| 1 | import re |
|---|
| 2 | import os |
|---|
| 3 | import urllib |
|---|
| 4 | |
|---|
| 5 | try: |
|---|
| 6 | from cStringIO import StringIO |
|---|
| 7 | except ImportError: |
|---|
| 8 | from StringIO import StringIO |
|---|
| 9 | |
|---|
| 10 | from trac import util |
|---|
| 11 | from trac.mimeview import * |
|---|
| 12 | from trac.wiki.api import WikiSystem |
|---|
| 13 | from trac.wiki.formatter import Formatter |
|---|
| 14 | from trac.util import shorten_line, to_unicode |
|---|
| 15 | |
|---|
| 16 | __all__ = ['wiki_to_latex'] |
|---|
| 17 | |
|---|
| 18 | def pretexfilter(s, escape=True): |
|---|
| 19 | """ |
|---|
| 20 | Escape LaTeX seqeuences in \a s that do not affect Wiki formatting |
|---|
| 21 | """ |
|---|
| 22 | s = s.replace('\\', '\textbackslash ') #<lbrace /><rbrace /> ?? |
|---|
| 23 | s = s.replace('{', '\\{{<rbrace />').replace('}', '\\<rbrace />{<rbrace />').replace('<rbrace />', '}') |
|---|
| 24 | s = s.replace('$', '\\${}') |
|---|
| 25 | s = s.replace('>', '\textgreater{}').replace('<', '\textless{}').replace('&', '\myamp{}') |
|---|
| 26 | s = s.replace('&', '\\&{}').replace('%', '\\mypercent{}') #etc |
|---|
| 27 | return s |
|---|
| 28 | |
|---|
| 29 | def gtotex(s, escape=True): |
|---|
| 30 | """ |
|---|
| 31 | Escape LaTeX seqeuences in \a s that may still be present after processing. |
|---|
| 32 | Favour pretexfilter() because the LaTeX formatting added _during_ processing need to be kept |
|---|
| 33 | """ |
|---|
| 34 | if not s: |
|---|
| 35 | return s |
|---|
| 36 | if escape: |
|---|
| 37 | #if s.find('<'): |
|---|
| 38 | # raise "Error" |
|---|
| 39 | s = re.sub(r'"([^,.) ])', r"``\1", s).replace('"', "''") #tex-ify quote character |
|---|
| 40 | s = s.replace('<', '\\textless{}').replace('>', '\\textgreater{}').replace('%', '\\mypercent{}') #should NEVER be any of these |
|---|
| 41 | s = s.replace('>', '\textgreater{}').replace('<', '\textless{}') |
|---|
| 42 | s = re.sub(r'^#', r'\\#{}', s) |
|---|
| 43 | s = re.sub(r'([^\\$])#', r'\1\\#{}', s) #replace hashes, if not already preceded by a backslash or $ |
|---|
| 44 | s = re.sub(r'^_', r'\\_{}', s) |
|---|
| 45 | s = re.sub(r'([^\\$])_', r'\1\\_{}', s) #replace underscores, if not already preceded by a backslash or $ |
|---|
| 46 | s = re.sub(r'^\^', '\\\\textasciicircum{}', s) |
|---|
| 47 | s = re.sub(r'([^$])^', r'\1\\\\textasciicircum{}', s) |
|---|
| 48 | return to_unicode(s) |
|---|
| 49 | |
|---|
| 50 | def gfilter_to_url(s, escape=True): |
|---|
| 51 | """ |
|---|
| 52 | Filter out things escaped by pretexfilter that \url is able to cope with |
|---|
| 53 | """ |
|---|
| 54 | if not s: |
|---|
| 55 | return s |
|---|
| 56 | s = s.replace('\\_{}', '_').replace('\\%{}', '%').replace('\\&{}amp;', '&') |
|---|
| 57 | return s |
|---|
| 58 | |
|---|
| 59 | def LatexURL(formatter, ns, target, text, escape=True): |
|---|
| 60 | """ |
|---|
| 61 | Create a LaTeX \url with a given namespace, target and link text |
|---|
| 62 | """ |
|---|
| 63 | if text == target: |
|---|
| 64 | return '\\url{' + (ns and ns + ':' or "") + gfilter_to_url(target) + '}'; |
|---|
| 65 | return '\\anchortext{' + (gtotex(text) or gtotex(target) or "broken") + '}\\footnote{\\url{' + (ns and ns + ':' or "") + gfilter_to_url(target) + '}}'; |
