htdocs/js/suggest.js is based on. The update is needed to decouple what is displayed in the drop-down from what is populated to the input field. {{{suggest.js}}} could be updated with this as well as the only place it is used, source:trunk/trac/versioncontrol/web_ui/changeset.py , if this is desired. A new !RequestHandler handles requests to {{{/users}}} to respond to AJAX request. This handler could be made to do more if requests were descriminated upon {{{req.get_header('X-Requested-With') == 'XMLHttpRequest'}}}, or the whole data could be added into the JS on the server side (which is not possible with the existing {{{suggest.js}}}). Several new javascript files are included, in addition to {{{autocomplete.js}}}. These could be consolidated and/or generated on the server side if that is more desirable.",Jeff Hammel next-dev-1.7.x Release,5211,Ticket - Wiki Integration enhancement/suggestion,ticket system,,enhancement,minor,,new,2007-04-24T17:36:50+02:00,2020-12-29T09:45:39+01:00,"Supposing that one used CamelCase to name their components, and that one then created wiki pages for their components; how about having the 'Component' column of reports created by the ticket system contain clickable links for each component that has a wiki page created for it. Also perhaps, in the case that components do not have a page already created; those components might have a link to create the wiki page. Thanks, Cullen",Cullen Newsom next-dev-1.7.x Release,4519,[PATCH] Feature: component owner ticket notification,notification,0.10.3,enhancement,normal,,new,2007-01-11T00:40:41+01:00,2020-12-29T09:45:39+01:00,"The following patch allows for component owner notification of ticket updates (even when the ticket is assigned to a different user than the component owner). This feature must be enabled in the trac.ini file with a line like this: {{{ [notification] always_notify_component_owner = true }}} We've been using this patch with Trac 0.10 and it has been working fine. I made the patch against this trunk file: http://svn.edgewall.org/repos/trac/trunk/trac/ticket/notification.py Let me know if there are any problems with the patch and I'll be happy to address them. Thanks for your consideration.",jenalias@… next-dev-1.7.x Release,3914,"ticket change notification should contain diff'ed text, not old and new",notification,0.10,enhancement,normal,,new,2006-10-11T15:23:04+02:00,2020-12-29T09:45:39+01:00,"it would be beneficial, if the notification mail on a ticket change containes a diff'ed output instead of the old and new text. especially when a description is long and just a little thing changed its quite a challenge to find out what changed. ",ThurnerRupert next-dev-1.7.x Release,1233,Descriptions of Components,ticket system,0.8,enhancement,major,,new,2005-02-28T15:52:00+01:00,2020-12-29T09:45:39+01:00,"One annoying problem we have with our existing bug-tracking system is that users often don't know what component to select for a ticket. They often end up choosing an inappropriate one. One could argue that better component names would minimise this, but I've often found users to be endlessly creative in their quest to misunderstand the (seemingly) obvious ;-) A nice feature of bugzilla is that the component selector label is a link to a page listing the components for the project, along with owner and description. (If you prefer the link could be a separate item, like a '?' icon.) (Currently I'm conducting a trial of trac, and I quite like it!)",Matthew Harrison next-dev-1.7.x Release,8989,Milestones should have a 'Back to Roadmap' link in the contextual navigation,roadmap,0.12dev,enhancement,normal,,new,2010-01-22T02:15:37+01:00,2020-12-29T09:45:39+01:00,"When navigating to a milestone from the roadmap, there should be a '''Back to Roadmap''' link displayed in the contextual navigation menu. This is not currently implemented in Trac `0.12dev`, as can be seen: milestone:0.12-multirepos This seems simple enough that I'll try to implemented it and provide a patch.",Ryan J Ollos next-major-releases Release,8813,"next_rev is slow, particularly in the direct-svnfs case",version control,0.11,defect,major,anonymous,new,2009-11-13T10:10:33+01:00,2023-03-23T22:27:14+01:00,"This is a continuation of #7490, which became too ""big"" to be usable. Please keep that other ticket closed. A lot of possible reasons for bad performance have been identified, some of them fixed, most being due to installation and configuration issues. Please refer to the TracPerformance checklist, and help maintain that wiki page with your findings. This ticket should be used to help identify the remaining performance bottlenecks, a part from those which are already known (see TracDev/Performance). ",Christian Boos next-major-releases Release,886,Add support for Master tickets,ticket system,devel,enhancement,major,,new,2004-11-05T12:44:10+01:00,2023-04-13T09:07:18+02:00,"It would be cool if you could create master tickets that other tickets refer to. Work could then be broken down into chunks (and assigned to different people), but the master ticket could only be closed when all referring tickets had been resolved. You could also have it so that referring tickets are excluded from the Roadmap ticket counts.",dmurphy25@… next-major-releases Release,11719,"Auto-preview breaks TicketExt plugin, possibily others",ticket system,1.0.1,defect,major,,new,2014-08-15T18:27:07+02:00,2021-07-03T20:18:35+02:00,"When auto-preview kicks in and new ticket content is returned via XHR, `trac/ticket/templates/ticket.html` uses jQuery's `.replaceWith()` function to update the page. This has the effect of dropping all the `display: none` added to the page on initial load by `TicketExt`. One solution is to pass pass the XHR response through the same filters as the initial page load so the same manipulation (e.g., setting display attributes) can be applied. Another is to change contents rather than replace the entire DOM subtree. This can be done with: {{{#!diff diff --git a/trac/ticket/templates/ticket.html b/trac/ticket/templates/ticket.ht index 353ef63..d0865c3 100644 --- a/trac/ticket/templates/ticket.html +++ b/trac/ticket/templates/ticket.html @@ -79,7 +79,14 @@ $(""#propertyform"").autoSubmit({preview: '1'}, function(data, reply) { var items = $(reply); // Update ticket box - $(""#ticket"").replaceWith(items.filter('#ticket')); + var newProps = $('table.properties td', items); + $('#ticket table.properties td').each(function(i) { + $(this).html(newProps.eq(i).html()); + }); + var newDescription = $('div.description div.searchable', items); + $('#ticket div.description div.searchable').html(newDescription.html()); + if (items.hasClass('ticketdraft')) + $('#ticket').addClass('ticketdraft'); // Unthread, unrevert and update changelog if (!$('#trac-comments-oldest').checked()) $('#trac-comments-oldest').click().change(); }}} though this is likely a partial fix. (This has also been discussed at th:#11923.)",chris.nelson.1022@… next-major-releases Release,10699,trac-admin - ticket export import,admin/console,0.12-stable,defect,normal,,new,2012-05-18T16:54:05+02:00,2019-08-27T19:38:16+02:00,"We are attempting to use trac on a large scale at an university. We have multiple projects and one main project where most tickets enter from. Everything is working out well but for one exception. If a ticket comes in on the main trac instance.. lets call it [Support].. some times we need to 'move' this ticket to another project.. There is some trac hacks.. DatamoverPlugin and TicketmoverPlugin.. However these are largely out of date and don't work. What I propose is this function be included in trac-admin as well as the API. Like the wiki.. wiki export [file] Export wiki page to file or stdout wiki import [file] Import wiki page from file or stdin Have a simple... ticket export [file] Export ticket to file or stdin. Note attachments will be ignored and whatever else that is tricky to handle... ticket import [file] Import ticket as new id from file or stdin. ticket append [file] Add more information to a ticket.. This will allow us to do something like... trac-admin /project/Support ticket export 1232 /tmp/1232 trac-admin /project/Webpage ticket import /tmp/1232 // output created ticket: 321 trac-admin /project/Webpage ticket append 321 ""Moved ticket from Support to Webpage"" Or create a simple module that uses API under the hood to do more or less the same thing. The export, import, and append parser could be based off the email2trac plugin. And perhaps while at it merge email2trac into trac since its almost a requirement for issue tracking systems. ",steverweber@… next-major-releases Release,10665,Search through all text fields,search system,,defect,normal,,new,2012-04-13T12:18:03+02:00,2023-03-29T15:28:04+02:00,"When I try to search for ''VerySpecificTermXYZ VerySpecificTermABC'' on [http://trac.edgewall.org/demo-0.13/search?q=VerySpecificTermXYZ++VerySpecificTermABC&noquickjump=1&ticket=on demo-0.13] Trac finds nothing, because '''VerySpecificTermXYZ''' is saved in '''description''' field and ''VerySpecificTermABC'' is saved in ''Release Notes'' in Test-Ticket [http://trac.edgewall.org/demo-0.13/ticket/600 #600]. It would be nice if the search string will be considered for all fields (not only for one field).",framay next-major-releases Release,10178,[PATCH] Allow TracQueries to order on multiple columns,query system,0.12dev,enhancement,normal,,new,2011-05-14T06:46:57+02:00,2021-11-10T22:37:39+01:00,"Currently trac only supports ordering on a single column. I've written a patch which provides the ability to sort on more than one column. This allows: {{{ [[TicketQuery(max=5,order=priority|somecolumn, desc=0|1, status=open)]] }}}",david@… next-major-releases Release,9936,CSS and JS files should be versioned,web frontend,0.12.1,enhancement,normal,,new,2010-12-17T10:30:45+01:00,2020-11-02T19:43:43+01:00,"When we recently upgraded a trac install a number of users has strange behaviour on the Custom Query page. It turned out that this was due to JS caching. Ideally Trac should use version numbering on the CSS and JS uris so that the cache is busted on upgrades.",peter.westwood@… next-major-releases Release,9648,[PATCH] Improve ticket fields with tooltips,ticket system,0.12,enhancement,normal,,new,2010-09-27T11:29:42+02:00,2017-03-07T02:56:26+01:00,"I've created a patch that allows a Trac admin to specify tooltips for arbitrary ticket fields in Trac: [[Image(tooltip.png)]] This way the admin can explain a certain field in more detail (for example the difference between priority and severity) or provide information about the syntax used in that field. The tooltips appear in the following places: * the ticket box on the top of an existing ticket page (only on labels, not on values) * in the ticket history for changes values * the ticket input fields (on labels and input fields) All labels get a help cursor if a tooltip is available. Tooltips are specified in the `trac.ini` under the new section `[ticket-hints]`. Just assign the tooltip value to the name of the field. The user may also provide the same tooltip in multiple languages by appending the language code to the field name. Works for built-in fields as well as for custom fields (this patch is a generalization of #899). Example: {{{ [ticket-hints] reporter = The reporter of this ticket owner = The owner of this ticket blockedby = Tickets preventing this one from closing. Accepts comma delimited ticket number (without #) complexity = The estimated complexity involved in fixing this ticked. complexity_de = Die geschätzte Komplexität zur Lösung des Tickets }}} I've decided to use the `trac.ini` for localization as it would be too much overhead to create translation files for just some tooltips. ",Sebastian Krysmanski next-major-releases Release,9643,[Patch] Ticket search by query with auto completion,search system,0.13dev,enhancement,major,,new,2010-09-24T08:11:33+02:00,2021-10-24T09:23:29+02:00,"I was inspired by [http://youtrack.jetbrains.net/issues/ YouTrack ticket search box] and spent the last couple of hours hacking together something similar. Its based on [http://www.jqueryui.com/demos/autocomplete jQuery UI's autocomplete] and uses the regular TracQuery syntax, but with auto complete for field names, operators and (multi) values. Its not as good-looking as YouTrac, but still, its really neat :-) The way I implemented this is by making the regular search box a bit bigger, and my functionality is trigged when 'query:' is written. I thought it would be better than cluttering the interface with a separate search box. I'm afraid I don't know how to create those cool egg files, so I just wrote it as a one .py file (it doesn't do much anyhow, just adds JavaScript/CSS/jQuery-UI and give JSON responses of field values). You can [http://www.codenition.com/wp-content/TicketQuerySearch.zip download it here], just place `TicketQuerySearch.py` and the `TicketQuerySearch` directory in the plugins folder. That's about it. just write 'query:' in the top-right search box, and start writing a regular TracQuery. You can move around with the arrows and use tab, enter or the mouse to choose an item from the list. === Currently it supports: === - Auto completes the field name (either at the very beginning or after an '&'). - When a valid field name is typed/chosen, or when just one field name option is left in the suggestions, it start suggesting search operators. Those operators are also filtered as you type them. - Once an operator is written, it suggests values. - When a valid value is typed/chosen, it suggest you to add more values after a pipe (`|`). - Also, in some contexts it lets you choose 'Another field...' and 'Another value...' which simply adds `&` or `|` to the query. === A couple more notes: === - I know some people would prefer this to be a plugin and not part of core, but I think it gives much better UX when searching for tickets. It can completely replace the Custom Query page in most cases, making it much faster and easier to search. You can search using the keyboard alone, or by using some simple mouse clicks. The current Custom Query interface is great and much more advanced than this, but for simple searches its an overkill - it takes too much clicks and time to make searches. - I posted this ticket as soon as I got a working version, so it wasn't tested enough yet. I want to hear what you have to say about the idea in general, and what you think about the way I implemented this. I assume changes and fixes will be needed. - I'm not quite sure if you would want to ship jQuery UI with Trac. I downloaded a version that includes only what's necessary for [http://www.jqueryui.com/demos/autocomplete Autocomplete], and it weights around ~120KB (20KB for JavaScript and the rest is CSS&images). If needed, I can write a stand-alone Autocomplete functionality, but it probably won't be as good as jQuery-UI's. - Most of the auto-complete is done on the client-side. When searching in values of some fields (like component), the client downloads a full list of possible values (only when required), searches them and caches them for the next time a search is needed on the same field. As there aren't usually many values, I thought it would be better to download it once instead of hitting the server with a request every time the user types something. - Support for custom fields is missing. - Maybe the '!query:' prefix can be removed so it'll always be enabled? It does raise some questions (when the form is submitted, and the user typed 'status=new', did he mean to search for tickets with that status or just 'status=new' as plain string in the regular search?), but I think its all solvable.",shesek next-major-releases Release,9574,add filters to TicketQuery from query string arguments,wiki system,0.13dev,enhancement,normal,,new,2010-08-18T21:54:08+02:00,2011-10-25T15:03:19+02:00,"I'm using wiki pages and the TicketQuery macro to create ticket summary/overview pages, as I find them much more flexible than reports as I can have multiple 'views' on the same page (for example, have 'Assigned to you', 'Awaiting assignment', 'In progress', etc - each shows top 10, has different filters and ordered by different columns, all in the same page. Creating a page like that isn't possible with reports). So - what I'm trying to do is create a per-project ('project' being a custom field I added) and per-components ticket overview using the Wiki. Instead of copying the wiki page for every project/component I have, I want to use something like `wiki/TicketSummary?project=foobar` and than read the query-string argument (`project`) to be used inside the TicketQuery macro and `[query:]`. I was trying to do it using IWikiSyntaxProvider and get_wiki_syntax (http://pastie.org/private/evw1r9jkhcmq3m6jjgrojw), by replacing ${foobar} with `req.args['foobar']`, which works - but not inside other macros. I think the best way is adding this functionality to core, by parsing custom arguments inside the TicketQuery macro and `[source:]` link, in a similar way to how $USER is replaced with the currently logged in user - but replacing them with query-string arguments instead. This will give users better control and flexibility in creating their own custom ticket overviews, which can replace the reports in some cases if those pages could be dynamic and show different content based on arguments.",shesek next-major-releases Release,9037,Incorporate footnotes into Trac (for use in documentation),wiki system,0.12dev,enhancement,normal,,new,2010-02-06T07:34:27+01:00,2023-06-11T16:46:44+02:00,"As I have tried to contribute a little bit to the documentation, it has occurred to me that it would be really nice for the sake of page layout and clarity to have footnotes available for use in the documentation. I have used the th:FootNoteMacro for quite some time and there does not seem to be any issues with it. There are currently not even any open tickets filed against the macro. Would it be possible to incorporate this macro into the next Trac distribution if we were to refactor the code to fit the style guidelines for Trac, write some unit tests, or whatever else might be necessary? Or is the preferred way to go about this through a direct incorporation into the wiki markup language, like I see is available with Restructured Text (as shown on WikiProcessors - see Example 2)? I [/search?q=footnote searched] around but did not find any similar suggestions.",Ryan J Ollos next-major-releases Release,8645,More readable header IDs,wiki system,0.11.4,defect,normal,,new,2009-09-07T20:19:51+02:00,2015-05-26T20:50:57+02:00,"The initial report was about invalid header IDs; they're actually not invalid in XHTML 1.0, but their readability can nevertheless be greatly improved. Re-targeting the ticket that way, see comment:5 for details. === Original Report Headers IDs generated from headers with non-ASCII characters keep/have them what is invalid: An ID attribute does not conform with the HTML document type specification. ID tokens must begin with a letter, and may contain only letters, digits, hyphens, underscores, colons, and periods. We could do something like [http://search.cpan.org/~sburke/Text-Unidecode-0.04/lib/Text/Unidecode.pm Text::Unidecode].",Mitar next-major-releases Release,8509,[PATCH] use precompiled regex for match_request,web frontend,0.11.5,defect,normal,,new,2009-07-23T20:42:45+02:00,2015-12-19T15:40:25+01:00,"This patch changes all the trac match_request calls to use precompiled regular expressions, avoiding multiple re compilations per request. ",mixedpuppy next-major-releases Release,7871,Add start date to a milestone,general,none,enhancement,major,,new,2008-12-07T17:51:17+01:00,2022-10-12T20:04:57+02:00,"It would be nice to have a start date for a milestone. With that, it's possible: * To filter on started/not started milestones on the Roadmap page * Create reports to show only tickets in currently active milestones * Sort on started date in the 'Milestone' dropdown on the ticket create and update pages. With a start date built-in, plugins could use that (for example the Scrum Burndown plugin) to draw their graphs. The Scrum Burndown plugin now has to arrange a column, while it could be useful for more plugins.",daan@… next-major-releases Release,7744,source:path@rev targeting a file below a copied dir may fail,version control/browser,0.11-stable,defect,minor,,new,2008-10-22T16:12:55+02:00,2020-09-23T06:08:21+02:00,"... when the file has not been modified since the copy and `rev` predates the copy operation itself. e.g. source:/branches/0.11-stable/contrib/bugzilla2trac.py@6820#L824 leads to: {{{ No node /branches/0.11-stable/contrib/bugzilla2trac.py at revision 6820 }}} Looking at source:/branches/0.11-stable/contrib/bugzilla2trac.py (without revision spec) works, and indeed shows a revision of r6820 for this file. This is happening because at r6820, there was no branches/0.11-stable. This behavior is actually ""correct"" if ...@6820 is supposed to be a peg revision in Subversion terminology, but it's annoying nonetheless, as the TracLinks syntax doesn't allow an alternative way of specifying the revision(*) and anyway I don't think it's a good idea to expose the peg revision vs. operative revision subtleties in Trac (http://svnbook.red-bean.com/en/1.5/svn.advanced.pegrevs.html). (*) well it does, but source:/branches/0.11-stable/contrib/bugzilla2trac.py?rev=6820 has the same problem",Christian Boos next-major-releases Release,7658,Option to disable automatic linking of CamelCase,wiki system,0.11-stable,enhancement,normal,,new,2008-09-22T13:32:30+02:00,2022-03-09T23:02:43+01:00,"It must have been brought up a million times by now. I can see that #476 almost addressed the problem for me, but didn't quite cut the mustard. Simply setting ignore_missing_pages isn't sufficient, as I'd like to see missing pages, but not automatic camelcase detection. Our naming convention uses camelcase and it's getting in the way. Trac is supposed to ""stay out of the way"" of developers. I'd just rather a wiki link was explicit. I'd like to see either a global option in the ini file to switch this off completely, or a more configurable option, something like that suggested here: http://trac.edgewall.org/ticket/476#comment:23 The fix suggested in http://trac.edgewall.org/ticket/476#comment:26 would do just fine. If it breaks my wiki pages like the default Trac* wiki, that's my problem, and I'm quite prepared to deal with that. It's such a simple fix, I can't see why anyone would have any objections. I'll hack my installation for now, but I'd rather not. Thanks, Dave.",dave@… next-major-releases Release,7014,"Trap ""ZipImportError: bad local file header.. "" and suggest server restart",general,0.11b1,defect,normal,,new,2008-03-18T00:18:39+01:00,2020-09-06T17:39:05+02:00,"==== How to Reproduce ==== While doing a GET operation on `/admin/general/plugin`, Trac issued an internal error. ''(please provide additional details here)'' Request parameters: {{{ {'cat_id': u'general', 'panel_id': u'plugin', 'path_info': None} }}} ==== System Information ==== || '''Trac''' || `0.11b1` || || '''Python''' || `2.5.1 (r251:54863, Mar 13 2008, 12:52:19) ` [[br]] `[GCC 3.4.6 [FreeBSD] 20060305]` || || '''setuptools''' || `0.6c8` || || '''SQLite''' || `3.3.17` || || '''pysqlite''' || `2.3.5` || || '''Genshi''' || `0.4.4` || || '''Pygments''' || `0.9` || || '''Subversion''' || `1.4.6 (r28521)` || ==== Python Traceback ==== {{{ Traceback (most recent call last): File ""/usr/local/lib/python2.5/site-packages/Trac-0.11b1-py2.5.egg/trac/web/main.py"", line 398, in _dispatch_request dispatcher.dispatch(req) File ""/usr/local/lib/python2.5/site-packages/Trac-0.11b1-py2.5.egg/trac/web/main.py"", line 195, in dispatch resp = chosen_handler.process_request(req) File ""/usr/local/lib/python2.5/site-packages/Trac-0.11b1-py2.5.egg/trac/admin/web_ui.py"", line 108, in process_request path_info) File ""/usr/local/lib/python2.5/site-packages/Trac-0.11b1-py2.5.egg/trac/admin/web_ui.py"", line 389, in render_admin_panel return self._render_view(req) File ""/usr/local/lib/python2.5/site-packages/Trac-0.11b1-py2.5.egg/trac/admin/web_ui.py"", line 503, in _render_view info = get_pkginfo(dist) File ""/usr/local/lib/python2.5/site-packages/Trac-0.11b1-py2.5.egg/trac/util/__init__.py"", line 228, in get_pkginfo pkginfo = email.message_from_string(dist.get_metadata('PKG-INFO')) File ""/usr/local/lib/python2.5/site-packages/setuptools-0.6c8-py2.5.egg/pkg_resources.py"", line 1137, in get_metadata return self._get(self._fn(self.egg_info,name)) File ""/usr/local/lib/python2.5/site-packages/setuptools-0.6c8-py2.5.egg/pkg_resources.py"", line 1195, in _get return self.loader.get_data(path) ZipImportError: bad local file header in /usr/local/lib/python2.5/site-packages/TracAccountManager-0.2dev_r3111-py2.5.egg }}} ",admin next-major-releases Release,5648,DefaultPermissionGroupProvider not doing its whole job,general,devel,defect,critical,,new,2007-07-03T00:31:08+02:00,2023-02-24T10:30:11+01:00,"[browser:trunk/trac/perm.py DefaultPermissionGroupProvider] doesn't really return the permission groups of which a user is a member; it only handles the ""anonymous"" and ""authenticated"" groups. All of the threading through user-defined groups is actually done by {{{DefaultPermissionStore}}}. While I'm sure that's efficient, it's a problem when the {{{permission_store}}} is replaced, as with [trachacks:wiki:TracForgePlugin TracForge], if it expects to be able to extend the existing permission group system. See [trachacks:browser:tracforgeplugin/branches/bewst/0.11-clearsilver/tracforge/admin/perm.py?rev=2355 TracForgeGroupsModule] for an example. This problem exists in 0.10-stable and 0.11 AFAICT. Of course, it's entirely possible that I've completely misinterpreted the meaning of ""group"" and the intention behind {{{DefaultPermissionStore}}} here :).",Dave Abrahams next-major-releases Release,4588,User Page: trac links to these pages,general,,enhancement,normal,,new,2007-01-24T22:11:55+01:00,2021-07-24T22:55:48+02:00,"When browsing tickets it would be nice, if I could click on the owner of the ticket and would be directed to his/her user specific page. I think this is interrelated to #2178.",triplem next-major-releases Release,4080,[Patch] Detect charset using enca or chardet,version control/changeset view,0.10,enhancement,normal,,new,2006-11-06T13:06:14+01:00,2015-12-14T14:37:03+01:00,"to detect charset in multiply charset repository http://gitorious.org/enca",mpv next-major-releases Release,3718,Trac should use HTTP 301 Moved when milestones get renamed,general,0.9.6,enhancement,major,,new,2006-09-14T06:11:44+02:00,2021-08-19T10:35:21+02:00," 1. Create a milestone ""Example"" 2. Add a ticket to the milestone 3. Subscribe to the RSS feed. 4. Edit the milestone, rename to ""Changed Milestone"" Expected behaviour: Trac 'knows' a milestone has been renamed, and does a 301 Moved response code. The client application / browser redirects, and the end user doesn't notice anything out of the ordinary Actual: Trac doesn't do this. milestone:""milestone name"" links break, feeds break, and so much more. Suggested: A simple entry to a 'milestone (previous) names' table would fix this - when a milestone isn't known, the table is checked, and a 301 is sent with an updated location. ",daniel.connor@… next-major-releases Release,3519,browse source - edit feature?,version control,0.9.6,enhancement,normal,,new,2006-08-04T22:40:37+02:00,2022-01-29T18:33:38+01:00,"It would be great if when browsing the source tree through the subversion repository, there was a way to pull up an ""Edit"" box to make quick edits to files and submit them to the subversion repository. This way I can give access to WIKI users to make edits to files within the subversion repository without them needing to set up full development environments.",anonymous next-major-releases Release,3470,improve handling of scoped repositories with copy history,version control,devel,defect,major,,new,2006-07-25T21:44:29+02:00,2015-05-26T20:50:57+02:00,"While testing r3015, the revisions are still not working properly. I removed trac 0.9.6, checked out and installed r3570. I ran all the required upgrade commands and resync'd, but trac still always shows one rev for the part of the svn repo moved from another part. Here's an example... the svn repo had proj1/trunk/java/src and proj2/trunk/websrc. I renamed websrc in proj2/trunk to src. Then I created proj1/trunk/web and moved src from proj2/trunk to proj1/trunk/web. When I make a change to web/src/file1 the revision is incremented, but all files in web/src show that revision too. If I point trac to the root of the repo and resync, things work better, but if I point to proj1, the revision for the moved code is always the latest revision for files that haven't been updated since the move. I'm guessing this is because trac is indexing only the current point in the history since the move.",posterday@… next-major-releases Release,3386,Embedding part of Wiki in other Wiki (e.g. Help: prefix for the TracGuide),wiki system,0.9.6,enhancement,normal,,new,2006-07-12T11:43:19+02:00,2015-05-26T20:50:57+02:00,"I'd like to be able to embed part of Wiki associated with some project in Wiki of other project, in order to reduce duplication of information and avoid using external links between wikis. By `embed` i mean they do share (part of?) namespace, and links between them are internal to wiki. The embedded part should be accessible, linkable and editable from both Wikies, without using any special syntax nor, possibly, addresses. Example: lets say i have: * /projects/FlagshipProject, containing: * wiki/Globals/AboutDevelopers/*, * wiki/Globals/AboutCompany/*, * wiki/Globals/OtherStuff; and * /projects/OtherProject. In Wiki of the later i'd like to embed /projects/FlagshipProject/wiki/Globals/*",dexen^^^goofy.pl next-major-releases Release,3332,Mimeview API v2,rendering,devel,enhancement,major,,new,2006-06-27T13:48:59+02:00,2022-10-19T18:03:49+02:00,"We recently introduced the `IContentConverter` interface, and there are some issues remaining with this: - the methods for that interface are quite complex, not really extensible - the usage of strings for model objects should be replace by using the corresponding classes - artificial differences from the `IHTMLPreviewRenderer`, and duplicated code (mainly between `Mimeview.render` and `Mimeview.convert_content`) Besides, there are other issues with the Mimetype API: - `content_to_unicode` needs to be deprecated/simplified as well - simplify the handling of mimetypes strings (the fact that they sometimes have the charset info, sometimes not) - in general, aim to simplify the code that uses the mimeview API Work in progress. A patch will be attached today or tomorrow for review. ",Christian Boos next-major-releases Release,3098,"Support ""Descript.ion"" files in Browse Source",version control/browser,0.9.5,enhancement,normal,,new,2006-05-01T18:46:20+02:00,2019-09-06T19:51:32+02:00,"As the summary says... Right now we hand edit a Wiki page describing each subdir and file. The `descript.ion` metadata file is a [http://jpsoft.com/ascii/descfile.txt quasi-standard] by JP-Soft (4DOS) that Browser might be able to support showing a short description of each file. ",trac@… next-major-releases Release,3021,wiki namespace support like the one in dokuwiki,wiki system,none,enhancement,major,,new,2006-04-11T17:47:04+02:00,2015-05-26T20:50:57+02:00,"It would be very great to have wiki namespace support, for sorting wiki pages and protect user to build too big pages. I think it's very important to support wiki namespaces. ---- As the topic is broad, it's worth to discuss it in a specific proposal page, see TracDev/Proposals/WikiNamespaces.",trac.ticket@… next-major-releases Release,2463,Mandatory fields in tickets,ticket system,0.9.2,enhancement,normal,,new,2005-12-09T11:18:09+01:00,2020-08-07T00:45:49+02:00,It would be great with ability to set some ticket fields as mandatory. These fields should be verified before allowing submit when creating a new ticket.,johan.regnell@… next-major-releases Release,2035,Changeset commenting,version control/changeset view,,enhancement,major,,new,2005-09-09T15:35:21+02:00,2022-09-08T16:44:10+02:00,"Commenting on changesets would be great, it would substitute for code review tools.",polarfox@… next-major-releases Release,1962,Due dates on tickets & email notification on overdue dates,ticket system,0.8.4,enhancement,minor,,new,2005-08-25T04:25:51+02:00,2021-06-03T22:03:37+02:00,"It maye be there already but i cannot find it. One things as a software project manager that drives me crazy about track is the inability to add a due date to tickets. Prioritization is all well and good, and most certainly handy, but as part of the timeline / roadmap and the ticket inspector you REALLY need to have a due date on a ticket. It'd also be fantastic for Trac to notify the allocated user that the ticket is now overdue and bug them on a daily / configurable timeframe. As i said, correct me if this is already in trac (please tell me where!!) but if it's not, this is a really string feature request & i think a definite must have. Thanks - trac is awesome & we;'re really starting to use it now :) Regards Owen",owen@… next-major-releases Release,1890,Can create tickets anonymously using the username of an authenticated user,general,0.8.4,defect,major,,new,2005-08-07T23:45:51+02:00,2022-03-01T05:06:05+01:00,"I can create tickets anonymously using usernames of registered users. This is a Bad Thing(TM) in that people can impersonate me on my Trac. Or, they could otherwise pretend to be me. Which, to some users, may be confusing and misleading. It also poses a security threat in that any random person can go in and meddle in my bugs and close at will because to be able to add a comment to a ticket, you have to have TICKET_MODIFY, which essentially means anonymous has TICKET_ADMIN (filing another bug for this, since I know that at least in my projects, I like two problems to be reported as... two problems...) ---- '''Current status of the discussion''': each change to a ticket must also record whether the user who did the change was authenticated or not. See more complete summary in comment:53.",ziggy@… next-major-releases Release,1386,Adding properties to Wiki Pages,wiki system,devel,enhancement,major,,new,2005-03-31T22:23:28+02:00,2021-12-30T03:07:32+01:00,"I think adding properties to Wiki pages would be a great feature. Such as owner, cc, rss feeds, and specifcally, properites. Some ideas for properties include * Publishing status - Published | Nonpublished[[br]] * Document Status - In Progress | Complete | Inconsistent | etc.. The idea for this is that I am in the process currently of adding content to the Wiki, however some items that I am documenting have not been completed either in terms of the documentation or having the items being documented having been completed. I would like to be able to have an owner associated with the Wiki, perhaps a list of people who want to subscribe to change notifications for a particular wiki page, and a report for instance that the owner could view to see which pages they have that are in various states a defined by my definition of the various properties. My definitions may need some refining but the idea still holds. Anyway, I would be happy to discuss this further and possibly try to create some patches to demonstrate what I'm discussing if interested. ",jon.brandenburg at theronyx.com next-major-releases Release,1113,Show milestone changes in timeline,timeline,0.8,enhancement,major,,new,2005-01-08T21:45:19+01:00,2021-11-27T07:16:12+01:00,It would be great to have the timeline noticing EVERY milestone change rather than only new milestones.,jens@… next-major-releases Release,1084,Percentage complete,ticket system,0.8,enhancement,normal,,new,2004-12-26T16:53:05+01:00,2023-02-11T22:54:18+01:00,"Im using Trac in my projet to track bugs, but also to maintaint a list of tasks to do. One thing that would be great is the possibility to have a ""percentage complete"" field. This way the task can be updated to reflect the percentage complete. Right now the task can only be ""0% complete"" or ""100% complete"". I mean a task is either finished or unfinished, which is not always the case.",cerel next-major-releases Release,53,plugin infrastructure for rendering diff HTML previews,version control/changeset view,devel,enhancement,normal,,new,2004-02-10T21:10:27+01:00,2023-02-26T06:09:59+01:00,"== Use Cases === Viewing image changes/patches in a changeset It would be pretty cool to be able to view image updates in the changeset viewer. An updated image should show the old version and the new version side by side, just like text diffs. === Filter out noisy changes Typical use case is for showing diffs on .po files, where the chunks corresponding to line number changes only are completely uninteresting and can be safely skipped. ",daniel next-major-releases Release,31,Bug dependencies/relations feature,ticket system,devel,enhancement,normal,,new,2004-01-29T05:53:39+01:00,2022-08-30T20:23:04+02:00,"A simple list of related bugs without logical/implicit constraints might be useful. It'd be nice to be able to relate bugs and issues in a useful manner. It would also thus allow for reports showing ""a bug and all related bugs"" in some fashion, which could be neat. This would also require checking for circular dependencies/relations. ",daniel next-major-releases Release,13279,Make cast method more robust,database backend,,enhancement,normal,,new,2020-04-30T04:43:37+02:00,2020-04-30T04:43:37+02:00,"Copied from [ticket:13078#comment:7]: > Another solution is using `CASE ... END` and regular expression operator. I think we could make `CAST(... AS int)` in PostgreSQL robust and same behavior in other databases. > > {{{ > trac=> SELECT (CASE > trac(> WHEN value IS NULL THEN NULL > trac(> WHEN value ~ '[-+]?[0-9]+$' THEN CAST(value AS int) > trac(> ELSE 0 END) AS cast, > trac-> value > trac-> FROM unnest(ARRAY[NULL, '42', '+123', '-789', 'text', '']) AS value > trac-> ; > cast | value > ------+------- > | > 42 | 42 > 123 | +123 > -789 | -789 > 0 | text > 0 | > (6 rows) > > }}} > > {{{#!diff > diff --git a/trac/db/postgres_backend.py b/trac/db/postgres_backend.py > index 3b5c9e92c..3775a9aa7 100644 > --- a/trac/db/postgres_backend.py > +++ b/trac/db/postgres_backend.py > @@ -272,7 +272,14 @@ class PostgreSQLConnection(ConnectionBase, ConnectionWrapper): > > def cast(self, column, type): > # Temporary hack needed for the union of selects in the search module > - return 'CAST(%s AS %s)' % (column, _type_map.get(type, type)) > + if type in ('int', 'int64'): > + fmt = ""(CASE WHEN %(column)s IS NULL THEN NULL "" \ > + ""WHEN %(column)s ~ '^[-+]?[0-9]+$' "" \ > + ""THEN CAST(%(column)s AS %(type)s)"" \ > + ""ELSE 0 END)"" > + else: > + fmt = 'CAST(%(column)s AS %(type)s)' > + return fmt % {'column': column, 'type': _type_map.get(type, type)} > > def concat(self, *args): > return '||'.join(args) > }}} ",Ryan J Ollos next-major-releases Release,13201,URL parameter defaults for milestone forms,general,,enhancement,normal,,new,2019-09-03T12:20:23+02:00,2019-09-09T09:50:13+02:00,"Ticket and wiki form defaults can be controlled by URL parameters: * [/newticket?summary=MyDefaultValue /newticket?summary=MyDefaultValue] * [/wiki/test?action=edit&text=MyDefaultValue /wiki/test?action=edit&text=MyDefaultValue] The same would be useful for milestones: * [/milestone/1.0?action=edit&description=MyDefaultValue /milestone/1.0?action=edit&description=MyDefaultValue] * [/admin/ticket/milestones?name=MyDefaultValue /admin/ticket/milestones?name=MyDefaultValue] But the default values are ignored there.",anonymous next-major-releases Release,13022,Reports can't be used with some realms,report system,,defect,minor,,new,2018-04-29T13:01:22+02:00,2018-04-29T22:52:28+02:00,"We can create [TracReports#Automaticallyformattedcolumns for non-ticket realms] like this: {{{#!sql SELECT name as id, 'milestone' as _realm FROM milestone }}} But this doesn't work for all realms: {{{#!sql SELECT value as id, 'repository' as _realm FROM repository WHERE name='name' }}} This SQL report above does not list any repositories, even though the following does list them: {{{#!sql SELECT value as id, 'repository' as _ignore FROM repository WHERE name='name' }}} (But here the rows link to the ticket realm.) The reason is a [source:trunk/trac/ticket/report.py@16445:592-594#L592 hardcoded check in Trac] that's wrong for some realms. It guesses that when realm is `repository`, the appropriate permission is probably `REPOSITORY_VIEW`. But no such permission exists, so all rows are hidden. (This problem was also [th:ticket:13424 reported to a plugin] where this check also fails.)",anonymous next-major-releases Release,13019,Make dropdowns searchable / filterable,web frontend,,enhancement,normal,,reopened,2018-04-27T07:41:31+02:00,2021-07-24T22:55:15+02:00,"Copied from [ticket:10012#comment:82]: > > About select boxes: > > > > Another interesting library I recently discovered is http://harvesthq.github.com/chosen/ ... > > Wow, impressive! Yes, looks like it's a must have. > > > Any arguments against including it in core? > Not from my side. [cboos] [https://harvesthq.github.io/chosen/#optgroup-support This would be especially useful] for milestone selection. [ticket:13018 Often] there are many milestones and finding a certain milestone can be difficult because [ticket:2130 they are sorted by due date]. Allowing to use the filtering of ''chosen'' would be very helpful and nice.",anonymous next-major-releases Release,12900,Hierarchy-specific PageTemplates,wiki system,,enhancement,normal,,new,2017-08-30T15:06:13+02:00,2019-09-06T19:51:32+02:00,"The PageTemplates feature uses: > wiki pages having a name starting with ''!PageTemplates/'' as templates. And the default selection: > will be ''blank page'', or ''!DefaultPage'' if ''!PageTemplates/DefaultPage'' exists. What's missing are templates that are only relevant to certain page hierarchy path prefixes, and a default template for those. For example when creating the page `wiki:Machines/NewMachine` the templates under `wiki:Machines/PageTemplates` should also be available. and `wiki:Machines/PageTemplates/DefaultPage` (or at least `wiki:PageTemplates/Machines/DefaultPage`) should be the default. ",anonymous next-major-releases Release,12854,"""rows"" can be used as custom field but rows=... condition in query doesn't work",ticket system,,defect,normal,,new,2017-06-27T05:43:08+02:00,2019-09-06T19:51:32+02:00,"Ticket with //rows// field can be created like this but query:status!=closed&rows=42 ignores `rows=42` condition. {{{#!ini [ticket-custom] rows = text }}} Because `rows` parameters is used for full rows feature in ticket query. See also wiki:TracQuery#Fullrows. I think we should add `'rows'` to `TicketSystem.reserved_field_names` because field named `rows` might be rarely used.",Jun Omae next-major-releases Release,12848,Add an ifnull function to database connection classes,database backend,,enhancement,normal,,new,2017-06-19T13:51:48+02:00,2019-09-06T19:51:32+02:00,"Copied from [ticket:12841#comment:2]: ---- One proposed change in [9f25cdfb3/rjollos.git]. When the group heading is NULL, it would be nice if `Milestone (empty)` was shown rather than just `(empty)`. We could add an `isnull` function for each connection class and modify the reports. For example, on SQLite for report {3} we'd have: {{{#!diff - 'Milestone '||milestone AS __group__, + 'Milestone '||ifnull(milestone,'(empty)') AS __group__, }}} I don't think it would be worth fixing up existing reports through a database upgrade step, but we could modify the reports in the database to affect new installations and add a note to TracUpgrade about how to manually modify reports.",Ryan J Ollos next-major-releases Release,12828,"Add a ""wiki set"" trac-admin command",wiki system,,enhancement,normal,,new,2017-06-05T19:49:52+02:00,2018-03-03T00:48:34+01:00,"Similar to the `repository set` command, {{{ repository set Set an attribute of a repository }}} it would be useful to have a `wiki set` command for setting the readonly attribute.",Ryan J Ollos next-major-releases Release,12810,Add a TicketReport macro,report system,,enhancement,normal,,new,2017-05-17T06:48:12+02:00,2022-03-28T19:27:46+02:00,"Add a `TicketReport` macro that takes the report `id` as the first argument, and also the following optional keyword arguments: * `order`: default column to sort by * `desc`: default sort direction * `max`: number of rows per page * Any uppercase report variable The argument names are consistent with the `TicketQuery` macro. `ReportModule._render_view` needs to be refactored by extracting the code that prepares the report data.",Ryan J Ollos next-major-releases Release,12786,Allow custom queries to be saved without granting REPORT_CREATE,report system,,enhancement,normal,,new,2017-04-27T00:25:06+02:00,2017-04-27T00:30:26+02:00,"In order to save a ticket query the user must have `REPORT_CREATE` and the `ReportModule` must be enabled. In the documentation, TracReports#TracReports, the user is ""encouraged"" to try the configuration with the `ReportModule` disabled. However, this will remove the ability to save queries. On the other hand, granting `REPORT_CREATE` allows the user to write SQL in a report, and inspect tables that you may not wish them to access (gmessage:trac-users:J23IZuJddzA/3clXkEJBAAAJ). Simple example: {{{#!sql SELECT sid,value FROM session_attribute WHERE name='email'; }}} We should at least allow the user to save a query without also having the ability to create a report with SQL. For the `ReportModule`, we may want to consider limiting the tables that a user can access with `REPORT_CREATE` and `REPORT_MODIFY`. `REPORT_ADMIN` could continue to access all tables. ---- Copied from [ticket:9982#comment:10]: One possible improvement would be to add a setting for disabling the creation and editing of //SQL// reports, which are the ones with the drawbacks mentioned in the TracReports page. The list of reports, the wiki syntax and the possibility to saves and edit the non-SQL queries would continue to work. ---- Copied from [ticket:9982#comment:13: I think we can make a generic trac.report module, with the usual api/model/web_ui files. The details of handling and rendering legacy SQL reports could stay in source:trunk/trac/ticket/report.py and source:trunk/trac/ticket/templates/ That way, it will be easy to disable trac.ticket.report.* and still have the report list of saved custom queries.",trac@… next-major-releases Release,12780,Save wiki expander state,wiki system,,enhancement,normal,,new,2017-04-19T10:02:35+02:00,2019-09-06T19:51:32+02:00,"Replying to [comment:3:ticket:12640 Christian Boos]: > As for the expander, ... the possibility to save as a UI preference the initial state of the wiki page, expanded or constrained. ",anonymous next-major-releases Release,12746,Add keyboard shortcut for blame from file view,version control/browser,,enhancement,normal,,new,2017-03-25T22:54:16+01:00,2017-03-28T06:21:58+02:00,"From the repository browser, the keyboard shortcut `a` leads to the annotation view. It would be nice to have the same from the file view. Line highlighting and anchor should be preserved when navigating to the blame view using either the keyboard shortcut or the contextual navigation link.",Ryan J Ollos next-major-releases Release,12741,Make WikiProcessors foldable,wiki system,,enhancement,normal,,new,2017-03-22T23:00:10+01:00,2019-09-06T19:51:32+02:00,"Discussed in #11550, extend WikiProcessors with features similar to: * [th:FoldMacroProcessorMacro] (see th:#13184 for description of how the macro fails to do its whole job) * [th:CollapsiblePlugin]",Ryan J Ollos next-major-releases Release,12717,Paginated attachment index,attachment,,enhancement,normal,,new,2017-03-10T19:56:21+01:00,2017-03-13T04:20:38+01:00,"It would be useful to have a `TitleIndex`-like listing of attachments at `/attachment/`. `/attachment` could contain links to each attachment realm, `/attachment/ticket`,`/attachment/wiki`, ... could contain indexes of attachments for each realm. See also [th:AllAttachmentsMacro].",Ryan J Ollos next-major-releases Release,12687,Provide additional hints when trac-admin repository commands are not provided a valid repository name,version control,,defect,normal,,new,2017-02-12T08:44:19+01:00,2017-02-12T08:44:19+01:00,"From reports on the MailingList and personal experience, it seems users mistakenly input ""repository "" rather than ""repository "" for the trac-admin repository commands: {{{#!sh $trac-admin tracenvs/proj-1.3 help repository repository add [type] Add a source repository repository alias Create an alias for a repository repository list List source repositories repository remove Remove a source repository repository resync [rev] Re-synchronize trac with repositories repository set Set an attribute of a repository repository sync [rev] Resume synchronization of repositories }}} We could provide additional hints, like the following: {{{#!sh $trac-admin tracenvs/proj-1.3 repository sync ""notfound"" 10 TracError: Repository ""notfound"" doesn't exist. Use 'repository list' to view the valid repository names. }}} {{{#!diff diff --git a/trac/versioncontrol/admin.py b/trac/versioncontrol/admin.py index 86294ac4f..0acd0e99a 100644 --- a/trac/versioncontrol/admin.py +++ b/trac/versioncontrol/admin.py @@ -128,7 +128,9 @@ class VersionControlAdmin(Component): reponame = '' repos = rm.get_repository(reponame) if repos is None: - raise TracError(_(""Repository \""%(repo)s\"" doesn't exist"", + raise TracError(_(""Repository \""%(repo)s\"" doesn't exist.\n"" + ""\nUse 'repository list' to view the "" + ""valid repository names."", repo=reponame or '(default)')) if rev is not None: repos.sync_changeset(rev) }}} Similar changes would need to be made for the other `trac-admin repository` commands. Alternatively we could list the repository names when the error occurs.",Ryan J Ollos next-major-releases Release,12685,"Add action menu for creating, deleting and modifying resources or parts thereof",general,1.3dev,enhancement,normal,,new,2017-02-11T15:06:50+01:00,2019-09-06T19:54:30+02:00,"With this menu, it would be easier to perform actions on resources (wiki page, ticket #10948, ticket comment #12671). The idea would be to remove from the contextual navigation the link what's not a navigation to another resource (e.g. wiki history is rather an action on the resource itself, show me the page's history), and to gather the bottom of the page buttons (Edit, Delete, Rename) into a contextual menu at the top of the resource, a bit like what we have now for the ticket comment. Copied from [ticket:12671#comment:2]: ---- Replying to [ticket:12671 Jun Omae]: > ... the buttons are shown only when cursor is hovered in property changes box. As the results, it is difficult to click the buttons. But also the whole

