#2318 closed defect (fixed)
Inconsistent display for Subversion move and copy
Reported by: | Matt Shin | Owned by: | Christian Boos |
---|---|---|---|
Priority: | normal | Milestone: | 0.9.1 |
Component: | version control/changeset view | Version: | 0.9 |
Severity: | normal | Keywords: | |
Cc: | Branch: | ||
Release Notes: | |||
API Changes: | |||
Internal Changes: |
Description
The Trac changeset view appears inconsistent following a restructure of our Subversion repository. I can demonstrate the problem by restructuring the following mini-directory structure:
test1/b test1/d test2/b test2/d test3/b test3/d
Create a new directory "test1/a", and move "test1/b" and "test1/d" to it:
svn mkdir test1/a svn mv test1/b test1/a svn mv test1/d test1/a
Create a new directory "test2/c" and move "test2/b" and "test2/d" to it:
svn mkdir test2/c svn mv test2/b test2/c svn mv test2/d test2/c
Create a new directory "test3/e" and move "test3/b" and "test3/d" to it:
svn mkdir test3/e svn mv test3/b test3/e svn mv test3/d test3/e
The changeset view displays the moves as follow:
(Added) test1/a (Moved) test1/a/b (moved from test1/b) (Removed) test1/b (Added) test2/c (Moved) test2/c/b (moved from test2/b) (Moved) test2/c/d (moved from test2/d) (Removed) test3/b (Removed) test3/d (Moved) test3/e/b (moved from test3/b)
While the operations on "test1", "test2" and "test3" are identical, alphabetical order appears to have some effect on what is displayed? (Clearly, the displays for "test1" and "test3" are both wrong.)
Attachments (1)
Change History (5)
comment:1 by , 19 years ago
Milestone: | → 0.9.1 |
---|---|
Owner: | changed from | to
Status: | new → assigned |
comment:2 by , 19 years ago
Resolution: | → fixed |
---|---|
Status: | assigned → closed |
comment:3 by , 19 years ago
cboos, in case you haven't thrown the test case away yet, how about attaching it as a patch here? Would be nice to get a regression test for this in, although it should of course be minimal.
by , 19 years ago
Attachment: | svn_fs-copyrename-test.patch added |
---|
Patch against r2520 adding a test case following this ticket's description
Reproduced and understood…