|---|
| 66 | |
|---|
| 67 | def WikiLatexURL(formatter, ns, target, text, escape=True): |
|---|
| 68 | """ |
|---|
| 69 | Create a LaTeX \url with a given namespace, target and link text |
|---|
| 70 | """ |
|---|
| 71 | if text == target: |
|---|
| 72 | return '\\anchortext{' + gtotex(target) + '} (\\S\\ref{sub:' + totexlabel(target) + '})' |
|---|
| 73 | return '\\anchortext{' + (gtotex(text) or gtotex(target) or "broken") + '} (\\S\\ref{sub:' + totexlabel(target) + '})\\footnote{\\url{' + (ns and ns + ':' or "") + gfilter_to_url(target) + '}}'; |
|---|
| 74 | |
|---|
| 75 | def totexlabel(s, escape=True): |
|---|
| 76 | """ |
|---|
| 77 | Create a valid LaTeX for \ref cross references |
|---|
| 78 | """ |
|---|
| 79 | if not s: |
|---|
| 80 | return s |
|---|
| 81 | return s.replace('&', '').replace('$', '').replace(' ', '-').replace('"', '').replace("'", '').replace('#', ':').replace('{', '').replace('}', '').replace('~', '-') |
|---|
| 82 | |
|---|
| 83 | class LatexFormatter(Formatter): |
|---|
| 84 | WRITE_SECTION = 0 |
|---|
| 85 | DELAY_SECTION = 1 |
|---|
| 86 | DONE_SECTION = 2 |
|---|
| 87 | |
|---|
| 88 | flavor = 'latex' |
|---|
| 89 | |
|---|
| 90 | def __init__(self, env, page, req=None, absurls=0, db=None, sectionlevel=""): |
|---|
| 91 | """ |
|---|
| 92 | Constructor |
|---|
| 93 | |
|---|
| 94 | \param sectionlevel set to null string to promote to section/subsection |
|---|
| 95 | """ |
|---|
| 96 | Formatter.__init__(self, env, req, absurls, db) |
|---|
| 97 | self.pagename = page.name |
|---|
| 98 | self.sublevel=sectionlevel |
|---|
| 99 | self.inmacro = False |
|---|
| 100 | self.beginsection = self.WRITE_SECTION |
|---|
| 101 | |
|---|
| 102 | def _get_link_resolvers(self): |
|---|
| 103 | return {'link':LatexURL, 'source':LatexURL, 'wiki':WikiLatexURL, 'ticket':LatexURL, 'changeset':LatexURL} #WikiSystem(self.env).link_resolvers |
|---|
| 104 | link_resolvers = property(_get_link_resolvers) |
|---|
| 105 | |
|---|
| 106 | def _get_db(self): |
|---|
| 107 | if not self._db: |
|---|
| 108 | self._db = self.env.get_db_cnx() |
|---|
| 109 | return self._db |
|---|
| 110 | db = property(fget=_get_db) |
|---|
| 111 | |
|---|
| 112 | def _get_rules(self): |
|---|
| 113 | return self.wiki.rules |
|---|
| 114 | rules = property(_get_rules) |
|---|
| 115 | |
|---|
| 116 | def replace(self, fullmatch): |
|---|
| 117 | for itype, match in fullmatch.groupdict().items(): |
|---|
| 118 | if match and not itype in self.wiki.helper_patterns: |
|---|
| 119 | # Check for preceding escape character '!' |
|---|
| 120 | if match[0] == '!': |
|---|
| 121 | return gtotex(pretexfilter(match[1:])) |
|---|
| 122 | if match[0:4] == "http": |
|---|
| 123 | return LatexURL(self, 'http', match, match) |
|---|
| 124 | elif match[0].isalpha(): |
|---|
| 125 | return WikiLatexURL(self, 'wiki', match, match) |
|---|
| 126 | if itype in self.wiki.external_handlers: |
|---|
| 127 | return self.wiki.external_handlers[itype](self, match, fullmatch) |
|---|
| 128 | else: |
|---|
| 129 | return getattr(self, '_' + itype + '_formatter')(match, fullmatch) |
|---|
| 130 | |
|---|
| 131 | def _tickethref_formatter(self, match, fullmatch): |
|---|