line as well, in particular the `comment:n` part which is just above the buttons. But I think you're right anyway, it's not as obvious as it should be. If you remember my idea presented recently on trac-dev about a ""page menu"" (gmessage:trac-dev:eVU0q8Tatgo/pokz_FIyEQAJ), I think it could be transposed here as well: we would have a permanent menu reminder character/icon (▤, ≣), which would show the list of actions for the comment when hovered / clicked. Something like: {{{#!html

comment:n Modified by Author

≣
}}} \\ ---- ",Christian Boos next-major-releases Release,12679,"Add ""View blame prior to this change"" link to blame view",version control/browser,,enhancement,normal,,new,2017-02-07T19:52:48+01:00,2019-09-06T19:54:30+02:00,"Blame view ([source:trunk/Makefile?annotate=blame example]) seems to be missing this very useful feature available on github ([https://github.com/edgewall/trac/blame/trunk/Makefile example]). On github each blame annotation line has a small icon with tooltip ""View blame prior to this change"" that links to the same blame page but at an older revision. This is allows quickly navigating through the history of the file. ",anonymous next-major-releases Release,12662,"TicketQuery macro, add title and select parameter for table format",query system,1.2,enhancement,normal,,new,2017-01-16T09:45:16+01:00,2019-09-06T19:51:32+02:00,"Currently the TicketQuery macro with format=table always has a titlebar like ""Results (1 - 10 of 55)"". Please make this an option: * title: true/false (defaults to **true**) Then it's annoying to have the page selection twice on top and on bottom. Please make that an option: * select: top/bottom/none (defaults to **bottom**) == Example of the current implementation of the macro: [[TicketQuery(max=3,order=modified,modified=30daysago..,desc=1,format=table,col=summary|modified|reporter|owner)]]",Massimo next-major-releases Release,12596,AuthzPolicy should recursively expand groups,general,,enhancement,normal,,new,2016-09-20T23:45:37+02:00,2016-09-20T23:45:37+02:00,AuthzPolicy does not allow groups to contain other groups. This is documented in TracFineGrainedPermissions#MissingFeatures. The request has also been raised in gmessage:trac-users:TGoX7YfG_xA/0uoB7AQTBrsJ.,Ryan J Ollos next-major-releases Release,12595,TracAdmin should label invalid permissions,admin/console,,enhancement,normal,,new,2016-09-20T22:44:55+02:00,2016-09-20T22:44:55+02:00,"In #10752 we added styling for invalid permissions on the //Permissions Admin// page. [[Image(Screen Shot 2016-09-20 at 13.38.27.png,100%)]] TracAdmin gives no indication if a permission is invalid: {{{#!sh $ trac-admin ../tracenvs/proj-milestonetest permission list user2 User Action ------------------------- user2 INVALID_PERMISSION user2 TICKET_APPEND user2 TICKET_CHGPROP user2 TICKET_CREATE user2 TICKET_MODIFY user2 WIKI_CREATE user2 WIKI_MODIFY Available actions: BROWSER_VIEW, CHANGESET_VIEW, CONFIG_VIEW, EMAIL_VIEW, FILE_VIEW, LOG_VIEW, MILESTONE_ADMIN, MILESTONE_CREATE, MILESTONE_DELETE, MILESTONE_MODIFY, MILESTONE_VIEW, PERMISSION_ADMIN, PERMISSION_GRANT, PERMISSION_REVOKE, REPORT_ADMIN, REPORT_CREATE, REPORT_DELETE, REPORT_MODIFY, REPORT_SQL_VIEW, REPORT_VIEW, ROADMAP_ADMIN, ROADMAP_VIEW, SEARCH_VIEW, TICKET_ADMIN, TICKET_APPEND, TICKET_BATCH_MODIFY, TICKET_CHGPROP, TICKET_CREATE, TICKET_EDIT_CC, TICKET_EDIT_COMMENT, TICKET_EDIT_DESCRIPTION, TICKET_MODIFY, TICKET_VIEW, TIMELINE_VIEW, TRAC_ADMIN, VERSIONCONTROL_ADMIN, WIKI_ADMIN, WIKI_CREATE, WIKI_DELETE, WIKI_MODIFY, WIKI_RENAME, WIKI_VIEW }}} I think that `trac-admin` properly lists `INVALID_PERMISSION` as being assigned to `user2`, and does not include `INVALID_PERMISSION` in the list of //Available actions//. However, there should be a label on `INVALID_PERMISSION`, such as the following: {{{#!sh $ trac-admin ../tracenvs/proj-milestonetest permission list user2 User Action ------------------------- user2 INVALID_PERMISSION * user2 TICKET_APPEND user2 TICKET_CHGPROP user2 TICKET_CREATE user2 TICKET_MODIFY user2 WIKI_CREATE user2 WIKI_MODIFY (*) Permission is no longer defined Available actions: BROWSER_VIEW, CHANGESET_VIEW, CONFIG_VIEW, EMAIL_VIEW, FILE_VIEW, LOG_VIEW, MILESTONE_ADMIN, MILESTONE_CREATE, MILESTONE_DELETE, MILESTONE_MODIFY, MILESTONE_VIEW, PERMISSION_ADMIN, PERMISSION_GRANT, PERMISSION_REVOKE, REPORT_ADMIN, REPORT_CREATE, REPORT_DELETE, REPORT_MODIFY, REPORT_SQL_VIEW, REPORT_VIEW, ROADMAP_ADMIN, ROADMAP_VIEW, SEARCH_VIEW, TICKET_ADMIN, TICKET_APPEND, TICKET_BATCH_MODIFY, TICKET_CHGPROP, TICKET_CREATE, TICKET_EDIT_CC, TICKET_EDIT_COMMENT, TICKET_EDIT_DESCRIPTION, TICKET_MODIFY, TICKET_VIEW, TIMELINE_VIEW, TRAC_ADMIN, VERSIONCONTROL_ADMIN, WIKI_ADMIN, WIKI_CREATE, WIKI_DELETE, WIKI_MODIFY, WIKI_RENAME, WIKI_VIEW }}}",Ryan J Ollos next-major-releases Release,12572,Add a download link to patch/diff renderer,wiki system,,enhancement,normal,,new,2016-08-31T01:25:06+02:00,2019-09-06T19:51:32+02:00,It would be nice to be able to download a patch that is posted inline of a comment. For example: comment:1:ticket:12546. We could add a download link to the patch renderer. I'd use it to grab a URL that could be passed to wget.,Ryan J Ollos next-major-releases Release,12570,Create a deprecated function decorator,general,,enhancement,normal,,new,2016-08-31T01:05:32+02:00,2019-09-06T19:51:32+02:00,"Since we've been deprecating and removing a lot of code, I thought it might be useful to add a `deprecated` decorator to `trac.util`. The decorator could log a message at the appropriate level, which might be `WARNING`. The message would indicate that a deprecated function, class or method has been called by a plugin.",Ryan J Ollos next-major-releases Release,12549,Provide option for blame to use merged history,version control,,enhancement,normal,,new,2016-07-28T21:31:40+02:00,2016-08-09T19:52:01+02:00,"`svn blame` has the `-g` option: {{{ -g [--use-merge-history] : use/display additional information from merge history }}} This can be useful for finding the changeset in which a line was changed, which is often more useful than the line in which the change was merged. For example, line 47 of [/browser/trunk/trac/core.py?rev=14978&annotate=blame] shows [13672], however that's a merge changeset and [13671] may be more interesting. `svn blame -g trac/core.py` on the trunk yields: {{{ G 13671 rjollos super(TracError, self).__init__(message) }}} while `svn blame trac/core.py` yields: {{{ 13672 rjollos super(TracError, self).__init__(message) }}} It looks like we could use `blame4` or `blame5` of `svn.client`, rather than [browser:trunk/tracopt/versioncontrol/svn/svn_fs.py@15012:865-866#L846 blame2]: {{{ svn_client_blame4(*args) svn_client_blame4(char const * path_or_url, svn_opt_revision_t peg_revision, svn_opt_revision_t start, svn_opt_revision_t end, svn_diff_file_options_t const * diff_options, svn_boolean_t ignore_mime_type, svn_boolean_t include_merged_revisions, svn_client_blame_receiver2_t receiver, void * receiver_baton, svn_client_ctx_t ctx, apr_pool_t pool) -> svn_error_t svn_client_blame5(*args) svn_client_blame5(char const * path_or_url, svn_opt_revision_t peg_revision, svn_opt_revision_t start, svn_opt_revision_t end, svn_diff_file_options_t const * diff_options, svn_boolean_t ignore_mime_type, svn_boolean_t include_merged_revisions, svn_client_blame_receiver3_t receiver, void * receiver_baton, svn_client_ctx_t ctx, apr_pool_t pool) -> svn_error_t }}} Btw, I wonder if it would be useful to have a `blame` TracLink, or extend the `browser` TracLink to allow blame to be specified.",Ryan J Ollos next-major-releases Release,12485,Add an ITicketCommentProvider interface,ticket system,,enhancement,normal,,new,2016-05-21T01:31:23+02:00,2018-02-26T13:10:13+01:00,"The design of the CommitTicketUpdater makes it difficult to retroactively add changeset comments (#11478). !CommitTicketUpdater comments added retroactively are not likely to be in the correct chronological order. It would be difficult to remove !CommitTicketUpdater comments if it's later decided to use a plugin such as [th:TracTicketChangelogPlugin] instead of having the changeset comments inline in the ticket changelog. The repository links are no longer correct if the repository name is changed (although this is generally true for TracLinks to the repository). I've also used the [https://wiki.jenkins-ci.org/display/JENKINS/Trac+Publisher+Plugin Jenkins Trac Publisher Plugin] which has similar issues. I've considered another approach. * Add an `ITicketCommentProvider` interface, that returns a comment along with a timestamp and possibly some other metadata. The ticket changelog would be created from data provided by `ITicketCommentProvider` implementations. * One of the `ITicketCommentProvider` methods could be `get_comment_filter`, so that it would be possible to have a filter such as //Show changesets// or //Show build results//. * The full history of a `ITicketCommentProvider` source would be available upon activating the plugin, and the comments would have the correct chronological ordering. * !CommitTicketUpdater would create a table associating tickets with changeset. As an aside, the table would also be useful for th:#12621. * !CommitTicketUpdater would provide comments through an implementation of `ITicketCommentProvider`. One difficult aspect might be, on activating the plugin it might be desired to remove all previous instances of comments containing `CommitTicketReferenceMacro` instances. This could be done in an upgrade script, or through a script provided in `contrib`.",Ryan J Ollos next-major-releases Release,12442,Support per-repository authz_file,version control,,enhancement,normal,,new,2016-04-04T21:43:39+02:00,2019-09-22T01:31:38+02:00,"Discussed in comment:5:ticket:11703 and gmessage:trac-users:sI9q_1OF24s/QpeUkYrWCgAJ, since Subversion 1.7 a per-repository `authz_file` is supported. We should continue to support a single `authz_file`, but also add support for the per-repository `authz_file`. References: * [https://subversion.apache.org/docs/release-notes/1.7.html#per-repos-authz 1.7 Release notes] * [http://svnbook.red-bean.com/nightly/en/svn.serverconfig.pathbasedauthz.html Getting started with path-based access control] * [http://svnbook.red-bean.com/nightly/en/svn.serverconfig.httpd.html#svn.serverconfig.httpd.authz.perdir Configuring per-directory access control]",Ryan J Ollos next-major-releases Release,12390,Attachment table primary key not unique when migrating from SQLite to MySQL,attachment,,defect,normal,,new,2016-03-07T09:22:36+01:00,2016-05-05T13:08:03+02:00,"We might just file this under the category of poor database schema, but I'm hoping to get feedback. I'm working on a migration from SQLite to MySQL and encountered the following error: {{{ Copying tables: attachment table... IntegrityError: (1062, ""Duplicate entry 'ticket-393-lv_pause_you_can_say_orderstatus_requestacatalog_cust' for key 'PRIMARY'"") }}} The length of each primary key is 111, which is calculated as: [browser:tags/trac-1.0.10/trac/db/mysql_backend.py@:149#L140]. Here is the schema: {{{#!mysql mysql> SHOW CREATE TABLE attachment; +------------+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+ | Table | Create Table | +------------+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+ | attachment | CREATE TABLE `attachment` ( `type` text COLLATE utf8_bin NOT NULL, `id` text COLLATE utf8_bin NOT NULL, `filename` text COLLATE utf8_bin NOT NULL, `size` int(11) DEFAULT NULL, `time` bigint(20) DEFAULT NULL, `description` text COLLATE utf8_bin, `author` text COLLATE utf8_bin, `ipnr` text COLLATE utf8_bin, PRIMARY KEY (`type`(111),`id`(111),`filename`(111)) ) ENGINE=InnoDB DEFAULT CHARSET=utf8 COLLATE=utf8_bin | }}} The issue is that a particular ticket has several attachments that differ only in one of the last characters, having a `.1` and `.2` suffix. The attachment filename length is greater than 111 characters. {{{#!sql SELECT LENGTH(filename),filename FROM attachment WHERE id=393 AND LENGTH(filename) > 111 ORDER BY filename; 127|lv_pause_you_can_say_orderstatus_requestacatalog_customerservice_or_placeanorder_lv_say_repeat_to_hear_the_options_again..2.wav 125|lv_pause_you_can_say_orderstatus_requestacatalog_customerservice_or_placeanorder_lv_say_repeat_to_hear_the_options_again..wav 120|lv_pause_you_can_say_placeaorder_orderstatus_requestacatalog_or_customerservice_say_repeat_to_hear_the_options_again.wav 121|lv_you_can_say_orderstatus_requestacatalog_customerservice_or_placeanorder_lv_say_repeat_to_hear_the_options_again..2.wav 119|lv_you_can_say_orderstatus_requestacatalog_customerservice_or_placeanorder_lv_say_repeat_to_hear_the_options_again..wav 114|lv_you_can_say_placeaorder_orderstatus_customerservice_or_requestacatalog_say_repeat_to_hear_the_options_again.wav }}} I haven't tested, but assume that if the organization had always been using MySQL the users would have seen `IntegrityError`s when adding some of these attachments. Would it make sense to change the database schema to use the hashed filename as a primary key instead of the filename?",Ryan J Ollos next-major-releases Release,12388,Help text output from TracAdmin should be wrapped,admin/console,,defect,normal,,new,2016-03-06T22:33:03+01:00,2016-03-06T22:33:03+01:00,"Discussed in th:#12691, the help text should be wrapped so that it fits in a column. {{{#!sh trac-admin - The Trac Administration Console 1.0.10 Usage: trac-admin [command [subcommand] [option ...]] Invoking trac-admin without command starts interactive mode. help Show documentation initenv Create and initialize a new environment attachment add Attach a file to a resource attachment export Export an attachment from a resource to a file or stdout attachment list List attachments of a resource attachment remove Remove an attachment from a resource changeset added Notify trac about changesets added to a repository changeset modified Notify trac about changesets modified in a repository component add Add a new component component chown Change component ownership component list Show available components component remove Remove/uninstall a component component rename Rename a component config get Get the value of the given option in ""trac.ini"" config remove Remove the specified option from ""trac.ini"" config set Set the value for the given option in ""trac.ini"" deploy Extract static resources from Trac and all plugins hotcopy Make a hot backup copy of an environment migrate Migrate to new environment with another database or another database without creating a new environment. milestone add Add milestone milestone completed Set milestone complete date milestone due Set milestone due date milestone list Show milestones milestone remove Remove milestone milestone rename Rename milestone permission add Add a new permission rule permission export Export permission rules to a file or stdout as CSV permission import Import permission rules from a file or stdin as CSV permission list List permission rules permission remove Remove a permission rule priority add Add a priority value option priority change Change a priority value priority list Show possible ticket priorities priority order Move a priority value up or down in the list priority remove Remove a priority value repository add Add a source repository ... }}}",Ryan J Ollos next-major-releases Release,12257,Add single-sign-on authentication as an optional component,web frontend,,enhancement,normal,,new,2015-11-27T05:47:28+01:00,2019-08-12T23:28:57+02:00,"After reviewing [th:SharedCookieAuthPlugin], #12251 and #8486, it looks like it would be fairly straightforward to add single-sign-on authentication as an optional component (in `tracopt.web.auth`). ",Ryan J Ollos next-major-releases Release,12212,Order timeline filters alphabetically by label,timeline,,enhancement,normal,,new,2015-09-23T00:23:05+02:00,2016-02-23T09:12:45+01:00,"The timeline filters seems to have a random ordering, determined by order in which the extensions points are discovered. It would be nice to have a predictable ordering and it seems logical to order alphabetically by label.",Ryan J Ollos next-major-releases Release,12210,Show username and email in tooltip when hovering over a full name,general,,enhancement,normal,,new,2015-09-22T22:34:29+02:00,2016-04-30T10:42:10+02:00,"When `[trac]` `show_full_names` is `True` the user's full name is shown. The username and email could also be shown in the tooltip when hovering over the full name, like th:UsernameDecoratePlugin does. The email address would be obfuscated or not shown depending on the settings and user permissions.",Ryan J Ollos next-major-releases Release,12165,List users included and excluded from notification,notification,,enhancement,normal,,new,2015-08-12T22:18:41+02:00,2017-05-05T18:19:43+02:00,"Bugzilla, after submitting changes for a ticket displays the following: {{{#!html
Changes submitted for bug 24
Email sent to:
remy@..., christian@..., jun@...
Excluding:
ryan@... peter@...
}}} Similarly, we could show a notice at the top of the page so that it's clear which users have been notified.",Ryan J Ollos next-major-releases Release,12162,Edit comment should be AJAX or standalone edit screen,ticket system,,enhancement,normal,,new,2015-08-09T12:06:55+02:00,2019-09-06T19:51:32+02:00,"As of today's trunk, editing a comment means the request still loads all the other comments, which on a big ticket is massively slow. I have yet to feel I need to take a look at other comments while editing my target. AJAX would be preferred, it's cleraly the fastest way to do it, but perhaps an incremental improvement could be achived much quicker by just patching out the other comments from the edit page load. ",lkraav next-major-releases Release,12121,Subversion copies could show peg revision in browser view,version control/browser,,defect,normal,,new,2015-07-14T21:51:09+02:00,2015-08-06T04:24:45+02:00,"Here is an example from the browser view: {{{#!html
  • Projects/Presto/Software/tags/urs_1.02.06 (copied) (copied from Projects/Presto/Software/branches/REL-1.xx-URS)
    • }}} The proposed change is to show the peg revision in the path: {{{#!html
  • Projects/Presto/Software/tags/urs_1.02.06 (copied) (copied from Projects/Presto/Software/branches/REL-1.xx-URS@10603)
    • }}}",Ryan J Ollos next-major-releases Release,12119,Options not written to trac.ini when --inherit option used,database backend,,enhancement,normal,,new,2015-07-13T01:56:43+02:00,2021-04-19T12:05:27+02:00,"Tested with the latest trunk r14154. The trac.ini file on the left was created using an empty inherited file {{{--inherit=`pwd`/global.ini}}}. The file on the right was created without specifying `--inherit`. It seems many options are missing from `trac.ini` when `--inherit` is used. It seems only `ConfigSection`s and options explicitly specified when creating the environment are written. In the case that a `ConfigSection` is specified in the inherited file it won't be written. The possible reason for not writing an option when there is an inherited file is to avoid having to delete the option from trac.ini when it is added to the inherited file. It seems inconsistent though to have only `ConfigSection`s in trac.ini. I'll have to investigate if the behavior changed after #11437. {{{#!diff --- trac/conf/trac.ini 2015-07-12 03:57:27.439652744 -0700 +++ trac2/conf/trac.ini 2015-07-12 03:58:12.643651198 -0700 @@ -1,7 +1,44 @@ # -*- coding: utf-8 -*- +[attachment] +max_size = 262144 +max_zip_size = 2097152 +render_unsafe_content = disabled + +[browser] +color_scale = enabled +downloadable_paths = /trunk,/branches/*,/tags/* +hide_properties = svk:merge +intermediate_color = +intermediate_point = +newest_color = (255, 136, 136) +oldest_color = (136, 136, 255) +oneliner_properties = trac:summary +render_unsafe_content = disabled +wiki_properties = trac:description + +[changeset] +max_diff_bytes = 10000000 +max_diff_files = 0 +wiki_format_messages = enabled + +[header_logo] +alt = (please configure the [header_logo] section in trac.ini) +height = -1 +link = +src = site/your_project_logo.png +width = -1 + [inherit] -file = /home/user/Workspace/test-inherit/global.ini +htdocs_dir = +plugins_dir = +templates_dir = + +[logging] +log_file = trac.log +log_format = +log_level = DEBUG +log_type = none [mainnav] admin.order = 8.0 @@ -20,13 +57,108 @@ logout.order = 2.0 prefs.order = 3.0 +[milestone] +default_retarget_to = +stats_provider = DefaultTicketGroupStatsProvider + +[mimeviewer] +max_preview_size = 262144 +mime_map = text/x-dylan:dylan,text/x-idl:ice,text/x-ada:ads:adb +mime_map_patterns = text/plain:README|INSTALL|COPYING.* +pygments_default_style = trac +pygments_modes = +tab_width = 8 +treat_as_binary = application/octet-stream,application/pdf,application/postscript,application/msword,application/rtf + +[notification] +admit_domains = +ambiguous_char_width = single +batch_subject_template = $prefix Batch modify: $tickets_descr +email_address_resolvers = SessionEmailResolver +email_sender = SmtpEmailSender +ignore_domains = +mime_encoding = none +sendmail_path = sendmail +smtp_always_bcc = +smtp_always_cc = +smtp_default_domain = +smtp_enabled = disabled +smtp_from = trac@localhost +smtp_from_author = disabled +smtp_from_name = +smtp_password = +smtp_port = 25 +smtp_replyto = trac@localhost +smtp_server = localhost +smtp_subject_prefix = __default__ +smtp_user = +ticket_subject_template = $prefix #$ticket.id: $summary +use_public_cc = disabled +use_short_addr = disabled +use_tls = disabled + [notification-subscriber] always_notify_cc = CarbonCopySubscriber always_notify_previous_updater = TicketPreviousUpdatersSubscriber always_notify_updater = TicketUpdaterSubscriber [project] +admin = +admin_trac_url = . +descr = My example project +footer = Visit the Trac open source project at
    http://trac.edgewall.org/ +icon = common/trac.ico name = My Project +url = + +[query] +default_anonymous_query = status!=closed&cc~=$USER +default_query = status!=closed&owner=$USER +items_per_page = 100 +ticketlink_query = ?status=!closed + +[report] +items_per_page = 100 +items_per_page_rss = 0 + +[revisionlog] +default_log_limit = 100 +graph_colors = #cc0,#0c0,#0cc,#00c,#c0c,#c00 + +[roadmap] +stats_provider = DefaultTicketGroupStatsProvider + +[search] +default_disabled_filters = +min_query_length = 3 + +[sqlite] +extensions = + +[svn] +authz_file = +authz_module_name = + +[ticket] +allowed_empty_fields = milestone,version +default_cc = +default_component = +default_description = +default_keywords = +default_milestone = +default_owner = < default > +default_priority = major +default_resolution = fixed +default_severity = +default_summary = +default_type = defect +default_version = +max_comment_size = 262144 +max_description_size = 262144 +max_summary_size = 262144 +preserve_newlines = default +restrict_owner = disabled +workflow = ConfigurableTicketWorkflow [ticket-workflow] accept = new,assigned,accepted,reopened -> accepted @@ -51,6 +183,63 @@ resolve.operations = set_resolution resolve.permissions = TICKET_MODIFY +[timeline] +abbreviated_messages = enabled +changeset_collapse_events = disabled +changeset_long_messages = disabled +changeset_show_files = 0 +default_daysback = 30 +max_daysback = 90 +newticket_formatter = oneliner +ticket_show_component = disabled +ticket_show_details = disabled + [trac] +auth_cookie_lifetime = 0 +auth_cookie_path = +auto_preview_timeout = 2.0 +auto_reload = disabled +backup_dir = db +base_url = +check_auth_ip = disabled database = sqlite:db/trac.db +debug_sql = disabled +default_charset = utf-8 +default_date_format = +default_dateinfo_format = relative +default_handler = WikiModule +default_language = +default_timezone = +genshi_cache_size = 128 +htdocs_location = +ignore_auth_case = disabled +jquery_location = +jquery_ui_location = +jquery_ui_theme_location = +never_obfuscate_mailto = disabled +permission_policies = ReadonlyWikiPolicy,DefaultPermissionPolicy,LegacyAttachmentPolicy +permission_store = DefaultPermissionStore +request_filters = +resizable_textareas = enabled +secure_cookies = disabled +show_email_addresses = disabled +show_ip_addresses = disabled +timeout = 20 +use_base_url_for_redirect = disabled +use_chunked_encoding = disabled +use_xsendfile = disabled +wiki_toolbars = enabled +xsendfile_header = X-Sendfile + +[versioncontrol] +allowed_repository_dir_prefixes = +default_repository_type = svn + +[wiki] +default_edit_area_height = 20 +ignore_missing_pages = disabled +max_size = 262144 +render_unsafe_content = disabled +safe_schemes = cvs,file,ftp,git,irc,http,https,news,sftp,smb,ssh,svn,svn+ssh +split_page_names = disabled }}}",Ryan J Ollos next-major-releases Release,12077,Ticket workflow should support negated permissions,ticket system,,enhancement,normal,,new,2015-05-17T01:38:13+02:00,2015-07-28T22:46:02+02:00,"This would support the user case of overlapping workflow actions in which it is redundant to show both actions to a user that has the higher permission. For example, consider a case in which any user with `TICKET_MODIFY` can reopen a ticket, but only a user with `TICKET_ADMIN` can reopen and change the owner of a ticket. The user with `TICKET_ADMIN` would see both the //reopen// and //reopen_and_assign// actions, which is redundant. {{{#!ini reopen.permissions = TICKET_MODIFY, !TICKET_ADMIN reopen_and_assign.permissions = TICKET_ADMIN }}}",Ryan J Ollos next-major-releases Release,12004,Replace uses of assert with exception-raising statements,general,,enhancement,normal,,new,2015-03-27T20:06:23+01:00,2015-11-12T00:37:39+01:00,"It was noted in comment:7:ticket:11295 that we should consider replacing uses of `assert`, which will be removed when running the Python interpreter in optimized (`-O`) mode, with `raises`. Here is an example: [browser:/tags/trac-1.0.5/trac/web/session.py@:242-244#L241]. There are a few possible paths we could pursue: 1. Replace all uses of `assert`. 1. Use `assert` to protect against coding errors, `raise` exceptions for errors that can be triggered by a user in the scenario of proper-functioning code. Whichever approach we take, it should be captured in TracDev/CodingStyle.",Ryan J Ollos next-major-releases Release,11992,Allow git commit hashes to be abbreviated to 7 characters,plugin/git,1.0.4,defect,normal,,new,2015-03-21T09:11:55+01:00,2022-08-07T15:41:34+02:00,"Currently an abbreviated git commit hash is only recognised by trac if it's at least 8 characters. But git itself defaults to abbreviating to 7 characters (if the 7 character abbreviation is unique in the repo) - to demonstrate this: {{{ olly@pippikin:~/git/xapian$ git log -1 --oneline 575eab1 docs/: Add a document about character encoding, as suggested by James Aylett in #550. }}} The fix is a single character change - I'll attach a patch in a moment. This seems to affect 1.0.4 and trunk.",olly@… next-major-releases Release,11985,Create a [database] section in trac.ini,database backend,,task,normal,,new,2015-03-17T23:54:59+01:00,2015-10-14T03:55:00+02:00,"Continuing with the theme of better categorizing the options in trac.ini (#11776, #11703, #11849, #11956, and less-so #11856, #11976), I'm considering the following: * `[trac]` `backup_dir` -> `[database]` `backup_dir` * `[trac]` `database` -> `[database]` `connection_uri` * `[trac]` `timeout` -> `[database]` `connection_timeout` * `[trac]` `debug_sql` -> `[database]` `debug_sql` * `[trac]` `pg_dump_path` -> `[postgresql]` `dump_path` * `[trac]` `mysqldump_path` -> `[mysql]` `dump_path` I'm considering putting the latter two in the `[database]` section, however note that an [TracIni#sqlite-section [sqlite]] section already exists. Careful review of the `[trac]` section may lead to some other options for re-categorization. Given the number of Option changes that are already made for milestone:1.2, it seems best to make these changes before that release.",Ryan J Ollos next-major-releases Release,11968,Contextual navigation link to log view from changeset view,general,,enhancement,normal,,new,2015-02-24T23:58:14+01:00,2015-02-25T02:44:54+01:00,"It would be beneficial to have a link to the //Revision Log// from the changeset view. The link could be placed in the contextual navigation, like we see on the //Browser// page.",Ryan J Ollos next-major-releases Release,11963,#!td processor inserts unwanted

    ...