| 132 | """ |
|---|
| 133 | This is from a patch for Trac 0.8 \todo how to call for 0.9+ (via link resolvers?) |
|---|
| 134 | """ |
|---|
| 135 | number = int(match[1:]) |
|---|
| 136 | cursor = self.db.cursor () |
|---|
| 137 | cursor.execute('SELECT summary,status FROM ticket WHERE id=%s', number) |
|---|
| 138 | row = cursor.fetchone () |
|---|
| 139 | if not row: |
|---|
| 140 | return '\\#%d (missing)' % (number) |
|---|
| 141 | else: |
|---|
| 142 | summary = gtotex(self.prefilter(util.shorten_line(row[0]))) |
|---|
| 143 | if row[1] == 'new': |
|---|
| 144 | return '\\#%d*\\footnote{%s \\emph{(new)}}' % (number, summary) |
|---|
| 145 | elif row[1] == 'closed': |
|---|
| 146 | return '\\sout{\\#%d}\\footnote{%s \\emph{(closed)}}'% (number, summary) |
|---|
| 147 | else: |
|---|
| 148 | return '\\#%d\\footnote{%s}' % (number, summary) |
|---|
| 149 | |
|---|
| 150 | def _bolditalic_formatter(self, match, fullmatch): |
|---|
| 151 | #italic = ('\\emph{', '}') |
|---|
| 152 | italic = ('\\bfseries\\itshape ', '') |
|---|
| 153 | italic_open = self.tag_open_p(italic) |
|---|
| 154 | tmp = '' |
|---|
| 155 | if italic_open: |
|---|
| 156 | tmp += italic[1] |
|---|
| 157 | self.close_tag(italic[1]) |
|---|
| 158 | tmp += self._bold_formatter(match, fullmatch) |
|---|
| 159 | if not italic_open: |
|---|
| 160 | tmp += italic[0] |
|---|
| 161 | self.open_tag(*italic) |
|---|
| 162 | return tmp |
|---|
| 163 | |
|---|
| 164 | def _unquote(self, text): |
|---|
| 165 | if text and text[0] in "'\"" and text[0] == text[-1]: |
|---|
| 166 | return text[1:-1] |
|---|
| 167 | else: |
|---|
| 168 | return text |
|---|
| 169 | |
|---|
| 170 | def _shref_formatter(self, match, fullmatch): |
|---|
| 171 | ns = fullmatch.group('sns') |
|---|
| 172 | target = self._unquote(fullmatch.group('stgt')) |
|---|
| 173 | return self._make_link(ns, target, match, match) |
|---|
| 174 | |
|---|
| 175 | def _make_link(self, ns, target, match, label): |
|---|
| 176 | if ns in self.link_resolvers: |
|---|
| 177 | return self.link_resolvers[ns](self, ns, target, label, False) |
|---|
| 178 | elif target.startswith('//') or ns == "mailto": |
|---|
| 179 | return self._make_ext_link(ns+':'+target, label) |
|---|
| 180 | else: |
|---|
| 181 | return gtotex(match) |
|---|
| 182 | |
|---|
| 183 | def _make_ext_link(self, url, text, title=''): |
|---|
| 184 | same = text == url |
|---|
| 185 | url = gfilter_to_url(url) |
|---|
| 186 | ref = totexlabel(url) |
|---|
| 187 | text, title = gtotex(text), gtotex(title) |
|---|
| 188 | if Formatter.img_re.search(url) and self.flavor != 'oneliner': |
|---|
| 189 | return '\\url{%s} (extimage - todo: make figure float with caption %s)' % ( |
|---|
| 190 | url, title or text) |
|---|
| 191 | if not url.startswith(self._local): |
|---|
| 192 | if same: |
|---|
| 193 | return '\\url{' + url + '}' |
|---|
| 194 | return '\\anchortext{%s}\\footnote{\\url{%s}}' % (text, url) |
|---|
| 195 | else: |
|---|
| 196 | if same: |
|---|
| 197 | return '\\url{' + url + '}' |
|---|
| 198 | return '\\anchortext{%s}\\footnote{ext: \\url{%s}}' % (text, url) |
|---|
| 199 | |
|---|
| 200 | def _make_relative_link(self, url, text): |
|---|
| 201 | same = text == url |
|---|