    ,wiki system,1.0.4,defect,normal,,new,2015-02-19T10:51:32+01:00,2021-05-18T02:37:47+02:00,"The table using the html processors has additional padding due to `

    ` elements that get inserted automatically. {{{ {{{#!th some }}} {{{#!td example }}} }}} {{{#!th some }}} {{{#!td example }}} In a long table, this vertical margin wastes too much space. I would prefer the compact style from the pipe syntax, but need to add css styling.",bastiK next-major-releases Release,11959,"Add custom fields to ""Group results by"" drop down in Custom Query page.",ticket system,1.0.1,enhancement,normal,,new,2015-02-12T19:21:21+01:00,2015-08-09T05:33:28+02:00,"Currently, the ""Group results by"" drop down in the ""Custom Query"" page only includes standard fields. No custom ticket fields appear there. It would be very useful to include custom fields there. ",anonymous next-major-releases Release,11946,Allow multiple permissions to be assigned in single submit,admin/web,,enhancement,normal,,new,2015-02-02T21:16:48+01:00,2015-03-02T22:35:15+01:00,"It is quick to remove multiple permissions from a subject because they can be selected and removed in a single submit. Assigning multiple permissions is not as quick. The Django admin page has been suggested on the Bloodhound mailing list as a good example to follow: [[Image(img6eR8.png,100%)]] [http://www.erichynds.com/examples/jquery-ui-multiselect-widget/demos/ jQuery multiselect widget] provides another possible implementation.",Ryan J Ollos next-major-releases Release,11943,Ticket delete notifications,notification,,enhancement,normal,,new,2015-01-31T21:16:14+01:00,2017-02-21T12:40:47+01:00,"Trac supports ticket deletion through the `ticket remove` TracAdmin command, and the `TicketDeleter` Component in `/tracopt/ticket`. It would be useful to receive notifications when tickets are deleted. The [th:AccountManagerPlugin] has a similar feature of sending notifications when users are removed. The subject and body are very simple, e.g. `Deleted User: tonyjha99`. The users to receive notifications are configured through the `[account-manager]` `account_changes_notify_address` option.",Ryan J Ollos next-major-releases Release,11939,Workflow graph should account for user/group permissions,ticket system,,defect,normal,,new,2015-01-27T09:11:12+01:00,2017-05-18T18:12:42+02:00,"The workflow actions have a [TracWorkflow#BasicTicketWorkflowCustomization permissions] attribute. Workflows can be configured such that different users will be configured with different actions. It would be useful to be able to visualize the workflow for a particular user or group using the [WikiMacros#Workflow-macro WorkflowMacro]. The !WorkflowMacro could be modified to accept a `username_or_group` argument. In the ticket workflow dialog the workflow for that user could be shown in a popover. Alternatively, it might be more useful to show all the workflow transitions but make it clear which transitions the current user is allowed to initiate.",Ryan J Ollos next-major-releases Release,11937,Improve suggestion of similarly named pages,wiki system,,enhancement,normal,,new,2015-01-26T23:59:49+01:00,2019-09-06T19:51:32+02:00,"Trac suggests similarly named pages when navigating to a non-existent page with a CamelCase name, but the suggestion is naive. Trac will suggest CamelCase if CamelCas is entered, but not if CameCase is entered. The code that generates suggestions is [browser:/tags/trac-1.0.3/trac/wiki/web_ui.py@:636-638#L618 here]. The aim of this ticket will be to improve the suggestions. The [browser:/tags/trac-1.0.3/trac/util/text.py@:703#L703 levenshtein_distance] function used by [browser:/tags/trac-1.0.3/trac/admin/api.py@:152#L140 AdminCommandManager] may be useful in implementing this ticket.",Ryan J Ollos next-major-releases Release,11919,Make diff paths available in the UI.,web frontend,1.0.1,enhancement,normal,,new,2015-01-14T19:49:02+01:00,2019-09-06T19:51:32+02:00,"In regards to this ticket: http://trac.edgewall.org/ticket/9824 If I want to compare a branch vs trunk. Eg Observing a junior/new hire's work. I have to go into the HTML - using a DOM inspector - and manually change the ""old path"" hidden form field to trunk. This could easily be made available in the interface for users. It would also make my life easier. I could probably contribute this change if it is seen as a valuable enhancement by others. When comparing trunk(old path) against a branch(new path) the revision must be set to the latest revision of the branch.",wrinkledcheese@… next-major-releases Release,11889,On-site notifications,notification,,enhancement,normal,,new,2014-12-21T19:45:00+01:00,2020-08-10T02:29:49+02:00,"''On-site notifications'' are displayed to users on the Trac site (while e.g. ''email notifications'' are sent by email). With [wiki:TracDev/Proposals/AdvancedNotification/INotificationDistributor INotificationDistributor] we can send notifications via email, XMPP or any other transport channel. We could also ""send"" notifications to the appropriate user account on the Trac site itself and store them until the user visits the site and gets notified.",Peter Suter next-major-releases Release,11887,List Ticket Attachments in TracNotification,notification,1.0-stable,enhancement,normal,,new,2014-12-19T18:53:31+01:00,2015-08-27T07:13:59+02:00,"I'd like to list the filenames and possibly URLs of a ticket's attachments in each notification email. I need to gain access to ticket attachments in the `trac_notify_email.txt`. I've been working on it for the past couple of days and am not sure if this is possible. Please excuse my limited experience working with Trac and Genshi. Is this something that can be done currently? If not maybe it's worth adding as a feature? Thanks",anonymous next-major-releases Release,11884,Notification distributor for XMPP,notification,,enhancement,normal,,new,2014-12-17T23:17:53+01:00,2019-09-06T19:51:32+02:00,The new [wiki:TracDev/PluginDevelopment/ExtensionPoints/trac.notification.api.INotificationDistributor INotificationDistributor] interface allows distributing notifications over different transport channels. So far we only have an email distributor. We could also provide XMPP support in `tracopt`.,Peter Suter next-major-releases Release,11879,Prevent ticket workflow hints from wrapping,ticket system,,enhancement,normal,,new,2014-12-12T20:46:30+01:00,2015-03-02T22:27:22+01:00,"It would be nice to prevent the text wrapping in the workflow hint that is seen even with the default workflow. One option would be to prevent line wrapping and truncate overflow with ellipsis. The entire hint could be shown in the `title` of the `span`. [[Image(20141212T104313.png,100%)]]",Ryan J Ollos next-major-releases Release,11866,Export information on /about page as wiki markup and plain text,general,,enhancement,normal,,new,2014-12-05T04:14:02+01:00,2017-08-30T08:40:55+02:00,"On the mailing list and when handling plugin issues on trac-hacks.org we are often asking users to post the information about their environment from the `/about` page. We could make this process more user friendly by exporting contents of the page as wiki markup for pasting into a ticket, and as plain text for posting on the mailing list. When the //Create// button on the error page is pressed a table of system info is created in the ticket description, so it seems that part of this functionality is already implemented: [browser:/tags/trac-1.0.2/trac/web/main.py@:582-590#L580].",Ryan J Ollos next-major-releases Release,11862,Allow keywords to be highlighted in code WikiProcessor,rendering,,enhancement,normal,,new,2014-12-03T16:13:43+01:00,2015-06-14T01:39:52+02:00,"The proposed enhancement adds a `keywords` WikiProcessor argument, for highlighting in bold the words or phrases specified in a comma-separated list. Related to SO:21775494.",Ryan J Ollos next-major-releases Release,11840,Add ISessionAttributeProvider interface,general,,enhancement,normal,,new,2014-11-19T06:22:05+01:00,2019-09-06T19:51:32+02:00,"It was described in comment:15:ticket:11612 how arbitrary data can be stored in session attributes by editing the client-side JavaScript in the browser. The proposed solution is to add a new interface for registering session attributes: {{{#!python class ISessionAttributeProvider(Interface): def get_session_attribute_names(): """"""Return a list of supported names in session attribute."""""" def validate_session_attribute(name, value): """"""Validate the given attribute."""""""" }}}",Ryan J Ollos next-major-releases Release,11826,Optimize ticket.query.execute,query system,1.1.2,enhancement,minor,,new,2014-11-13T19:19:19+01:00,2021-10-21T21:53:55+02:00,"The ticket.query.execute function could be optimised somewhat... {{{ if self.max == 0: self.has_more_pages = False self.offset = 0 else: self.has_more_pages = True self.offset = self.max * (self.page - 1) }}} source:trunk/trac/ticket/query.py?rev=13234#L110 {{{ self.num_items = self._count(sql, args) if self.num_items <= self.max: self.has_more_pages = False if self.has_more_pages: }}} source:trunk/trac/ticket/query.py?rev=13234#L295 if''' self.max = 0''' this: {{{self.num_items = self._count(sql, args)}}} is useless since {{{if self.has_more_pages:}}} will always be False. So I suggest adding something like: {{{ if self.has_more_pages: // This line self.num_items = self._count(sql, args) if self.num_items <= self.max: self.has_more_pages = False if self.has_more_pages: }}} That would save one sql-query which in worst case scenario could double(?) the execution time I suppose",anonymous next-major-releases Release,11746,Generate list of valid default_handlers in ExtensionOption documentation,general,,enhancement,normal,,new,2014-09-08T11:06:28+02:00,2016-08-30T23:39:30+02:00,"The list of valid `[trac]` `default_handlers` in the `ExtensionOption` docstring should be dynamically generated. The list should includes all `IRequestHandlers` for enabled plugins with `is_valid_default_handler = True`. First mentioned in comment:description:ticket:11519. Similarly, it would be useful to dynamically generate the list of allowed values for `[trac]` `mainnav` and `metanav`. More generally, the feature would be useful for any option with allowed values that can be extended by plugins.",Ryan J Ollos next-major-releases Release,11730,Don't quickjump if multiple words in search phrase,search system,,enhancement,normal,,new,2014-08-24T03:28:01+02:00,2015-06-01T00:48:22+02:00,"I noticed when seaching for a ticket that `TracLinks wiki diff` will result in a quickjump to the TracLinks page. When multiple words are used for search, I assume in most cases the user does not want to quickjump.",Ryan J Ollos next-major-releases Release,11722,Batch ticket delete,query system,,enhancement,normal,,new,2014-08-17T21:06:48+02:00,2015-03-02T22:27:22+01:00,The batch ticket delete feature request mentioned in th:#8074 could be useful. Users should be required to have `TICKET_BATCH_MODIFY` and `TICKET_DELETE` (or perhaps even have a `TICKET_BATCH_DELETE` permission).,Ryan J Ollos next-major-releases Release,11713,Repositories can still be added when no IRepositoryConnectors enabled,admin/web,1.0-stable,defect,normal,,new,2014-08-09T18:55:27+02:00,2019-10-22T20:50:41+02:00,"When there are no `IRepositoryConnector`s enabled, the //Manage Repositories// page is still shown, including the //Add Repositories// fieldset. The allowed type is //(default)//. I propose adding a notice that //No repository connectors are enabled//, and to not allow a user to add a new repository. With the following configuration in //trac.ini// and the `MercurialConnector` not enabled, the repository is displayed on the table with no indicator that it is not navigable. {{{#!ini [repositories] hg-plugin.dir = /home/user/Workspace/mercurial-plugin hg-plugin.type = hg }}} On the other hand, the //Browser Source// tab is not present and when navigating to `/browser`, the following text is shown in the **Last Change** column: //TracError: Unsupported version control system ""hg"": Can't find an appropriate component, maybe the corresponding plugin was not enabled?//. We could display something similar in the table on the //Manage Repositories// page. Also, when `SubversionConnector` is not enabled, the following is still shown on the //Add repositories// form: [[span(Default: svn, class=hint)]]",Ryan J Ollos next-major-releases Release,11702,Add a macro for listing ticket Components,ticket system,,enhancement,normal,,new,2014-08-03T08:23:47+02:00,2022-11-02T17:52:58+01:00,"As discussed in comment:9:ticket:1233, it would be nice to have a macro for listing ticket system //Components//. comment:25:ticket:1233 suggests having a link to the query page from an //open tickets// column generated by the macro. The `TicketQuery` macro might be the most similar in term of existing macros, given its output //format//s that we might want to pattern after. Macros for //Milestones//, //Versions// and //Ticket enums// might follow after this one. A macro for listing ticket enums could be used on the TracTicketTriage#StatusandResolution page. The macro would be useful on the TracTicketTriage, where we currently do not list the components. Addition of this macro would deprecate [th:ComponentsProcessorMacro].",Ryan J Ollos next-major-releases Release,11683,Dynamic table sorting in Admin tables,admin/web,,enhancement,normal,,new,2014-07-18T00:10:26+02:00,2021-10-28T08:09:28+02:00,"Allow the **Components**, **Milestones**, **Repositories** and **Versions** (and possible **Enum**) tables to be sorted. This was mentioned in #10994. For inspiration, see th:#10876 and th:TableSorterPlugin.",Ryan J Ollos next-major-releases Release,11676,Reusable commenting module,general,,enhancement,major,,new,2014-07-14T21:22:10+02:00,2014-12-29T11:53:53+01:00,"Trac so far only supports comments on tickets, but not on other resources like: * Wiki pages (#4279) * Changesets (#2035) * Source files * Attachments * Milestones * Resources provided by external plugins I would like to implement a reusable ''comment'' module, that (similar to the existing ''attachment'' module) allows us (and plugin writers) to easily add a commenting feature to all kinds of resources. === Existing plugins th:AddCommentMacro, th:WikiCommentsPlugin, th:CodeCommentsPlugin, various CodeReview plugins === Related previous ideas [wiki:TracDev/Proposals/AdvancedWikiOperations#/generic%20comment AdvancedWikiOperations], [wiki:TracDev/Proposals/DataModel#/comment DataModel], [wiki:TracObjectModelProposal#/comment TracObjectModelProposal], [wiki:TracDev/ContextRefactoring#/generic%20comment ContextRefactoring], GenericTrac, GenericTrac/Brainstorm, CodeReview/Annotations ",Peter Suter next-major-releases Release,11637,Hide versions and components,ticket system,,enhancement,normal,,new,2014-06-02T07:43:02+02:00,2019-12-05T03:55:09+01:00,"Closed milestones are only visible on the ticket form for users that possess `TICKET_ADMIN`. However, all Versions and Components are visible for all users, and the only way to hide them is to delete them. These lists can grown very long, and it would be nice to narrow the list that is seen from the ticket form, while keeping the entries on the Query form.",Ryan J Ollos next-major-releases Release,11633,Allow CommitTicketReferenceMacro to be used in tickets that aren't referenced in the changeset message,ticket system,,enhancement,normal,,new,2014-05-31T19:12:08+02:00,2016-05-10T01:10:22+02:00,"It was discussed in gdiscussion:trac-users:sCVEU-z8y_A that it would be useful to allow the !CommitTicketReferenceMacro to be used from tickets that the changeset does not reference. Currently the message //The changeset message doesn't reference this ticket// is displayed for that use case. TODO: Document on CommitTicketUpdater page.",Ryan J Ollos next-major-releases Release,11627,Backport Bloodhound search API,search system,,enhancement,normal,,new,2014-05-28T21:17:23+02:00,2015-05-17T01:27:29+02:00,"Bloodhound has an enhanced search API (bh:wiki:/Proposals/BEP-0004) and a [https://pypi.python.org/pypi/Whoosh/ Whoosh] plugin. A plugin for [http://lucene.apache.org/solr/ Apache Solr] is being developed: bh:wiki:/Proposals/BEP-0014. This ticket will explore whether the API, and possibly the plugins as well, can be backported.",Ryan J Ollos next-major-releases Release,11619,Option to show whitespace and tab characters in file and diff view,version control/browser,,enhancement,normal,,new,2014-05-20T00:46:48+02:00,2016-03-23T15:15:20+01:00,"The TortoiseSVN diff view has the option to show whitespace and tab characters. It would be useful to have an option to do the same in the Trac file and diff views. [[Image(TsvnDiffView.png)]]",Ryan J Ollos next-major-releases Release,11539,Remove deprecated methods in trac.wiki.formatter,general,,enhancement,normal,,new,2014-03-12T23:34:57+01:00,2015-03-02T22:27:22+01:00,"The pre-0.11 formatting utilities `wiki_to_html`, `wiki_to_oneliner` and `wiki_to_outline` could be removed on the trunk, but it looks like we need a replacement for `wiki_to_outline` (`format_to_outline`?): [browser:/trunk/trac/wiki/formatter.py@12551:1604-1640#L1604].",Ryan J Ollos next-major-releases Release,11508,"Add a ""permission copy"" command to TracAdmin",admin/web,,enhancement,normal,,new,2014-02-28T21:30:55+01:00,2017-09-10T02:31:26+02:00,"As mentioned in comment:4:ticket:11099, it would be nice to have a `permission copy` command to complement the form that was added to the Permission admin page.",Ryan J Ollos next-major-releases Release,11452,Implement a set_field workflow action,ticket system,,enhancement,normal,,new,2014-01-21T08:13:21+01:00,2015-07-28T10:01:15+02:00,"As mentioned on the [https://groups.google.com/d/msg/trac-users/lxDpEoiF8ws/VvQXnM0YmNMJ mailing list], it would be nice to have a `set_field` action that would couple a specified field to a workflow action in the same way that `set_owner` and `set_resolution` couple the owner and resolution fields to the workflow action. ",Ryan J Ollos next-major-releases Release,11446,Cummulative diff view for non-contiguous changesets,version control/changeset view,,enhancement,normal,,new,2014-01-15T01:17:13+01:00,2019-08-20T11:11:09+02:00,"When linking to a non-contiguous range of revisions from a TracLink such as [2,4:5], multiple //Show changes// buttons are displayed, and it doesn't seem to be possible to view a single diff encompassing all the changes. However, the latter view is usually what I'm looking for.",Ryan J Ollos next-major-releases Release,11398,Existing attachments can be overwritten without warning when uploading through trac-admin,admin/console,,enhancement,normal,,new,2013-12-13T18:50:27+01:00,2019-09-06T19:51:32+02:00,"For example, both commands return success: {{{#!sh $ trac-admin $env attachment add wiki:WikiStart file1 $ trac-admin $env attachment add wiki:WikiStart file1 }}} We could perhaps prompt with an ""Are you sure you want to overwrite the existing file?"".",Ryan J Ollos next-major-releases Release,11378,SQLite backend does not use auto-incrementing primary keys,database backend,,defect,normal,,new,2013-11-28T06:29:45+01:00,2021-08-19T18:16:39+02:00,"When using the SQLite backend, Trac does not actually create auto-incrementing columns when generating `CREATE TABLE` statements for Table objects that contain a `Column(..., auto_increment=True)`. The [http://trac.edgewall.org/browser/trunk/trac/db/sqlite_backend.py#L118 relevant code is in the sqlite_backend's _to_sql method]: {{{ #!python if column.auto_increment: ctype = ""integer PRIMARY KEY"" }}} According to [http://www.sqlite.org/autoinc.html the SQLite documentation], this will assign a unique value to each existing row -- but a newly created row might reuse a value that was previously used by a deleted row. In order to prevent reuse of deleted values, the column must instead be created with type ""integer PRIMARY KEY AUTOINCREMENT."" Specifically, the behavior only differs in cases when a deleted row happens to have the largest existing value for its primary key. To see this behavior in Trac: 1. Start up an environment using the SQLite backend 2. Enable the `tracopt.ticket.deleter.ticketdeleter` component 3. Log in as an admin 4. Create a new ticket (`#1`) 5. Create a second new ticket (`#2`) 6. Delete ticket `#2` 7. Create a new ticket The newly created ticket will be `#2`, instead of `#3`. If you had instead deleted ticket `#1` in step 6 above, the newly created ticket in step 7 would be `#3`. I believe this behavior only occurs with the SQLite backend. This does not only affect tickets, and does not only affect Trac core -- any plugin that defines a database table with a `Column(..., auto_increment=True)` is also affected. For example I found out about this while looking into a !TracPastePlugin bug, th:#4265, which exhibits the same behavior -- the ID of a deleted paste will be reused by a new paste, if and only if the deleted paste is the most recently created, and the site is using the sqlite backend. The SQLite documentation does explain that using the AUTOINCREMENT keyword means slower INSERTs: > The behavior implemented by the AUTOINCREMENT keyword is subtly different from > the default behavior. With AUTOINCREMENT, rows with automatically selected ROWIDs > are guaranteed to have ROWIDs that have never been used before by the same table in > the same database. And the automatically generated ROWIDs are guaranteed to be > monotonically increasing. These are important properties in certain applications. But if > your application does not need these properties, you should probably stay with the > default behavior since the use of AUTOINCREMENT requires additional work to be done > as each row is inserted and thus causes INSERTs to run a little slower. Was the omission of AUTOINCREMENT a deliberate decision, or is this a bug?",ejucovy@… next-major-releases Release,11353,Hierarchical tree of repositories,version control/browser,1.0.1,enhancement,normal,,new,2013-11-05T06:49:28+01:00,2015-03-05T14:43:47+01:00,"For a big project it would be really nice to organize git repositories into the hierarchical tree. Like [[Image(http://dside.dyndns.org/projects/screenshots/trac-repotree-small.png)]] I have hacked in a small patch for myself, but it will be really nice to have a proper implementation upstream. http://dside.dyndns.org/projects/patches.dir/trac-ds-repotree.diff",csa@… next-major-releases Release,11323,WikiProcessor argument to limit line width,wiki system,,enhancement,normal,,new,2013-10-10T00:41:50+02:00,2014-03-09T18:36:07+01:00,"The th:EmailProcessorMacro can be used to limit the width of text in a `pre` block. It would be generally useful to add an argument to WikiProcessors for limiting the width of the text. The argument could be `wrap` or `cols` and should allow for a numeric argument for number of columns (`wrap=79` or `cols=79`). Additionally, we might consider ""auto-wrapping"" columns based on the width of the screen.",Ryan J Ollos next-major-releases Release,11301,intermittent failure with notification test,ticket system,1.0-stable,defect,normal,,new,2013-09-15T18:32:54+02:00,2013-09-17T00:41:15+02:00,"The usual `make test` after a recent pull on trunk gave me the following: {{{ ERROR: test_from_author (trac.ticket.tests.notification.NotificationTestCase) Using the reporter or change author as the notification sender ---------------------------------------------------------------------- Traceback (most recent call last): File ""c:\Trac\repos\trunk\trac\ticket\tests\notification.py"", line 369, in test_from_author ticket.save_changes('noemail', 'More changes') File ""c:\Trac\repos\trunk\trac\ticket\model.py"", line 387, in save_changes old_db_values[name], db_values.get(name, ''))) File ""c:\Trac\repos\trunk\trac\db\util.py"", line 121, in execute cursor.execute(query, params) File ""c:\Trac\repos\trunk\trac\db\util.py"", line 54, in execute r = self.cursor.execute(sql_escape_percent(sql), args) File ""c:\Trac\repos\trunk\trac\db\sqlite_backend.py"", line 78, in execute result = PyFormatCursor.execute(self, *args) File ""c:\Trac\repos\trunk\trac\db\sqlite_backend.py"", line 56, in execute args or []) File ""c:\Trac\repos\trunk\trac\db\sqlite_backend.py"", line 48, in _rollback_on_error return function(self, *args, **kwargs) IntegrityError: columns ticket, time, field are not unique ---------------------------------------------------------------------- Ran 1212 tests in 13.759s FAILED (errors=1) }}} Twice in a row... also on 1.0-stable, then not anymore: I could do 10 runs without any errors. Wonder if it's a timing issue or something like that. The errors all happened with Python 2.7 but I don't think this has anything to do with it. Looking at the code, I find a succession of 4 `save_changes` calls, without much done in between. It might simply be that the 3rd `save_changes` (line 369) happens within the same microsecond(?) than the 2nd `save_changes` (line 361). This raises the interesting question if we should try to fix this for the tests only or if it would be useful to introduce a small delay in general for the `save_changes` so that this API becomes ""timing"" safe. For example, here's a very simple way to reproduce this error: {{{#!diff Index: trac/ticket/tests/model.py =================================================================== --- trac/ticket/tests/model.py (revision 12049) +++ trac/ticket/tests/model.py (working copy) @@ -110,6 +110,8 @@ ticket2 = self._create_a_ticket() ticket2.insert() ticket2['summary'] = 'Bar' + ticket2['foo'] = 'New value 0' + ticket2.save_changes('santa', 'this is my comment 0') ticket2['foo'] = 'New value' ticket2.save_changes('santa', 'this is my comment') return ticket2 }}} But ""fixing"" this is not straightforward, as experimenting shows that it's not obvious how to pick the appropriate value for the delay. On Windows, adding a call to `time.sleep(0.001)` did the trick, but 0.0001 didn't. I wonder if 0.001 would be good enough for all platforms and versions of Python. Then of course, using ""time"" as part of the key was also not the best idea we ever had...",Christian Boos next-major-releases Release,11263,AuthzPolicy should allow restricting access to only the most recent version of a resource,general,1.0-stable,enhancement,normal,,new,2013-08-02T03:56:46+02:00,2017-05-01T00:16:52+02:00,"It would be useful to be able to restrict access to only the most recent version of a resource. This was discussed in comment:8:ticket:8976. One suggestion is to use the syntax `@HEAD`. Another possibility would be to use `@None`, since by convention `None` refers to the most recent version of a resource. What would be the most logical syntax to use? Currently, if the user is restricted to accessing the most recent version of the using a rule such as {{{#!ini [wiki:WikiStart@11] anonymous = WIKI_VIEW }}} the user will only be able to access that page if `?version=11` is included in the URL. Navigating to `/wiki/WikiStart` will result in a //Forbidden// error. Another request from comment:8:ticket:8976 is to hide the //History// contextual navigation item and prevent navigating to `/wiki/WikiPage?action=history` when the user only has access to the most recent version of a resource. This is potentially more difficult because we have a way to check if the user has access to a particular resource, but it is less straightforward to check which versions of a resource the user has access to, and performance considerations may come into play.",Ryan J Ollos next-major-releases Release,11213,Updating Saved Query results does not save column order,query system,1.0.1,defect,normal,,new,2013-06-19T21:04:28+02:00,2017-04-27T11:39:42+02:00,"When using a saved query (using TracQuery TracLinks syntax), using the ""Update"" button to update the results does not save the report's column ordering (left-to-right ordering), but reverts to default ordering. Reproduction: 1. Create a new saved query with TracQuery TracLinks syntax. 2. Customize the column order of the results in the query; potentially add custom fields to the column results 3. Give query a name and save 4. Run saved query 5. '''Note column ordering of results respects what was specified in TracQuery syntax 6. '''Click ""Update"" 7. '''Note that column ordering no longer respects what was specified in the saved report and appears to be the default column ordering''' Untested: Not sure if this would change reproduction, but all my saved queries also have a report= where n is the SQL ID of the saved report.",achittur@… next-major-releases Release,11206,Reduce join in query process,query system,1.0dev,enhancement,normal,,new,2013-06-17T09:23:05+02:00,2014-03-08T23:01:17+01:00,"Hi, We are running Trac with 10k tickets, and some modules need query through TracQuery a lot, but we found backend MySQL's load is heavy. From ''show processlist'', we found join query of ""copy to tmp table"" action between tables may cost a lot of time. We debugged the SQL below: {{{ SELECT COUNT(*) FROM (SELECT t.id AS id,t.owner AS owner,t.type AS type,t.status AS status,t.priority AS priority,t.resolution AS resolution,t.time AS time,t.changetime AS changetime,t.summary AS summary,priority.value AS priority_value FROM ticket AS t LEFT OUTER JOIN enum AS priority ON (priority.type='priority' AND priority.name=priority) WHERE ((COALESCE(t.summary,'')='querystr')) ORDER BY COALESCE(priority.value,'')='',CAST(priority.value AS signed),t.id) AS foo }}} After profiling the query: Copying to tmp table cost 99.355% percentage of whole time. If `((COALESCE(t.summary,'')='querystr'))` changed to `t.summary = ""querystr""`, this copy action will not take that much time. And if count only 0, problem still happens, if not count, problem not exists, maybe we can directly select the result with count. Or break this query into `t.summary = '' or t.summary is null` if query string is empty. Another problem is 'priority' column are regarded as required fields even I deselect priority column, thus join will be need every query. Maybe in default, join on priority can be removed, or deselect priority column will disable it. Thanks! V.E.O",VEOVUSH@… next-major-releases Release,11185,improve speed of TicketQuery for format=table,query system,1.0-stable,enhancement,major,,new,2013-05-13T19:11:22+02:00,2019-08-12T23:28:57+02:00,"The `format=table` was added in 0.11 merely as a proof-of-concept to see how Genshi templates could be reused in different contexts, in this case within a WikiMacro. That proved to work well, except for the performance side of things (see e.g. #10879). A query like `[[TicketQuery(status=closed,format=list)]]` on this site (corresponding to [[TicketQuery(status=closed,format=count)]] tickets) would take several seconds to complete successfully, while `[[TicketQuery(status=closed,format=table)]]` fails after several (more) seconds with: {{{ Error: Macro TicketQuery(status=closed,format=table) failed }}} and in the log we see: {{{ [pid 9964 139956728231680] 2013-05-13 18:46:25,397 Trac[formatter] ERROR: Macro TicketQuery(status=closed,format=table) failed: Traceback (most recent call last): File ""/usr/local/virtualenv/1.1dev/lib/python2.6/site-packages/Trac-1.1.2dev_r11794-py2.6.egg/trac/wiki/formatter.py"", line 765, in _macro_formatter return macro.ensure_inline(macro.process(args)) File ""/usr/local/virtualenv/1.1dev/lib/python2.6/site-packages/Trac-1.1.2dev_r11794-py2.6.egg/trac/wiki/formatter.py"", line 380, in ensure_inline text = _markup_to_unicode(text) File ""/usr/local/virtualenv/1.1dev/lib/python2.6/site-packages/Trac-1.1.2dev_r11794-py2.6.egg/trac/wiki/formatter.py"", line 107, in _markup_to_unicode markup = stream.render('xhtml', encoding=None, strip_whitespace=False) File ""/usr/local/virtualenv/1.1dev/lib/python2.6/site-packages/Genshi-0.7dev_r1134-py2.6.egg/genshi/core.py"", line 183, in render return encode(generator, method=method, encoding=encoding, out=out) File ""/usr/local/virtualenv/1.1dev/lib/python2.6/site-packages/Genshi-0.7dev_r1134-py2.6.egg/genshi/output.py"", line 57, in encode return _encode(''.join(list(iterator))) MemoryError: }}} I think we could approach this particular issue as a way to start a performance enhancement process which will lead to eliminate the Genshi bottleneck, by only generating the relevant data in JSON format (here result rows content), and have some client-side Javascript code for doing the actual rendering (here building the HTML table). This presents a lot of implementation possibilities, but in a first step I believe simple jQuery based building should be good enough, there's probably no need of client-side templating or additional framework. I think that solving #10373 will also be a prerequisite. Once it works for this use case, we could think about how to make the normal `/query` pages benefit from the same performance boost (getting rid of the `query_results.html` template). Then, once we have a robust solution in place, gradually extend that to other ""Genshi-intensive"" places (diffs, source browser, timeline, etc.).",Christian Boos next-major-releases Release,11163,Option to set HTML title.,wiki system,0.12.2,enhancement,minor,,new,2013-04-16T05:01:26+02:00,2015-11-30T22:30:50+01:00,"Currently no option to set HTML title. At preset, title seems to be the same as the page name. The problem with this is the title can end up like ""section/subsection/subsubsection/mypage"" which isn't very handy. I've tried using the html processor, as follows; !{{{ #!html Test Title }}} but no luck there either.",anonymous next-major-releases Release,11142,Add download link in attachment section,attachment,,enhancement,normal,,new,2013-04-04T11:24:40+02:00,2019-09-06T19:51:32+02:00,"Currently to download an attachment you need to view it, scroll down and then click on the plain link. Most of the cases I don't want to view it, but rather want to have it. Please add a direct download link in the attachment section of wiki pages (and tickets). Would save a lot of time on user side and does not really have any negative side effects.",Dirk Stöcker next-major-releases Release,11141,Add ConfigSection to all sections,general,,enhancement,normal,,new,2013-04-03T17:33:20+02:00,2015-12-14T05:28:27+01:00,"After I modified the [[http://trac-hacks.org/wiki/TracIniAdminPanelPlugin|TracIniAdminPlugin]] to show the description for sections I noticed that a proper description is very helpful. Please add a !ConfigSection entry for each section, not only for the default-empty ones. Also this would improve [wiki:TracIni].",Dirk Stöcker next-major-releases Release,11133,Color overdue dates red,general,,enhancement,normal,,new,2013-03-26T17:53:42+01:00,2015-09-10T17:34:04+02:00,"Create config option for 'due_date_fields' in reports. These dates are colored red if in the past. This implementation assumes dates are stored in a text custom field in ISO format (i.e., yyyy-mm-dd). ",chris.nelson.1022@… next-major-releases Release,11130,Add an inlinelinkbutton class,general,,enhancement,normal,,new,2013-03-26T13:43:10+01:00,2019-09-06T19:51:32+02:00,"The buttons plugins add to ticket pages (Create, Delete, etc.) have a range of plugin-specific styles. It would be nice to unify them. Some plugins use a button, others use a link styled as a button. The existing inline buttons style had a relative font size but depending on where the button was in the page, it could have a different appearance. This patch makes the text size consistent. The new `inlinelinkbutton` style makes links with that style look like inline buttons. ",chris.nelson.1022@… next-major-releases Release,11119,Indicate enabled or disabled status of plugins on web admin panel,admin/web,,enhancement,normal,,new,2013-03-17T19:35:04+01:00,2015-03-02T22:27:22+01:00,"There are occasions when I want to be able to determine what plugins are enabled from the Web Admin page, and it is tedious to expand each folded section and inspect the checkboxes. Initially, I started thinking about a solution similar to what is present for other folded sections, such as attachments, for which a number is shown next to the heading to indicate the number of attachments. What might be more useful in this case, is to change the background color of the `form` depending on whether the plugin is enabled or disabled. Maybe that is not obvious enough though. [th:hasienda] has done some nice work on the [th:AccountManagerPlugin] lately using icons guide the user through the configuration process (see th:#8930 and [http://trac-hacks.org/attachment/ticket/8930/20130215_acct_mgr-cfg_wizard_p6_dev.png this] attachment). That got me thinking that maybe a check or x icon could be used to indicate a plugins enabled or disabled status (with enabled being defined as at least one component being enabled). ",Ryan J Ollos next-major-releases Release,11078,Perform fine-grained permission checks on resource in get_navigation_items,wiki system,,enhancement,normal,,new,2013-02-23T18:00:27+01:00,2017-05-21T01:36:16+02:00,"First mentioned in comment:1:ticket:11067, fine-grain permissions checks should be performed in `get_navigation_items`, taking into account the resource that may be specified in the `[mainnav]` section through the `href` option. See comment:4:ticket:11067 for additional suggestions.",Ryan J Ollos next-major-releases Release,11061,"TypeError: Git object id must be byte or a text string, not: int",plugin/git,,defect,minor,,new,2013-02-13T08:10:33+01:00,2023-03-01T05:44:59+01:00,"We should be able to cope with git-svn converted repositories which might contain svn style TracLinks (like our own repository). ... but this ticket is more an opportunity to congratulate Jun for his effort with the [repos:jomae.git@pygit2] branch ;-) We should also add libgit2/pygit2 versions to the system info. In my case, it's 0.17.0 and 0.17.3 on OpenSuSE 0.12.2. Btw, I've also tried on Windows with the bleeding edge (build of latest from yesterday), but got many errors, the API has evolved quite a bit since 0.17.3. ---- ==== How to Reproduce ==== While doing a GET operation on `/log/trunk/`, Trac issued an internal error. ''(please provide additional details here)'' Request parameters: {{{ {'path': u'/trunk/', 'revs': u'11577-11612'} }}} User agent: `Mozilla/5.0 (X11; Linux x86_64; rv:15.0) Gecko/20100101 Firefox/15.0` ==== System Information ==== || '''`Trac`''' || `1.1.2dev` || || '''`Babel`''' || `0.9.6` || || '''`Docutils`''' || `0.9` || || '''`Genshi`''' || `0.6 (without speedups)` || || '''`Pygments`''' || `1.5` || || '''`pysqlite`''' || `2.6.0` || || '''`Python`''' || `2.7.3 (default, Apr 14 2012, 08:58:41) [GCC]` || || '''`pytz`''' || `2012b` || || '''`setuptools`''' || `0.6` || || '''`SQLite`''' || `3.7.12.1` || || '''`jQuery`''' || `1.8.3` || ==== Enabled Plugins ==== ==== Python Traceback ==== {{{ Traceback (most recent call last): File ""/home/cboos/trac/trunk/trac/web/main.py"", line 497, in _dispatch_request dispatcher.dispatch(req) File ""/home/cboos/trac/trunk/trac/web/main.py"", line 214, in dispatch resp = chosen_handler.process_request(req) File ""/home/cboos/trac/trunk/trac/versioncontrol/web_ui/log.py"", line 113, in process_request rev = unicode(repos.normalize_rev(rev)) File ""/home/cboos/trac/trunk/tracopt/versioncontrol/git/git_fs.py"", line 520, in normalize_rev commit = self._get_commit(rev, raises=False) File ""/home/cboos/trac/trunk/tracopt/versioncontrol/git/git_fs.py"", line 445, in _get_commit git_object = self.git_repos[rev] TypeError: Git object id must be byte or a text string, not: int }}}",Christian Boos next-major-releases Release,11049,TracD on Windows 8 (x64),web frontend/tracd,1.0,defect,normal,,new,2013-02-05T13:16:24+01:00,2019-09-06T19:51:32+02:00,"I am encountering numerous problems on Windows 8 (x64) running TracD with Python 2.7. In Internet Explorer 10, loading Trac pages is erratic, with first load often missing style/layout or missing the header logo. Pressing back then forward several times may resolve this problem as if little bits of the page are building up in the cache over time. (The logo will load if right-clicked to Show Picture). With the !WikiExtras hack, the [!WikiIcons] documentation fails to load all icons in the page, but will load them eventually with back/forth as above (or by right-click Show Picture). However, hitting the Refresh button will crash TracD/Python. The Trac service may stop responding randomly, at which point it is reported that the Trac database needs to be upgraded. (Running an upgrade allows the wiki to display, but all macros are now unrecognized and the custom logo becomes undefined, so this is probably spontaneous corruption). These problems are also evident with Google Chrome, so it does not appear to be IE10 specific. I should note that IE10 is aggressive - with a tracking protection list turned on, the Trac header logo is blocked. Compatibility settings on executables has not helped, nor has making exceptions in !Antivirus/Firewall. An identical setup was working fine in Windows 7 (x64). ",paulcoddington@… next-major-releases Release,10983,PATCH (and RFC) for IQueryPreprocessor extension point,ticket system,0.11.6,enhancement,normal,,new,2012-12-13T18:27:55+01:00,2018-06-14T19:47:14+02:00,"I'm working in 0.11.6 and I don't expect this to be adopted there but I'd like to get feedback on structure, naming, etc. so I can make this patch more acceptable before porting to an active version. In my PM work, I find it desirable to query tickets based in special ways like asking for all the tickets required by a ticket (and their predecessors, etc.) or all the descendants of a ticket. My plugin supports this with `goal=id|id|...` and `root=id|id|...` but every place I want this support, I have to modify plugin source to use my preprocessor or query wrapper. For example, I did this locally for Estimation Tools so I could do a workload chart of all the work in a ticket's predecessors. By adding an extension point to the core query system, all ticket queries can benefit from the PM logic. For example, you could do `[[TicketQuery(goal=self)]]` in a ticket to list all the ticket's predecessors. This patch defines and uses IQueryPreprocessor to accomplish this extensible query system. Then I implement this interface in my PM plugin: {{{ #!py class PMQueryHelper(Component): implements(IQueryPreprocessor) def __init__(self): self.pm = TracPM(self.env) # IQueryPreprocessor methods # Return the list of constraints handled by process_constraints() def custom_constraints(self): return ['root', 'goal', 'scheduled'] # Turn custom constraints (e.g., root) into standard contraints (e.g., id) # # @param req the Trac web request object, may be None # @param constraints hash indexed by contraint name (e.g., 'root', # 'id'). Each entry is a list of values. # # @return updated constraints. def process_constraints(self, req, constraints): # constraint values are lists, TracPM.preQuery() expects # pipe-delimited strings. options = {} for constraint in self.custom_constraints(): if constraint in constraints: options[constraint] = ""|"".join(constraints[constraint]) del constraints[constraint] ids = list(self.pm.preQuery(options, req)) if len(ids): if 'id' not in constraints: constraints['id'] = [] for tid in ids: if tid not in constraints['id']: constraints['id'].append(tid) return constraints }}} and the above `TicketQuery()` works.",chris.nelson.1022@… next-major-releases Release,10915,Provide example repository paths,admin/web,1.1.1dev,enhancement,normal,,new,2012-10-19T01:28:22+02:00,2015-03-02T22:27:22+01:00,"It's not clear what path to use when adding a repository without reading the documentation. It would be nice to have some example paths (like `/home/john/test/.git` or `svn://test.com/svn/trunk`) in the interface: [[Image(repository-add.png)]] I was also not able to locate the proper path format for Git repositories in the documentation, and the logs suggested rather vaguely what path I should be using instead: {{{ Trac[test-trac:PyGIT] ERROR: entry '.git' found in '/home/john/repository' -- maybe use that folder instead... }}}",Thijs Triemstra next-major-releases Release,10914,Changeset columns overlapping,version control/changeset view,,defect,normal,,new,2012-10-19T01:20:03+02:00,2015-03-02T22:27:22+01:00,"The columns with changeset information are overlapping (for a Git repository), as shown in this screenshot: [[Image(git-changeset.png)]] This is on Ubuntu 12.04 with Firefox.",Thijs Triemstra next-major-releases Release,10905,Enabled CommitTicketUpdater only for some repositories,version control,0.12.3,enhancement,normal,,new,2012-10-15T14:49:47+02:00,2017-09-09T00:45:15+02:00,"I am using CommitTicketUpdater on a Trac installation with multiple (git) repositories. The problem I am having is that for some of those repositories we are using GitHub for ticketing. And because CommitTicketUpdater listens to all comments, when somebody reference ticket with GitHub ticket number, a wrong reference is made in Trac. Would it be possible to enable CommitTicketUpdater only for some repositories?",Mitar next-major-releases Release,10866,"custom search query cannot do ""keywords contains x and y""",query system,,defect,normal,,new,2012-09-25T06:50:08+02:00,2014-08-22T15:10:19+02:00,"The custom search query (eg: on this site) cannot search for tickets containing both of two keywords. If you enter a keyword search for ""x"", then use the ""And"" dropdown to add a second keyword ""y"" to the search, the GUI indicates that it will search for ""x or y"", and indeed, the found tickets are those that contain one of more of ""x"" and ""y"". ",ian.wood@… next-major-releases Release,10764,refactor changeset search,version control,1.0dev,defect,major,,new,2012-07-12T16:38:43+02:00,2015-07-11T07:41:34+02:00,"Currently it's done at the level of the `ChangesetModule`, assuming all the data is in the database's `revision` table! (see [source:trunk/trac/versioncontrol/web_ui/changeset.py@11093#L1143]) Obviously this won't work for non-cached repositories. The current implementation needs to be moved to the cached repository and each repository should be able to provide results (or not, e.g. on a per repository basis depending on some settings, like don't search that non-cached linux repo, as it's too big).",Christian Boos next-major-releases Release,10757,Use language specific starting day for date picker,i18n,0.13dev,enhancement,normal,,new,2012-07-03T11:11:35+02:00,2014-10-14T17:04:01+02:00,"It would be nice, if the date picker introduced with ticket #10245 would use the language specific starting day of a week. In Germany for example the starting day of a week is Monday (and not Sunday, like in US). Alternatively it could be a user-specific preference.",framay next-major-releases Release,10755,[PATCH] CommitTicketUpdater does not allow user to define what state to jump to,ticket system,0.12.3,enhancement,normal,,new,2012-07-02T22:27:07+02:00,2022-03-16T23:13:24+01:00,"The CommitTicketUpdater will only ever close a ticket. In general, the closed state is for when no more work is to be done, aka ''after'' QA has verified the fix. Trac should allow configuration of what state name the ticket will get updated to.",patrick.little next-major-releases Release,10741,"more readable ""closed"" style",general,1.0dev,defect,normal,,new,2012-06-25T22:37:55+02:00,2015-05-26T20:50:57+02:00,"The current strikethrough style is barely readable, as already mentioned in the past, like in #9981. If people can get used to it for numbers (but what about #8080 for example?), it's really a bother for milestones ([milestone:0.11.8 a milestone that didn't turn into a release]) or comments ([ticket:9981#comment:5 as was said before]). We just revived the subject in #10012 (see [ticket:10012#comment:42 comment:42]), with several new interesting proposals, notably the one from Peter which is to append an unicode '!CHECK MARK' character to the ticket (✓).",Christian Boos next-major-releases Release,10739,Give Trac a sidebar widget provider,web frontend,1.0dev,enhancement,normal,,new,2012-06-25T13:34:28+02:00,2015-08-08T10:50:03+02:00,"I think it's about time. I searched ""sidebar"" through tickets and am not seeing any discussion about this topic throughout the years. First, let's take wiki for example: * there's plenty of horizontal room on the screen even with minimum resolutions (1024x) * all room is currently filled with text... * but usability guidelines say text readability starts to suffer at >600px width So there's plenty of room available. Second, ticket view: * we simply have a centered ticket view, although having much better width for usability (compared to wiki), there's A LOT of whitespace filling the edges * whitespace is monstrous with larger screens * '''most importantly''' key pieces of information are mostly out of sight - i foresee launchpad style participant list, various statistics about the ticket, etc other useful things Third, search view: * faceted browsing, with ebay style checkboxes controlling search results I'll stop here with the examples at the moment. th:FullBlogPlugin already has useful sidebars self-implemented. rjollos has also worked on th:TicketSidebarProviderPlugin. I hereby propose a universal sidebar area and some standardized Provider interfaces for plugin authors to fill them with their widgets if they so desire. Sidebars should of course be '''optional''' for the user. I am '''not''' proposing that core start creating new widgets for the sidebar, at least not right off the bat. Perhaps some single widget as a learning example. Whatever comes over time, comes. I think many of the useful things people create can be adopted easily due to widgets usually being very focused and therefore small in code size. Also, I just lately ran into a [http://blog.jackcheng.com/post/25160553986/the-slow-web clean blog design with a position:fixed sidebar] and absolutely loved the feeling of control there. Just a design experiment thought. [http://wordpress.org WordPress] for one, which I've lately worked with extensively, has a very polished widget system (and a lot more that Trac should adopt over time.. except PHP!) :>. This surely has affected my thinking about this. If any of the people interested in this want to have access to a demo site to get a feel of how it works there, I can easily provide access.",lkraav next-major-releases Release,10736,env.get_known_users() should also return session.last_visit,general,0.12.3,enhancement,normal,,new,2012-06-25T00:14:53+02:00,2016-04-20T16:14:32+02:00,"I'm trying to build up something really lightweight to get people visible in my projects without #2456 and various hardcore TH plugins. I'd like to make use of the last_visited attribute to for example only show recently logged in people. But it turns out I gotta jump through hoops to get to it. For example th:AccountManagerPlugin goes straight SQL here: th:source:accountmanagerplugin/trunk/acct_mgr/api.py@11554#L277 {{{ accountmanagerplugin/trunk/acct_mgr/api.py:277: def last_seen(self, user = None): }}} In 1.0dev source:/trunk/trac/env.py@11073#618 we have: {{{ 618 def get_known_users(self, cnx=None): 619 """"""Generator that yields information about all known users, 620 i.e. users that have logged in to this Trac environment and 621 possibly set their name and email. 