| 202 | url = gfilter_to_url(url) |
|---|
| 203 | ref = totexlabel(url) |
|---|
| 204 | text = gtotex(text) |
|---|
| 205 | if Formatter.img_re.search(url) and self.flavor != 'oneliner': |
|---|
| 206 | return '\\url{%s} (relimage - todo: make figure float with caption %s)' % ( |
|---|
| 207 | url, text) |
|---|
| 208 | if not url.startswith(self._local): |
|---|
| 209 | if same: |
|---|
| 210 | return '\\url{' + url + '}' |
|---|
| 211 | return '\\anchortext{%s}\\footnote{\\url{%s}}' % (text, url) |
|---|
| 212 | else: |
|---|
| 213 | if same: |
|---|
| 214 | return '\\url{' + url + '}' |
|---|
| 215 | return '\\anchortext{%s}\\footnote{\\url{%s}} or Subsection \\ref{sub:%s}' % (text, url, ref) |
|---|
| 216 | |
|---|
| 217 | def _bold_formatter(self, match, fullmatch): |
|---|
| 218 | return self.simple_tag_handler('\\textbf{', '}') |
|---|
| 219 | |
|---|
| 220 | def _italic_formatter(self, match, fullmatch): |
|---|
| 221 | return self.simple_tag_handler('\\textit{', '}') |
|---|
| 222 | |
|---|
| 223 | def _underline_formatter(self, match, fullmatch): |
|---|
| 224 | if match[0] == '!': |
|---|
| 225 | return gtotex(pretexfilter(match[1:])) |
|---|
| 226 | else: |
|---|
| 227 | return self.simple_tag_handler('\\underbar{', |
|---|
| 228 | '}') |
|---|
| 229 | |
|---|
| 230 | def _strike_formatter(self, match, fullmatch): |
|---|
| 231 | if match[0] == '!': |
|---|
| 232 | return gtotex(pretexfilter(match[1:])) |
|---|
| 233 | else: |
|---|
| 234 | return self.simple_tag_handler('\\sout{', '}') |
|---|
| 235 | |
|---|
| 236 | def _subscript_formatter(self, match, fullmatch): |
|---|
| 237 | if match[0] == '!': |
|---|
| 238 | return gtotex(pretexfilter(match[1:])) |
|---|
| 239 | else: |
|---|
| 240 | return self.simple_tag_handler('$_{', '}$') |
|---|
| 241 | |
|---|
| 242 | def _superscript_formatter(self, match, fullmatch): |
|---|
| 243 | if match[0] == '!': |
|---|
| 244 | return gtotex(pretexfilter(match[1:])) |
|---|
| 245 | else: |
|---|
| 246 | return self.simple_tag_handler('$^{', '}$') |
|---|
| 247 | |
|---|
| 248 | def _inlinecode_formatter(self, match, fullmatch): |
|---|
| 249 | return '\\texttt{%s}' % gtotex(fullmatch.group('inline')) |
|---|
| 250 | |
|---|
| 251 | def _inlinecode2_formatter(self, match, fullmatch): |
|---|
| 252 | return '\\texttt{%s}' % gtotex(fullmatch.group('inline2')) |
|---|
| 253 | |
|---|
| 254 | def _htmlescape_formatter(self, match, fullmatch): |
|---|
| 255 | return match == "&" and "\\&{}" or \ |
|---|
| 256 | match == "<" and "\\textless{}" or "\\textgreater{}" |
|---|
| 257 | |
|---|
| 258 | def _macro_formatter(self, match, fullmatch): |
|---|
| 259 | name = fullmatch.group('macroname') |
|---|
| 260 | if name in ['br', 'BR']: |
|---|
| 261 | if len(self.current_line) < 7: |
|---|
| 262 | return '\\vspace{1ex}' |
|---|
| 263 | return '\\\\' |
|---|
| 264 | if name in ['TracGuideToc']: |
|---|
| 265 | return |
|---|
| 266 | if name in ['PageOutline']: |
|---|
| 267 | if self.beginsection == self.WRITE_SECTION: |
|---|
| 268 | self.beginsection = self.DELAY_SECTION |
|---|
| 269 | return '\\tableofcontents{}' |
|---|
| 270 | self.inmacro = True |
|---|
| 271 | s = '\\begin{verbatim}\n' + Formatter._macro_formatter(self, match, fullmatch) + '\n\\end{verbatim}\n' |