622 623 This function generates one tuple for every user, of the form 624 (username, name, email) ordered alpha-numerically by username. }}} Why couldn't this function also include the apparently orphanized last_visit attribute right there? Can't be a significant performance penalty, can it? Oh yeah, going through trac.web.session.!SessionAdmin source:/trunk/trac/web/session.py@11012#L398 doesn't seem to make much sense either. Or...?",lkraav next-major-releases Release,10733,Renamed files and prev/next links don't work well,version control,,defect,normal,,new,2012-06-20T21:37:43+02:00,2012-06-23T12:46:52+02:00,"source:trunk/trac/versioncontrol/web_ui/browser.py@10924#L427 creates links for the previous, next and latest revision of a given file. This does not take name changes into account: the link will always be for the created_path member value, which is described in #3340. But in the presence of a rename, both the preceding and the next revision of a file might be in a different location. So Repository.previous_rev and next_rev should probably be allowed to not only return a revision but a new path as well. Perhaps you could allow them to return a tuple, either checking the actually returned value using isinstance or breaking backwards compatibility. Or you could introduce a new method. Up to you. Current api at source:trunk/trac/versioncontrol/api.py@11059#L873 Perhaps this would help getting rid of created_path altogether, as it would now be up to the repository backend to decide what path to return.",Martin.vGagern@… next-major-releases Release,10682,Show the path of changesets in log view,version control/log view,,enhancement,normal,,new,2012-04-30T16:41:08+02:00,2015-05-26T20:50:57+02:00,"Would be great if the path of the changed file is visible in the log view. If multiple files were changed, the highest common path of all changed files should be showed. I know this feature from WebSVN and it's very helpful to survey which parts of the project were actually changed.",anonymous next-major-releases Release,10670,Image macro references inconsistent with attachment links,wiki system,0.13dev,enhancement,normal,,new,2012-04-17T23:47:11+02:00,2015-03-02T22:27:22+01:00,"The format of the reference for [WikiMacros#Image-macro Image macro] is not aligned with that of `attachment:` links. External reference to attachment is stated as ""`attachment:file:module:id`"". It would be nice if one could copy an attachment link to the Image macro by just ripping away ""`attachment:`"" Now, one has to swap parts of the attachment reference to suit the Image macro: `attachment:file:module:id` > `[[Image(module:id:file)]]` This is annoying/confusing. My view is that the Image macro is fault since both `attachment:` and `comment:` links use the same pattern. ''(Hard to accept, I imagine, since that would break existing image references to external resources.)''",Mikael Relbe next-major-releases Release,10633,TicketQuery 'clist' format,wiki system,,enhancement,normal,,new,2012-03-19T19:11:52+01:00,2018-07-13T12:29:32+02:00,"I had a need for a short delimited list that was more than 'compact' and 'list'; but less than 'table'. I am typically using this in a sidebar to get a current list of tickets with more useful information. The coding may be horid, as i am not a python programmer most days; but it works and gives a very useful function if you want to add it. {{{ Index: trac/ticket/query.py =================================================================== --- trac/ticket/query.py (revision 1322) +++ trac/ticket/query.py (working copy) @@ -1332,6 +1332,24 @@ else: alist = [ticket_anchor(ticket) for ticket in tickets] return tag.span(alist[0], *[(', ', a) for a in alist[1:]]) + + elif format == 'clist': + if query.group: + return tag.div( + [(tag.p(tag_('%(groupvalue)s %(groupname)s tickets:', + groupvalue=tag.a(v, href=href, class_='query', + title=title), + groupname=query.group)), + tag.dl([(tag.dt(ticket_anchor(t)), + tag.dd(' - '.join(map((lambda x:t[x].__str__()),kwargs['col'].split('|'))) )) for t in g], + class_='wiki compact')) + for v, g, href, title in ticket_groups()]) + else: + return tag.div(tag.dl([(tag.dt(ticket_anchor(ticket)), + tag.dd(' - '.join(map((lambda x:ticket[x].__str__()),kwargs['col'].split('|'))) )) + for ticket in tickets], + class_='wiki compact')) + else: if query.group: return tag.div( }}}",branson@… next-major-releases Release,10599,"GitPlugin: Show branches in timeline, changesets, browser",plugin/git,1.0dev,enhancement,normal,,new,2012-03-04T15:18:35+01:00,2017-11-23T13:50:35+01:00,"''Created as part of the [ticket:10594 move of GitPlugin]. Tickets originally reported for th:GitPlugin: th:#2633, th:#7860, th:#8522'' Quoting [th:ticket:2633 jkp@...]: > It would be best if you could see which branch a commit was made against in both the changeset view and the timeline view. -> worksforme, but really needs to be improved, see comment:4 Quoting [th:ticket:7860 shogsbro]: > For Git repos with multiple branches it would be useful to be able to browse the branch -> [comment:3 worksforme] Quoting [th:ticket:8522 trac@...]: > I'd like browse source to take me to the same [branch] by default each time. -> [comment:2 duplicate]",Peter Suter next-major-releases Release,10579,"Do not use ""anonymous"" updater's email in ticket notification From",notification,,enhancement,normal,,new,2012-02-23T18:27:36+01:00,2012-02-23T18:57:29+01:00,"Originally an additional suggestion to #8360 in [comment:11:ticket:8360 comment:11]: With the ""improvement"" mentioned in [comment:9:ticket:8360 comment:9], it would be possible for a malefactor to use a Trac system to send emails (new tickets or comments) that would appear to come from someone else. Suggestion: The new `smtp_from_author` feature should only use email addresses from authenticated users.",Mark Potter next-major-releases Release,10525,TicketQuery macro broken for some keywords-based queries,query system,0.13dev,defect,major,,new,2012-01-14T06:09:22+01:00,2012-01-15T18:51:53+01:00,"Example: 1. What I want- A list of tickets that contain the keyword `svnauthz` and also does not contain the keyword `permission` 1. How I think I should do it- `[[TicketQuery(keywords~=svnauthz&keywords!~=permission, format=table, col=keywords)]]` 1. The result: [[TicketQuery(keywords~=svnauthz&keywords!~=permission, format=table, col=keywords)]] == The problem Note that #4503 is on the list, although it has the `permission` keyword. In fact, the result is identical to `[[TicketQuery(keywords~=svnauthz, format=table, col=keywords)]]`. I have observed the same issue in 0.12.3dev, and was able to debug this into the trac.ticket.query.Query class. When executing the query, the keywords constraint looks like {'keywords': ['~svnauthz', '!~permission']}, but in `get_clause_sql` the `neg` and `mode` variables are set only according to the first item (e.g. `'~svnauthz'`), so the generated SQL comes out messed up w.r.t the second constraint. Sorry for not suggesting a patch, but this code is quite complex...",Itamar Ostricher next-major-releases Release,10492,bash tab completion extension for trac-admin,admin/console,,enhancement,normal,,new,2011-12-06T04:13:00+01:00,2019-10-22T23:59:01+02:00,"This is a wishlist: it would be nice to have a bash tab-completion snippet that could complete the trac-admin subcommands directly in bash. A baseline implementation could just be a hard-coded list; a fancier approach could interrogate the selected trac-admin instance to get a list of enabled sub-commands (which might be different due to installed plugins, enabled/disabled components, etc). This feature request was originally reported to debian as http://bugs.debian.org/631259",dkg@… next-major-releases Release,10455,A repository hides a directory which has the same name in default repository,version control/browser,0.12-stable,defect,normal,,new,2011-11-09T18:43:13+01:00,2011-11-10T10:13:26+01:00,"=== Reproduce 1. Create a repository with `trac/versioncontrol/tests/svnrepos.dump` and adds as default repository. The root directory has `trunk`, `branches`, `tags` directories in the repository. 2. Create another repository and add it with repository name being `tags`. 3. Launch tracd and view `/browser` page. 4. Expand `tags` directory in default repository. It will show second repository instead of `tags` in default repository.",Jun Omae next-major-releases Release,10425,[PATCH] Add support for configuring a default set of columns to show in a query.,query system,0.12dev,enhancement,normal,,new,2011-10-20T23:57:02+02:00,2017-04-27T11:42:00+02:00,"At my work, we are using trac to view our bugs and tasks, but one of our customer uses JIRA, so we export our issues from JIRA and import them in trac as a tickets. To keep track of the issues, we have a custom field called 'jira' that contains the upstream id. As the upstream id is quite important, we want to include the column in the set of default columns for a query. The attached patch introduces two new config directives, one for setting the default order in the query (rather than hardcoding it to be 'priority', still a fallback through...), the second is a list of columns to automatically show unless the user specified a set of columns. Examples: Using this [query:owner=$USER link] results in the following set of columns shown in the browser (verified here on t.e.o): * Ticket * Summary * Owner * Type * Status * Priority * Milestone With my patched applied and the following in the trac.ini {{{ [query] default_cols = id,time,changetime,priority }}} would result in the following columns shown in the browser: * Ticket * Created * Modified * Priority",Torbjörn Svensson next-major-releases Release,10423,Destroy trac.ini at race condition.,admin/web,0.12.2,defect,normal,,new,2011-10-20T07:46:30+02:00,2014-06-05T08:47:51+02:00,"May be destroy trac.ini when several users change trac.ini from Trac Admin page. When change trac.ini, config module is executed save() method and parse_if_needed method in order. However, If several users change trac.ini, Happen race condition between save() method and parse_if_needed method. A) self._section and self.parser_sectios is cleared by if_parse_needed method, B) save() method read into ""section"" (local value of save() method) from self.parser._sections(into self.sections()) Unable to check process that search the change option value when A and B execute at the same time. As a result, destroyed trac.ini by save() method. I have fix this bug using RLock object of threading module. {{{ #!diff Index: trac/config.py =================================================================== --- trac/config.py +++ trac/config.py @@ -24,6 +24,9 @@ from trac.util.text import printout, to_unicode, CRLF from trac.util.translation import _, N_ +import threading +lock=threading.RLock() + __all__ = ['Configuration', 'Option', 'BoolOption', 'IntOption', 'FloatOption', 'ListOption', 'ChoiceOption', 'PathOption', 'ExtensionOption', 'OrderedExtensionsOption', 'ConfigurationError'] @@ -214,48 +217,53 @@ return # Only save options that differ from the defaults - sections = [] - for section in self.sections(): - section_str = _to_utf8(section) - options = [] - for option in self[section]: - default_str = None - for parent in self.parents: - if parent.has_option(section, option, defaults=False): - default_str = _to_utf8(parent.get(section, option)) - break - option_str = _to_utf8(option) - current_str = False - if self.parser.has_option(section_str, option_str): - current_str = self.parser.get(section_str, option_str) - if current_str is not False and current_str != default_str: - options.append((option_str, current_str)) - if options: - sections.append((section_str, sorted(options))) - - # At this point, all the strings in `sections` are UTF-8 encoded `str` try: - fileobj = AtomicFile(self.filename, 'w') - try: - fileobj.write('# -*- coding: utf-8 -*-\n\n') - for section, options in sections: - fileobj.write('[%s]\n' % section) - for key_str, val_str in options: - if to_unicode(key_str) in self[section].overridden: - fileobj.write('# %s = \n' % key_str) - else: - val_str = val_str.replace(CRLF, '\n') \ - .replace('\n', '\n ') - fileobj.write('%s = %s\n' % (key_str, val_str)) - fileobj.write('\n') - finally: - fileobj.close() - self._old_sections = deepcopy(self.parser._sections) - except Exception: - # Revert all changes to avoid inconsistencies - self.parser._sections = deepcopy(self._old_sections) - raise - + lock.acquire() + sections = [] + for section in self.sections(): + section_str = _to_utf8(section) + options = [] + for option in self[section]: + default_str = None + for parent in self.parents: + if parent.has_option(section, option, defaults=False): + default_str = _to_utf8(parent.get(section, option)) + break + option_str = _to_utf8(option) + current_str = False + if self.parser.has_option(section_str, option_str): + current_str = self.parser.get(section_str, option_str) + if current_str is not False and current_str != default_str: + options.append((option_str, current_str)) + if options: + sections.append((section_str, sorted(options))) + + + # At this point, all the strings in `sections` are UTF-8 encoded `str` + try: + fileobj = AtomicFile(self.filename, 'w') + try: + fileobj.write('# -*- coding: utf-8 -*-\n\n') + for section, options in sections: + fileobj.write('[%s]\n' % section) + for key_str, val_str in options: + if to_unicode(key_str) in self[section].overridden: + fileobj.write('# %s = \n' % key_str) + else: + val_str = val_str.replace(CRLF, '\n') \ + .replace('\n', '\n ') + fileobj.write('%s = %s\n' % (key_str, val_str)) + fileobj.write('\n') + finally: + fileobj.close() + self._old_sections = deepcopy(self.parser._sections) + except Exception: + # Revert all changes to avoid inconsistencies + self.parser._sections = deepcopy(self._old_sections) + raise + finally: + lock.release() + def parse_if_needed(self, force=False): if not self.filename or not os.path.isfile(self.filename): return False @@ -263,13 +271,17 @@ changed = False modtime = os.path.getmtime(self.filename) if force or modtime > self._lastmtime: - self._sections = {} - self.parser._sections = {} - self.parser.read(self.filename) - self._lastmtime = modtime - self._old_sections = deepcopy(self.parser._sections) - changed = True - + try: + lock.acquire() + self._sections = {} + self.parser._sections = {} + self.parser.read(self.filename) + self._lastmtime = modtime + self._old_sections = deepcopy(self.parser._sections) + changed = True + finally: + lock.release() + if changed: self.parents = [] if self.parser.has_option('inherit', 'file'): }}} And I confirmed fix at Windows platform Apache and tracd.",kawamuramst next-major-releases Release,10413,[PATCH] Descriptive component names in trac.log,general,,enhancement,normal,,new,2011-10-15T08:41:52+02:00,2015-03-02T22:27:22+01:00,"trac.log doesn't contain full component names and as a result it is hard to filter information relevant to plugins. Consider the following excerpt: {{{ 2011-10-15 01:24:47,716 Trac[main] DEBUG: Dispatching 2011-10-15 01:24:47,744 Trac[svn_fs] DEBUG: Subversion bindings imported 2011-10-15 01:24:47,747 Trac[api] INFO: Synchronized '' repository in 0.03 seconds 2011-10-15 01:24:47,747 Trac[web_ui] DEBUG: authenticate() method called 2011-10-15 01:24:47,747 Trac[session] DEBUG: Retrieving session for ID '' 2011-10-15 01:24:47,748 Trac[web_ui] DEBUG: user_locked: False 2011-10-15 01:24:47,753 Trac[chrome] DEBUG: Prepare chrome data for request 2011-10-15 01:24:47,755 Trac[web_ui] DEBUG: authenticate() method called }}} Here: - Trac[main] is probably trac.core.main - Trac[web_ui] is acct_mgr.web_ui - Trac[session] is ??? I'd propose to change that default format to replace Trac[] with full module name (i.e. __name__) ",anatoly techtonik next-major-releases Release,10409,Link to list of supported macros/processors on error,rendering,0.12.2,enhancement,normal,,new,2011-10-14T17:30:38+02:00,2011-10-14T21:38:06+02:00,"The message below can link to pages with supported macros and processors. {{{ #!noteX some error }}}",anatoly techtonik next-major-releases Release,10396,"[PATCH] New extension point interface to reformat the ""replying to"" block in a comment",ticket system,0.12-stable,enhancement,normal,,new,2011-10-05T12:13:27+02:00,2015-12-19T16:53:00+01:00,"Sometimes you don't want the original text in a response on a ticket in the ""Replying to"" line. Instead some sort of custom formatting (e.g. abbreviation, reference no.) could be used preferably. Solution: A new extension point ITicketCommentFormatter defined in ticket/api.py. The function quote_original() in ticket/web_ui.py calls the plugins, that implement the new extension point. So it is possible to reformat the ""replying to"" block in a comment, to leave it unchanged or to drop it at all. ",juliak next-major-releases Release,10381,"Use full width for custom textarea fields, and make them collapsible",ticket system,0.12.2,enhancement,normal,,new,2011-09-29T11:14:09+02:00,2015-03-03T15:34:58+01:00,"The content of custom textareas might get very big, so it would be better to place these fields in full width after description field. Furthermore it would be nice, if these fields are collapsable, so you could close them, if you do not need them at the moment.",anonymous next-major-releases Release,10376,Opening a link in a new window?,wiki system,,enhancement,normal,,new,2011-09-27T15:54:08+02:00,2016-12-19T15:54:19+01:00,"What code is necessary for me to create a link that will open in a new window, as opposed to opening within the existing window? I am including a link on a new webpage on gsa.gov that will take the customer away from the GSA site and onto a non-government site. ",jennifer.johns@… next-major-releases Release,10372,Search in wiki should prioritize title matches and whole word matches above partial word matches,search system,,enhancement,normal,,new,2011-09-26T04:35:52+02:00,2019-08-10T06:52:08+02:00,"I am searching for ""tea"" in my wiki. I have a page whose title is ""!TeaNotes"", whose contents is a list of my notes for various teas where the word ""tea"" appears on every single line that is not a heading (13 total occurrences). I get 6 pages of results for ""tea"". This wiki page appears on page 5 of 6 of search results. Right above it are completely irrelevant pages including the word ""instead"" (which includes ""tea"") with the ""tea"" part of it highlighted. When results are shown from wiki pages, the order of them should be: 1. Whole words matched in title 2. Whole words matched in text 3. Partial words matched in title 4. Partial words matched in text The order of 2 and 3 may be switched, or maybe they should be a single bracket, but any way this is done would be an improvement over the current situation. This particular case with tea illustrates the problem of finding stuff in the wiki very clearly, but the lack of a sensible ordering of results is a problem whenever I try to find stuff in the wiki in just about any trac installation I'm using. TitleIndex is pretty much the only way to search for things in the wiki if you have an idea of the page title (or know that you're searching for something which appears in the title).",oleg@… next-major-releases Release,10350,e.g. [#1234] in svn commit 5678 comment generates wrong Trac link,wiki system,0.12.2,enhancement,minor,,new,2011-09-09T12:08:04+02:00,2020-07-14T19:19:11+02:00,"The effect is that the Trac generated link will be http://......./changeset/5678#1234 which on clicking does not lead to the ticket 1234. The link should rather be http://....../ticket/1234. IMHO the link generator should ignore the brackets if it sees the #. The whole problem is because some people are used to brackets instead of parenthesis when they write additional comments in their texts. Actually, they should write {{{ (#1234) }}}, but I think you should also be able to stand the one with brackets.",fbrettschneider@… next-major-releases Release,10325,Way to delete ticket comments versions,ticket system,0.12.2,enhancement,normal,,new,2011-08-20T23:03:11+02:00,2011-08-20T23:58:49+02:00,"In a similar way that it is possible to delete just wiki page versions, it would be great to be able to delete also ticket comments versions. Currently it is possible only to delete the comment as a whole, with all versions. Similar maybe also for summary versions?",Mitar next-major-releases Release,10323,Enable sorting in TitleIndex macro,wiki system,,enhancement,normal,,new,2011-08-19T10:05:15+02:00,2021-11-19T05:32:16+01:00,"The goal is to be allowed to order the result of the macro `TitleIndex` in the same way that `RepositoryIndex` with the addition of the parameters `order` and `desc`. * `Order` possible values: [title|date] * `Desc` possible values: [0|1], same repositoryindex. For example: {{{ [[TitleIndex(order=title, desc=1)]] }}}",carlos.lopezperez@… next-major-releases Release,10317,"log?rev=N&stop_rev=N-1&limit=2 only shows a single revision, but should show two revisions",version control/log view,0.11.7,defect,normal,,new,2011-08-17T18:40:12+02:00,2015-08-05T10:45:04+02:00,"To reproduce: * Go to http://trac.edgewall.org/log/trunk?rev=10769&stop_rev=10768&limit=2 Only a single revision is shown. If you instead go to either of these URLs: * http://trac.edgewall.org/log/trunk?rev=10769&stop_rev=10768&limit=3 * http://trac.edgewall.org/log/trunk?rev=10769&limit=2 …two revisions are shown. Since ""limit"" corresponds to the ""Show at most N revisions per page"" form field, it seems like limit=2 should allow 2 revisions to be shown, even when combined with rev and stop_rev.",adam@… next-major-releases Release,10311,improve ticket change hints in custom query,query system,,defect,normal,,new,2011-08-11T08:02:47+02:00,2012-08-02T14:00:12+02:00,"Steps to reproduce: 1. query tickets owned by user A and B group by owner 2. reassign one ticket found in step 1 to user C 3. query tickets owned by user A and B again, with the same constraint as in step 1 Expect result, tickets are shown group by user A and B In fact, user C with the ticket just assigned to him shown too. After a quick look at the source code, found it's the cached ""query_tickets"" session attribute caused the issue. If query constraints keep the same but ticket changed within time delta, query result may be wrong. ",haterw@… next-major-releases Release,10305,Incomplete URL sent to Akismet API,web frontend,,defect,normal,,new,2011-08-09T08:49:30+02:00,2015-05-26T20:50:57+02:00,"Akismet integration wasn't working for me (I hope I've fixed it now), in combination with lighttpd and fastcgi: the log just says ""invalid API key."" I'm using Trac 0.11 but I see what I think is the problem in the 0.12 version of [source:plugins/0.12/spam-filter-captcha/tracspamfilter/filters/akismet.py]: when calling the Akismet Rest API, the spam filter plugin passes {{{req.base_url}}} for the ""blog"" parameter. This does the wrong thing when {{{TRAC_BASE_URL}}} is not a full URL to the Trac instance, with protocol and hostname. (In my case it was set to ""/development""). In retrospect, this is probably just my mistake in configuring lighttpd but given how sparse its documentation is, I probably won't be the last user to get that wrong. And Akismet's API gives no details beyond ""invalid,"" so debugging this is little fun. (In my case it was about two days, a support session, thirty spam bugs, and a tcpdump trace of fun. ☺) Assuming there's no good way to make up a host name to send to Akismet if {{{base_url}}} contains none, it'd be helpful if the plugin could detect this particular problem and log it. Or if it turns out to be __almost__ possible to construct a full URL: Akismet support tells me they normalize protocol and hostname, so subtleties such as https vs. http or a ""www."" prefix don't matter. (By the way, love the interactive ticket preview at the bottom of the newticket page! Hadn't seen that before.)",Jeroen Vermeulen next-major-releases Release,10286,Way to query all tickets with attachments,query system,0.12.1,enhancement,normal,,new,2011-07-26T13:10:32+02:00,2020-07-14T19:19:20+02:00,"Trac should provide a way to query all tickets with attachments, or even better, with a given name/description in an attachment. Or maybe even attachment size(s) (individual or combined).",Mitar next-major-releases Release,10279,Sort admin page milestone table by descending due date,admin/web,,defect,minor,,new,2011-07-22T13:58:53+02:00,2022-07-08T10:52:32+02:00,"Currently, the list of milestones in the web admin seems to be sorted by ascending due date. If you have a lot of milestones, that's kind of a pain: the old milestones are likely to be the least interesting. Simply sorting by descending due date probably makes for a better experience (even though it will show far-future milestones before near-future ones). For a more refined solution, split the table in two parts: a list of open milestones, by ascending due date, and a list of closed milestones, by descending due date.",Dirkjan Ochtman next-major-releases Release,10202,TracQuery fails to handle checkbox custom fields,query system,0.12-stable,defect,major,,new,2011-05-31T12:14:38+02:00,2011-05-31T12:45:22+02:00,"Define a custom field {{{ col_private = checkbox col_private.label = Private col_private.value = 0 col_private.order = 10 }}} If you now create a new TracQuery via webpage, select this custom field for filter and set filter value to 'No', all previous created tickets are filtered out. Ok, accepting that the old tickets do not have a ""Private"" value set, we edit the `query:`-String and replace `col_private=0` by `col_private!=1`. But now all tickets are listed, including all tickets set as ""Private"". The macro TicketQuery ""fails"" too using `col_private=0` for filtering, but can handle `col_private!=1` properly. ||Trac ||0.12.2|| ||Babel ||0.9.6|| ||Genshi ||0.6|| ||pysqlite ||2.6.3|| ||Python ||2.4.3 (#1, Sep 3 2009, 15:37:37) [GCC 4.1.2 20080704 (Red Hat 4.1.2-46)]|| ||pytz ||2011e|| ||setuptools ||0.6c11|| ||SQLite ||3.3.6|| ||Subversion ||1.4.2 (r22196)|| ||jQuery: ||1.4.2||",anonymous next-major-releases Release,10201,[Patch] WikiProcessors lacks detailed help,wiki system,0.13dev,enhancement,normal,,new,2011-05-29T18:32:23+02:00,2015-12-19T16:52:05+01:00,"I've recently discovered the detailed help-feature of WikiMacros, which is activated by adding a question sign (`?`) after a macro name. Example: {{{ [[TitleIndex?]] }}} This will show the detailed help, instead of the result. This is very handy during side-by-side edit to quickly find out the details of the macro. This feature does **not** work for WikiProcessors, which would be a-nice-to-have feature, and natural since a processor really is a macro. Example: {{{ {{{ !#TitleIndex? }}} }}} This yields the normal output and not the help information as intuitively expected due to the behaviour of WikiMacros. It would also be nice if all in-built processors (`div`, `span`, `table` etc) are capable of providing help information that could be brought up during side-by-side editing.",Mikael Relbe next-major-releases Release,10185,"Roadmap view doesn't cache ""Show completed milestones""",roadmap,0.12.2,defect,normal,,new,2011-05-16T15:59:06+02:00,2015-09-10T10:19:58+02:00,"When enabling ""Show completed milestones"" it isn't cached, so when calling Roadmap page again it isn't enabled. Same for ""Hide milestones with no due date"".",anonymous next-major-releases Release,10183,How to browser sources which folders have been changed and/or removed?,version control/browser,0.12.2,defect,normal,,new,2011-05-16T12:36:21+02:00,2015-05-26T20:50:57+02:00,"When I try to browser sources which folders have been changed and/or removed, there is an error message: No node ""/path/to/oldfile"" at revision ""head revision number"". Could I browser these kind of sources through Trac browser? Thank you!",zheng.bangyou@… next-major-releases Release,10182,Unit tests do not run in their own sandbox,general,0.13dev,enhancement,normal,,new,2011-05-16T12:16:37+02:00,2012-07-11T07:54:18+02:00,"Unit tests use the system temporary directory to create sub dirs and files. This leads to two distinct issues: 1. Two test sessions running at once on the same host may collide, as they create, update and remove files and directories within the same top directory 1. If the temporary directory already contains files or dirs from a previous test session, the current test session fails within the setUp() methods. This issue causes the Python `unittest` framework not to call the clean up tearDown() methods, leaving the faulty directory unfixed. Any subsequent unit test session keeps failing until manual removal of the directory is performed. A unique temporary directory could be created and used for every single test session, to avoid those issues. ",Emmanuel Blot next-major-releases Release,10168,trac generates error 500 in case no prermission to conf/trac.ini,general,0.12.2,defect,normal,,new,2011-05-05T19:02:18+02:00,2011-05-06T09:01:26+02:00,"Hi, I just spend several hours trying to debug a trac problem. Which can be easily reproduced. I got ... {{{ Trac Error TracError: The Trac Environment needs to be upgraded. Run ""trac-admin /srv/trac upgrade"" }}} Then I realized that by changing the owner of the config file conf/trac.ini it got to work. To reproduce this do: {{{ root@Apollon:/srv/trac # chown root conf/trac.ini root@Apollon:/srv/trac # ll conf/trac.ini -rw------- 1 root apache 6,8K 5. Mai 18:18 conf/trac.ini root@Apollon:/srv/trac # service httpd restart httpd beenden: [ OK ] httpd starten: [ OK ] }}} It does now not work. {{{ root@Apollon:/srv/trac # chown apache conf/trac.ini root@Apollon:/srv/trac # service httpd restart httpd beenden: [ OK ] httpd starten: [ OK ] }}} Now it works again. I am using Trac 0.12.3dev-r10683.",mail@… next-major-releases Release,10167,"Column ""Date"" in Change History of a Wiki Page",wiki system,0.12.2,enhancement,minor,,new,2011-05-05T18:35:29+02:00,2011-05-05T22:47:42+02:00,"This column should by labelled ""Age"" (and the css class should be ""age""), since it actually shows the age of the changes (... days '''ago'''), not the date.",wilfred.berger@… next-major-releases Release,10162,Add a REPOS kind of Node,version control,0.12.1,defect,normal,,new,2011-05-03T00:34:39+02:00,2020-04-01T17:09:59+02:00,"How to Reproduce While doing a GET operation on `/changeset/26e036576133f07b15eeb4363bf993b67bf9eacb/tp`, Trac issued an internal error. This only happens for a changeset that only updates git submodules, for which ""get_content_length"" returns None (because there is no associated git blob). (The Python error is ""unsupported operand type(s) for +"") This is similar to the old ticket #3287 here. Request parameters: {{{ { new : u 26e036576133f07b15eeb4363bf993b67bf9eacb , new_path : u /tp } }}} User agent: `Mozilla/5.0 X11; U; Linux x86_64; en-US; rv:1.9.2.17 Gecko/20110428 Fedora/3.6.17-1.fc14 Firefox/3.6.17` System Information || `Trac` || `0.12.1` || || `Docutils` || `0.7` || || `Genshi` || `0.6` || || `GIT` || `1.7.3.4` || || `psycopg2` || `2.2.2` || || `Pygments` || `1.3.1` || || `Python` || `2.6.6 r266:84292, Mar 6 2011, 17:53:48 ` br ` GCC 4.3.4 ` || || `pytz` || `2010o` || || `setuptools` || `0.6` || || `Subversion` || `1.6.16 r1073529 ` || || `jQuery` || `1.4.2` || Enabled Plugins || `BatchModify` || `0.8.0-trac0.12` || || `FilteredTicketQuery` || `N/A` || || `timingandestimationplugin` || `1.0.6` || || `TracGit` || `0.12.0.6dev` || Python Traceback {{{ Traceback most recent call last : File /usr/lib64/python2.6/site-packages/trac/web/main.py , line 511, in _dispatch_request dispatcher.dispatch req File /usr/lib64/python2.6/site-packages/trac/web/main.py , line 237, in dispatch resp chosen_handler.process_request req File /usr/lib64/python2.6/site-packages/trac/versioncontrol/web_ui/changeset.py , line 352, in process_request self._render_html req, repos, chgset, restricted, xhr, data File /usr/lib64/python2.6/site-packages/trac/versioncontrol/web_ui/changeset.py , line 586, in _render_html diff_bytes _estimate_changes old_node, new_node File /usr/lib64/python2.6/site-packages/trac/versioncontrol/web_ui/changeset.py , line 537, in _estimate_changes return old_size new_size TypeError: unsupported operand type s for : NoneType and NoneType }}}",dmarti next-major-releases Release,10152,"Custom Query - ""and"" becomes ""or""",query system,0.12.2,defect,normal,,new,2011-04-27T10:38:56+02:00,2017-03-07T07:00:53+01:00,"In ''custom query'' when you have a field {{{ Description contains ... }}} click ""and"" > ""Description"" (combobox to the left) it changes to {{{ Description contains ... or ... }}} It is not intuitive if you have the choice between ""and"" and ""or"" and decide for ""and"", you get just another variant of ""or"". Expected: {{{ Description contains ... (and) Description contains/doesn't contain ... }}} ",bastik next-major-releases Release,10150,Plugin installation errors should be shown on admin page,admin/web,,enhancement,normal,,new,2011-04-26T22:13:49+02:00,2011-04-26T23:13:25+02:00,"The last few times I've tried to install a plugin, I've made some sort of stupid mistake. Either I got an egg for the wrong version of Python, or I've right-click copied what I thought was the url from trac-hacks, but was actually the html page with the link to the attached egg. (Then I wget the ""egg"" and I'm none the wiser until I investigate it). And when I put it in the plugins directory and reload trac... nothing happens. I have to do the investigation myself. It would be nice if, whatever it is that makes trac decide to reject the egg, would be displayed in the same screen as the list of plugins. Thanks!",orblivion@… next-major-releases Release,10125,CommitTicketUpdater should call manipulators and listeners,ticket system,0.12-stable,enhancement,normal,,new,2011-04-10T21:35:04+02:00,2015-05-26T20:50:57+02:00,"I don't use this hook myself, but I frequently get questions about it on IRC and try to help out. One very frequent request is some or handling for additional fields or related plugins that need to act on what gets updated on the ticket. Example can be also adjusting estimates or time spent on fixing bugs, or updating some other related fields and structures. Anyway, I quickly scanned the source and see that we don't call ticket system manipulators or listeners? That would provide the hooks to either change fields directly together with commit-update, or be handle updating of related information after-the-fact via listener. Any reason not to call manipulators and listeners?",osimons next-major-releases Release,10101,Broken permissions for changeset view,version control/changeset view,0.12,defect,normal,,new,2011-03-26T13:10:34+01:00,2019-09-06T19:51:32+02:00,"I installed Trac 0.12 some time ago and created me as admin user (TRAC_ADMIN). I added SVN respository, synchronization and everything works fine. After some time, I added a new user and added him some permissions. But when I try to view a changeset logged as this new user, I get weird results - only some added files (2-3 from ~20 new files) with not completed paths (e.g. ballance instead of user_accounts/ballance). Diffs are not shown. Sometimes I get only: Message: (No files) It is ok when I log in as admin user. After some time playing with permissions, when I delete all permissions, then I add TRAC_ADMIN permission to the new user, then changesets are ok. But when I try to add only CHANGESET_VIEW permission (or all other permissions except TRAC_ADMIN), then it is not working. I have to add (or add, remove, add) TRAC_ADMIN permission to the new user, then all required permissions and finally remove TRAC_ADMIN permission if I want CHANGESET_VIEW to work without admin permission.",bruce@… next-major-releases Release,10091,Exception when formatting Mercurial changeset link,plugin/mercurial,0.12.2,defect,normal,,new,2011-03-19T19:22:39+01:00,2015-05-26T20:50:57+02:00,"When trying to format `[cset:34c7144cba5a]` or `[123:34c7144cba5a]` when having multiple Mercurial repositories but no default repository fails with the following exception: {{{ UnboundLocalError: local variable 'errmsg' referenced before assignment }}} Stack trace: {{{ File ""/usr/lib/python2.6/dist-packages/trac/ticket/templates/ticket_change.html"", line 61, in ${wiki_to_html(context, change.comment, escape_newlines=preserve_newlines)} File ""/usr/lib/python2.6/dist-packages/trac/wiki/formatter.py"", line 1497, in format_to_html return HtmlFormatter(env, context, wikidom).generate(escape_newlines) File ""/usr/lib/python2.6/dist-packages/trac/wiki/formatter.py"", line 1452, in generate escape_newlines) File ""/usr/lib/python2.6/dist-packages/trac/wiki/formatter.py"", line 1240, in format result = re.sub(self.wikiparser.rules, self.replace, line) File ""/usr/lib/python2.6/re.py"", line 151, in sub return _compile(pattern, 0).sub(repl, string, count) File ""/usr/lib/python2.6/dist-packages/trac/wiki/formatter.py"", line 1158, in replace replacement = self.handle_match(fullmatch) File ""/usr/lib/python2.6/dist-packages/trac/wiki/formatter.py"", line 1154, in handle_match return internal_handler(match, fullmatch) File ""/usr/lib/python2.6/dist-packages/trac/wiki/formatter.py"", line 540, in _lhref_formatter return self._make_lhref_link(match, fullmatch, rel, ns, target, label) File ""/usr/lib/python2.6/dist-packages/trac/wiki/formatter.py"", line 581, in _make_lhref_link fullmatch) File ""/usr/lib/python2.6/dist-packages/trac/wiki/formatter.py"", line 590, in _make_link fullmatch) File ""build/bdist.linux-x86_64/egg/tracext/hg/backend.py"", line 438, in _format_link title=errmsg, rel=""nofollow"") }}}",Mitar next-major-releases Release,10059,prop.rendered not properly set in browser.html,version control/browser,0.12.2,defect,normal,,new,2011-03-02T18:04:35+01:00,2012-01-19T19:40:14+01:00,"Hi, despite having the wiki_properties set to a property that is rendered correctly as wiki syntax, the prop.rendered in browser.html is not set. This causes everything to be displayed not in the py:when case of a span and a div.",anonymous next-major-releases Release,10047,Add RSS support for Roadmap,roadmap,,enhancement,normal,,new,2011-02-26T14:48:53+01:00,2014-07-31T15:37:56+02:00,"Hello, i'm requesting RSS support for the ""roadmap"" data. Something like the RSS output for timeline: http://trac.edgewall.org/timeline?format=rss Not working currently: http://trac.edgewall.org/roadmap?format=rss This would be usefull to display the list of milestones with dates in an rss-reader or in another website. Best regards, Jens",vain@… next-major-releases Release,10032,Wiki page mouse hover tooltip,wiki system,0.11.7,enhancement,minor,,new,2011-02-17T12:58:02+01:00,2011-02-17T15:28:28+01:00,"When I am currently in a comment (or even on the timeline), the WikiPages and CamelCase words are highlighted in red. But they do not show tooltips when the mouse hovers over them. It is really nice when the mouse hovers over ticket numbers like #1 it will show the summary of the ticket. This would be nice if it works with WikiPages and CamelCase words as well.",francoishill11@… next-major-releases Release,10027,TicketQuery groups without translation,ticket system,0.12.2,defect,minor,,new,2011-02-14T15:49:28+01:00,2011-02-25T11:11:46+01:00,"I'm using Trac 0.12.2 with German language. On my WikiStart page I have a TicketQuery with ""order=milestone"". The group title results in ""test milestone Tickets:"". Due to my language settings I would expect something like ""test Meilenstein Tickets:"" as the German word ""Meilenstein"" is associated with the English ""Milestone"". Without having it tested, I assume it's the same e.g. for component (""Komponente"") and all these normally translated default fields. (see [source:/tags/trac-0.12.2/trac/ticket/query.py#L1338 here])",the3illings@… next-major-releases Release,9979,[PATCH] Single changeset resync fails if changesets are missing,version control,0.12-stable,defect,normal,,new,2011-01-15T00:28:50+01:00,2011-01-18T17:04:55+01:00,"When running {{{repository resync repo rev}}} the operation fails if the revision you are trying to resync is missing from the revisions table. Attached patch ""fixes"" that, maybe not in the most elegant or correct way. Improvements welcome. ",barry@… next-major-releases Release,9978,Guard against pkg_resources.resource_filename errors,general,0.13dev,defect,normal,,new,2011-01-14T20:25:40+01:00,2016-11-03T07:08:11+01:00,"During a trac-svn-hook resync (`trac-admin ... changeset added /var/svn/trac 10433`): {{{ [pid 1052 139942713288416] 2011-01-14 18:46:06,498 Trac[env] INFO: -------------------------------- environment startup [Trac 0.13dev-r10426] ------------------------ -------- [pid 1052 139942713288416] 2011-01-14 18:46:09,926 Trac[console] ERROR: Exception in trac-admin command: Traceback (most recent call last): ... File ""build/bdist.linux-x86_64/egg/trac/versioncontrol/admin.py"", line 95, in _do_changeset_added rm.notify('changeset_added', reponame, revs) File ""build/bdist.linux-x86_64/egg/trac/versioncontrol/api.py"", line 620, in notify repositories = self.get_repositories_by_dir(dir) ... File ""build/bdist.linux-x86_64/egg/tracext/hg/backend.py"", line 210, in __init__ locale_dir = pkg_resources.resource_filename(__name__, 'locale') ... File ""/usr/local/virtualenv/0.13dev/lib/python2.5/site-packages/setuptools-0.6c9-py2.5.egg/pkg_resources.py"", line 887, in extraction_error raise err ExtractionError: Can't extract file(s) to egg cache [Errno 17] File exists: '/var/www/.python-eggs/TracMercurial-0.13.0.1dev_r10432-py2.5.egg-tmp/tracext/hg/locale' }}} So a duplicate of #9107 for the TracMercurial plugin. We can fix that particular occurrence, but I wonder if it wouldn't make sense to add a wrapper for the pkg_resources stuff (in `trac.utils.dist`?) and use that. That wrapper should be ""process safe"" not only thread-safe, so using a lock won't cut it. Maybe trapping such errors and restarting the call //once// after a random amount of time between .3 to 1s would be enough.",Christian Boos next-major-releases Release,9967,Allow setting option values through Option objects,general,0.12.1,enhancement,normal,,new,2011-01-08T13:57:33+01:00,2017-09-01T16:51:32+02:00,"Trac currently allows the user to read an option's value through a `Option` object, like in the following code: {{{ #!python class MyComponent(Component): my_option = Option('my_section', 'my_option') def my_func(self): # retrieving option value my_option_value = self.my_option }}} When calling `my_func()`, `my_option_value` will be filled with the current option values (as specified in `trac.ini`). Now, while retrieving works this way, one can't ''set'' the option value this way. {{{ #!python class MyComponent(Component): my_option = Option('my_section', 'my_option') def my_func(self): # setting option value # will raise an AttributeError self.my_option = 'test' self.config.save() }}} I've created a patch that enables this functionality. I've done some tests with it and it seems to work for `Option` and all of its subclasses.",Sebastian Krysmanski next-major-releases Release,9947,Unexpected resolver result for TracLinks in wiki page within hierarchy,wiki system,0.12-stable,defect,normal,,new,2010-12-26T23:10:46+01:00,2014-03-14T00:55:54+01:00,"Current wiki parser seems unable to translate links like {{{WikiPage}}}, {{{[../WikiPage]}}} or even {{{[wiki:WikiPage]}}} to the expected target, a least if used to reference an equally named page from within a wiki subpage: See [wiki:TranslationDe/TracWiki@1], where I tried to reference {{{[wiki:TracWiki]}}} by typing just TracWiki at the top of the page but got the equivalent of {{{[wiki:TranslationDe/TracWiki]}}} instead, what is plain wrong. I tried all of the aforementioned syntax and came up with {{{[/wiki/TracWiki]}}} as valid workaround to get the expected «top-hierarchy page» for a fixed revision in [wiki:TranslationDe/TracWiki@2].",Steffen Hoffmann next-major-releases Release,9938,Improve on static resources,web frontend,0.13dev,enhancement,normal,,new,2010-12-18T20:06:31+01:00,2011-02-25T21:11:22+01:00,"I know there are multiple ways of externally optimizing servicing static resources from a trac environment. However, for must users these are way off their limits, so here is a different approach that is directly built into trac. By default, the chrome component is the instance where all template providers will be evaluated for servicing static resources for a given prefix and filename /aka path. As of now, requests for these files will be processed by the chrome component's request handler, which then will call req.send_file for sending the file. Since this requires the whole bunch of the default request dispatcher component, along with multiple other instances of components being either instantiated or referenced, overall servicing latency is rather high, especially for resources that lead to a 404 or 304 response. My proposition is now to introduce a `StaticResourceCache` component at the web.main level, which will then be used by both the chrome component for registering resources and the main _dispatch_request routine. The cache will be filled lazily, so for each resource currently not in the cache, the default request dispatcher will call chrome to process the request normally. As soon as the resource is in the cache, the _dispatch_request function will intercept the call and directly handle the static resource request. In my local environment, this has lowered response times by about 100...500ms. As an example, for a given page I did multiple reloads and viewing the results in firebug's network monitor. Without the optimizations, response times for static resources vary from 13ms to >200ms. With the optimizations, response times for static resource become much more reproducible, varying also from 13ms to >200ms. The peaks, however, are now less frequent and overall response times are well below the 60ms mark. A nice side effect is that, the overall time for fetching a simple page has dropped from >560ms .. 1,2s (without optimizations) to 420ms..1,2s (with the optimizations). Again, here the peaks are less perceptible, especially when reloading the same page during edit/save/preview cycles. You can find the source for that patch under http://github.com/axnsoftware/trac-bugs-n-features. I will name you the branch as soon as I have the ticket id. ",Carsten Klein next-major-releases Release,9868,Impossible to see the changelog of a branch using Mercurial,version control/log view,,defect,normal,,new,2010-11-24T21:37:43+01:00,2011-04-18T02:16:55+02:00,"Using Mercurial Plugin: * browse source * select a branch * go to revision log Due to the nature of how it is implemented, it only shows in the first row the tip of such branch and the rest of revisions belongs to all branches. ",Ismael de Esteban next-major-releases Release,9843,Non existing Wiki pages should return 404.,general,,defect,minor,,new,2010-11-17T14:43:39+01:00,2010-11-17T15:12:46+01:00,"When a request is made to a non-existent wiki pages (for example: http://trac.edgewall.org/wiki/doesnotexist), a 200 OK code is returned. Shouldn't it return 404? ",AP² next-major-releases Release,9807,Wrong sections are collapsed by default!,ticket system,,enhancement,normal,,new,2010-11-10T01:38:35+01:00,2015-03-19T15:20:49+01:00,"Collapsing the sections in a ticke is great - the trouble is that the section that is used 99% of the time is the ""modify"" section which is pre-collapsed. An ideal solution would be to make this system configurable (or even user configurable). It is going to pee-off my users to now have to press that extra button... petty I know.... but its the truth.",morgand next-major-releases Release,9775,implement get_path_history in the versioncontrol cache,version control/log view,0.12.1,defect,normal,,new,2010-11-05T10:08:57+01:00,2015-05-26T20:50:57+02:00,"While doing a GET operation on `/log/oppimappi`, Trac issued an internal error. '''Steps to reproduce:''' * Click ""Browse Source"" * Select a line with a git repository and click hex label in the ""Rev"" column (View Revision Log) * Select ""Show only adds and deletes"" in ""Revision Log Mode:"" and hit ""Update"" * ""Trac detected an internal error:"" and ""NotImplementedError:"" (note that it seems that the error string is missing the end of error message) I'd guess that this is caused by an unimplemented feature in TracGit. Notice that ""Stop on copy"" and ""Follow copies"" do not cause errors (though I guess, they don't do anything with TracGit because everything is labelled as ""Modified""). I think the ""Summary"" for this ticket is bad but I'll leave it as is to make this issue easier to find from the ""Internal Error""'s ""search for similar issues"" link. The string ""oppimappi"" is the repository name in my trac environment. Request parameters: {{{ {'action': u'stop_on_copy', 'limit': u'100', 'mode': u'path_history', 'path': u'/oppimappi', 'rev': u'f55567eb3a36f3189e3212678aedf5cfd7e0d0c2', 'stop_rev': u''} }}} User agent: `Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.2.12) Gecko/20101027 Ubuntu/10.04 (lucid) Firefox/3.6.12` ==== System Information ==== || '''`Trac`''' || `0.12.1` || || '''`Babel`''' || `0.9.5` || || '''`Genshi`''' || `0.6` || || '''`GIT`''' || `1.7.0.4` || || '''`pysqlite`''' || `2.4.1` || || '''`Python`''' || `2.6.5 (r265:79063, Apr 16 2010, 13:09:56) ` [[br]] `[GCC 4.4.3]` || || '''`RPC`''' || `1.1.0-r9360` || || '''`setuptools`''' || `0.6` || || '''`SQLite`''' || `3.6.22` || || '''`Subversion`''' || `1.6.6 (r40053)` || || '''`jQuery`''' || `1.4.2` || ==== Enabled Plugins ==== || '''`TracGit`''' || `0.12.0.5dev` || || '''`TracXMLRPC`''' || `1.1.0-r9360` || ==== Python Traceback ==== {{{ Traceback (most recent call last): File ""/usr/local/lib/python2.6/dist-packages/Trac-0.12.1-py2.6.egg/trac/web/main.py"", line 511, in _dispatch_request dispatcher.dispatch(req) File ""/usr/local/lib/python2.6/dist-packages/Trac-0.12.1-py2.6.egg/trac/web/main.py"", line 237, in dispatch resp = chosen_handler.process_request(req) File ""/usr/local/lib/python2.6/dist-packages/Trac-0.12.1-py2.6.egg/trac/versioncontrol/web_ui/log.py"", line 163, in process_request for old_path, old_rev, old_chg in history(): File ""/usr/local/lib/python2.6/dist-packages/Trac-0.12.1-py2.6.egg/trac/versioncontrol/web_ui/log.py"", line 119, in history for h in repos.get_path_history(path, rev): File ""/usr/local/lib/python2.6/dist-packages/Trac-0.12.1-py2.6.egg/trac/versioncontrol/cache.py"", line 386, in get_path_history limit) File ""/usr/local/lib/python2.6/dist-packages/Trac-0.12.1-py2.6.egg/trac/versioncontrol/api.py"", line 888, in get_path_history raise NotImplementedError NotImplementedError }}} ",Mikko Rantalainen next-major-releases Release,9738,"Add ""attach file"" button at the end of the ticket",ticket system,0.12dev,enhancement,normal,,new,2010-10-31T21:39:09+01:00,2014-08-02T06:47:32+02:00,"Add ""attach file"" button at the end of the ticket as some (novice) users have problems finding a way to attach a file. They are used from wiki that button is at the bottom and in ticket also a change ticket form is at the bottom. So it is not really reasonable to imagine that there is one button up there (if you missed it). Even the anchor link to attachments does not help much. So I believe it would be useful to simply add ""attach file"" at the bottom, too.",Mitar next-major-releases Release,9735,TicketQuery adds reporter by default,ticket system,0.12dev,defect,normal,,new,2010-10-31T02:01:36+01:00,2011-10-29T18:28:35+02:00,`TicketQuery` macro adds reporter row by default if using `rows` option. It seems this is [source:trunk/trac/ticket/query.py#L430 hard-coded]. It would be great to be able to disable this feature for those who wants more concise output.,Mitar next-major-releases Release,9734,Redirecting to non-HTTP URLs not possible,general,,defect,normal,,new,2010-10-30T20:12:35+02:00,2011-04-13T12:07:21+02:00,"Redirecting to non-HTTP URLs is not possible because of the [source:trunk/trac/web/api.py#L369 check in the code] which assumes only `http` or `https` based URLs. Of course redirect should work for any valid URL. I also doubt there is any reason why Trac itself should try to fix non-absolute URLs. This is browser's work to do.",Mitar next-major-releases Release,9707,search ticket does not work when query for both subject keyword and comment keyword,search system,0.12,defect,major,,new,2010-10-25T00:16:24+02:00,2010-10-25T09:23:39+02:00,"If I [http://trac.edgewall.org/search?q=optionally+anticipate&noquickjump=1&ticket=on search tickets] for query ""optionally anticipate"", this should yield #1329 in the results, as the subject for this ticket contains ""optionally"" and one of the comments contains ""anticipate"". But the search contains 0 results as you can see.",tejas81@… next-major-releases Release,9683,Make htdocs_location more useful,web frontend,0.13dev,enhancement,normal,,new,2010-10-10T20:53:25+02:00,2011-09-15T13:25:44+02:00,"At the occasion of the discussion on r10203 and r10204, osimons pointed me at the limitations of the [TracIni#htdocs_location-option htdocs_location] and the fact that it was sometimes pretty useless as such, as it only coped with the core trac resources (those below `common/`). You end up having to add rewrite rules in the web server if you want to serve the static resources of the plugins, those exported along with Trac core ones when using `trac-admin deploy

    `. This got me thinking if we couldn't enhance the `htdocs_location` internal rewrite mechanism so that it would take care of all the chrome/ resources (except for the `site/` and `shared/` ones). By using a new setting, `htdocs_location_prefix`, we could control the restriction that we currently apply on resources starting with `common/`. So the default value for that new setting will be `common/`, but we would also support the empty value for serving everything but `site/` and `shared/` prefixes, and *, serving everything including `site/` and `shared/`. Testing here on [/demo-0.13], seems to work well, but OTOH, there are no plugins ;-) Feedback welcomed. ",Christian Boos next-major-releases Release,9674,[PATCH] trac.ini parameterization support,general,0.12,enhancement,normal,,new,2010-10-06T23:00:58+02:00,2015-05-26T20:50:57+02:00,"I'm running trac using the {{{trac.env_parent_dir}}} option. It would be useful to have a variable that can be expanded within an inherited trac.ini file, which indicates the subfolder for the trac instance using it. (Of course, other variables could also be conceived of). For example, given the following layout of trac environments: {{{ /trac - /env1 - /env2 - /env3 }}} A common trac.ini file could be created that references a {{{%{project}}}} variable, which is interpreted to be env1, env2, or env3, depending on which environment is inheriting the configuration. The following is all the instances in a common trac.ini file for which I found the variable useful (in my setup, at least): {{{ [account-manager] authentication_url = https://some.domain.com/trac/%{project}/login [authz_policy] authz_file = /var/auth/trac.%{project}.authz [notification] smtp_from = trac-%{project}@some.domain.com [trac] authz_file = /etc/apache2/auth/authz/trac.%{project}.authz base_url = https://some.domain.com/trac/%{project}/ [project] url = https://some.domain.com/trac/%{project}/ }}} If it were just one field that needed customization, I'd override it in the individual trac.ini files of each environment. Or, if it were a set of fields within a common section, I could use ConfigParser's support for {{{%variable}}} within that section (1). But, since it's many fields across many sections (as shown above), I went ahead and looked into creating a patch. (1) A relevant Python patch I ran across while looking for solutions: http://bugs.python.org/issue9876",sahendrickson@… next-major-releases Release,9673,Some TracIni settings should be user configurable,general,0.13dev,enhancement,normal,,new,2010-10-06T20:49:07+02:00,2015-05-26T20:50:57+02:00,"Many aspects of Trac are configurable, via the TracIni settings. A lot of these switches are presentation or slight behavior tweaks that are largely a matter of personal taste and preferences. As those are TracIni settings, this has the effect that the project owners' preferences are enforced, not necessarily the same as those of their users. It should however not be hard to allow //some// of these options to be tailored to the end users' needs. As suggested in #3360, we could extend our Option descriptors with an extra `user` parameter. For example: {{{#!python section = Section('ticket', N_('Ticket Settings')) chronological_order = BoolOption('ticket', 'chronological_order', default=True, user=N_(""Show Change History in Chronological Order""), doc=N_("""""" Controls in which direction the Change history is displayed. ''(since 0.13)'' """""") }}} We could use this parameter to pass the option summary that the user will see in a new preference panel. We could either have one user panel, with field sets grouping options from a given section, or we could have one panel per section (but then we need to fix the layout of panel tabs to allow for having lots of panels). For getting back the values, we could think about having a helper function `userpref(env, req, section, option)`, or something smarter that would fetch the value first from the `req.session` and only if not found pick the environment's default. A few tickets are concerned by this new feature: [[TicketQuery(keywords=~userpreferences,format=table)]] ",Christian Boos next-major-releases Release,9655,Source code navigation in browser,version control/browser,,enhancement,normal,,new,2010-09-30T23:09:34+02:00,2015-05-26T20:50:57+02:00,"It would be very cool if when viewing source code in Trac source browser, it will render the language elements (e.g. functions, methods, classes, module imports / includes) as links to where they are defined (like most IDEs). The lacking of this is the only reason that my team uses !SlickEdit for code-reviews, and not Trac browser. (well, this and the ability to annotate the reviewed code with comments, but that's another enhancement request). Now, I assume this is a heavy feature (language awareness?!), but thought I'd raise it anyway. Maybe it can be done partially using other libraries (I guess Pygments is doing something similar, syntax-analysis or something of that sort).",Itamar Ostricher next-major-releases Release,9653,tracd --base-path matches any string of the same length,web frontend/tracd,0.12,defect,minor,,new,2010-09-29T23:35:37+02:00,2010-10-06T21:40:00+02:00,"Say I start `tracd` with a a script that looks like: {{{ #!/bin/bash export ENV=/srv/trac-alpha tracd -r -p 8081 -s --basic-auth=trac-alpha,$ENV/htpasswd,example.com --base-path=trac/ $ENV }}} Then any 4-character URL works as the base path. For example visiting http://example.com:8081/trac/timeline shows the timeline, as does http://example.com:8081/1234/timeline or any other 4-character string. Visiting http://example.com:8081/123/timeline gives an error page with ""No handler matched request to timeline"" and visiting http://example.com:8081/123/timeline gives an error page with ""No handler matched request to 5/timeline""",kamil@… next-major-releases Release,9646,[PATCH] Highlight selected line when #L123 is used in source browser,version control/browser,0.13dev,enhancement,normal,,new,2010-09-26T18:51:40+02:00,2015-12-19T16:40:59+01:00,"When viewing a file via the source browser and #L is used, it would be nice if it'll visually show the line that was passed, especially when its near the end so its not scrolled down right to that line. Also, when it does and scroll the page up and down, its nice to still see the original line you were linked to.",shesek next-major-releases Release,9642,Resizable textarea for the attachment description,attachment,,enhancement,minor,,new,2010-09-24T00:24:23+02:00,2010-10-01T23:06:00+02:00,"When uploading attachments, you can add a one-line file description. It would be nice if it'll also be possible to add some longer text in a textarea, that'll be displayed after the short description in the ticket comment. Currently, I'll usually upload the attachment with a short description and add another separate comment with a longer description, which is a bit annoying.",shesek next-major-releases Release,9638,Submit forms with CTRL+Enter in textareas,web frontend,,enhancement,normal,,new,2010-09-22T17:30:14+02:00,2019-09-06T19:51:32+02:00,"Can you add support for submitting forms when pressing CTRL+enter in a textarea, in the same way they're submitted when enter is pressed in a regular input box? Its a nice feature that makes it easier to submit forms, as no mouse or tabbing is needed (//when I found out Eclipse supports that when writing commit messages, I thought it was the best invention since sliced bread! //), which is quite easy to implement using jQuery: {{{ $('textarea').keypress(function(e){ if (e.ctrlKey && e.keyCode == 13) { $(this).closest('form').find('input[type=submit][name=submit]').click(); } }); }}} This snipped would click an input with name=submit and type=submit (the one that submits the changes, without a preview) in the same form as the textarea, whenever CTRL+enter are pressed in a textarea. If you feel it'll make people submit tickets without previewing them first, it can be changed to only send ticket comments directly and use the preview for other forms: {{{ $('textarea').keypress(function(e){ if (e.ctrlKey && e.keyCode == 13) { $(this).closest('form').find('input[type=submit]'+(this.id==='comment'?'[name=submit]':'')).eq(0).click(); // When it isn't textarea#comment, the first submit button is used (which is the 'preview' button) } }); }}} P.S. When I first tried to implement this on my Trac, my first attempt was with `$('textarea').closest('form').submit();` which didn't work as the name used for the submit button [http://jibbering.com/faq/names/ is considered unsafe]. This is also mentioned on [http://api.jquery.com/submit/#notes-0 jQuery documention], and there's [http://yura.thinkweb2.com/domlint/ DOMLint] to check for that kind of issues. Yes, I was also surprised that 'submit' isn't a valid name for submit input and I'm not quite sure if it should really be fixed.",shesek next-major-releases Release,9554,plain text download of wiki pages reveals comments not meant for public consumption,wiki system,0.12dev,enhancement,normal,,new,2010-08-07T00:16:20+02:00,2015-05-26T20:50:57+02:00,"I find this one critical in that one might prototype content in the wiki that is not meant to be released to the public yet. In order to reconstruct the scenario, 1. create a new wiki page 2. enter some content 3. enter {{{ {{{ #!comment TBD Critical information that must not yet be leaked to the public. }}} }}} 4. download as plain text in the downloaded plain text you can see also the comment with content that was not yet meant to be consumed by the public. ",Carsten Klein next-major-releases Release,9550,Ticket should be divisible,ticket system,,enhancement,normal,,new,2010-08-06T04:51:22+02:00,2015-05-26T20:50:57+02:00,"Sometimes tickets are entered into the ticket system that contain more than one issue. These tickets should be diversible. When something like this happens I create new tickets with the parts and assign them to their owners. The all new ticket ids are put into the 'parent'. It would be nice if trac could help with this task and manage a parent/child relationship between tickets. When childs of a ticket are created the parent ticket should be set on hold until all children are resolved. The parent could show a progress bar or something. Notifications for children should be send to their subscribers and the subscribers of the parent ticket.",stephan@… next-major-releases Release,9539,Attachments do not change Last Modified Time,attachment,0.12-stable,defect,major,,new,2010-08-02T20:59:15+02:00,2010-08-03T00:48:44+02:00,"Adding attachments/files to a ticket does not change the value of Last Modified Time. I am using the RPC API to find recently updated Tickets. Since attachments do not change the modified time value, I never know if a ticket had new attachments.",kageex@… next-major-releases Release,9537,to implement cherry-picking for wiki pages,wiki system,,enhancement,normal,,new,2010-08-01T16:25:42+02:00,2015-05-26T20:50:57+02:00,"partially motivated by #1114 Would be nice to have an ability to pick any version from wiki revision history and ''copy'' it, making it the latest version. This way we don't need to delete anything (as mentioned in #1114). Instead, we keep full history intact, but just bring the version we need to the ""head"".",team@… next-major-releases Release,9526,Fine Grained Permission possible realms and paths format are not documented.,wiki system,0.12,defect,minor,,new,2010-07-27T10:09:14+02:00,2019-09-06T19:51:32+02:00,"I'm trying to configure authz_policy.py (TracFineGrainedPermissions) for things other than Wiki Pages. I can see that this is possible and working, but I'm waisting my time discovering ""paths and realms"" for every single thing I want to manage via authz_policy. All the samples and documentation I've found describe configuration for wiki: realms *only* and while this is probably most common usage I'd like to create a ticket to expand the documentation in this topic.",dpc@… next-major-releases Release,9510,Action block stays visible when no actions available,ticket system,0.11-stable,defect,normal,,new,2010-07-20T12:20:58+02:00,2015-03-02T22:27:22+01:00,"The action block on the ticket page is visible (empty) even if there are no actions available. I've implemented a custom workflow for our trac site, removing the ""leave"" action for users that were having just that option (senseless). In that case though, the action block is still there, empty (senseless and ugly now :)).",fredck@… next-major-releases Release,9478,Concurrent ticket modifications is still possible,ticket system,0.12,defect,minor,,new,2010-07-02T14:02:03+02:00,2015-07-28T12:31:04+02:00,"Before saving a ticket modification, we try to detect if the page was not already modified by another user (r5449). However this check is not done within a single transaction, and therefore it can be ineffective if two changes happen concurrently. See ticket:9235#comment:10 for an example (or `comments:10` actually ;-) ).",Christian Boos next-major-releases Release,9472,Workflow actions not translated,i18n,0.12,enhancement,minor,,new,2010-06-29T14:32:34+02:00,2013-11-23T01:54:34+01:00,"I'm administering an installation of Trac 0.12, Python 2.5.4, with multi-lingual users. Though most of Trac is translated correctly, the ''Action'' menu (under ''Modify Ticket'') still lists the main options in English. This is true for all languages that I have tried, namely Japanese, Hebrew and German.",sajjad@… next-major-releases Release,9443,"Dropdown list on ""View tickets""",report system,0.12dev,enhancement,normal,,new,2010-06-17T08:59:20+02:00,2010-09-06T15:06:21+02:00,"If you work frequently with tickets it is a pain to go from ""view ticket"" to a new page where you then have to pick the propper sorting. An enhancement would be if this list drops down when the mouse is over ""view tickets"" in the menu. Also, when a ticket is modified, there is no button to go back to the ticket list that you came from. This could also be fixed with this enhancement.",marc.rawer@… next-major-releases Release,9343,"Because Trac support for multiple repos, each repo with a related encoding.",version control,0.12dev,enhancement,minor,,new,2010-05-15T14:50:12+02:00,2017-04-16T22:35:39+02:00,"So I recommend for each repos, set an distinct default charset for it. ",anonymous next-major-releases Release,9310,Allow cookies be set for textarea heights upon resize and restore upon document load,general,0.12dev,enhancement,trivial,,new,2010-05-06T23:46:43+02:00,2013-11-15T04:26:51+01:00," This basically involves incorporating the jquery plugin for handling cookies. The patch is rather simple. See the attached jquery_cookies.js which must be included with trac/htdocs/js/. Here is the original project page for reference: http://plugins.jquery.com/project/cookie {{{#!diff Index: htdocs/js/resizer.js =================================================================== --- htdocs/js/resizer.js (Revision 57) +++ htdocs/js/resizer.js (Arbeitskopie) @@ -5,6 +5,15 @@ var textarea = $(this); var offset = null; + function getCookieName(textarea) { + if (textarea.attr(""id"")) { + return ""trac."" + textarea.attr(""id""); + } + else if (textarea.attr(""name"")) { + return ""trac."" + textarea.attr(""name""); + } + } + function beginDrag(e) { offset = textarea.height() - e.pageY; textarea.blur(); @@ -18,6 +27,12 @@ } function endDrag(e) { + // set cookie for permanently storing information + // on the height of the resizable, if possible + name = getCookieName(textarea); + if (name) { + $.cookie(name, textarea.height(), ( new Date() ).getTime() + 1000000); + } textarea.focus(); $(document).unbind('mousemove', dragging).unbind('mouseup', endDrag); } @@ -27,5 +42,15 @@ .parent().append(grip); grip.style.marginLeft = (this.offsetLeft - grip.offsetLeft) + 'px'; grip.style.marginRight = (grip.offsetWidth - this.offsetWidth) +'px'; + + // restore height from cookie, if avail + var height = 0; + var name = getCookieName(textarea); + if (name) + { + height = $.cookie(name) + } + textarea.height(Math.max(32, height)); }); }); + Index: web/chrome.py =================================================================== --- web/chrome.py (Revision 57) +++ web/chrome.py (Arbeitskopie) @@ -522,6 +522,7 @@ add_link(req, 'help', req.href.wiki('TracGuide')) add_stylesheet(req, 'common/css/trac.css') add_script(req, 'common/js/jquery.js') + add_script(req, 'common/js/jquery_cookies.js') # Only activate noConflict mode if requested to by the handler if handler is not None and \ getattr(handler.__class__, 'jquery_noconflict', False): }}} BTW Thanks for incorporating this, it makes using trac a lot more easier. ",Carsten Klein next-major-releases Release,9281,Allow IAttachmentManipulator to easily check on the uploaded attachment's file data,attachment,0.12dev,enhancement,normal,,new,2010-04-29T21:48:04+02:00,2010-05-05T22:11:50+02:00," Currently, both the filename and the file data is not available to the manipulator, unless it would reimplement part of the logic found in AttachmentModule#_do_save. _do_save already does some verification on the uploaded file, which I presume is present in some temporary location on the local filesystem. My proposal would be to also pass in the upload file object associated with that file to the manipulator so that it is able to verify its contents, for example scanning it with an antivirus scanner and so on. See also ticket #9280. ",Carsten Klein next-major-releases Release,9248,Trac as reST editor,wiki system,0.12dev,enhancement,normal,,new,2010-04-19T21:15:02+02:00,2011-02-25T15:52:39+01:00,"Trac markup is nice, but it is not suitable for documentation that should be distributed in various formats. For example, for Bitten reST is a better option as it could transformed by Sphinx into various stuff. It could be nice if Trac could maintain documentation in specially designated wiki namespace as reST docs, so that these docs can be then synced back with version control system. If it is possible to create a 1:1 symmetrical Trac to reST markup mapper (Genshi?), even partial - Trac could be used as online editor for Bitten docs, which will surely improve community participation.",anatoly techtonik next-major-releases Release,9214,expandable blockquotes,wiki system,,enhancement,normal,,new,2010-04-10T12:19:49+02:00,2015-05-26T20:50:57+02:00,"Would be nice to have an ability to configure blockquotes as ""expandable by jQuery"". Sometimes the content of blockquote is very large, say, 1000 lines, and a wiki page would look ugly with such a long text. Making a blockquote ""expandable"" would help, for example: {{{ {{{+expandable #!xml ...very long xml... }} } }}} Maybe such feature exists already, but I didn't find it. Thanks.",team@… next-major-releases Release,9205,Make the timeline threaded,timeline,0.12dev,enhancement,normal,,new,2010-04-07T22:42:51+02:00,2010-07-01T11:52:56+02:00,"From a little dicussion in [http://groups.google.com/group/trac-users/browse_thread/thread/377dfea4090c7a85 trac-users], it came out that it might be a good idea to have a threaded timeline with the possibility to expand/collapse threads. This would make the timeline less noisy, especially on busy timelines. What I mean here is something like (expanded view): {{{ 04/07/10 Ticket #123 created by Ticket #122 created by Ticket #100 closed by Ticket #100 updated by Ticket #100 created by Ticket #121 created by }}} Or a collapsed view: {{{ 04/07/10 Ticket #123 created by Ticket #122 created by + Ticket #100 closed by (2 more events) Ticket #121 created by }}} And if you click on the plus sign at !#100, the view gets expanded. We already have this in TracBrowser. The user should be able to configure this behaviour in her/his preferences.",r.sokoll@… next-major-releases Release,9172,Trac diff highlighting start and end points,version control/changeset view,0.11-stable,enhancement,minor,,new,2010-03-25T12:05:57+01:00,2017-04-16T17:22:25+02:00,"At the moment the changeset diff viewer correctly shows the difference between two files, however the start and end points are oddly positioned. So for example this diff view shows the deletion starting at the closing ""}"" of the previous section: [[Image(diffoddstartend.png)]] In complex code this can be offputting, as it is really the second declaration that has simply been removed (including it's bracketed {…} section). However as the diff catches the closing }, you are then left with a closing } at the end of the diff. Whilst technically still a correct diff, other systems (such as fisheye) are able to show the diff in a more intuitive way, by highlighting the removed section instead, mockup example: [[Image(diffnewstartend.png)]] It would be a nice enhancement to change the diff viewer so it is more in line with removing the section rather than clipping the previous closing bracket.",c.jones@… next-major-releases Release,9168,milestone URL persistence,roadmap,0.10.5,enhancement,major,,new,2010-03-24T15:30:22+01:00,2017-03-17T01:06:42+01:00,"I've been struggling with Milestone URL persistence. It a maintenance nightmare changing milestone names. Our implementation of Trac oversees a large number of disparate university groups each with multiple milestones. Each group has their own wiki pages linking to their milestones as well as milestones of other related groups. I see ticket #364 mentioned this (jamesm about 6 years ago) >I realize its ugly, but I think that the ability to link to the milestone outweighs the benefits of a pretty-url. I would say go with a fixed integer (like tickets do) and let the user assign a name/descrption to the fixed integer. I think that the linking throughout Trac/outside Trac is more important. I agree with this statement, permanent ID's, just like tickets, elegantly solve this issue. Further, a milestone with a fixed ID having a title and description field would be highly useful. Macros (such as MilestoneDate) could pull the title into a wiki for display letting the milestone names change independently of milestone links without losing any descriptive benefits. I checked newer versions of trac hoping this was implemented, but I see no evidence * http://trac.edgewall.org/demo-0.12/milestone/Feed%20the%20dogs Link to a milestone I created but then changed the name * http://trac.edgewall.org/demo-0.12/milestone/My%20name%20WILL%20change * http://trac.edgewall.org/demo-0.12/milestone/My%20name%20changed ",john.williams@… next-major-releases Release,9149,Conf option for showing previous query instead of default query,ticket system,0.12dev,enhancement,minor,,new,2010-03-19T12:38:50+01:00,2010-06-23T17:53:27+02:00,"Trac stores query_href into session. It would be nice to have option for using that value as a default value. {{{ [query] default_query = ${session_query_href} }}}",anonymous next-major-releases Release,9141,'Retarget associated tickets' does not respect [milestone-groups] config,roadmap,0.11.6,defect,normal,,new,2010-03-16T08:44:03+01:00,2015-07-28T09:49:49+02:00,"Hi, first of all thanks for a great application. We recently completed a large milestone with 300 associated tickets. We use a custom [milestone-groups] section to accomodate QA after the ticket has been marked closed by the developer. {{{ [milestone-groups] active = * active.css_class = open active.label = in progress active.order = 1 closed = closed,verified closed.order = 0 closed.overall_completion = true closed.query_args = group=resolution }}} After closing the milestone, all tickets with verified status were moved to the next milestone, leaving us with ~20 tickets on the completed one. 'Retarget associated tickets' should respect the [milestone-groups] section.",jakob.gruber@… next-major-releases Release,9123,WikiFormatting in link text does not work,wiki system,0.12dev,enhancement,normal,,new,2010-03-09T10:42:09+01:00,2015-05-26T20:50:57+02:00,"Try to use a link which has a text that makes use of WikiFormatting. Like this [SandBox link to `SandBox`] or this [source:/ link to the ''root'' of the Trac sources]. I expect the link texts to be rendered in a way so that they contain “`SandBox`” and “''root''”, respectively, not “!`SandBox`” and “!''root!''”. It is impossible to use markup consistently in, for example, a ticket comment when you also like to link to files, other tickets, revisions, etc., and you use monospace highlighting for file names, or have some other convention or personal preference for using markup. I have successfully been able to reproduce this shortcoming in the sandbox in the [/demo-0.11/wiki/SandBox 0.11] and [/demo-0.12/wiki/SandBox 0.12] demo installations.",n-roeser@… next-major-releases Release,9091,Use RawConfigParser instead of ConfigParser in the Configuration component,general,0.12dev,enhancement,normal,,new,2010-02-25T20:45:15+01:00,2015-03-02T22:31:46+01:00,"The use of `ConfigParser` in the configuration layer causes issues when using `%(foo)s` values in the config file. The th:IrcLogsPlugin does this and causes a traceback when trying to write the sample config during upgrade. Trac itself doesn't use the interpolation feature of the `ConfigParser` nor do we advertize the interpolation feature of the `ConfigParser`. As the interpolation feature is of dubious value, I think that we should replace `ConfigParser` with `RawConfigParser`. This should also give us case sensitivity in the section fields.",John Hampton next-major-releases Release,9085,Notification system silently drops failures to unicode email addresses,notification,,enhancement,major,,new,2010-02-25T03:07:42+01:00,2016-10-07T15:52:53+02:00,"See #9082 for some background. Currently, email addresses with unicode in them are silently discarded when sending ticket updates. While not trying to deliver the email is correct, it's a little misleading. If the user enters a unicode email address in the `CC` list, then it appears to get added (indeed it does) but they will never receive any emails. Unfortunately, there is no indication as to why the email will never be delivered. An admin can look in the log and see the email addresses which have been discarded, but they first have to know what they are looking for. Additionally, the same thing occurs for addresses in the `smtp_always_cc` and `smtp_always_bcc` fields. We could show a warning indicating that the email wasn't delivered to the problem addresses. The problem with the `CC` list, is that the invalid address is put on the `CC` list, and so anytime an update occurred, the warning would be shown. An administrator would then need to clean out the `CC`list. The problem is even more complicated if a user has the invalid email address in their preferences and simply puts their username on the `CC` list (or as the owner/reporter/updater of a ticket). An admin would then need to clear out the email address in the user's session. I think the solution is to warn when an email can't be delivered, as well as stricter input validation. Input validation would need to be in: Preferences:: when entering email address `CC` list:: would need to differentiate between entering an email address and a username. This might be really hard when taking into account `use_short_addr` and `smtp_default_domain` `smtp_always_cc` and `smtp_always_bcc`:: Don't know how to properly validate these, or whether it's worth it. Populating these with invalid values should be readily apparent to the admin.",John Hampton next-major-releases Release,8967,Partial conversion of files when calling trac.mimeview.api.Mimeview.convert_content,rendering,0.11.5,defect,major,,new,2010-01-13T19:32:33+01:00,2010-03-12T21:26:24+01:00," In the aforementioned function there is a branch that is used to guess the content type of the input file. In order to do that `max_preview_size` bytes are read from the beginning of the file, and then the file is not used anymore. This means that only that first block is converted rather than the whole file. The exact location of the code I am talking about is [source:trunk/trac/mimeview/api.py@8899:648-652#L648 here] ",olemis+trac@… next-major-releases Release,8961,[PATCH] Linkable lines on diff view,wiki system,0.12dev,enhancement,minor,,new,2010-01-12T14:44:39+01:00,2015-12-19T16:38:37+01:00,Lines are not linkable in diff view. This patch modifies diff_div.html so that lines are linkable.,niko.sten@… next-major-releases Release,8952,trac-admin hotcopy should (be able to) copy contents of symbolic links,admin/console,0.11-stable,enhancement,normal,,new,2010-01-10T15:50:22+01:00,2010-12-14T18:22:42+01:00,"I have some symlinks in the my trac project folder and when I run {{{trac-admin myproject hotcopy}}} it copies the symlink. Is it possible to make a copy of the folder that includes the content of those symlinks? That way I don't need to worry about those when I make a backup.. The {{{htdocs}}} and {{{templates}}} folders in my environment are symlinks as well (to checkouts of svn repositories) that contain the templates and config files for my trac project, so they're essential to be part of the hotcopy. If you don't want to do it by default for some reason, is it possible to add an option that makes this possible? The test below shows the output of hotcopy contains a symbolic link to {{{folder -> ../folder}}}. {{{ #!sh thijs@home:~/Sites$ mkdir test thijs@home:~/Sites$ cd test/ thijs@home:~/Sites/test$ ls -l thijs@home:~/Sites/test$ trac-admin myproject initenv Creating a new Trac environment at /Users/thijs/Sites/test/myproject Trac will first ask a few questions about your environment in order to initialize and prepare the project database. [snip] thijs@home:~/Sites/test$ ls -l total 0 drwxr-xr-x 11 thijs staff 374 10 Jan 14:31 myproject thijs@home:~/Sites/test$ mkdir folder thijs@home:~/Sites/test$ cd myproject/ thijs@home:~/Sites/test/myproject$ ln -s ../folder thijs@home:~/Sites/test/myproject$ cd .. thijs@home:~/Sites/test$ trac-admin myproject hotcopy mycopy Hotcopying /Users/thijs/Sites/test/myproject to mycopy ... Hotcopy done. thijs@home:~/Sites/test$ ls -l mycopy total 24 -rw-r--r-- 1 thijs staff 98 10 Jan 14:31 README -rw-r--r-- 1 thijs staff 27 10 Jan 14:31 VERSION drwxr-xr-x 2 thijs staff 68 10 Jan 14:31 attachments drwxr-xr-x 4 thijs staff 136 10 Jan 14:31 conf drwxr-xr-x 3 thijs staff 102 10 Jan 14:35 db lrwxr-xr-x 1 thijs staff 9 10 Jan 14:35 folder -> ../folder drwxr-xr-x 2 thijs staff 68 10 Jan 14:31 htdocs drwxr-xr-x 2 thijs staff 68 10 Jan 14:31 log drwxr-xr-x 2 thijs staff 68 10 Jan 14:31 plugins drwxr-xr-x 3 thijs staff 102 10 Jan 14:31 templates }}}",Thijs Triemstra next-major-releases Release,8949,add sortable tables to wiki,wiki system,,enhancement,normal,,new,2010-01-10T00:37:19+01:00,2015-03-01T03:54:53+01:00,"See http://en.wikipedia.org/wiki/Help:Sorting for Wikipedia's approach to sortable tables. An approach that does not require JavaScript is preferred, even if it causes the page to reload, but any approach will do. Useful trac-hacks related to this ticket may include: http://trac-hacks.org/wiki/FlatTableMacro http://trac-hacks.org/wiki/MultiSortReportPatch http://trac-hacks.org/wiki/WikiTableMacro http://trac-hacks.org/wiki/SqlQueryMacro A work-around is also appreciated.",anonymous next-major-releases Release,8945,[PATCH] Send Notification Email with local timezone,notification,,enhancement,normal,,new,2010-01-08T12:52:42+01:00,2015-03-02T22:31:46+01:00,"Currently, Trac sends out notifications with an UTC Date header. I would like to see my server's local timezone (patch attached) or even better the users' configured timezone. Related: http://trac.edgewall.org/changeset/1512",hwyss@… next-major-releases Release,8735,wiki import should maintain page hierarchy,wiki system,none,enhancement,normal,,new,2009-10-12T17:46:55+02:00,2010-06-24T18:45:14+02:00,"When importing wiki pages via {{{ trac-admin ENV wiki import }}} it would be great if the directory structure in the import directory was maintained as the page hierarchy in the wiki. Alternatively, it could be useful if an import directory could be placed in some specific page in the wiki page hierarchy. ",anonymous next-major-releases Release,8729,add Latest Comment option to custom query,report system,none,enhancement,normal,,new,2009-10-09T06:14:55+02:00,2022-09-03T15:04:54+02:00,"it would be great if Latest Comment was offered as a column in TracQuery. or as an option under ""Show Under Each Result"", the way Description is offered. this would be a quick way to see updates on tickets rather than having to click each one to open it up and read. the entire Comments field is usually way too long. Description is not continually updated with the latest status. we tried adding our own [TracTicketsCustomFields custom field] for status notes but it was cumbersome to have to keep updating that in addition to the comments field. in addition to giving people a quick view of what's going on in tickets, with a little more care by commenters, it could also be used in status reports for clients or management.",jhn@… next-major-releases Release,8728,add Closed Date to list of columns offered in custom query,query system,,enhancement,normal,,new,2009-10-09T05:46:07+02:00,2021-01-27T12:28:36+01:00,"the Created and Modified columns are already offered by TracQuery. it would be very useful to also be able to add the Closed date to a report. note that the Closed date on a ticket is not the same as the Last Modified date on a closed ticket. there may be follow-up communications long after a ticket has already been closed. we often get requests from upper management or from clients for reports such as how many tickets were closed in a particular month. it is extremely tedious to search all closed tickets that were modified during or after that time, and then see when the ticket was actually closed. another use of the Closed date column would be to run metrics such as average length of time that a ticket was open. if this column was available, it would take a quick export to excel and some simple calculations. this data would be accessible with custom sql but i think there are enough common use cases that it would be worthwhile to make it part of the default package. i recall seeing many tickets related to custom date reports and this enhancement might provide enough for a lot of people. it may not be that simple since i see in TracReports that the time and changetime columns live on the ''ticket'' table, but not the closed time. but the TracTimeline does show when tickets are closed, so hopefully that means it wouldn't be too hard to get at the data.",jhn@… next-major-releases Release,8672,Filter ticket queries by user groups,ticket system,none,enhancement,normal,,new,2009-09-15T15:20:27+02:00,2015-03-30T18:37:03+02:00,"We had a need to filter the results of a TicketQuery by user groups (from the permission system) to keep query results from becoming overwhelming. The attached patch does this by expanding a 'username' like '@software' in the owner field of a query to all the users with that role. This is a very rough implementation for now (the group functionality should be provided by the permissions system and there should be a way to list groups on the Custom Query page) but I'm putting it up anyway in the hope that the idea is useful. We use this like {{{ [[TicketQuery(status=blocked,order=owner,owner=@software)]] }}} ",joshua.hoke@… next-major-releases Release,8670,Ticket owner list should not depend on TICKET_MODIFY permission,ticket system,none,enhancement,normal,,new,2009-09-15T00:11:24+02:00,2012-02-10T22:50:27+01:00,"It's nice to be able to use `restrict_owner` in the `[ticket]` section of trac.ini to yield a select box (dropdown) for the ""Assign to"" field. But the list contains everyone who has a profile and has TICKET_MODIFY privileges. It would be very, very nice to have a restrict_group option after it, which would allow me to define a group with a given privilege constellation (e.g. developers), and then only have their names appear as candidates.",fewayne@… next-major-releases Release,8640,Enable headline numbering,wiki system,0.11.5,enhancement,normal,,new,2009-09-06T10:52:36+02:00,2016-04-08T21:00:48+02:00,"For larger wiki pages with a lot sections it's often difficult to differentiate the sections ""level"" (i.e. h1 - h6). It would be helpful if headline numbering could be enabled, e.g.: == 1. First