|---|
| 272 | #self.inmacro = False |
|---|
| 273 | return s |
|---|
| 274 | |
|---|
| 275 | # args = fullmatch.group('macroargs') |
|---|
| 276 | # try: |
|---|
| 277 | # macro = WikiProcessor(self.env, name) |
|---|
| 278 | # return macro.process(self.req, args, 1) |
|---|
| 279 | # except Exception, e: |
|---|
| 280 | # self.env.log.error('Macro %s(%s) failed' % (name, args), |
|---|
| 281 | # exc_info=True) |
|---|
| 282 | # return system_message('Error: Macro %s(%s) failed' % (name, args), e) |
|---|
| 283 | |
|---|
| 284 | def _heading_formatter(self, match, fullmatch): |
|---|
| 285 | match = match.strip() |
|---|
| 286 | self.close_table() |
|---|
| 287 | self.close_paragraph() |
|---|
| 288 | self.close_indentation() |
|---|
| 289 | self.close_list() |
|---|
| 290 | self.close_def_list() |
|---|
| 291 | |
|---|
| 292 | depth = min(len(fullmatch.group('hdepth')), 5) |
|---|
| 293 | heading = match[depth + 1:len(match) - depth - 1] |
|---|
| 294 | |
|---|
| 295 | star = '' |
|---|
| 296 | moresub = '' |
|---|
| 297 | |
|---|
| 298 | if depth > 1 and self.sublevel == '': |
|---|
| 299 | moresub = 'sub' |
|---|
| 300 | if depth > 1 and self.sublevel != '' or depth > 2: |
|---|
| 301 | star = '*' |
|---|
| 302 | |
|---|
| 303 | anchor = text = heading |
|---|
| 304 | sans_markup = re.sub(r'</?\w+(?: .*?)?>', '', text) |
|---|
| 305 | |
|---|
| 306 | #check if valid LaTeX label |
|---|
| 307 | i = 1 |
|---|
| 308 | anchor = anchor_base = anchor.encode('utf-8') |
|---|
| 309 | while anchor in self._anchors: |
|---|
| 310 | anchor = anchor_base + str(i) |
|---|
| 311 | i += 1 |
|---|
| 312 | self._anchors.append(anchor) |
|---|
| 313 | try: |
|---|
| 314 | ##\todo gtotex on the next line should be latex_oneliner |
|---|
| 315 | return '\\' + self.sublevel + moresub + 'subsection' + star + '{\\label{anchor:%s}%s}' % (totexlabel(anchor), gtotex(text)) |
|---|
| 316 | except: |
|---|
| 317 | return '\\' + self.sublevel + moresub + 'subsection' + star + '{Bad Unicode}' |
|---|
| 318 | |
|---|
| 319 | def _indent_formatter(self, match, fullmatch): |
|---|
| 320 | depth = int((len(fullmatch.group('idepth')) + 1) / 2) |
|---|
| 321 | list_depth = len(self._list_stack) |
|---|
| 322 | if list_depth > 0 and depth == list_depth + 1: |
|---|
| 323 | self.in_list_item = 1 |
|---|
| 324 | else: |
|---|
| 325 | self.open_indentation(depth) |
|---|
| 326 | return '' |
|---|
| 327 | |
|---|
| 328 | def _last_table_cell_formatter(self, match, fullmatch): |
|---|
| 329 | return '' |
|---|
| 330 | |
|---|
| 331 | def _table_cell_formatter(self, match, fullmatch): |
|---|
| 332 | self.open_table() |
|---|
| 333 | self.open_table_row() |
|---|
| 334 | if self.in_table_cell: |
|---|
| 335 | return ' & ' |
|---|
| 336 | else: |
|---|
| 337 | self.in_table_cell = 1 |
|---|
| 338 | return '' |
|---|
| 339 | |
|---|
| 340 | def close_indentation(self): |
|---|
| 341 | self.out.write(('\\end{quote}' + os.linesep) * self.indent_level) |
|---|
| 342 | self.indent_level = 0 |
|---|
| 343 | |
|---|
| 344 | def open_indentation(self, depth): |
|---|
| 345 | if self.in_def_list: |
|---|
| 346 | return |
|---|
| 347 | diff = depth - self.indent_level |
|---|
| 348 | if diff != 0: |
|---|