    headline == === 1.1. First

    headline === === 1.2. Second

    headline === == 2. Second

    headline == === 2.1. First

    headline === ==== 2.1.1. First

    headline ==== === 2.2. Second

    headline === == 3. Third

    headline == This could be implemented as a user preference so that each user could decide whether he/she wants headline numbering or not. Also note that

    headlines (i.e. `=`) should not be numbered as they most likely only appear once on each page.",Sebastian Krysmanski next-major-releases Release,8637,"""Add to CC"" for anonymous with TICKET_APPEND permissions",ticket system,0.11.5,enhancement,major,,new,2009-09-04T16:25:46+02:00,2014-07-31T15:41:40+02:00,"Currently when `always_notify_updater` is set to `false` an commenter/updater has to put himself/herself to the CC list of a ticket to be notified about future changes. However, to do so the user is required to have `TICKET_MODIFY` permissions. If an administrator only want the users to be able to comment on tickets (i.e. only grants `TICKET_APPEND` permissions) the user has no way of being notified about future changes. So it would be nice to activate this checkbox even when the user only has `TICKET_APPEND` permissions.",Sebastian Krysmanski next-major-releases Release,8607,Option to skip notification to subscribers when a ticket is updated,notification,0.11.5,enhancement,normal,,new,2009-08-27T05:15:50+02:00,2015-11-05T13:20:33+01:00,"Consider the following scenarios for editing a ticket: 1. Changing a minor typo in the ticket '''Description''' 1. Adding some entries to the '''Keywords''' field 1. Changing the '''Milestone''', '''Component''', '''Type''', etc ... 1. Adding an attachment, when the attachment was mentioned in the previous message, therefore all users can expect it will be posted to the ticket within minutes and don't need to see another email telling them an attachment was added. Users on my team receive a lot of email from Trac each day, and it is frustrating and Trac becomes less effective when they receive email notifications when someone is just cleaning up or making some minor changes to a ticket. Rather than setting up a lot of options in ''trac.ini'' that determine who and when a subscriber is notified of changes, I propose that a checkbox is added to the right of '''Submit Changes'''. When the '''Don't Notify Subscribers''' checkbox is selected, notifications emails are not sent on a ticket update. [[Image(SubmitChangesButton.png)]] The question then arises, what permission should be required for this checkbox to be present. Perhaps someone running a Trac instance does not want users to be able to change a ticket without, for instance, the ticket's owner receiving a notification. I propose two options for the [notification] section of trac.ini. 1. `optional_notify_authenticated = {true, false}` 1. `optional_notify_anonymous = {true, false}` Option 1 would show/hide the '''Don't Notify Subscribers''' checkbox for authenticated users. Option 2 would show/hide the '''Don't Notify Subscribers''' checkbox for anonymous users. Users with `TICKET_ADMIN` permission would always see the checkbox. ",Ryan J Ollos next-major-releases Release,8605,Allow non-authenticated users to replace their attachments,attachment,0.12dev,enhancement,normal,,new,2009-08-27T00:42:28+02:00,2021-08-27T13:35:43+02:00,"#8592 fixed replacing attachments for authenticated users. Non-authenticated users are currently denied replacing any attachments, unless they have `ATTACHMENT_DELETE` permission. The goal of this enhancement is to allow non-authenticated users to replace their own attachments, without having `ATTACHMENT_DELETE` permission. The main use case is to allow correcting a bad upload shortly after the fact. [comment:9:ticket:8592 This comment] gives a possible solution, by structuring the `author` field as `anonymous:` for unauthenticated users. Removing one's own attachments could be enabled in the same way.",Remy Blank next-major-releases Release,8559,New Report: Tickets of Interest,report system,0.11.4,enhancement,normal,,new,2009-08-10T23:01:18+02:00,2015-03-02T22:27:22+01:00,"When one visits a Trac system as a user of a product and not as a developer of the product, one often wants to search for those tickets of personal interest. Such tickets would be defined as those tickets where the user is the Reporter '''OR''' in the CC (or maybe additionally as a commenter). * It would be nice to see such a report added here. * Additionally, it would also be nice to see this added as one of the default reports thus would likely to be found on other Trac installations. * This report is easy to write as an SQL based report. However, I do not see how one could implement this as a Custom Query report. Thus it is currently not possible to add such a report when the SQL report system is fully deprecated. Also it is not possible for a user to implement this as a custom query that they could save as a book mark.",mpotter@… next-major-releases Release,8541,[patch] recursive auto-expansion of folders containing only one sub-directory,version control/browser,0.12dev,enhancement,trivial,,new,2009-08-02T15:32:30+02:00,2015-12-19T16:36:40+01:00,"... like Vista's explorer does, for example. The Mercurial repository browser also does something similar, it shows all those nested directories on the same line. It's quite handy for Java packages, where the actual files are often nested in a chain of single entry directories... Attaching the patch here, in case other people find it useful. ",Christian Boos next-major-releases Release,8526,SubWiki link & SubMapTxt page,wiki system,0.12dev,enhancement,normal,,new,2009-07-28T01:34:52+02:00,2010-06-24T18:25:49+02:00,"There are several nice features added to support SubWiki in Trac to categorize simliar wiki pages in hierarchical tree structure however linking to SubWiki from an arbitary Trac page can be really tedious if the level of depth become more than one. If I have a SubWiki page Page/SubPage/SubSubPage/SubSubSubPage and it should be explicitly specified using full path to the SubWiki page. {{{ [wiki:Page/SubPage/SubSubPage/SubSubSubPage] or [wiki:Page/SubPage/SubSubPage/SubSubSubPage ""This is SubSubSubPage""] }}} I can use InterWiki and InterMapTxt to create shortcuts to SubWiki pages under Page/SubPage/SubSubPage/. {{{ Define an entry in InterMapTxt === SubSubPage http://localhost/trac/wiki/Page/SubPage/SubSubPage/$1 # Categorized under SubSubPage }}} and then use it as shown below {{{ SubSubPage/SubSubSubPage or [SubSubPage/SubSubSubPage ""This is SubSubSubPage""] }}} It worked almost but a problem is that it does not create internal Trac database update (as it is treated as external link) which means Trac does not know the relationship between pages. i.e. [th:BackLinksMacro BackLinksMacro] can not display the back link to SubWiki page if InterWiki is used. SubWiki macro support simliar to InterWiki using customizable SubMapTxt (like InterMapTxt) would be of great use for those who using SubWiki intensively to categorize simliar wiki pages. This macro can possibly add /SubSubPage handler which can be utilzed in maintab or navtab to create direct access to categorized SubWiki pages list from Trac menu. (Simliar to [th:HierWikiPlugin#SubWiki HierWikiPlugin#SubWiki] mcaro) I created a Requset-a-Hack ticket [th:#5569] in Trak-Hacks site as well.",Jae next-major-releases Release,8519,internal SQL string matches are always case insensitive,database backend,none,defect,normal,,new,2009-07-27T14:56:05+02:00,2021-03-07T11:53:11+01:00,"Our `like()` helper actually translate into doing an `ILIKE` or similar. There are several places where this is not desired, we would actually need a case sensitive search. The idea is to rename the current `like()` to `ilike()` and add a real `like()` helper. See #8424 for the original ticket and implementation hints for SQLite which doesn't have a case-sensitive LIKE.",Christian Boos next-major-releases Release,8507,Developer WSGI Profiling and Debugging,general,none,enhancement,normal,,new,2009-07-23T19:34:24+02:00,2015-05-26T20:50:57+02:00,"I have a wsgi setup for development that includes the ability to use a couple different methods of profiling as well as debugging using DBGP (a debugger protocol for dynamic langauges, used in Komodo, Eclipse DLTK, and a bunch of other editors/ides). I hope this makes it easy for others who are interested in profiling and/or debugging trac. A couple small changes would make it useful for any wsgi application. ",Shane Caraveo next-major-releases Release,8482,Proposal: Plugin update notification,web frontend,0.11.1,enhancement,trivial,,new,2009-07-15T08:09:49+02:00,2010-05-04T00:30:24+02:00,"I'm using a trac (0.11.1) server in a project environment with a couple of plugins. Some of them work fine, some of them still have bugs and some of them had to be patched to get them working properly. However, most of them are hardly under development. So I thought that it would be nice to have a 'check for update' function for each or all of the plugins. Wouldn't it make sense to add such a feature to future Trac versions which compares the version or changeset number of the used plugin against it's source repo? (Maybe also with an automatic function from the web frontend)",the3illings@… next-major-releases Release,8426,Unable to print any of the admin pages,admin/web,0.11.4,defect,normal,,new,2009-06-29T20:09:25+02:00,2013-10-17T18:54:32+02:00,"From any of the admin pages (e.g. [/admin/general/perm], [/admin/accounts/users], [/admin/ticket/components]) am unable to generate a useful printout. When one does so, I get the page title and Administration navigate box, but no useful information that was on the page. I assume the is a CSS issue.",mpotter@… next-major-releases Release,8403,Encourage users to select meaningful values for 'required' fields,ticket system,none,enhancement,normal,,new,2009-06-19T23:14:06+02:00,2009-06-26T15:57:26+02:00,"When an enum field, in particular, is marked as 'required', the blank option is removed from the select element, and the first option is selected by default, if no other default value is set. The problem with this is that it often leads to users submitting tickets with those fields left in their default values. It doesn't force them to select a meaningful value. A decent workaround is to have a catchall (like 'general' for the Component field) as the default value for the field. But I've had use cases where we want to make sure the user has paid some attention to the field. The desired behavior here would be to have something like 'Select a Component...' as the first option. If this option is selected when the form is submitted, a warning would be displayed, and the ticket would not be saved.",ebray next-major-releases Release,8387,Need method to view source code of RST files,rendering,0.11.4,enhancement,normal,,new,2009-06-15T22:32:34+02:00,2012-02-10T22:52:02+01:00,"As discussed in WikiRestructuredText, files in the repository are parsed using RST if the Subversion property `svn:mime-type` is set to `text/x-rst` (for example - source:trunk/INSTALL). This is a very useful feature in general. However, on occasions I would like to be able to view the RST file as a text file complete with line numbers. For example, this would allow me to highlight a set of lines in the file. Links are already provided at the bottom of the page for ""Original Format"" and ""Plain Text"". Could an additional link be provided for ""Source""? This would also apply to files containing wiki markup which have had `svn:mime-type` set appropriately (can't remember what you have to set this to and I can't find it in the documentation :-( ).",David Matthews next-major-releases Release,8380,Search In Report / Search In Query functionality,search system,0.11.4,enhancement,normal,,new,2009-06-13T00:36:50+02:00,2009-06-26T17:42:55+02:00,"It would be nice to be able to use the search function, restricted to the tickets returned in a report or query.",Denis de Bernardy next-major-releases Release,8371,need to view ticket-list of only tickets with unanswered updates,report system,0.11.1,enhancement,normal,,new,2009-06-10T17:35:07+02:00,2012-02-10T22:52:45+01:00,"Trac is so good that one's database grows rapidly. I need to be able to view a filtered ticket-list, showing only those tickets which: - were created by me - the last update was by someone else, not me ie tickets which have queries I need to answer. We can't see how to do this. Tony",carrt7@… next-major-releases Release,8361,cannot find file trac-digest.py,general,0.11,defect,normal,,new,2009-06-08T13:15:42+02:00,2019-09-06T19:51:32+02:00,"I've installed trac 0.11 with python 2.5.4 on windows 2008 server. i've followed all the instructions in method2 (TracOnWindows) with no errors. however i'm not able to do the following 1. cannot setup admin, {{{#!dos python trac-digest.py -u adm -p adm >> c:\digest.txt }}} gives the error - cannot fild trac-digest.py please suggest how i can rectify this issue. 2. even when i setup a project environment is gives the error - please see the attached screen shot. Please suggest how i can get Trac up and running ",shruti.lalwani@… next-major-releases Release,8335,Create People page with user profiles and status of who's doing what.,general,none,enhancement,normal,,new,2009-05-31T13:46:33+02:00,2015-05-26T20:50:57+02:00,"As you may see, the idea is fairly simple -- let the first page be a page with user profiles that reflects a project's team structure. Let each profile contain user name, avatar, role in a team (dev, qa, etc), status (idle, working, blocked, etc), and the last few messages/events from the timeline made by user.",Alex Radzie next-major-releases Release,8328,Repeat SCRIPT_NAME fixup using SCRIPT_URL after environment selection,web frontend,0.11-stable,enhancement,normal,,new,2009-05-27T23:02:14+02:00,2015-05-26T20:50:57+02:00,"Ticket #2299 introduced a very useful feature to guess the SCRIPT_NAME from the SCRIPT_URL. However, this works less than optimal in some scenarios using an env_parent_dir. '''Example:''' suppose you want to serve multiple project home pages using multiple virtual hosts from apache, and suppose further that, in order to save memory e.g. on a virtual server, you want to serve all Trac requests from a single FastCGI process. You can configure the process to serve from a parent of all trac environments, and add a line like this to the individual vhosts: {{{ ScriptAlias /trac/ /path/to/trac.fcgi/my.environment.name/ }}} This works well enough, except for the links. The link detection checks that the SCRIPT_URL (e.g. `/trac/wiki/PageName`) ends with the PATH_INFO (e.g. `/my.environment.name/wiki/PageName`) - which is not the case. Subsequently, the first part of the PATH_INFO is used to choose the environment, and moved from PATH_INFO to SCRIPT_NAME, causing a wrong script name of `/trac/my.environment.name`. This can easily be solved by doing the same fixup ''after'' the environment has been chosen. In that case, PATH_INFO only contains `/wiki/PageName` which allows correct detection of `/trac` as the SCRIPT_NAME for that environment, not all environments. In order to produce correct link in the listing of all available projects (in setups providing access to this list, not in a vhost like the one above), the fixup before should probably be kept in place as well, or be moved to send_project_index. Because it has to be executed from two places, it might make sense to factor it into a separate function. A patch for this is easy. If you want me to write one, I'll do so. For now, I've simply moved the whole block dealing with script_url to after the env_path selection, as I don't need the project listing. So the concept is proven.",Martin.vGagern@… next-major-releases Release,8219,[PATCH] Forward additional parameters from /report to /query,report system,0.11-stable,enhancement,normal,,new,2009-04-20T11:25:10+02:00,2015-12-19T16:32:13+01:00,"Calling an URL like this {{{ http://myserver.example/trac/report/10 }}} redirects to {{{ http://myserver.example/trac/query?col=x&status=new&...&report=10 }}} Now, it would be nice, if additional parameters could be passed to the (short) report URL and not be forgotten while redirecting to the query, e.g. {{{ http://myserver.example/trac/report/10?format=csv&col=description }}} would redirect to {{{ http://myserver.example/trac/query?col=x&status=new&...&col=description&report=10&format=csv }}} Ideally, these parameters would be given precedence over those from the report definition (where applicable - i.e. `col` parameters would simply be added, like in the example).",thomas.moschny@… next-major-releases Release,8165,[PATCH] csv download doesn't process 'description' field,report system,0.12dev,defect,minor,,new,2009-03-27T12:33:46+01:00,2015-03-02T22:27:22+01:00,"In a report that includes the ticket 'description' field, the text is processed by the wiki processor, but when I choose to download the report by clicking on 'comma-delimited-text', the 'description' field in the csv file is *not* processed. E.g. I have a macro which displays its output correctly in the onscreen report, but just shows ""![[MyMacro()]]"" in the csv file.",pellatt@… next-major-releases Release,8146,Unified diff in changeset view differs from exported unified diff,version control/changeset view,0.12dev,defect,normal,,new,2009-03-21T20:59:58+01:00,2021-01-11T14:45:35+01:00,"In the changeset view the unified diff provided by clicking on Unified is different from the one which can be downloaded. View when selecting **Unified** in the changeset view: {{{ Index: trunk/setup.py =================================================================== --- a/trunk/setup.py +++ b/trunk/setup.py @@ -2,5 +2,5 @@ # -*- coding: utf-8 -*- # -# Copyright (C) 2003-2008 Edgewall Software +# Copyright (C) 2003-2009 Edgewall Software # All rights reserved. # @@ -71,5 +71,5 @@ install_requires = [ 'setuptools>=0.6b1', - 'Genshi>=0.5' + 'Genshi>=0.6dev-r960' ], extras_require = { }}} Content downloaded as a //Unified Diff//: {{{ Index: trunk/setup.py =================================================================== --- trunk/setup.py (revision 7699) +++ trunk/setup.py (revision 7890) @@ -2,5 +2,5 @@ # -*- coding: utf-8 -*- # -# Copyright (C) 2003-2008 Edgewall Software +# Copyright (C) 2003-2009 Edgewall Software # All rights reserved. # @@ -71,5 +71,5 @@ install_requires = [ 'setuptools>=0.6b1', - 'Genshi>=0.5' + 'Genshi>=0.6dev-r960' ], extras_require = { }}} here the diff of the two diffs : {{{ #!diff --- C:\Users\Absynthe\Downloads\diff-trunk_trac_versioncontrol_web_ui_changeset.py-from-r7626-to-r7890.diff sam. mars 21 12:38:50 2009 +++ C:\Users\Absynthe\Downloads\diff-trunk_setup.py-from-r7699-to-r7890.diff sam. mars 21 12:37:11 2009 @@ -1,7 +1,7 @@ Index: trunk/setup.py =================================================================== ---- a/trunk/setup.py -+++ b/trunk/setup.py +--- trunk/setup.py (revision 7699) ++++ trunk/setup.py (revision 7890) @@ -2,5 +2,5 @@ # -*- coding: utf-8 -*- # }}} As you can see paths in the index of the first diff are a bit custom.",anonymous next-major-releases Release,8094,Preview who will be notified,notification,none,defect,normal,,new,2009-02-28T21:12:42+01:00,2018-06-10T10:58:33+02:00,"Dear Trac people, as you see in http://core.trac.wordpress.org/ticket/9208 one never knows if Trac will send email or not. Note, I am talking about bugs one is the reporter of. I am not an administrator. To be ''doubly'' sure to get mail, one must always click the CC box. Otherwise, well, most of the time it will send email about others' changes to the bug, but sometimes not. P.S., the corresponding Bugzilla checkbox looks like ""Add jidanni@jidanni.org to CC list"" Which never needs to be checked for the reporter to get mail. So my question is should I add my CC to the rest of http://core.trac.wordpress.org/query?reporter=jidanni&col=id&col=cc&order=id ? I can only do a few a day before my wrists get tired.",jidanni@… next-major-releases Release,7990,Can't view plain text wiki links,wiki system,0.11.2.1,enhancement,normal,,new,2009-01-20T21:27:14+01:00,2015-05-26T20:50:57+02:00,"You can no longer view the plain text version of a wiki page in your browser by selecting the plain text link - instead you are forced to download. I understand the security concerns but, having read #7603, I expected to be able to enable the old behaviour by setting {{{ [wiki] render_unsafe_content = yes }}} However, this doesn't work. Is it meant to? Admittedly the description of this option in TracIni#wiki-section doesn't say this will work. However, wouldn't it make sense to make this consistent with the same option from the `browser` and `attachment` sections?",David Matthews next-major-releases Release,7929,make links clickable in annotate view,version control/browser,none,enhancement,minor,,new,2009-01-05T03:13:19+01:00,2009-02-24T15:52:53+01:00,"In the 'annotate' view, the first column contains for each source row a hyperlink to an associated changeset. The problem is that if javascript is enabled, the 'href' attribute gets removed, destroying the link. I would like to be able to open the linked page in a new browser tab. But that's currently not possible, because... All mouse clicks are captured and used to trigger ajax-ish retrieval of the linked page and display it in an overlay, which finally contains a clickable link. Also, each associated row gets highlighted blue as a side effect, which is cpu-intensive. This makes it somewhat annoying. A solution was proposed to me by 'coderanger_' on IRC. instead of removing the href, we should capture and reject clicks on the (the click handler is on the table cell itself, not the ) Hopefully the handler would only catch LMB clicks, leaving MMB ('open in new tab') or at least RMB ('context menu') functionality intact. ",theultramage@… next-major-releases Release,7898,[PATCH] Keep NavigationContributor Ordering,general,0.11-stable,enhancement,normal,,new,2008-12-18T21:27:40+01:00,2010-11-26T15:26:26+01:00,"I have a small patch which applies against both 0.11-stable (r7236) branch and the trunk (r7768). It updates the use of INavigationContributors, so if one contributor returns more then one navigation item the order they are returned in is retained when they reach the main/meta navigation bar. This is of course over-ridden by the *_order settings in the configuration file. The justification for this patch is that some plugins may wish to dynamicly generate navigation items which are perhaps based on user supplied wiki content. The order of the navigation items could be significant, so its bad to throw it away. Since the items are generated dynamically it makes adding the order to the trac.ini file impractical. The patch includes a unit test, and an update the INavigationContributor interface documentation. I also think ticket #3695 has a little bit of bearing on this.",anonymous next-major-releases Release,7884,Ticket should contain a field for the expected time to complete/fix it,ticket system,none,enhancement,normal,,new,2008-12-12T18:26:14+01:00,2017-05-08T11:14:10+02:00,"This would give a number of benefits, and among them: * a (super)user that has the rights to change the priority of the tickets can rearrange them based on the time to handle them * a milestone could show an expected time to be completed as the sum of the time of each ticket",dusty@… next-major-releases Release,7872,"[Patch] Styling for ""Powered by"" footer element",general,none,enhancement,minor,,new,2008-12-08T14:42:18+01:00,2015-12-15T10:39:24+01:00,"This patch for Trac design alters theme.html and CSS to allow plugins add new ""Powered by"" blocks into footer with a simple jQuery. {{{ $('

    Powered by TracHacks

    ').appendTo(""#footer""); }}} ",techtonik next-major-releases Release,7867,"h,l keyboard shortcuts",version control/browser,none,enhancement,normal,,new,2008-12-05T14:41:51+01:00,2015-03-02T22:27:22+01:00,"A request to add h and l to move up and down (or better say left and right) the directory tree structure expanding/collapsing leaves as needed. There is also a bug in j,k shortcuts that they stop work after is pressed.",techtonik next-major-releases Release,7850,[patch] Update default_ options in trac.ini when enum value is updated/deleted,ticket system,devel,enhancement,normal,,new,2008-12-01T18:49:08+01:00,2023-04-19T18:08:14+02:00,"I had a user complained that he deleted a component, but that the component was still showing up as the default when creating new tickets. I explained that this is the expected behavior, since the default_component is still set to that component, causing it to be displayed. But he pointed it that it's somewhat user-hostile to keep a default_component setting for a component that no longer exists, and I'm inclined to agree. Likewise the default_ setting should be updated when an enum value is updated.",ebray next-major-releases Release,7793,[PATCH] Mail notifications for a ticket with multiple owners / reporters,notification,0.12dev,enhancement,normal,,new,2008-11-06T19:51:00+01:00,2018-06-10T16:42:54+02:00,"When a new ticket is created in a component with multiple owners (i.e., the component owner field contains ""user1,user2""), currently no owner gets a notification, because this field is not parsed and thus no corresponding e-mail address can be determined. I find it useful to have a component have multiple owners that are each equally responsible or interested in new tickets. A small diff to trac/ticket/notification.py which fixes the issue for me is included. === Note Trac doesn't currently support multiple owners, therefore this feature depends on #2662.",jsch@… next-major-releases Release,7747,timeline rss-feeds should be able to show diffs of wiki changes,timeline,0.10.4,enhancement,normal,,new,2008-10-22T21:04:36+02:00,2014-08-24T12:39:38+02:00,"I use rss-feeds to get the latest news about wiki (and other) changes of my project via Thunderbird. Viewing the rss message of a wiki change I only see the whole ''actual'' state of that wiki page, which may be even overturned. I’d prefer to see the diffs made on that wiki page directly without having to click through to it (and that even on an other program (Firefox)). Therefore I suggest to add a rss_wiki_show_diff (or so) option to the [timeline] section of the trac.ini; or to change that behavior at the whole. The page the feed shall link to shall be http://wiki.neo-layout.org/wiki/“wiki_page”?action=diff&version=“new_version” (be careful! new_version doesn’t need to mean the up-to-date version, but the one meant by this special change.)",m.p.roppelt@… next-major-releases Release,7699,TracQuery documentation on Query Language operators is incorrect,general,0.11-stable,defect,normal,,new,2008-10-07T21:17:23+02:00,2015-05-26T20:50:57+02:00,"The documentation at [wiki:TracQuery#QueryLanguage] regarding the operators is incorrect. It says: {{{ The available operators are: || '''`=`''' || the field content exactly matches the one of the values || || '''`~=`''' || the field content contains one or more of the values || || '''`^=`''' || the field content starts with one of the values || || '''`$=`''' || the field content ends with one of the values || All of these operators can also be negated: || '''`!=`''' || the field content matches none of the values || || '''`!~=`''' || the field content does not contain any of the values || || '''`!^=`''' || the field content does not start with any of the values || || '''`!$=`''' || the field content does not end with any of the values || }}} However, these operators don't seem to work in a query. In fact, when selecting their equivalent from the custom query dropdown it seems the generated operators are in fact the other way around, with a leading `=` instead of a trailing one. It seems it really should be: {{{ The available operators are: || '''`=`''' || the field content exactly matches the one of the values || || '''`=~`''' || the field content contains one or more of the values || || '''`=^`''' || the field content starts with one of the values || || '''`=$`''' || the field content ends with one of the values || All of these operators can also be negated: || '''`=!`''' || the field content matches none of the values || || '''`=!~`''' || the field content does not contain any of the values || || '''`=!^`''' || the field content does not start with any of the values || || '''`=!$`''' || the field content does not end with any of the values || }}}",kamil@… next-major-releases Release,7690,Add doctype supporting mathml,rendering,,enhancement,normal,,new,2008-10-03T13:37:28+02:00,2015-07-15T18:26:08+02:00,"I've been trying to add support for mathml in the Trac wiki. We work on the development of algorithms for imaging and the inclusion of equations into tickets and the wiki would be extremely useful. I think that we can do this by using the commands to embed html into the wiki markup then cutting and pasting mathml from an editor. The issue is that genshi doesnt support a suitable doctype and there is also no way of setting the doctype on a per project basis Ideally, I would like the following definition to be available so that math code can be pasted into the wiki {{{#!xml ]> }}}",Tim Kerby next-major-releases Release,7680,Broken Error Reporting?,rendering,0.11.1,defect,normal,,new,2008-09-27T21:05:30+02:00,2012-02-10T22:55:04+01:00,"The following comes up as the entire literal contents of the page I'm viewing. It seems to me to be broken in several ways: 1. It's not valid HTML 2. Trac's nice built-in traceback display is bypassed. 3. It appears to be generating an error while reporting an error. In this case the original error is lost and the system tells the user almost nothing she needs to know in order to debug the problem. Is that really the intended behavior? {{{ #!python Traceback (most recent call last): File ""/usr/lib/python2.5/site-packages/Trac-0.11.1-py2.5.egg/trac/web/api.py"", line 339, in send_error 'text/html') File ""/usr/lib/python2.5/site-packages/Trac-0.11.1-py2.5.egg/trac/web/chrome.py"", line 702, in render_template if not req.session or not int(req.session.get('accesskeys', 0)): File ""/usr/lib/python2.5/site-packages/Trac-0.11.1-py2.5.egg/trac/web/api.py"", line 169, in __getattr__ value = self.callbacks[name](self) File ""/usr/lib/python2.5/site-packages/Trac-0.11.1-py2.5.egg/trac/web/main.py"", line 257, in _get_session return Session(self.env, req) File ""/usr/lib/python2.5/site-packages/Trac-0.11.1-py2.5.egg/trac/web/session.py"", line 150, in __init__ self.get_session(req.authname, authenticated=True) File ""/usr/lib/python2.5/site-packages/Trac-0.11.1-py2.5.egg/trac/web/session.py"", line 164, in get_session super(Session, self).get_session(sid, authenticated) File ""/usr/lib/python2.5/site-packages/Trac-0.11.1-py2.5.egg/trac/web/session.py"", line 56, in get_session (sid, int(authenticated))) File ""/usr/lib/python2.5/site-packages/Trac-0.11.1-py2.5.egg/trac/db/util.py"", line 50, in execute return self.cursor.execute(sql_escape_percent(sql), args) File ""/usr/lib/python2.5/site-packages/Trac-0.11.1-py2.5.egg/trac/db/util.py"", line 50, in execute return self.cursor.execute(sql_escape_percent(sql), args) ProgrammingError: current transaction is aborted, commands ignored until end of transaction block }}}",dave@… next-major-releases Release,7660,Be able to search for commenters by name,ticket system,0.11.1,enhancement,normal,,new,2008-09-23T00:57:03+02:00,2009-10-26T15:35:12+01:00,"Somewhat related/started by http://trac.edgewall.org/ticket/7581 On the Pidgin instance of Trac I would like to be able to search (via the query module) for tickets that have been commented by myself...so search for text such as: ""...ago by bernmeister..."" Or really, just search for ""bernmeister"". At the moment there are a lot of Pidgin tickets I've touched and I want to search across all tickets to tell me which ones I've touched that are still NEW (or PENDING or CLOSED or ...). BTW, Pidgin's Trac instance does NOT have ticket reports enabled. What say you (please)?",anonymous next-major-releases Release,7608,Private ticket permission users can get ticket counts that include tickets they're not allowed to view,report system,0.11.1,defect,normal,,new,2008-09-08T18:19:21+02:00,2014-05-24T19:30:05+02:00,"A user with private tickets permissions can still query how many tickets meet criteria even if they can't see the tickets listed or view them. For instance they can type in query?status=!closed&priority=critical to get a count of how many critical defects are open. I originally wrote this ticket on the !PrivateTicketsPlugin ([http://trac-hacks.org/ticket/3674 #3674]) but heard that it's a problem in Trac core.",jevans next-major-releases Release,7573,base_url for many projects is painful,general,0.12dev,enhancement,normal,,new,2008-08-28T22:35:21+02:00,2016-09-08T15:48:46+02:00,"Setting the base_url in hundreds of ini files, and changing them with a hostname change is painful. I've attached a patch for a parent_base_url trac.ini setting that can be used in a parent trac.ini. ",doki_pen next-major-releases Release,7419,Custom ticket status: customised ordering of statuses,ticket system,0.11-stable,enhancement,normal,,new,2008-07-09T17:52:49+02:00,2019-12-03T03:44:58+01:00,"We have a large number of custom ticket statuses now, thanks to 0.11 - it would be nice if I could define a sort order in the config file (similar to how you can do so in a [TracIni#milestone-groups-section ""[milestone-groups]""] section) so that when ordering tickets by status, they appear in my defined order.",trac+jon@… next-major-releases Release,7371,Use authentication variables populated by Apache,web frontend,,enhancement,normal,,new,2008-06-24T02:57:54+02:00,2015-12-17T23:17:47+01:00,"Documentation here: http://httpd.apache.org/docs/2.2/mod/mod_authnz_ldap.html#exposed If the display name and email are part of the LDAP query, the AUTHENTICATE_DISPLAYNAME and AUTHENTICATE_MAIL environment variables will be set. Presumably other Apache auth modules may also set these variables. The attached patch will check for the existence of the variables and set the corresponding profile data. ",john.barstow@… next-major-releases Release,7196,restrict ticket properties to add/edit only/...,ticket system,,enhancement,normal,,new,2008-05-05T10:16:39+02:00,2015-05-26T20:50:57+02:00,"would be nice to have the ability to define if a property should be readonly/disabled/not mentioned at all, if in ticket add-mode/edit-mode... so e.g. the milestone field could be set to edit-mode, so the creator is not able to choose the value for it... this is the intended behavior on t.e.o. also the priorty field should be set by a developer, so maybe also a restriction to some usertypes would be great.. such restriction would also make sense for other fields and also in other scenarios. this would improve the user-experience and help also the developer team, because of less misfilled tickets",Markus.Staab next-major-releases Release,7176,Workflow actions indistinguishable from user changes,ticket system,devel,defect,normal,,new,2008-04-30T03:17:34+02:00,2015-05-26T20:50:57+02:00," - someone writes a workflow action that modifies a user-modifiable ticket field - user selects that action, and previews the change - user selects a different action - user previews or submits changes The changes made by the workflow action will still be present; they will not be reverted. Internally, the code does not differentiate between user changes and workflow changes across a preview. The correct behavior here, particularly in cases where the user modifies the same fields as the workflow, is open for debate. ",Eli Carter next-major-releases Release,7150,"""contrib/workflow/showworkflow"" is KDE centric",general,devel,defect,trivial,,new,2008-04-23T00:25:33+02:00,2015-05-26T20:50:57+02:00,"I reckon this should call evince where kpdf is not available (like on Gnome desktops, which is a pretty common occurance). Here's a patch: {{{ Index: trac/contrib/workflow/showworkflow =================================================================== --- trac/contrib/workflow/showworkflow (revision 6880) +++ trac/contrib/workflow/showworkflow (working copy) @@ -24,4 +24,8 @@ dot -T ps -o ""$ps"" ""$dot"" && ps2pdf ""$ps"" ""$pdf"" || exit 1 dot -T png -o ""$png"" ""$dot"" -kpdf ""$pdf"" +if [ -n ""`which kpdf 2>/dev/null`"" ] ; then + kpdf ""$pdf"" +else + evince ""$pdf"" +fi }}} ",trac@… next-major-releases Release,7106,File attachment 'added' time is incorrect,attachment,devel,defect,major,,new,2008-04-11T17:30:24+02:00,2020-04-30T09:56:23+02:00,"Found in r6814. The added time shown in the ticket is correct: ''MTM1.log (1.6 kB) - added by admin '''2 days''' ago.'' but when you show the file, the added time is incorrect: ''File MTM1.log, 1.6 kB (added by admin, '''84 seconds''' ago)'' The tool-tip timestamp is the same (and correct) on both.",robert.nadler@… next-major-releases Release,7073,notify dup subscribers,notification,,enhancement,normal,,new,2008-04-03T15:35:05+02:00,2015-05-26T20:50:57+02:00,"if a set of users is on a ticket and that ticket is marked as dup of another, notifications should now be done to both tickets subscribers.",tracbugs@… next-major-releases Release,7069,'SearchResults doesn't indicate closed by etc..,search system,,enhancement,minor,,new,2008-04-02T18:15:36+02:00,2020-04-15T10:00:15+02:00,"would be nice to have the same information about a ticket and it's current state in the SearchResult view, as in the timeline. In the timeline you can also see the user who has closed a ticket,.. and in the SearchResult not. Maybe it would also be interessting who was the last user who has edited the ticket and also when he did it..",Markus.Staab next-major-releases Release,7055,"Add ""toggle line numbers"" JavaScript to source browser",rendering,devel,enhancement,normal,,new,2008-03-27T21:26:03+01:00,2019-09-06T19:51:32+02:00,"At the moment when you copy-n-paste code snippets from Trac's source browser (I often do) it also copies the line numbers, which is almost never what you want. To get code without the line numbers, you have to scroll down and click the ""Plain Text"" version, find your place again, and then do the copy and paste. It'd be nice if the Trac source file browser had a little JavaScript link, called ""Toggle line numbers"", that toggled the line numbers on and off, making copy-n-paste really clean and simple. (The toggle link could be placed in the table cell to the right of the ""Line"" heading cell at the top-left of the table.) See the further discussion (agreement?) on the trac-users group: http://groups.google.com/group/trac-users/browse_frm/thread/71da6e9b5554b89c It shouldn't be much code, especially with jQuery -- Ted Gifford posted one possible solution: http://groups.google.com/group/trac-users/msg/846b21eb4e550e0a?dmode=source Also see a somewhat related discussion on ticket:5779 about a ""toggle line numbers"" link in the diff viewer. They decided against it there, as normally you shouldn't need to copy-n-paste just a few lines of a diff or patch. (I find it's much more common with full source files, so believe the arguments against it for the diff viewer don't hold for the source viewer.) BTW, not that I'd use it much, but I really like the operation of the ""Tabular/Unified"" button for the diff viewer in the new version of Trac. See an example of it at changeset:6599. ---Ben Hoyt (benhoyt.com)",benhoyt next-major-releases Release,7044,RSS feed on Sourcechanges,version control/browser,,enhancement,minor,,new,2008-03-25T19:51:30+01:00,2015-05-26T20:50:57+02:00,"In 'Browse Source' add RSS feed on changes for a specific directory. This would be usefull for developers interested only in a part of the sourcecode changes. If a directory has changes in sourcecode it should give a new RSS feed. ",allarddijk@… next-major-releases Release,7043,save custom queries,report system,,enhancement,minor,,new,2008-03-25T17:41:01+01:00,2015-05-26T20:50:57+02:00,It would be very helpful to be able to return to custom queries. Especially when adding a ticket you can't get back to the query that you were working off of.,dmeyers@… next-major-releases Release,7026,[Patch] Query module column reordering,report system,devel,enhancement,normal,,new,2008-03-20T22:29:37+01:00,2017-05-19T23:03:34+02:00,"This is a patch I came up with at PyCon specifically because some of my users need it, but I think something like it could go into Trac core. First of all it solves the problem of the poor scalability of the UI for selecting columns to display, which is pretty unusable in situations where there are something on the order of 50 custom fields (which I've seen, and it's ugly). I also needed the ability to let users reorder the columns, which this solves. I'm not sure how the Trac devs will feel about this. It replaces the 'col=' query arguments with arguments in the form 'col#=', where # is the order that that column should be displayed in. This works fine for URL query strings. But I haven't updated the TracQuery syntax to support column reordering. Other than that it seems to work fine, and includes support for non-!JavaScript users.",ebray next-major-releases Release,6986,MySQL schema changes for performance,general,0.11rc1,enhancement,major,,new,2008-03-10T21:44:57+01:00,2016-03-29T15:31:35+02:00,"We have been dealing with extremenly poor Trac performance when using MySQL as the backend database. The root cause is the use of 'text' fields in many columns that should preferably be varchar. We finally decided to make some schema data changes to improve performance. As a result of these changes, we have substantial performance improvements. For an example ,the following query is executed very often by Trac: {{{ SELECT DISTINCT s.sid, n.value, e.value FROM session AS s LEFT JOIN session_attribute AS n ON (n.sid=s.sid and n.authenticated=1 AND n.name = 'name') LEFT JOIN session_attribute AS e ON (e.sid=s.sid AND e.authenticated=1 AND e.name = 'email') WHERE s.authenticated=1 ORDER BY s.sid; }}} Our execution time dropped from 8 seconds for this query to approximately 0.1 seconds. When timing this, there were about 6000 rows in the session table, and 20k rows in session_attribute. The only data type changes me made were to convert several columns from text to varchar, primarily colums used often in joins and where clauses. We also added indexes on ticket.{priority,milestone,component} I am attaching three files: - trac_schema_before is a MySQL dump of our table definitions before the changes (Trac 0.10.4 bas schema) - trac_schema_after is a MySQL dump of our table definitions after the changes. - trac_schema_diffs shows the changes we made. I recommend that these changes or similar ones be merged into a future Trac version. Thanks, mike ",mikeyp@… next-major-releases Release,6958,prevent double activation of the same functionality (component sub-classing),general,0.10.4,enhancement,normal,,new,2008-03-07T11:23:04+01:00,2015-05-26T20:50:57+02:00,"we activated trac tags including the wiki module. it was possible to activate the standard trac wiki module too. the effect was that in the timeline two ""include wiki changes"" appeared (see http://groups.google.com/group/trac-users/browse_thread/thread/bfdff087ad1eddf8). it would be great having some mechanism preventing such a behaviour. ",ThurnerRupert next-major-releases Release,6858,Improve intra-line diff highlighting,version control/changeset view,,enhancement,minor,,new,2008-02-20T11:06:38+01:00,2015-05-26T20:50:57+02:00,"Looking at this one: http://trac.edgewall.org/wiki/TracInstall?action=diff&version=231 {{{ #!diff --- TracInstall +++ TracInstall @@ -43,7 +43,7 @@ * [http://www.postgresql.org/ PostgreSQL] * [http://initd.org/projects/psycopg2 psycopg2] or [http://pypgsql.sourceforge.net/ pyPgSQL] - '''Warning''': Postgresql-8.3 uses a strict type checking mechanism. To use trac with the 8.3 Version of Postgresql, you will need the latest version of trac. + '''Warning''': PostgreSQL 8.3 uses a strict type checking mechanism. To use Trac with the 8.3 Version of PostgreSQL, you will need the latest version of Trac. === For MySQL === }}} I think we should be able to do much better, something like: {{{ #!html

    • TracInstall

      old new
      4343* [http://www.postgresql.org/ PostgreSQL]
      4444* [http://initd.org/projects/psycopg2 psycopg2] or [http://pypgsql.sourceforge.net/ pyPgSQL]
      4545
      46 '''Warning''': Postgresql-8.3 uses a strict type checking mechanism. To use trac with the 8.3 Version of Postgresql, you will need the latest version of trac.
      46 '''Warning''': PostgreSQL 8.3 uses a strict type checking mechanism. To use Trac with the 8.3 Version of PostgreSQL, you will need the latest version of Trac.
      4747
      4848=== For MySQL ===
      4949
    }}} That would be especially useful for Wiki text diffs, where it's common to have one line for a full paragraph, but the clarity of code diffs would probably benefit of this enhancement as well. A somewhat related enhancement would be to have an heuristic for triggering the intra-line diff highlighting also when the base and new block length differ, as it's common to have common lines changed together with lines added/removed, in the same chunk. ",Christian Boos next-major-releases Release,6807,Option to skip revisions with lots of added files,version control/changeset view,,enhancement,normal,,new,2008-02-11T09:35:43+01:00,2015-05-26T20:50:57+02:00,"On the real production systems with several millions LOC, it should be possible to skip the revisions that contain initial import of the sources into svn. Such a revisions could contain thousands of files and to trac it takes a hours to process such a revision, probably because of the indexing or code diff calculation. The request is to have an option to skip such a revision, i.e. either make them invisible to trac somehow, or make trac know to take the special care of such a cases. Ideally, trac could detect such a revisions by itself and proceed according to some config. The huge revisions could be detected by the size of the revision file, for example. ",moisei next-major-releases Release,6783,Diff Tabular / Unified switcher highlights the wrong way around?,version control/changeset view,devel,enhancement,normal,,new,2008-02-07T14:10:55+01:00,2015-05-26T20:50:57+02:00,"When viewing a diff, the selected alternative is bolded (and not linked) and the alternative option that can be selected is greyed out (with link). Shouldn't it be the other way around? It is very confusing, and contrary to common UI practice. Or could it be adjusted like the appearance of small mini tabs or something where both look like valid choices, but where current choice can still be clearly seen? ",osimons next-major-releases Release,6638,Allow file exclusions in diffs,version control/changeset view,,enhancement,normal,,new,2008-01-12T12:44:12+01:00,2015-05-18T20:07:03+02:00,"Trac should allow us to set file types (by extension) that should not be diff-ed. A lot of large auto generated xml-based files are showing up in changesets and they really hurt performance and make Trac difficult to use. ",anonymous@… next-major-releases Release,6515,Trac detected an internal error: Session ID not set,general,,defect,normal,,new,2007-12-17T09:12:16+01:00,2015-06-09T23:41:15+02:00,"Oops… Trac detected an internal error: If you think this really should work and you can reproduce it, you should consider reporting this problem to the Trac team. Go to http://trac.edgewall.org/ and create a new ticket where you describe the problem, how to reproduce it. Don't forget to include the Python traceback found below. TracGuide — The Trac User and Administration Guide Python Traceback {{{ Traceback (most recent call last): File ""/usr/local/lib/python2.5/site-packages/trac/web/main.py"", line 406, in dispatch_request dispatcher.dispatch(req) File ""/usr/local/lib/python2.5/site-packages/trac/web/main.py"", line 184, in dispatch req.session = Session(self.env, req) File ""/usr/local/lib/python2.5/site-packages/trac/web/session.py"", line 48, in __init__ self.get_session(sid) File ""/usr/local/lib/python2.5/site-packages/trac/web/session.py"", line 90, in get_session self.bake_cookie() File ""/usr/local/lib/python2.5/site-packages/trac/web/session.py"", line 56, in bake_cookie assert self.sid, 'Session ID not set' AssertionError: Session ID not set }}}",franz.x.maier@… next-major-releases Release,6492,[PATCH] Change icons on timeline,timeline,0.10-stable,enhancement,normal,,new,2007-12-12T17:57:59+01:00,2015-12-19T15:52:49+01:00,"would be nice to have icons which are easier to distinguish .. for example, having a star or something like that on the lower right corner (like the green tick in the closed-ticket-icons)",Markus.Staab next-major-releases Release,6365,Show trac:summary (oneliner_properties) property at the directory level,version control/browser,devel,enhancement,minor,,new,2007-11-16T23:40:34+01:00,2015-05-26T20:50:57+02:00,"If !trac:summary is set on `trunk/prop` I think it would be good to see it shown in the listing for `trunk`, e.g.: {{{ Name Size Rev Age Last Change addins 7 3 Something prop Proposal files 9 2 Something else zap.py 11k 3 23 Imported }}} if for example !trac:summary was set to 'Proposal files' on prop. ",TerryBrown next-major-releases Release,6330,'Back to query' when no query has been used,ticket system,devel,defect,normal,,new,2007-11-12T12:59:29+01:00,2015-05-26T20:50:57+02:00,"I'm not sure whether its a feature or a bug: When a ticket is entered in the search box with the quickjump syntax, the ticket page contains the usual ""previous ticket"", ""back to query"" and ""next ticket"" links. The ""back to query"" link is quite unexpected, as no query has been used to reach the ticket page, and the default report ({6}) seems to come a bit out of the blue to me. I would have expected that in this case (no query, direct access to the ticket) the 'Back to query' link not to be shown, and that 'previous ticket' and 'next ticket' links lead to the immediate existing tickets, ''i.e.'' for ticket #''N'', 'previous ticket' jumps to ticket #''N,,-1,,'' and 'next ticket' jumps to ticket #''N,,+1,,'', whatever their status (closed or not) ",Emmanuel Blot next-major-releases Release,6249,Customized field labels for standard fields,ticket system,,enhancement,minor,,new,2007-10-26T14:56:12+02:00,2015-05-26T20:50:57+02:00,I'd like to be able to easily modify the labels on standard fields.,anonymous next-major-releases Release,6226,Value for limit size of logfile,general,0.10.4,enhancement,normal,,new,2007-10-22T11:07:37+02:00,2015-05-26T20:50:57+02:00,"It would be great if TRAC would able rotate logfiles exceeding a size limit. It means: I want able to say don't let logfiles get bigger than 10MB, if size is reached, rotate them ''count'' times. So if ''count'' is set to 5, then it will get rotated up to log.5. This would very handy when running in debug mode when something goes wrong not often and so debug-logging must run a long time and this stuff shouldn't go into syslog. ",trac@… next-major-releases Release,6219,Excessive time taken by notification can lead to db timeout?,notification,,defect,major,,new,2007-10-20T23:32:02+02:00,2016-05-10T13:52:54+02:00,"All database activites are taking a long time - creation of a ticket always times out though the ticket gets saved properly and now I am consistently getting this error while hitting any page that makes a database call. {{{ Traceback (most recent call last): File ""/usr/lib/python2.4/site-packages/trac/web/main.py"", line 406, in dispatch_request dispatcher.dispatch(req) File ""/usr/lib/python2.4/site-packages/trac/web/main.py"", line 183, in dispatch req.perm = PermissionCache(self.env, req.authname) File ""/usr/lib/python2.4/site-packages/trac/perm.py"", line 263, in __init__ self.perms = PermissionSystem(env).get_user_permissions(username) File ""/usr/lib/python2.4/site-packages/trac/perm.py"", line 227, in get_user_permissions for perm in self.store.get_user_permissions(username): File ""/usr/lib/python2.4/site-packages/trac/perm.py"", line 109, in get_user_permissions db = self.env.get_db_cnx() File ""/usr/lib/python2.4/site-packages/trac/env.py"", line 203, in get_db_cnx return DatabaseManager(self).get_connection() File ""/usr/lib/python2.4/site-packages/trac/db/api.py"", line 76, in get_connection return self._cnx_pool.get_cnx(self.timeout or None) File ""/usr/lib/python2.4/site-packages/trac/db/pool.py"", line 107, in get_cnx raise TimeoutError('Unable to get database ' TimeoutError: Unable to get database connection within 20 seconds }}} ",pranavbhasin@… next-major-releases Release,6173,"Allow accounts to be disabled, such that they no longer receive email notifications",notification,devel,enhancement,major,,new,2007-10-12T18:59:05+02:00,2015-06-23T17:44:12+02:00,"When a developer or other Trac user becomes inactive in a project, it would be nice to be able to simply 'lock' their account, preserving as much information as possible. Because otherwise they'll be receiving email notifications for every bug they've reported, etc. Workarounds, and why they're undesirable: * Modify every ticket they've reported. Error-prone and tedious, plus minor data loss. * Modify the configuration of the mail server to alias the login to /dev/null. Not always practical, and sometimes you still want non-automated emails to go through to them. * {{{DELETE FROM session WHERE sid=:developername AND var_name='email';}}} Someone needs to look this code up, and if the schema changes it might break. Also, encouraging people to mess with the database internals is probably a bad idea in the long term. If there's a way for Trac to determine whether the auth system has locked an account, relying on that would probably be the most elegant solution.",anonymous next-major-releases Release,6160,Enhanced Query for All Tickets by Milestone + RFE for a new automatically detected datatype,report system,0.10.3,enhancement,normal,,new,2007-10-09T22:47:17+02:00,2015-05-26T20:50:57+02:00,"The below query orders all milestones and tickets by their respective due date or the time of creation in case of the ticket, as was implemented before. It would be nice to have a DUE datatype being mapped to the datetime datatype in order for displaying the due date in the `__group__` header along with the `t.milestone` name. TIA Carsten ----------------- {{{ SELECT p.value AS __color__, t.milestone ||' Release' AS __group__, t.id AS ticket, t.summary, t.component, t.version, t.type AS type, (CASE status WHEN 'assigned' THEN t.owner||' *' ELSE t.owner END) AS owner, t.time AS created, m.due as date, t.changetime AS _changetime, t.description AS _description, t.reporter AS _reporter FROM ticket t LEFT JOIN enum p ON p.name = t.priority AND p.type = 'priority' LEFT JOIN milestone m ON m.name = milestone WHERE status IN ('new', 'assigned', 'reopened') ORDER BY (milestone IS NULL),m.due, p.value, t.type, time }}}",Carsten Klein next-major-releases Release,6158,"Use separate, multiple CSS classnames to enable more versatile stylesheet customization",report system,,enhancement,normal,,new,2007-10-09T19:47:50+02:00,2015-05-26T20:50:57+02:00,"Customizing CSS for trac reports is a bit difficult at the moment because semantically independent classnames are concatenated together in a single class. For example, ""color2-odd"" reflects that a row is both at an odd index, and that it should have color2 applied to it. Thus, changing the even/odd color style requires defining at least 12 styles (6 colors X 2 even/odd values), as seen in the current report.css. === Generic Color Styling === If, instead, the classname were specified as ""color_N odd"", (where N indicated the cell value) then CSS rules could be specified like so: {{{ .even {...} .odd {...} .color_1 {...} .color_2 {...} ... .color_6 {...} }}} === Cell-Specific Styling === Furthermore, individual cells should have a class value that reflects both the cell name and cell value. For example, cells showing ticket priority should have a class of ""priority priority_N"" (where N is the priority value). E.g. the markup should look like this: E.g. a cell could have class=""priority priority_''N''"", instead of just class=""priority"" === Backward Compatibility === Note that existing stylesheets can continue to be supported by simply including the existing stylenames as part of the 'class' attribute in the markup. To illustrate what this might look like for all of this ... {{{ ... ... ... ... ... ... }}} ",robert@… next-major-releases Release,6154,better formatted rss feed,ticket system,0.10.4,enhancement,normal,,new,2007-10-09T01:18:19+02:00,2015-05-26T20:50:57+02:00,"would be nice to have more contents and details in the rss feed. The Bucktracker Bugdar has a nice layout for RSS Feeds see e.g. (with a rss reader) http://www.bluestatic.org/bugs/syndicate.php",markus.staab@… next-major-releases Release,5999,From ticket to wiki : add links to ticket/comment metadata when appropriate,ticket system,,enhancement,normal,,new,2007-09-10T11:24:56+02:00,2015-05-26T20:50:57+02:00,"In a ticket, we have several metadata that relates to something existing in the Trac system: milestone, component, etc. It would be helpful to have these fields link to their actual wiki page (or to a template-based url, defined in the configuration file): * the milestone field would redirect to the milestone page, * the component would redirect to the component view in the milestone page, Plus, in case these would exist as well in the near future: * each keyword could redirect to a search page looking for every item with this keyword, * each person (reporter, owner, on cc list, anyone who adds to the ticket) could redirect to a user page summarizing done and pending work in the whole Trac. While not adding direct context to a ticket, it would make it easier to access data related to it, without a need for a manually typed search or several clicks.",romain.dalverny@… next-major-releases Release,5997,Clean name-spaces for CSS for 0.11+?,general,devel,enhancement,normal,,new,2007-09-09T20:05:01+02:00,2016-02-06T11:34:26+01:00,"I suspect that it may be too late for this, but I think this is a low-risk change. Trac uses CSS all over the place. This is good. The problem is that I need to pass through some content that *also* uses CSS, and there is a problem of name space collision in the choices of names for the class attribute. Since Trac is cutting over the template system, now is an opportune time to look at CSS issues if there is a desire to do so. How difficult would it be for Trac to adopt the convention that all class attributes and all id attributes that are emitted from the trac templates would begin with ""trac-"" so that they are less likely to collide with CSS within content? In our case, we cannot simply import the incoming content once and have done; it is a reference manual that we need to maintain externally via our version control system. Our goal is to simultaneously make it available within the Wiki for commenting and feedback. I considered re-writing the class attributes on our content using XSLT. It is doable, but it is probably beyond the skills of most XSLT authors.",shap@… next-major-releases Release,5979,Add more option to filters with drop-downs in custom query,query system,devel,enhancement,normal,,new,2007-09-05T17:08:14+02:00,2018-10-16T12:44:28+02:00,"Adding a ""contains"" option to filters with drop-downs in custom query would be really helpful. The hard part is that it would have to ignore the drop-down and use a text input field. It is really necessary for people overloading the Component field, due to the fact we don't have component/subcomponent capability. ie. How do you search for all of component ""blob"" when your choices are ""blob: bit"", ""blob: bot"", ""blob: bat"", ""thingie: subdodad""? ",asloan12@… next-major-releases Release,5856,Enterprise workflow enhancements for request info,general,devel,enhancement,minor,,new,2007-08-10T12:50:28+02:00,2015-05-26T20:50:57+02:00,"I'm trying to use the enterprise workflow and I stumbled upon some obstacles. For example: A new ticket is created and assigned to a user. The user finds he needs more information so he presses request info. The ticket gets state infoneeded_new and the only actions available is provide_info and leave. Suggestion: 1. Make it possible to assign a user when requesting more information. If in the infoneeded_new state I would like it to be automatically assigned to the reporter 2. It should alse be able to reassign a ticket in infoneeded_new state. 2. Can be fixed by adding the following to the workflow: {{{ reassign_infoneeded_new = infoneeded_new -> infoneeded_new reassign_infoneeded_new.name = reassign reassign_infoneeded_new.operations = set_owner reassign_infoneeded_new.permissions = TICKET_MODIFY }}} 1. Looked easy but I found something unwanted in default_workflow.py. Apparently set_owner adds a ""to"" between the name and the assignee. If I modify requestinfo_new to: {{{ requestinfo_new = new -> infoneeded_new requestinfo_new.name = request info from requestinfo_new.operations = set_owner requestinfo_new.permissions = TICKET_MODIFY }}} Then the output will be: request info from to Isn't it better to remove the ""to"" from default_workflow.py and let the user configure the whole string in trac.ini instead? Cheers / Erik",kirean@… next-major-releases Release,5850,Search Results Per Page Variable,search system,0.10.3,enhancement,normal,,new,2007-08-09T20:05:18+02:00,2015-05-26T20:50:57+02:00,"Currently when I want to set the results per page from a search, I modify the Search.py ""RESULTS_PER_PAGE"" variable. This however effects all of my sites. It would be very useful if I could set this in trac.ini. Something like results_per_page = 20 within the [search] section. Additionally, I would write this into WebAdmin and have it web administer-able.",jchan@… next-major-releases Release,5670,[patch] Refactoring of general `NotifyEmail` class to make it easier to subclass,notification,devel,enhancement,normal,,new,2007-07-08T12:06:01+02:00,2017-03-11T22:40:53+01:00,"The `NotifyEmail` class in `trac.notification` is a great starting-point for any custom email classes. I have made a simple class that does my special things in `notify()` and `send()`, but where 98% of the logic it is left to `NotifyEmail` and `Notify` classes to handle the practical things. Among other things I use my custom Email-sending class to send registration confirmations and similar system mail. However, for any kind of reuse of this class there is one snag: Notification needs to be enabled for the project. So, if you want to send you own mail or have some other module that sends email there is no easy way to do this based on general classes without copy & paste of logic logic from original class (in notify() as it would need to be 'skipped' in the call-chain). My system email sending new passwords based on 'forgotten password' form should not depend on ticket notification being enabled. As it is now I have to trick the class in my notify() by enabling notification, calling `NotifyEmail.notify()`, and then setting it back at the end... Not pretty, and potentially a source of application misbehavior. So much for the argument of why. Here is a small patch that extracts the enable-check logic into a method that can be easily overridden for any subclass of `NotifyEmail`: {{{ #!python Index: trac/trac/notification.py =================================================================== --- trac/trac/notification.py (revision 5803) +++ trac/trac/notification.py (working copy) @@ -209,11 +209,14 @@ self._charset.output_charset = 'ascii' else: raise TracError, 'Invalid email encoding setting: %s' % pref + + def notification_enabled(self, resid): + return self.config.getbool('notification', 'smtp_enabled') def notify(self, resid, subject): self.subject = subject - if not self.config.getbool('notification', 'smtp_enabled'): + if not self.notification_enabled(resid): return self.smtp_server = self.config['notification'].get('smtp_server') self.smtp_port = self.config['notification'].getint('smtp_port') }}} It should not be able to break any existing code anywhere as far as I can deduct, and would be a great bonus for reuse of Trac code.",osimons next-major-releases Release,5641,Allow enums to be used in ticket custom value sets,general,0.10.4,enhancement,normal,,new,2007-07-02T00:22:19+02:00,2015-05-26T20:50:57+02:00,"I am writing a plugin that just adds a custom ticket field. It is to hold a list of clients to which a ticket ""belongs"" but that is not really important. The possible values for this could be quite long and using the {{{.options}}} value of the .ini file doesn't seem appropriate. So I thought a sensible route would be to use the enum table to hold the values and write a simple extension to TracWebAdmin to add another editing page page based on !AbstractEnumAdminPage type (I'm not sure if this is still part of Trac 0.11 with merged TracWebAdmin tho' :)) Anyway, this patch allows easy use of the enum table and easy ways to hook up custom fields to use that list. This would not really be of use to people unless they create a plugin to admin the actual list too but it's still a useful feature IMO and the patch itself is fairly unobtrusive, so hopefully you can accept it. The changes just allow a {{{.options}}} value of ""enum:name"" where name is the name you give to your enum and this will load the values from the database rather than parsing a pipe-separated list. The other change was for when an enum name is updated (name changed) and which then updates the ticket table to reflect said changes. I just added a check so that it can update the ticket_custom table instead. Hope this is acceptable. Not adding my mail here as I don't want it spammed but I will check back for updates. Patches are against 0.10.4 ",Colin Guthrie next-major-releases Release,5566,Postgresql admin security issue,admin/console,0.10.3,defect,normal,,new,2007-06-21T01:45:45+02:00,2017-06-12T04:47:13+02:00,"Trac accepts postgres as a database backend using a string such as: postgres://user:password@localhost:port/database The problem is anyone with read access to this trac directory basically has full administrator priveleges, because they can just issue the command ""trac-admin trac_project permission myself TRAC_ADMIN"" Ideally I'd like to grant project administrators with the ability to customize their trac environment and delegate permissions to other users as necessary. (I keep a master authz file to share amongst various projects) But it seems like 'trac-admin' should really be using the local users authentication credentials. And there should be a seperate set of credentials (that I can keep in a protected file) for the trac web interface. This problem most likely extends to mysql and any other database connection type. SQLite is one of the few exceptions since it's entirely file based. I'm using trac-0.10.3.1-2.fc6 and subversion-1.4.3-2.fc6 (on Fedora Core 6, obviously). ",dannys@… next-major-releases Release,5535,[inherit] not working for me (error on missing config),general,devel,defect,major,,new,2007-06-16T07:39:35+02:00,2015-05-26T20:50:57+02:00,"plugins in the inherited plugin directory don't show up in my administration, etc. They work fine if I move them into the local trac env's plugins directory. local and global confs attached.",Dave Abrahams next-major-releases Release,5526,Support LIKE and wildcards in custom queries,report system,,enhancement,normal,,new,2007-06-15T08:50:35+02:00,2019-08-23T12:45:50+02:00,"For many fields (component, version, milestone, ...) only ""is"" and ""is not"" are present as options in the filter combo box. For my application I would need at least ""like"" (processed as SQL's LIKE) in addition to that. Without this capability I cannot currently dispense with the report modules (which, as I read, is being phased out in favor of the query module).",koegl@… next-major-releases Release,5506,Confusing formatting for attachments in Timeline,timeline,devel,defect,minor,,new,2007-06-12T23:10:01+02:00,2019-02-13T19:50:07+01:00,"{{{ #!html Click on '    Ticket #1234' at the attachment line on Timeline page opens attachment page and not actual ticket as expected. }}} Simple solution can be done with JavaScript (attached). ",anonymous next-major-releases Release,5441,Timeline shows incomplete information about status changes for customized workflow,timeline,devel,defect,normal,,new,2007-06-07T05:39:14+02:00,2015-03-02T22:27:22+01:00,"Currently 'Timeline' shows details about status transitions only for few standard values ('new', 'reopened' and 'closed') and everything else shows as ""''status'' changed"", which is not very convenient if custom workflow has more states. We could have something like records in ticket's change history. Sample patch is attached. More generic solution (e.g. show changes in custom fields similarly) could be even better...",anonymous next-major-releases Release,5390,"Zip Archive: aliases, symlinks",version control/browser,0.10.3,defect,normal,Jun Omae,assigned,2007-05-30T04:13:24+02:00,2022-05-19T14:00:43+02:00,"Downloading source from the Trac Browsers ""Zip Archive"" generates Mac aliases or symlinks as files instead of links.",anonymous next-major-releases Release,5246,[PATCH] Use permission system to store groups for authz access control,version control/browser,0.10-stable,enhancement,normal,,new,2007-05-02T00:11:50+02:00,2015-12-19T15:48:35+01:00,"We're currently using LDAP to store our users and groups. Our problem with using Subversion-style authz files for restricting access in Trac is that we'd have to define group memberships again in the authz file itself - thus we'd have to manage group memberships twice: Once in LDAP, once in the authz file (we're currently NOT using an authz file for Subversion itself for exactly this reason - we're doing it the ""hard way"" via Apache configuration instead). Basically, this issue is #4224 the other way 'round. In this context, it would be great if the authz module could simply use the built-in permission system of Trac to retrieve the groups a user belongs to - and not the authz file. I've written a patch that adds a configuration option authz_use_perm_groups that's false by default, resulting in the current behaviour. If set to true, Trac will not care about the groups section of the authz file and use the PermissionSystem to retrieve the group names instead (currently by fetching all lower-case permissions for the current user and stripping an eventual @ in front of the group name). I don't know much about the internal design of Trac so my code is probably quite ugly - but at least it works. Feel free to find a nicer solution. :-) Please note that this patch would interfere with #4997 since Subversion itself does not implement any of this.",christian.seiler@… next-major-releases Release,5177,"norobots option for: raw source code, rss, old source code",version control/browser,0.10.3,enhancement,normal,,new,2007-04-19T11:24:50+02:00,2015-05-26T20:50:57+02:00,"it would be beneficial if there would be an option to prevent search engines from spidering: * rss * raw contents if the contents is displayed beautifully rendered in trac * old versions out of svn via e.g. norobots or similar. ",ThurnerRupert next-major-releases Release,5031,Queries ought to support less/great than for ordered values (i.e. priorities and severities),ticket system,0.10.2,enhancement,normal,,new,2007-03-27T10:48:40+02:00,2014-02-19T19:06:36+01:00,I.e. expand the [is|is not] operator set a bit.,macke@… next-major-releases Release,5001,offer possibility to inform users of new releases and changes,notification,,enhancement,normal,,new,2007-03-23T11:15:06+01:00,2015-01-13T18:50:21+01:00,"I can see that more and more Open Source projects are now using Trac as their development tool which is nice. There's only one thing I dislike about it: Most Trac sites don't have a possibility to inform users of changes or new releases of the software. I have noticed that some sites offer a RSS-feed. But sites like sourceforge.net or http://www.berlios.de offer a automated email notification which is very convenient for system admins etc. Maybe some RSS2news solution could be implementet. then rss could automatically also be send via email. Thanks for addressing this.",timmichelsen@… next-major-releases Release,4931,notification with SMTP through SSL is not supported (fix included),notification,0.10.3,defect,normal,,new,2007-03-11T14:53:03+01:00,2015-05-26T20:50:57+02:00,"Sending mail notifications does not work with SSL SMTP server. Python hangs (so does not raise an Error) on source:/trunk/trac/notification.py@4316#L278. This could be fixed using the ssmtplib from http://aleph-null.tv/downloads/ssmtplib.py The method 'beginSend' on source:/trunk/trac/notification.py@4316#L277.should be like: {{{ #!python def begin_send(self): if self._use_tls: self.server=ssmtplib.SMTP_SSL(self.smtp_server, self.smtp_port) else: self.server=ssmtplib.SMTP(self.smtp_server, self.smtp_port) if self.user_name: self.server.login(self.user_name, self.password) }}} ",info@… next-major-releases Release,4887,Option for changeset viewer to show added and deleted files inline,version control/changeset view,0.10.3,enhancement,normal,,new,2007-03-04T23:16:11+01:00,2015-05-26T20:50:57+02:00,"The unified diff correctly includes new files. However the web changeset view does not. For an example, see this change set on r4848 trac. It is missing the contents of trunk/trac/htdocs/js/ie_pre7_hacks.js.",paul.wants.no.spam@… next-major-releases Release,4854,UI Wart: Log viewer navigation links should be duplicated at the bottom of the page,version control/log view,devel,enhancement,minor,,new,2007-02-26T23:08:55+01:00,2015-12-23T19:21:31+01:00,"As the summary says. Generally you scroll down the page to see the changesets, then if you don't find the changeset you're after you have to scroll back up to move to the next set.",Alec Thomas next-major-releases Release,4788,show symlink info in source browser,version control/browser,devel,enhancement,normal,,new,2007-02-19T17:17:07+01:00,2015-05-26T20:50:57+02:00,"Currently the source browser doesn't seem to display if a given file in the svn repository is a symlink. In the browser of the directory containing a symlink, it could be something like {{{ (→ target) }}} right after the name of the file. And in the browser of the symlink itself it could be an extra line like {{{ Symbolic linc to target }}} where ""target"" would be a link to the appropriate file (if the target is also in the repository)",Sergei Haller next-major-releases Release,4778,simplified wikification of svn log messages,wiki system,devel,enhancement,normal,,new,2007-02-19T01:11:11+01:00,2015-05-26T20:50:57+02:00,"Is it possible to simplify the wikification of svn log messages? The problem I get is if we have some math content (like {{{ x^5+x^4+x }}} which will show like x^5+x^4+x. There are some other things as well, which we don't actually need in or project, as the log messages are and will always be plain text. The only thing I'd like to keep is the recognition of revision numbers in the log messages with the appropriate links. Is it hard to achieve this? (Would be ok if it is in the source, i.e. not configurable by admin/user) Thanks.",Sergei Haller next-major-releases Release,4733,Option for all login links to use HTTPS,general,,enhancement,normal,,new,2007-02-14T06:13:37+01:00,2015-05-26T20:50:57+02:00,"It would be nice if trac could automatically use the https protocol in creating links to login with an appropriately configured web server. I have tried a number of mod_rewrite rules and LocationMatch directives, but have not been successful in getting logins for multiple trac projects to be re-directed to the secure server for processing.",jeffcorbets@… next-major-releases Release,4612,Changeset URL to contain the revision in the query string.,version control,0.10.3,enhancement,normal,,new,2007-01-28T21:00:56+01:00,2016-01-18T23:36:05+01:00,"I’m using bzr as version control backend provided by the trac+bzr plugin. Since each branch has its own revision numbers, the revisions as seen by trac have the format: {{{ path/to/branch,revno }}} !ChangesetModule.match_request will match 'path' to the revision and 'to/branch,revno' as the path. trac+bzr works around this problem by URL escaping the revision string but the result is highly unreadable and not at all user-friendly. Attached is a very simple patch I’m using to make trac gracefully handle the above scenario. I have added a configuration switch for the new format because I expect some people would like to keep the old URLs valid. The solution is simply to change the changeset URLs from: {{{ /changeset/revision/path }}} to: {{{ /changeset/path?new=revision }}} The latter is also totally consistent with URLs of log and browser: {{{ /log/path?rev=old_rev&stop_rev=new_rev /browser/path?rev=revision }}} The patch also allows a new changeset notation: {{{ changeset:path@revision }}} Clearly the {{{changeset:revision/path}}} notation cannot be used when revision can contain a slash. ",Peter Dimov next-major-releases Release,4590,There is no interface for providing conversion over files or changeset like in wiki IContentConverter,version control/browser,,enhancement,normal,,new,2007-01-25T10:59:28+01:00,2015-05-26T20:50:57+02:00,It will be useful. But now i should modify Trac sources.,mou@… next-major-releases Release,4549,"[PATCH] Ticket std/custom fields improvements - sql based values, order by, labels, value evals, ...",ticket system,0.10,enhancement,normal,,new,2007-01-16T14:05:36+01:00,2017-01-02T16:49:48+01:00,"Hi, First to say, I find Trac very useful and programmed nicely (appologies for my English). Revision 5 is trac 0.10. Ticket fields std/custom improvements: * field values (combo) filled with sql * customized order of fields * std fields - customized label/default values * values - evaluation of variables TRAC.INI with samples: {{{ std fields default_* - default values order_* - order by (not applicable for some fields - e.g. ticket title) label_* - labels custom fields req_who.options - sql filling the fields req_when.value = $(today) - variable evaluation *.order }}} I plan to implement following - but the question is when :( * obligatory/optional/hidden field definition in ini * init ticket callback - can change field definition (value, type, order ...) * save ticket callback - can check/reject/change field values * ticket initialization * add to ticket/ticket_change ""status by user"" (like unread ticket, ticket, ...) can be used also as internal mail ... instead of sending the mail mark ticket/ticket_change to be read by some users * customized home page for the user - will show number of unread tickets newly * ticket ""factory"" for periodical tickets - e.g. one ticket generated for every month It would be nice to have customized statuses, so in the future trac could be some kind of workflow system (add allowed state-transitions/persons). Each status could have extra fields, e.g. if status is ""fixed"" then field could be revision number. Hope this will be useful. Best regards from Croatia, Robert",trebor74hr@… next-major-releases Release,4468,better support for relative name handling,wiki system,0.10.3,defect,normal,,new,2006-12-28T06:30:09+01:00,2019-11-12T03:50:48+01:00,"relative names should have support for: * getting page name from the relative name * getting page href (self.env.href.wiki) from relative name this will allow something like: {{{ [[Include(./RelativePage)]] }}}",ittayd@… next-major-releases Release,4412,batch copy/rename wiki pages,wiki system,0.10.3,enhancement,normal,,new,2006-12-17T11:22:09+01:00,2015-05-26T20:50:57+02:00,"==== Use cases ===== cross cutting wiki versions If I have a product in version 1.x and I'm about to release 2.0, there are probably many changes to the wiki. However, I'd like to still retain a view of the 1.x wiki, for users of 1.x. Moreover, the 1.x wiki may continue to change with releases to the 1.x stream.",ittayd@… next-major-releases Release,4398,Inconsistent search for single-digit tickets,search system,0.10,defect,minor,,new,2006-12-14T23:11:41+01:00,2015-05-26T20:50:57+02:00,"If you search for #1 in the search field, you will jump to that ticket. If you try to search for #1 on the search page, you get an error that the search term is too short. (So you'll get that error on the first 9 tickets of the project.) It is also useful to search for tickets that reference a ticket number; perhaps the search page should do that instead. ",eli.carter@… next-major-releases Release,4298,tickets linked to multiple versions,ticket system,0.10.2,enhancement,critical,,new,2006-11-30T09:00:23+01:00,2016-04-10T12:08:45+02:00,"As far as I understand, bug 2945 dealt only with description changes for requirement issues. I have a different scenario. Say I have a project with versions 1.0, 1.0.1, 1.0.2 and 1.1. a bug is found in 1.0. it is fixed in some way in 1.0.2 and in another way in 1.1 What I'd like to have is the ability to see all open bugs in 1.0 or 1.0.1 (the above bug included), in 1.0.2 (the above bug is fixed, and the comment describing the fix affects 1.0.2) and 1.1 (the bug is fixed in another way) I think it means only the ability to associate status changes with a version. maybe also priority and severity (in the above scenario, the fix in 1.0.2 may have been partial, not exactly closing the bug for the 1.0.x stream, but lowering the priority)",ittayd@… next-major-releases Release,4292,ROADMAP_VIEW / MILESTONE_VIEW privilege,roadmap,0.10.2,defect,minor,,new,2006-11-29T14:48:43+01:00,2021-08-26T03:00:12+02:00,"roadmap.py is checking for ROADMAP_VIEW, which will only work as long as one keeps the ROADMAP_VIEW permission for anonymous that's set up by db_default.py. Once you delete that, nobody without WIKI_ADMIN privileges can look at the roadmap, because you can create MILESTONE_VIEW privs to your hearts content but they'll be ignored. ",dave@… next-major-releases Release,4286,Accepts email addresses without validation,notification,0.10.2,defect,normal,,new,2006-11-28T20:41:57+01:00,2015-05-26T20:50:57+02:00,"trac accepts email addresses without validation several different places -- on the New Ticket screen, on the settings screen, etc. This is bad for several reasons. For starters, it could be abused by spammers. They could put the email address of a victim in the ""Your email"" box on the New Ticket screen, and submit a ticket to any Trac instance configured to send notifications of new tickets to the submitter. It could also be used by a miscreant to subscribe an unwitting victim to notifications, as an annoyance. trac should never send any emails that aren't verified opt-in emails.",John Goerzen next-major-releases Release,4270,Ticket notification emails contain wiki formatting,notification,0.10.2,defect,normal,,new,2006-11-27T18:52:01+01:00,2017-03-12T11:21:24+01:00,"Hi, when I send an email of a ticket, through the CC list, I find that it sends the wiki formatting as well, since it's a text email, the normal text should be displayed and the wiki formatter should strip out the formatting before the email is sent. In order to get acceptable results from the ticket display, an example is like thus 1. an item !CamelCase[[br]] 2. another item [[br]] but in the email, instead of seeing what I have above, I see this {{{ 1. an item with !CamelCase[[br]] 2. another item[[br]] }}} so of course, people reading the emails, are wondering what all the {{{[[br]]}}} are for and it makes some emails kinda unreadable. this should be remedied",chris.alex.thomas@… next-major-releases Release,4224,Using SVN AUTHZ groups to map to group permissions in Trac,admin/web,none,enhancement,normal,,new,2006-11-21T00:13:46+01:00,2015-05-26T20:50:57+02:00,"I have an authz file that I use with Apache/SVN that has a definition of some groups. e.g. G_developers, G_admin, etc I have the same named groups in Trac created. I want Trac to automatically consider the usernames in the AUTHZ file to have the same permissions as the groups that match. How do I do this?",leighmcc@… next-major-releases Release,4160,Broken InterWiki Link,version control/log view,0.10,defect,normal,,new,2006-11-14T17:37:43+01:00,2015-05-26T20:50:57+02:00,"The following InterWiki link behaves strangely: || '''Syntax''' || '''Behavior''' || || `[tps-t6:1234567]` || Displays as-is, not hyperlinked. || || `[tps-t6:1234567 test]` || Displays `test` hyperlinked to a working URL. || || `tps-t6:1234567` || Displays `tps-t6:1234567` hyperlinked to a working URL. || If I change the InterWiki link to tps instead of tps-t6, all of the above syntaxes generate a hyperlink to a working URL as expected. ",jeff.wise@… next-major-releases Release,4130,Ticket Component Interface Fails to Properly Check for Existing Component,ticket system,0.10,defect,normal,,new,2006-11-11T00:33:08+01:00,2015-05-26T20:50:57+02:00,"When using the ticket.model.Component class, it will raise a `TracError` if passed the name of a non-existent component on instantiation. If the component is instantiated with an existing component name, it will assert an error if one tries to call insert(). However, if one instantiates the object with out a name, and then sets the name and owner after the fact, and then calls insert(), no checking of duplicate rows is done. Possible borkage: {{{ #!python c = Component(env) c.name = 'component1' c.owner = 'sombody' c.insert() }}} The API should be modified to either raise a TracError in the case in question, and/or provide a method for discovering whether or not a Component exist. Personally, I think that the following would be desired behavior: {{{ #!python c = Component(env, 'new component') c.insert() }}} Where the insert() is a noop if the component already exists. The `exists` property could also reflect whether or not the Component instance references an existing component or not. {{{ #!python c = Component(env, 'new component') if not c.exists: c.insert() }}} I find this preferable to: {{{ #!python try: c = Component(env, 'new component') except TracError: c = Component(env) c.name = 'new component' c.insert() }}} ",pacopablo next-major-releases Release,4048,Diff between repositories,version control,devel,enhancement,normal,,new,2006-11-01T21:21:39+01:00,2017-07-11T12:46:51+02:00,"Trac allows to see the files of the project (in any Versioning system) but it were nice if trac allows to: 1. commit on-line files and directories, 2. comment the files, directories and changes of these, 3. see diffs between two repositories, etc. Probably it's a long term enhancement (version 0.20...) Thank