| 349 | self.close_paragraph() |
|---|
| 350 | self.close_indentation() |
|---|
| 351 | self.close_list() |
|---|
| 352 | self.indent_level = depth |
|---|
| 353 | self.out.write(('\\begin{quote}' + os.linesep) * depth) |
|---|
| 354 | |
|---|
| 355 | def _list_formatter(self, match, fullmatch): |
|---|
| 356 | ldepth = len(fullmatch.group('ldepth')) |
|---|
| 357 | depth = int((len(fullmatch.group('ldepth')) + 1) / 2) |
|---|
| 358 | self.in_list_item = depth > 0 |
|---|
| 359 | type_ = ['ol', 'ul'][match[ldepth] == '*'] |
|---|
| 360 | self._set_list_depth(depth, type_) |
|---|
| 361 | return '' |
|---|
| 362 | |
|---|
| 363 | def _definition_formatter(self, match, fullmatch): |
|---|
| 364 | tmp = '' |
|---|
| 365 | if not self.in_def_list: |
|---|
| 366 | tmp += '\\begin{description}\n' |
|---|
| 367 | |
|---|
| 368 | definition = match[:match.find('::')] |
|---|
| 369 | if definition[0] == ' ': |
|---|
| 370 | definition = definition[1:] |
|---|
| 371 | ##\todo next line should be latex_oneliner |
|---|
| 372 | tmp += '\\item[%s] ' % gtotex(definition) |
|---|
| 373 | |
|---|
| 374 | self.in_def_list = True |
|---|
| 375 | return tmp |
|---|
| 376 | |
|---|
| 377 | def close_def_list(self): |
|---|
| 378 | if self.in_def_list: |
|---|
| 379 | self.out.write('\\end{description}\n') |
|---|
| 380 | self.in_def_list = False |
|---|
| 381 | |
|---|
| 382 | def _hl_to_ll(self, l): |
|---|
| 383 | if l == 'ul': |
|---|
| 384 | return 'itemize' |
|---|
| 385 | else: |
|---|
| 386 | return 'enumerate' |
|---|
| 387 | |
|---|
| 388 | def _set_list_depth(self, depth, type_): |
|---|
| 389 | current_depth = len(self._list_stack) |
|---|
| 390 | diff = depth - current_depth |
|---|
| 391 | self.close_table() |
|---|
| 392 | self.close_paragraph() |
|---|
| 393 | self.close_indentation() |
|---|
| 394 | self.check_section() |
|---|
| 395 | if diff > 0: |
|---|
| 396 | for i in range(diff): |
|---|
| 397 | self._list_stack.append(type_) |
|---|
| 398 | self.out.write('\\begin{' + self._hl_to_ll(type_) + '}\n') |
|---|
| 399 | self.out.write('\\item ') |
|---|
| 400 | elif diff < 0: |
|---|
| 401 | for i in range(-diff): |
|---|
| 402 | tmp = self._list_stack.pop() |
|---|
| 403 | self.out.write('\\end{' + self._hl_to_ll(tmp) + '}\n') |
|---|
| 404 | if self._list_stack != [] and type_ != self._list_stack[-1]: |
|---|
| 405 | tmp = self._list_stack.pop() |
|---|
| 406 | self._list_stack.append(type_) |
|---|
| 407 | self.out.write('\n\\end{%s}\n\\begin{%s}\n\\item ' % (self._hl_to_ll(tmp), self._hl_to_ll(type_))) |
|---|
| 408 | if depth > 0: |
|---|
| 409 | self.out.write('\\item ') |
|---|
| 410 | # diff == 0 |
|---|
| 411 | elif self._list_stack != [] and type_ != self._list_stack[-1]: |
|---|
| 412 | tmp = self._list_stack.pop() |
|---|
| 413 | self._list_stack.append(type_) |
|---|
| 414 | self.out.write('\n\\end{%s}\n\\begin{%s}\n\\item ' % (self._hl_to_ll(tmp), self._hl_to_ll(type_))) |
|---|
| 415 | elif depth > 0: |
|---|
| 416 | self.out.write('\\item ') |
|---|
| 417 | |
|---|
| 418 | def close_list(self): |
|---|
| 419 | if self._list_stack != []: |
|---|
| 420 | self._set_list_depth(0, None) |
|---|
| 421 | |
|---|
| 422 | def open_paragraph(self): |
|---|
| 423 | if not self.paragraph_open: |
|---|
| 424 | self.out.write(os.linesep) |
|---|
| 425 | self.paragraph_open = 1 |
|---|
| 426 | |
|---|
| 427 | def close_paragraph(self): |
|---|
| 428 | if self.paragraph_open: |
|---|
| 429 | while self._open_tags != []: |
|---|
| 430 | self.out.write(self._open_tags.pop()[1]) |
|---|
| 431 | self.out.write(os.linesep) |
|---|
| 432 | self.paragraph_open = 0 |
|---|
| 433 | |
|---|
| 434 | def open_table(self): |
|---|
| 435 | if not self.in_table: |
|---|
| 436 | self.close_paragraph() |
|---|
| 437 | self.close_indentation() |
|---|
| 438 | self.close_list() |
|---|
| 439 | self.close_def_list() |
|---|
| 440 | self.in_table = 1 |
|---|
| 441 | self.out.write('\\begin{tabular}{|l|l|l|l|l|l|l|l|l|} \\hline' + os.linesep) |
|---|
| 442 | |
|---|
| 443 | def open_table_row(self): |
|---|
| 444 | if not self.in_table_row: |
|---|
| 445 | self.open_table() |
|---|
| 446 | self.in_table_row = 1 |
|---|
| 447 | self.out.write(os.linesep) |
|---|
| 448 | |
|---|
| 449 | def close_table_row(self): |
|---|
| 450 | if self.in_table_row: |
|---|
| 451 | self.in_table_row = 0 |
|---|
| 452 | if self.in_table_cell: |
|---|
| 453 | self.in_table_cell = 0 |
|---|
| 454 | self.out.write('~\\\\ \\hline' + os.linesep) |
|---|
| 455 | |
|---|
| 456 | def close_table(self): |
|---|
| 457 | if self.in_table: |
|---|
| 458 | self.close_table_row() |
|---|
| 459 | self.out.write('\\end{tabular}' + os.linesep) |
|---|
| 460 | self.in_table = 0 |
|---|
| 461 | |
|---|
| 462 | def handle_code_block(self, line): |
|---|
| 463 | if line.strip() == '{{{': |
|---|
| 464 | self.in_code_block += 1 |
|---|
| 465 | if self.in_code_block == 1: |
|---|
| 466 | self.out.write('\\begin{verbatim}' + os.linesep) |
|---|
| 467 | else: |
|---|
| 468 | self.out.write(line + os.linesep) |
|---|
| 469 | elif line.strip() == '}}}': |
|---|
| 470 | self.in_code_block -= 1 |
|---|
| 471 | if self.in_code_block == 0: |
|---|
| 472 | self.out.write('\\end{verbatim}' + os.linesep) |
|---|
| 473 | else: |
|---|
| 474 | self.out.write(line + os.linesep) |
|---|
| 475 | else: |
|---|
| 476 | self.out.write(line + os.linesep) |
|---|
| 477 | |
|---|
| 478 | def preamble(self): |
|---|
| 479 | self.secname = self.pagename |
|---|
| 480 | self.out.write('\\documentclass{article}' + os.linesep) |
|---|
| 481 | self.out.write('\\usepackage{url}' + os.linesep) |
|---|
| 482 | self.out.write('\\usepackage{ulem}' + os.linesep) |
|---|
| 483 | # self.out.write("""\def\dotuline{\\bgroup |
|---|
| 484 | # \\ifdim\\ULdepth=\\maxdimen % Set depth based on font, if not set already |
|---|
| 485 | # \\settodepth\\ULdepth{(j}\\advance\\ULdepth.4pt\\fi |
|---|
| 486 | # \\markoverwith{\\begingroup |
|---|
| 487 | # \\advance\\ULdepth0.08ex |
|---|
| 488 | # \\lower\\ULdepth\\hbox{\\kern.15em .\\kern.1em}% |
|---|
| 489 | # \\endgroup}\\ULon}""") |
|---|
| 490 | self.out.write("""\\usepackage[pagebackref, bookmarks, pdfpagelabels, |
|---|
| 491 | citecolor=black, linkcolor=black, anchorcolor=blue, filecolor=black, |
|---|
| 492 | bookmarksopen=true, plainpages=false, colorlinks=true, bookmarksnumbered=true, |
|---|
| 493 | pdftitle={%s}, |
|---|
| 494 | pdfauthor={userid}]{hyperref} |
|---|
| 495 | """ % self.pagename) |