you very much",anonymous next-major-releases Release,4022,Provide Etag and/or Last-Modified HTTP headers to ease cache control,wiki system,0.10,defect,normal,,new,2006-10-24T21:32:30+02:00,2015-05-26T20:50:57+02:00,"Trac serves pages without any Etag or Last-Modified header, which makes cache management rather troublesome, and increases the server load. This is particularly important for RSS feeds, which might be checked relatively often. (In one of the comments for #663 it was mentioned that conditional GET was planned for 0.9, but one of Trac setups I'm checking on is based on 0.10 and the functionality seems to be missing.) I think most pages could be served with a Last-Modified header containing the date of the last change (for wiki and ticket pages) or the date of the most recent thing changed (for RSS feeds). ",Giuseppe Bilotta next-major-releases Release,3911,"implement an object system (xwiki, roundup)",general,0.10,enhancement,major,,new,2006-10-11T11:30:16+02:00,2021-01-21T17:51:40+01:00,"in xwiki, i can define objects (set of properties) and then have pages to create new objects, and displaying them, and of course, querying them. this is a great way of structuring documents. e.g., i can create a FAQ object with 'question' and 'answer' fields, then form a page that shows all questions and then all items.",anonymous next-major-releases Release,3906,Request for a files search (no indexing just plain find like search),version control/browser,0.10,enhancement,normal,,new,2006-10-11T03:15:59+02:00,2018-07-08T12:48:00+02:00,"I would like a /usr/bin/find like file search to look for a specific filename in the repository recursively (ie, some*.pdf) I've seen the RepoSearch hack but I would like a simple file search (perhaps in Browse source or in Search directly with a files checkbox) that would let me search for a file or files in particular strictly based on the files in the repository. This is a simple feature, I'm surprised it hasn't been implemented yet ;) perhaps a simple hack would be in order...",sineer@… next-major-releases Release,3841,commit after adding attachments to tickets,attachment,0.10,defect,minor,,new,2006-10-02T17:23:53+02:00,2015-05-26T20:50:57+02:00,"I had the problem after adding attachments I lost my other changes to the ticket: either I had to click the back button in my browser twice or go to the ticket again (now with attachments) and add other changes. Wouldn't it be better in a ticket after adding an attachment to go back to the ticket view and be able to either click ""Submit changes"" or ""Cancel"" as usual (including the fresh attachement)?? ",anonymous next-major-releases Release,3749,recent ticket-comments macro,ticket system,0.9.6,enhancement,normal,,new,2006-09-18T23:14:26+02:00,2020-05-19T19:23:14+02:00,I think it would be nice to have a page that lists recent ticket comments to see what tickets are being actively discussed.,Gary Wilson next-major-releases Release,3737,"login-id, trac-nick",general,devel,enhancement,normal,,new,2006-09-16T20:30:15+02:00,2015-05-26T20:50:57+02:00,"as also noted in #3667 there are cases where the username may change. also login mechanisms coming via basic auth may provide some userid which is not easy to remember. therefor it would be great to have more fields for a user, i.e. a username, userid, a nick, and maybe the nick is displayed. ",ThurnerRupert next-major-releases Release,3689,"patch for mimeviewer: autodetect charset by ""# -*- coding: windows-1251 -*-""",version control/browser,0.10b1,enhancement,minor,,new,2006-09-09T07:54:32+02:00,2015-12-19T15:46:40+01:00,"***** api.py.bak2 {{{ #!python return utf return self.default_charset }}} ***** api.py {{{ #!python return utf m = re.search(r'-\*-\s+coding:\s*(\S+)\s+-\*-', content[:100]) if m: return m.group(1) return self.default_charset }}} *****",igor daniloff next-major-releases Release,3584,"[PATCH] Zip file names should match the latest version of their contents, not of the whole repository",version control/browser,devel,enhancement,minor,,new,2006-08-19T12:14:16+02:00,2015-05-26T20:50:57+02:00,"The original ticket appears as th:ticket:558. I don't know which solution is the best to implement such a feature (renaming feature as Alec suggested, ...), but the current implemention is a real issue on multi-project host servers such as TracHacks. ",Emmanuel Blot next-major-releases Release,3511,quick access to favorite reports/queries,report system,,enhancement,normal,,new,2006-08-04T03:24:12+02:00,2015-05-26T20:50:57+02:00,"Users should be able to define ""favorite"" reports and save queries in their session for quick access. The ""View Tickets"" menu item could be enhanced with a menu that drops down when hovering over it to display this list of reports and queries.",Matthew Good next-major-releases Release,3320,Custom fields can't be hidden,ticket system,devel,enhancement,minor,,new,2006-06-26T16:16:35+02:00,2015-05-26T20:50:57+02:00,"Though this is already fixed in the new workflow branch, it would be nice to fix this for 0.10. All that needs to be done is add to following [source:trunk/trac/ticket/api.py@3444#L161 here]: {{{ #!python 'skip': config.getbool(name + '.skip', False) }}}",coderanger@… next-major-releases Release,3302,Add Permission Change Hooks to Trac,general,devel,enhancement,normal,,new,2006-06-22T02:04:42+02:00,2015-05-26T20:50:57+02:00,"summary says most of it, attaching a patch. basically, allow plugins to listen for permission changes (i've written an svn_authz generating plugin that would like to know when to rebuild the file)",trac@… next-major-releases Release,3273,Add REGEXP support to SQL queries,report system,devel,enhancement,minor,,new,2006-06-15T16:54:53+02:00,2016-01-18T23:26:42+01:00,"SQLite allows the use of a REGEXP search ![1]. Such a feature would be useful in cases such as parsing multiple CCs, multiple asignees on a ticket ![2], etc. All that has to be done is to create a user function on the connection ![3]. I've attached a diff of what (I think) will enable this (I am using r3406 currently). The patch will add a user function in SQLite immediately after the connection is made. The user function is called regexp. (See [attachment:trac_sqlite_regexp_r3406.patch]) References: 1. [http://www.sqlite.org/lang_expr.html]: What is currently allowed in SQLite expressions 2. #2662: Using REGEXP in SQLite to parse multi-user assignments 3. [http://initd.org/pub/software/pysqlite/doc/usage-guide.html#creating-user-defined-functions] pysqlite documentation to add a user-defined function",david.l.jones@… next-major-releases Release,3251,A documentation about components should be added,wiki system,0.9.6,enhancement,minor,,new,2006-06-11T21:31:37+02:00,2015-05-26T20:50:57+02:00,"I'm trying to update the list of components for my ticket system, but I've been unable to find any way in either the gui or configuration that will allow me to do this.",dcarter@… next-major-releases Release,3236,Support for a merge change type in version control API,version control,devel,enhancement,normal,,new,2006-06-08T03:36:04+02:00,2015-05-26T20:50:57+02:00,"Trac currently supports ''add'', ''edit'', ''delete'', ''copy'' and ''move'' change types for files in a changeset. However, it would also be nice if Trac supported some of the ''merge'' change type for changes that involve merging changes from other files. I realise that Subversion doesn't currently implement merge tracking natively (''yet''), but I have been working on a [http://www.perforce.com/ Perforce] plugin for Trac (similar to the one described in #257 but based on [http://pyperforce.sourceforge.net/ PyPerforce]) and would like to be able to get Perforce's merge/integration information displayed through Trac. Perforce supports a number of different merge/integration types: * '''branch from''' - An exact copy of a file to a new location. Equivalent to Trac's ''copy'' or ''move'' depending on whether the source file was also deleted in the same changeset. * '''add from''' - A copy of a file to a new location that has been modified before checkin (related to #3172). * '''copy from''' - A merge operation that ignores any changes in the target file, effectively replacing the target with the source. * '''ignore from''' - A merge operation that ignores any changes in the source file, keeping the target file's changes. * '''merge from''' - A merge operation that merges the changes from both source and target files. * '''edit from''' - A merge operation that merges the changes from both source and target files and contains original user edits. * '''delete from''' - A merge of a deleted source file into the target file, deleting the target file. Basically a delete operation with an associated source file. See the [http://www.perforce.com/perforce/doc.052/manuals/cmdref/integrated.html#1040665 p4 integrated] documentation for more info on the Perforce integration model. ",lewisbaker@… next-major-releases Release,3162,LDAP support for e-mail notification,notification,0.9.5,enhancement,normal,,new,2006-05-17T17:38:38+02:00,2021-06-22T01:21:14+02:00,"When using trac with LDAP authorization it would be nice if LDAP is also used for getting the e-mail address of logged in users (this feature should probably be settable using some trac.ini parameters). I do not know howto read trac.ini parameters so I added a few lines to Notify.py Maybe somebody can use this. ",remco.wouts@… next-major-releases Release,3080,Custom field sorts only as text,general,0.9.5,enhancement,normal,,new,2006-04-24T18:50:46+02:00,2015-05-26T20:50:57+02:00,"I've created a custom field to enable me to manually order tickets. I've used integers as the value, but it seems to be sorting as text. For instance, the order shows up like 1, 10, 11, 2, 3, 4, 5, 6, 7, 8, 9 Would like to be able to sort numerical values through custom fields. ",aroach@… next-major-releases Release,3022,Remove ROADMAP_ADMIN,roadmap,0.9.4,enhancement,minor,,new,2006-04-11T19:10:24+02:00,2022-06-18T19:27:08+02:00,Tiny ticket to track the removal of `ROADMAP_ADMIN` (alias of `MILESTONE_ADMIN`) in the next release of Trac.,Emmanuel Blot next-major-releases Release,2961,custom comment fields,ticket system,0.9.4,enhancement,normal,,new,2006-03-30T21:41:12+02:00,2018-02-26T12:59:51+01:00,"I'd like a function similar to TracTicketsCustom, but for comments. I'm looking to integrate time tracking into Trac to eliminate some double entry, but unlike all the folks at #710, I'd like my ""hours spent"" field to be on the comment, so we can track who is putting in the time when, and not just have one field on the ticket everyone keeps updating.",Ryan@… next-major-releases Release,2827,Preview attachments in main ticket (or wiki) window,attachment,devel,enhancement,normal,,new,2006-03-02T16:58:56+01:00,2023-02-27T12:54:55+01:00,"It will be useful if Trac shows previews for attachments in main ticket window. This can be implemented using the Trac plugin architecture in order to allow extending the methods for generating previews for attachments. Taking into account overall performance of the system, the architecture should allow to run generating of previews in background when an attachment is added. As a part of standard distribution, a displaying of scaled down images can be implemented: * The size of scaled picture can be configured; * There is an excellent [http://www.pythonware.com/products/pil/ Python Imaging Library] that will do the work. Similar approach can be used in other places of Trac, e.g. previewing attachments in wiki.",pkou at ua.fm next-major-releases Release,2733,Push some more behaviour to the vc backend,version control,0.9.3,enhancement,normal,,new,2006-02-08T21:26:43+01:00,2015-05-26T20:50:57+02:00,"In order to get my new TracDarcs plugin working, I had to apply [http://artiemestieri.tn.it/~lele/projects/trac-darcs/trac-api-2866.patch this patch] that 1. Fixes #2731 1. Adds an ""hash"" field to the `revision` table 1. Moves the actual insertion of the cache data from the `CachedRepository` class to the `Changeset` one Point 3 allows the backend to augment the cached data as needed, simply by overriding a `Changeset` method (see [http://artiemestieri.tn.it/~lele/projects/trac-darcs/tracdarcs/changeset.py for example]). Of course, I could have reimplemented the whole `sync()` machinery in a subclass of `CachedRepository`, but DRY approach suggests that's not the right way. ",Lele Gaifax next-major-releases Release,2702,markdown support,wiki system,0.9.3,enhancement,major,,new,2006-01-31T19:55:45+01:00,2017-04-04T01:55:59+02:00,"It'd be nice to see support for [http://daringfireball.net/projects/markdown/ Markdown] with [http://daringfireball.net/projects/smartypants/ SmartyPants] support. Both are available as PHP libraries, so it shouldn't be difficult to include.",anonymous next-major-releases Release,2695,[PATCH] Pretty icons for various file types.,version control/browser,0.9.3,enhancement,normal,,new,2006-01-31T06:37:51+01:00,2015-12-19T15:46:12+01:00,"I'd like to propose using different icons for various file-types in the browser view. The Silk icon set at http://www.famfamfam.com/lab/icons/silk/ is ""licensed under a [http://creativecommons.org/licenses/by/2.5/ Creative Commons Attribution 2.5 License] and contains good icons for a number of file-types, including: * http://www.famfamfam.com/lab/icons/silk/icons/folder.png a folder icon * http://www.famfamfam.com/lab/icons/silk/icons/folder_delete.png folder ""deny"" * http://www.famfamfam.com/lab/icons/silk/icons/page_white.png a default file icon * http://www.famfamfam.com/lab/icons/silk/icons/page_white_delete.png file ""deny"" * http://www.famfamfam.com/lab/icons/silk/icons/page_white_acrobat.png Acrobat/PDF * http://www.famfamfam.com/lab/icons/silk/icons/page_white_actionscript.png Action Script * http://www.famfamfam.com/lab/icons/silk/icons/page_white_c.png C * http://www.famfamfam.com/lab/icons/silk/icons/page_white_cplusplus.png C++ * http://www.famfamfam.com/lab/icons/silk/icons/page_white_csharp.png C# * http://www.famfamfam.com/lab/icons/silk/icons/page_white_code.png Code/Markup * http://www.famfamfam.com/lab/icons/silk/icons/page_white_coldfusion.png Cold Fusion * http://www.famfamfam.com/lab/icons/silk/icons/page_white_excel.png Excel * http://www.famfamfam.com/lab/icons/silk/icons/page_white_flash.png Flash * http://www.famfamfam.com/lab/icons/silk/icons/page_white_gear.png Settings/Config (gear) * http://www.famfamfam.com/lab/icons/silk/icons/page_white_h.png .h * http://www.famfamfam.com/lab/icons/silk/icons/page_white_paint.png Paint * http://www.famfamfam.com/lab/icons/silk/icons/page_white_php.png PHP * http://www.famfamfam.com/lab/icons/silk/icons/page_white_ruby.png Ruby * http://www.famfamfam.com/lab/icons/silk/icons/page_white_visualstudio.png Visual Studio * http://www.famfamfam.com/lab/icons/silk/icons/page_white_word.png Word * http://www.famfamfam.com/lab/icons/silk/icons/page_white_world.png World/Globe * ...and others. Even if no one likes my idea for the different file types, the basic folder, file, and deny icons could be used. I've started hacking this together in my own Trac install, but I'm not a python developer, so I'm not sure if it's worth anything to anyone else. I'll be posting a first-attempt patch in just a moment.",Curtis Buys next-major-releases Release,2662,assign tickets to multiple users,ticket system,0.9.3,enhancement,normal,,new,2006-01-25T13:36:02+01:00,2021-09-19T18:37:53+02:00,"It would be very nice, if I could assign a ticket to more than one user. I set the restrict owner-parameter to false and typed in two users. I tried with different syntax, but the trac system did not notice the ticket to belong to the them. Would be fine...",mala next-major-releases Release,2656,Make built-in documentation independent from wiki,wiki system,0.9.3,enhancement,major,,new,2006-01-24T11:13:55+01:00,2019-09-16T01:54:34+02:00,"I noticed that there are various links to the built-in documentation pages in the wiki. We use the wiki for internal documentation and development matters and therefore anonymous users don't have rights to access it. So it would be nice if the online documentation pages would be independant from the wiki, because anonymous users can't view them if they have no right to access the wiki and it is annoying if you have those links to is everywhere, which lead you (as an anonymous user) to pages which say ""Access denied"". A quick workaround would be to only display the links if you have rights to access the wiki. Best regards Florian Lahr",flo@… next-major-releases Release,2639,Add userid/author in RSS feeds for repository commit information,timeline,0.10rc1,enhancement,normal,,new,2006-01-19T20:39:27+01:00,2015-05-26T20:50:57+02:00,"We have an interest in using the RSS feed feature of Trac. Besides the issues described in #2580, which I am very happy to see resolved, we would also like to see the changeset author's name in the RSS feed information. For instance, when subscribing to the Timeline's RSS feed, you would see a title such as: '''{{{Changeset [614]: Fixed reset procedure.}}}''', with a body of '''{{{Fixed reset procedure.}}}'''. There is apparently no author information, although this is a very useful part of the Timline page itself. It appears to me that the revison log RSS feed is also hobbled in this way.",Dan next-major-releases Release,2636,browsing a directory with many files takes a long time,version control/browser,0.9.3,enhancement,normal,,new,2006-01-19T13:05:15+01:00,2015-05-26T20:50:57+02:00,"when browsing a repository browser with a lot of files (tested with 3370) it takes a long time (minutes) before the files are shown in the browser.[[BR]] '''enhancement:''' It would be better to limit the amount of files processed, as to process them in blocks (configurable, default 250?) tested with gentoo-linux on a P4 2.4GHz 512MB RAM with apache and fastcgi (5 preforks) as interface to trac (0.9.3) (all processing power was available to trac at that moment)[[BR]] client is windows xp with firefox 1.5.0.1 ",joramagten@… next-major-releases Release,2625,Add option to send HTML mail messages,notification,0.11-stable,enhancement,normal,,new,2006-01-18T01:06:19+01:00,2023-09-15T04:02:49+02:00,"It would be nice if Trac was able to send notification emails in some sort of rich format, for example HTML. This way one would not have to use fixed fonts in the mail client.",tomek.piatek@… next-major-releases Release,2588,Option to show the latest completed milestone,roadmap,devel,enhancement,minor,,new,2006-01-12T02:28:44+01:00,2018-12-16T15:45:29+01:00,"It would be nice if another option were added to the ""show already completed milestones"" option: it would display the latest completed milestone - or the n last completed milestones, so that users can see the latest milestone achievement and features even if this milestone has just been reached/completed. Up to now, the only option to show the recently completed milestone is to display all milestones, which is far too coarse selection as it can lead to display a very long list of old and ""useless"" info. Many users want to browse the latest provided features in the days that follow a milestone completion.",Emmanuel Blot next-major-releases Release,2469,timeline should show my own tickets in bold,timeline,0.9b2,enhancement,major,,new,2005-12-09T20:23:19+01:00,2015-05-26T20:50:57+02:00,"Everyday when I start working, I check the Timeline to see what's up. Someone may have changed or commented on a ticket I'm working on. So, the Timeline should display my own tickets in bold (or other visual clue). btw, thanks for the great product. we started using it in october, and all the team is pleased.",rodolfo.borges@… next-major-releases Release,2465,"Add ""custom query"" functionality for changesets",version control/changeset view,none,enhancement,critical,,new,2005-12-09T11:54:06+01:00,2015-05-26T20:50:57+02:00,"I think it would be really useful to have the ability to quickly define queries on changesets, similarily to what can be done on tickets (ie. with a GUI allowing filters on changeset comments, author, file path and commit dates).[[BR]] This page should also implement RSS feed (as Ticket custom queries), to allow notification of specific changes. [[BR]] I suppose that the implementation would be similar to the ticket 'custom query' page (but I havn't looked in depth at the code, not being a Python developer... yet). I leave it to the TRAC people to set milestone, priority, etc. Many Thanks for this great software BTW.",guillaume.tardif@… next-major-releases Release,2464,Conditional fields in tickets,ticket system,0.9.2,enhancement,major,,new,2005-12-09T11:34:48+01:00,2021-08-19T21:34:04+02:00,"Not necessarily all possible fields should be presented to the user. There could be preconditions attached to fields, for checking if they're applicable at all and for checking if the current user can actually set/modify them. Then, as an advanced case, instead of being an all or nothing situation, the possible content of one field could be dependent on the ticket state: see #1299, #2752 and [#comment:13]. === Use cases === ==== Ability to have different ticket fields for different ticket types ==== ''(original one)'' When using quite different types of tickets (bug-report, requirement, etc), there is a need for different fields depending on the ticket type. This might add just the flexibility required for complying with existing development processes. Great feature in particular when support for master and child tickets is planned, trac could be more of a requirements management system beyond the issue database approach. See also duplicates: #4028, #4643 ==== Abilty to have different ticket fields for different ticket components ==== As for ticket types, but for components. See #2752 ==== Hide the ''Keywords:'' field ==== See #3281 ==== Hide some fields for new tickets ==== See #1580. This could be achieved quite simply by putting the ticket `id` in the condition. ",johan.regnell@… next-major-releases Release,2456,Implement API for user management,general,0.9.2,enhancement,major,,new,2005-12-08T02:06:33+01:00,2015-05-26T20:50:57+02:00,"I'm adding this ticket to keep track of the on-going discussion about creating a new 'user detail' interface to Trac. The new interface would be something like: {{{ #!python class IUserProvider(Interface): def get_known_users(attrs=('name', 'email'), limit=None): pass def get_user_attribute(user, attr): pass def get_supported_attributes(): pass }}} Related ML thread: * http://lists.edgewall.com/archive/trac/2005-December/005771.html ",Emmanuel Blot next-major-releases Release,2382,WebAdmin: resync from the web interface,version control,0.9,enhancement,normal,,new,2005-11-20T10:24:21+01:00,2015-05-26T20:50:57+02:00,"It would be nice to be able to resynchronize from the web interface. I've written a WebAdmin component that does this for any repository that supports the 'sync' method. (Should we have a WebAdmin component for tickets?)",Manuzhai next-major-releases Release,2141,Upload more than one attachment at a time,attachment,devel,enhancement,normal,,new,2005-09-28T16:41:04+02:00,2015-07-14T20:29:06+02:00,It would be nice to be able to upload and delete multiple attachments.,anonymous next-major-releases Release,2130,Sort Milestones in Roadmap and NewTicket alphabetically (as an option),roadmap,,enhancement,normal,,new,2005-09-27T13:29:18+02:00,2021-07-24T22:54:41+02:00,"I use Trac for different components with different version number. The milestones are sorted in the sequence I added them plus a sort over the date that is set. It would be very nice to get them optional sorted alphabetically. The milestone listbox in the ""New Ticket"" looks now as the following example: * ComponentA 1.0 * ComponentB 2.1 * ComponentC 0.3 * ComponentC 0.4 * ComponentA 1.1 * ComponentB 2.3 I want to have an option in the ""Roadmap"" to sort it alphabetically and my be as default in the ""New Ticket"" settings. Then it should look like the following: * ComponentA 1.0 * ComponentA 1.1 * ComponentB 2.1 * ComponentB 2.3 * ComponentC 0.3 * ComponentC 0.4",jan.linnenkohl@… next-major-releases Release,2036,Inconsistent sub-menu on Browse Source comparing with View Tickets,report system,0.9b1,defect,minor,,new,2005-09-09T19:44:51+02:00,2015-05-26T20:50:57+02:00,"Menu ''View Tickets'' contains two sub-items: * ''Available Reports'' * ''Custom Query'' An item are not clickable when active. In opposite, ''Browse Source'' contains one sub-item that changes its value depending on the active selection: * ''Revision Log'' * ''View Latest Revision'' This needs to be unified in either way.",pkou at ua.fm next-major-releases Release,1804,[Patch] Strip off realm from usernames when using Kerberos,general,0.8.4,defect,normal,,new,2005-07-19T15:48:02+02:00,2018-12-31T09:58:38+01:00,"When using Kerberos authentication, Trac will use USER@REALM as the username. For example, the default template will show at the top ""logged as ME@REALM.EXAMPLE.COM"". It would be great to tell Trac about a default realm (in the example, REALM.EXAMPLE.COM) which, when present, would be stripped off. The simplest way to do this would be to offer a new attribute, shortauthname, so that templates can be tweaked to use that rather than trac.authname. The next step would be to change the codebase and the templates to use the short name when applicable. I am not familiar with the Trac internals, so I have no idea how hard or invasive that would be.",nutello@… next-major-releases Release,1796,Stylesheet for Docutils output,rendering,0.8.4,enhancement,minor,,new,2005-07-16T00:07:20+02:00,2015-05-26T20:50:57+02:00,"{{{ #!rst Some output elements (e.g. line blocks or admonitions) of reStructuredText look really crappy without a stylesheet. May I suggest you store tools/stylesheets/default.css from the Docutils tarball at /trac_common/css/docutils.css and add ``@import url(/trac_common/css/docutils.css)`` in front of the other import statements? The Docutils stylesheet shouldn't interfere with any elements outside the Docutils-generated HTML code. }}}",Felix Wiemann next-major-releases Release,1520,Highlight TracLinks in browser,wiki system,0.8.1,enhancement,normal,,new,2005-05-08T11:45:46+02:00,2015-05-26T20:50:57+02:00,"It would be nice to highlight TracLinks in browser file view. For example to display ticket links when viewing source:trunk/ChangeLog@latest. //This is already possible by specifying text/x-trac-wiki as the mimetype for this kind of files// Another example would be to wiki format the inline Wiki syntax present in comments (this needs some cooperation from the syntax highlighter, e.g. the Pygments renderer).",shad next-major-releases Release,1513,Multi-language support in the wiki,wiki system,,enhancement,normal,,new,2005-05-06T16:14:51+02:00,2015-05-26T20:50:57+02:00,"Let trac be the first-ever multi-lingual wiki on the market! Have the ability to translate wiki articles into different languages. A user can select their language when they view the wiki. We have looked around and there isnt a wiki out there that handles this. And if you dont have time to do it, do you mind having some volunteers help you out? :)",Paul Baranowski next-major-releases Release,1329,Ticket Search should (optionally) search only non-closed tickets,ticket system,0.8.1,enhancement,major,,new,2005-03-21T09:19:37+01:00,2015-05-26T20:50:57+02:00,"One of the popular uses for searching (I believe) is to search the ticket database if an issue has already been filed. It would help if the search would only search tickets that have not been closed. This request is similar to Ticket #584 but I would like a way to filter the results so that only open tickets are displayed in the search result. It does not help much if you need to browse loads of resolved tickets to find the one open ticket you are looking for. ''fixed typos''",Norbert Unterberg next-major-releases Release,1242,[ER] Generalized automatic cross-references in Trac,wiki system,0.8,enhancement,critical,,new,2005-03-03T11:13:58+01:00,2022-11-16T06:31:00+01:00,"== Problem == Trac provides a fast, easy and flexible way to relate Trac objects together, by the use of TracLinks in the WikiFormatting. However, there's no automatic support for back-linking. This lack has been noticed in several circumstances: * BackLinks for Wiki pages (tickets #611 and #646, the latter containing a [http:/trac/ticket/646#change_4 note] which also highlights the genericity of the problem) * Related Changesets for a given Ticket (ticket #508) * Ticket dependencies and other relationships (tickets #31 and #226) Instead, the backlinking is done manually. For example, when closing a ticket with a resolution of {{{duplicate}}}, the convention is to comment the closing of the ticket by saying ''this is a duplicate of #xyz'' and to mention the existence of the duplicate ticket on the original #xyz ticket. With a generalized backlinking facility, this would cease to be necessary. == Solution == What I would propose is that every relationship that gets detected while parsing new wiki content is recorded. Then, each wiki, changeset and ticket page could contain a '''See also:''' section at the end, listing the other objects that contain TracLinks targeting it, along with a bit of context, to catch the semantic of the link as well. ---- Example given: Consider the following wiki !OutstandingPatches page: {{{ = Outstanding Patches page = I think that the patches #187, #909, #919, #944 and #948 are great! }}} Then, each time this page is edited, the content of an {{{XREF (source, target, context}}} table would be cleared for this page: {{{ delete from XREF where source = 'wiki:OutstandingPatches' }}} and when the content is processed, each time a TracLink is detected, the table would be filled: {{{ insert into XREF values ('wiki:OutstandingPatches', 'ticket:187', 'I think that the patches #187, #909, #919, #944 and #948 are great!') }}} Then, while browsing !#187, one could easily get the backlinks: {{{ select source, context from XREF where target='ticket:187' }}} in order to create the following output at the end of the ticket page: {{{ #!html
    Backlinks: OutstandingPatches
    }}} ---- Of course, I'm willing to provide a patch for that, but I would like to get some feedback before. -- Christian",Christian Boos next-major-releases Release,1132,Use Subversion (or the source repository) for Trac's data as well,general,none,enhancement,critical,,new,2005-01-14T18:02:17+01:00,2015-05-26T20:50:57+02:00,Use Subversion for data repository like [http://subwiki.tigris.org/ subwiki],twenim@… next-major-releases Release,978,Specify InterTrac context for a repository,version control,devel,enhancement,normal,,new,2004-11-22T11:22:12+01:00,2022-07-19T03:21:36+02:00,"This topic was already mentioned on the list of features for the 2.0 Roadmap, so I thought I should created a ticket for it in order to comment on that topic. Actually, [http://svk.elixus.org SVK] ''is'' supported, out of the box, because a SVK repository is just a regular Subversion repository. Simply configuring a Trac environment to use the SVK repository just works. However, starting from this basic level of support, several SVK-specific enhancements could be made: * show the list of mirrors * when displaying a changeset, show if it is one generated by SVK (like those created while synchronizing a mirrored repository) * annotate branches with their merge tickets * ... ",Christian Boos next-major-releases Release,918,[patch] Custom Ticket Fields should support the multiple selection type,ticket system,devel,enhancement,normal,,new,2004-11-12T19:42:28+01:00,2018-05-12T08:56:20+02:00,"I propose an addition to the TracTicketsCustomFields. Currently, there is support for the simple selection field: {{{ .type = select }}} I would like to have the {{{multi}}} type as well, for multiple selection fields. The usage in {{{trac.ini}}} for {{{multi}}} would be the same as it is for {{{select}}}. I implemented it in a crude way (see the accompanying patch). I'm not entirely pleased with the patch: * the support for query is partial (i.e. it won't work) * the display of the fields is not as neat as I would like A better solution would involve a change in the database: drop the unicity constraint on the {{{(id,name)}}} pair for the {{{ticket_custom}}} table. With that, the query support would be easier to add. ",Christian Boos next-major-releases Release,791,"login/logout, authentication, and authorization",general,devel,enhancement,normal,,new,2004-09-30T06:13:49+02:00,2022-08-07T15:41:21+02:00,"I'm kind of combining a couple of things from #599 and #788 here. The main issue is that the browser caches the authentication information, so you can't log out unless you close the browser (or with Mozilla you can use certain extensions to clear the information). A related issue is that in situations where authentication is enforced to access the project, users should never be logged in anonymously. I've already gotten a patch working that will force users who logged out to log back in, allowing them to login as a different user. I'm going to try to extend this to allow users to disable anonymous access and handle these setups more appropriately.",Matthew Good next-major-releases Release,756,trac-admin initenv might prompt for a wiki-default path,admin/console,devel,enhancement,normal,,new,2004-09-10T16:19:22+02:00,2015-05-26T20:50:57+02:00,"Trac-admin prompts for default template path, but not for default wiki page path. The following patch supports wiki-default path on command line and prompts for such a path when {{{initenv}}} command is invoked. === Proposed patch === {{{ #!diff Index: scripts/trac-admin =================================================================== --- scripts/trac-admin (revision 915) +++ scripts/trac-admin (working copy) @@ -1,4 +1,4 @@ -#!/usr/bin/env python +#!/usr/bin/python # -*- coding: iso8859-1 -*- __author__ = 'Daniel Lundin , Jonas Borgström ' __copyright__ = 'Copyright (c) 2004 Edgewall Software' @@ -427,7 +427,7 @@ ## Initenv _help_initenv = [('initenv', 'Create and initialize a new environment interactively'), - ('initenv ', + ('initenv ', 'Create and initialize a new environment from arguments')] def do_initdb(self, line): @@ -460,6 +460,13 @@ dt = trac.siteconfig.__default_templates_dir__ prompt = 'Templates directory [%s]> ' % dt returnvals.append(raw_input(prompt) or dt) + print + print ' Please enter location of Trac initial wiki pages.' + print ' Default is the location of the site-wide wiki pages installed with Trac.' + print + dw = trac.siteconfig.__default_wiki_dir__ + prompt = 'Wiki pages directory [%s]> ' % dw + returnvals.append(raw_input(prompt) or dw) return returnvals def do_initenv(self, line): @@ -470,18 +477,21 @@ project_name = None repository_dir = None templates_dir = None + wiki_dir = None if len(arg) == 1: returnvals = self.get_initenv_args() project_name = returnvals[0] repository_dir = returnvals[1] templates_dir = returnvals[2] - elif len(arg)!= 3: + wiki_dir = returnvals[3] + elif len(arg)!= 4: print 'Wrong number of arguments to initenv %d' % len(arg) return else: project_name = arg[0] repository_dir = arg[1] templates_dir = arg[2] + wiki_dir = arg[3] from svn import util, repos, core core.apr_initialize() pool = core.svn_pool_create(None) @@ -498,6 +508,9 @@ or not os.access(os.path.join(templates_dir, 'ticket.cs'), os.F_OK): print templates_dir, ""doesn't look like a Trac templates directory"" return + if not os.access(os.path.join(wiki_dir, 'WikiStart'), os.F_OK): + print wiki_dir, ""doesn't seem to contain initial Wiki pages"" + return try: print 'Creating and Initializing Project' self.env_create() @@ -516,7 +529,7 @@ # Add a few default wiki pages print ' Installing wiki pages' cursor = cnx.cursor() - self._do_wiki_load(trac.siteconfig.__default_wiki_dir__,cursor) + self._do_wiki_load(wiki_dir,cursor) print ' Indexing repository' sync.sync(cnx, rep, fs_ptr, pool) }}} ",eblotml@… next-major-releases Release,695,Keywords for wiki pages,wiki system,0.7.1,enhancement,major,,new,2004-08-11T16:49:42+02:00,2021-11-26T13:29:21+01:00,"It would be nice to see a keywords field for wiki pages, same as the one for tickets. This list of manually chosen keywords can serve multiple purposes: * for searching and ranking, grouping, ... * to automatically generate the ""See also"" list of links at the bottom of a page, listing all other pages that share 1 or more of the keywords. Currently this seems to be done manually, so if a wiki page changes such that it should not be related anymore, then one would have to hunt down and change all other pages that ""See also"" to it. ",mario at ruggier.org next-major-releases Release,611,Backlinks,wiki system,0.7.1,enhancement,major,,new,2004-07-11T14:52:44+02:00,2021-04-07T13:43:14+02:00,"A nice feature would be automatic backlink generation for wikipages. if there is intrest in this sugested enhancement i would like to contribute in coding this feature (wih a little guidance) keep up the good work!",siegfried@… next-major-releases Release,515,Collapsible diff display,version control/changeset view,0.7.1,enhancement,normal,,new,2004-06-04T23:14:55+02:00,2023-02-26T06:25:47+01:00,It would be nice to have a button on each diff section to collapse each the diff section/file. In this way you could display only the file diff you are interested in rather than having to display all the diffs.,jamesm@… next-major-releases Release,425,Wiki page name need to support digits and periods,wiki system,0.8,enhancement,normal,,new,2004-05-19T15:00:40+02:00,2015-03-02T22:31:46+01:00,"Restricting Wiki page names to only alphabetic chars makes it very have to create pages that discuss MilestoneV1.0 etc Is there any problem with allowing digits and period? ",barry.scott@… next-major-releases Release,221,Creating TR for multiple components,ticket system,,enhancement,critical,,new,2004-04-02T02:41:40+02:00,2020-09-09T13:20:24+02:00,"Its sometimes necessary to fix a bug in both a branch release and the mainline trunk. It would be very useful to have a method by which a ticket can be assigned to multiple releases ---- Note: see rather #4298 for the association to multiple versions (or milestones). This ticket is focused on the association to multiple '''components''' (cf. comment:10)",banerjed@… next-major-releases Release,13223,"""All Tickets Reported by Me"" should mention if a ticket is open or closed",ticket system,,enhancement,trivial,,new,2019-11-17T19:57:48+01:00,2019-11-17T20:28:47+01:00,"""{9} All Tickets Reported by Me "" should mention if a ticket is open or closed. Powered by Trac 1.2.2 Quite important, vs. other items it mentions. Do add a column.",jidanni@… next-major-releases Release,13101,Unified diff in git format?,version control/changeset view,,enhancement,minor,,new,2018-11-06T12:52:25+01:00,2018-11-06T13:38:25+01:00,"Is the diff format of ""Download in other formats: Unified Diff"" in changeset view configurable? I'm no expert on the various diff formats. wikipedia:Diff#Unified_format says: > There are some modifications and extensions to the diff formats that are used and understood by certain programs and in certain contexts. For example, some revision control systems—such as Subversion—specify a version number, ""working copy"", or any other comment instead of or in addition to a timestamp in the diff's header section. > > Some tools allow diffs for several different files to be merged into one, using a header for each modified file that may look something like this: > {{{ > Index: path/to/file.cpp > }}} Trac includes such `Index:` headers. Many tools (GNU `patch`, `hg import`, `git apply`) require a `-p0` switch to understand this. It would be convenient if Trac could also create diffs in a different format that does not require remembering special switches. [https://git-scm.com/docs/diff-format git diff-format] typically uses a different format(?) with `a/` / `b/` prefixes and ""extended header lines"". [https://www.mercurial-scm.org/doc/hgrc.5.html#diff Mercurial] also supports ""git extended diff format"".",anonymous next-major-releases Release,11977,Attachment title in search results does not render closed ticket with strikethrough,attachment,,defect,normal,,new,2015-03-09T05:17:47+01:00,2015-07-11T07:53:03+02:00,"Ticket #10833 is closed, but the attachment title does not indicate that. [[Image(NotStrikethrough.png)]]",Ryan J Ollos next-major-releases Release,11816,Some comment TracLinks are incorrectly displayed as missing,wiki system,,defect,minor,,new,2014-11-11T02:39:16+01:00,2022-09-23T13:44:11+02:00,"It can be seen in comment:3:ticket:11813 that the following two TracLinks are incorrectly displayed as missing links. * comment:2:ticket:956 ([/ticket/956#comment:2]) * comment:1:ticket:1752 ([/ticket/1752#comment:1]) Recent changes to comment TracLinks were made in #10742.",Ryan J Ollos next-major-releases Release,11493,Add config option to show changeset files in ticket's comment,ticket system,1.0-stable,enhancement,minor,,new,2014-02-21T02:50:45+01:00,2017-09-09T00:45:38+02:00,"I want to confirm changeset files in comment related to the ticket. It's useful to find a particular file related to the ticket. I made a patch to do this. How does that sound? {{{ #!ini [ticket] commit_ticket_reference_show_files = true }}} [[Image(show-changeset-files-in-comment.png)]]",t2y next-major-releases Release,11013,Unable to hotcopy SQLite database on Windows if >1Gb,admin/console,0.12,defect,normal,,new,2013-01-10T10:16:41+01:00,2013-01-23T09:19:40+01:00,"System Information: - Trac 0.12 - Windows XP 64bits SP2 (should be reproduceable on any version of Windows) How To Reproduce: Assuming your Trac env use a large (>1Gb) SQLite database, when you use ""trac-admin hotcopy ENV"" (while Trac is running), the copy of the database fails at 1Gb. ''Why :'' On Windows, when a SQLite DB is locked (eg, during a transaction), database file has an exclusive lock between 1Gb and 1Gb + 512bytes (see http://www.mail-archive.com/sqlite-users@sqlite.org/msg08077.html), so hotcopy fails to read it. ''Proposition :'' When SQLite backend is used, skip the database file during the hotcopy process and use something like ""sqlite3 trac.db .backup trac.db.bak"" instead.",pdecourcel@… next-major-releases Release,10945,"horizontal scrollbar in ""editing wiki"" textareas",wiki system,,enhancement,minor,,new,2012-11-08T16:15:04+01:00,2015-11-18T12:51:12+01:00,"Hi there, Could you please add an horizontal scrollbar to ""Editing wiki"" textareas i.e. could you please change {{{