|---|
| 496 | self.out.write('\\newcommand{\\anchortext}[1]{\def\ULthickness{.2pt}\\underbar{#1}\def\ULthickness{.4pt}} %this does not appear to work properly' + os.linesep) |
|---|
| 497 | self.out.write('\\newcommand{\\mypercent}{\\%{}}' + os.linesep) |
|---|
| 498 | self.out.write('\\newcommand{\\myamp}{\\&{}}' + os.linesep) |
|---|
| 499 | self.out.write('\\setlength\\parskip{\\medskipamount}' + os.linesep) #this is more like the HTML, and probably what people expect |
|---|
| 500 | self.out.write('\\setlength\\parindent{0pt}' + os.linesep) |
|---|
| 501 | |
|---|
| 502 | self.out.write('\\begin{document}' + os.linesep) |
|---|
| 503 | #self.out.write('\\' + self.sublevel + 'section{\\label{sub:' + self.secname + '}' + self.secname + '}' + os.linesep) |
|---|
| 504 | |
|---|
| 505 | def check_section(self): |
|---|
| 506 | if self.beginsection == self.WRITE_SECTION: |
|---|
| 507 | self.out.write('\\' + self.sublevel + 'section{\\label{sub:' + self.secname + '}' + self.secname + '}' + os.linesep) |
|---|
| 508 | self.beginsection = self.DONE_SECTION |
|---|
| 509 | elif self.beginsection == self.DELAY_SECTION: |
|---|
| 510 | self.beginsection = self.WRITE_SECTION |
|---|
| 511 | |
|---|
| 512 | def format(self, text, out, escape_newlines=False): |
|---|
| 513 | self.out = out |
|---|
| 514 | self._open_tags = [] |
|---|
| 515 | self._list_stack = [] |
|---|
| 516 | |
|---|
| 517 | self.in_code_block = 0 |
|---|
| 518 | self.in_table = 0 |
|---|
| 519 | self.in_def_list = 0 |
|---|
| 520 | self.in_table_row = 0 |
|---|
| 521 | self.in_table_cell = 0 |
|---|
| 522 | self.indent_level = 0 |
|---|
| 523 | self.paragraph_open = 0 |
|---|
| 524 | |
|---|
| 525 | self.preamble() |
|---|
| 526 | |
|---|
| 527 | for line in text.splitlines(): |
|---|
| 528 | self.current_line = line |
|---|
| 529 | # Handle code block |
|---|
| 530 | if self.in_code_block or line.strip() == '{{{': |
|---|
| 531 | self.handle_code_block(line) |
|---|
| 532 | continue |
|---|
| 533 | # Handle Horizontal ruler |
|---|
| 534 | elif line[0:4] == '----': |
|---|
| 535 | self.close_paragraph() |
|---|
| 536 | self.close_indentation() |
|---|
| 537 | self.close_list() |
|---|
| 538 | self.close_def_list() |
|---|
| 539 | self.close_table() |
|---|
| 540 | self.out.write('{\\normalsize \\vspace{1ex} \\hrule width \\columnwidth \\vspace{1ex}}' + os.linesep) |
|---|
| 541 | continue |
|---|
| 542 | # Handle new paragraph |
|---|
| 543 | elif line == '': |
|---|
| 544 | self.close_paragraph() |
|---|
| 545 | self.close_indentation() |
|---|
| 546 | self.close_list() |
|---|
| 547 | self.close_def_list() |
|---|
| 548 | continue |
|---|
| 549 | |
|---|
| 550 | if escape_newlines: |
|---|
| 551 | line += ' [[BR]]' |
|---|
| 552 | self.in_list_item = False |
|---|
| 553 | # Throw a bunch of regexps on the problem |
|---|
| 554 | try: |
|---|
| 555 | result = re.sub(self.rules, self.replace, line) |
|---|
| 556 | except TypeError, (e): |
|---|
| 557 | result = "\\textless Bad Unicode on this line \\textgreater" |
|---|
| 558 | try: |
|---|
| 559 | result += '% ' + e |
|---|
| 560 | except: |
|---|
| 561 | pass |
|---|
| 562 | |
|---|
| 563 | if not self.in_list_item: |
|---|
| 564 | self.close_list() |
|---|
