1 # This file is part of lyx2lyx
2 # -*- coding: utf-8 -*-
3 # Copyright (C) 2006 José Matos <jamatos@lyx.org>
4 # Copyright (C) 2004-2006 Georg Baum <Georg.Baum@post.rwth-aachen.de>
6 # This program is free software; you can redistribute it and/or
7 # modify it under the terms of the GNU General Public License
8 # as published by the Free Software Foundation; either version 2
9 # of the License, or (at your option) any later version.
11 # This program is distributed in the hope that it will be useful,
12 # but WITHOUT ANY WARRANTY; without even the implied warranty of
13 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
14 # GNU General Public License for more details.
16 # You should have received a copy of the GNU General Public License
17 # along with this program; if not, write to the Free Software
18 # Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
20 """ Convert files to the file format generated by lyx 1.5"""
26 from parser_tools import find_re, find_token, find_token_backwards, find_token_exact, find_tokens, find_end_of, get_value, find_beginning_of, find_nonempty_line
27 from LyX import get_encoding
30 ####################################################################
31 # Private helper functions
33 def find_end_of_inset(lines, i):
34 " Find end of inset, where lines[i] is included."
35 return find_end_of(lines, i, "\\begin_inset", "\\end_inset")
37 def find_end_of_layout(lines, i):
38 " Find end of layout, where lines[i] is included."
39 return find_end_of(lines, i, "\\begin_layout", "\\end_layout")
41 def find_beginning_of_layout(lines, i):
42 "Find beginning of layout, where lines[i] is included."
43 return find_beginning_of(lines, i, "\\begin_layout", "\\end_layout")
45 # End of helper functions
46 ####################################################################
50 # Notes: Framed/Shaded
53 def revert_framed(document):
54 "Revert framed notes. "
57 i = find_tokens(document.body, ["\\begin_inset Note Framed", "\\begin_inset Note Shaded"], i)
61 document.body[i] = "\\begin_inset Note"
69 roman_fonts = {'default' : 'default', 'ae' : 'ae',
70 'times' : 'times', 'palatino' : 'palatino',
71 'helvet' : 'default', 'avant' : 'default',
72 'newcent' : 'newcent', 'bookman' : 'bookman',
74 sans_fonts = {'default' : 'default', 'ae' : 'default',
75 'times' : 'default', 'palatino' : 'default',
76 'helvet' : 'helvet', 'avant' : 'avant',
77 'newcent' : 'default', 'bookman' : 'default',
79 typewriter_fonts = {'default' : 'default', 'ae' : 'default',
80 'times' : 'default', 'palatino' : 'default',
81 'helvet' : 'default', 'avant' : 'default',
82 'newcent' : 'default', 'bookman' : 'default',
83 'pslatex' : 'courier'}
85 def convert_font_settings(document):
86 " Convert font settings. "
88 i = find_token_exact(document.header, "\\fontscheme", i)
90 document.warning("Malformed LyX document: Missing `\\fontscheme'.")
92 font_scheme = get_value(document.header, "\\fontscheme", i, i + 1)
94 document.warning("Malformed LyX document: Empty `\\fontscheme'.")
95 font_scheme = 'default'
96 if not font_scheme in roman_fonts.keys():
97 document.warning("Malformed LyX document: Unknown `\\fontscheme' `%s'." % font_scheme)
98 font_scheme = 'default'
99 document.header[i:i+1] = ['\\font_roman %s' % roman_fonts[font_scheme],
100 '\\font_sans %s' % sans_fonts[font_scheme],
101 '\\font_typewriter %s' % typewriter_fonts[font_scheme],
102 '\\font_default_family default',
105 '\\font_sf_scale 100',
106 '\\font_tt_scale 100']
109 def revert_font_settings(document):
110 " Revert font settings. "
113 fonts = {'roman' : 'default', 'sans' : 'default', 'typewriter' : 'default'}
114 for family in 'roman', 'sans', 'typewriter':
115 name = '\\font_%s' % family
116 i = find_token_exact(document.header, name, i)
118 document.warning("Malformed LyX document: Missing `%s'." % name)
121 if (insert_line < 0):
123 fonts[family] = get_value(document.header, name, i, i + 1)
124 del document.header[i]
125 i = find_token_exact(document.header, '\\font_default_family', i)
127 document.warning("Malformed LyX document: Missing `\\font_default_family'.")
128 font_default_family = 'default'
130 font_default_family = get_value(document.header, "\\font_default_family", i, i + 1)
131 del document.header[i]
132 i = find_token_exact(document.header, '\\font_sc', i)
134 document.warning("Malformed LyX document: Missing `\\font_sc'.")
137 font_sc = get_value(document.header, '\\font_sc', i, i + 1)
138 del document.header[i]
139 if font_sc != 'false':
140 document.warning("Conversion of '\\font_sc' not yet implemented.")
141 i = find_token_exact(document.header, '\\font_osf', i)
143 document.warning("Malformed LyX document: Missing `\\font_osf'.")
146 font_osf = get_value(document.header, '\\font_osf', i, i + 1)
147 del document.header[i]
148 i = find_token_exact(document.header, '\\font_sf_scale', i)
150 document.warning("Malformed LyX document: Missing `\\font_sf_scale'.")
151 font_sf_scale = '100'
153 font_sf_scale = get_value(document.header, '\\font_sf_scale', i, i + 1)
154 del document.header[i]
155 if font_sf_scale != '100':
156 document.warning("Conversion of '\\font_sf_scale' not yet implemented.")
157 i = find_token_exact(document.header, '\\font_tt_scale', i)
159 document.warning("Malformed LyX document: Missing `\\font_tt_scale'.")
160 font_tt_scale = '100'
162 font_tt_scale = get_value(document.header, '\\font_tt_scale', i, i + 1)
163 del document.header[i]
164 if font_tt_scale != '100':
165 document.warning("Conversion of '\\font_tt_scale' not yet implemented.")
166 for font_scheme in roman_fonts.keys():
167 if (roman_fonts[font_scheme] == fonts['roman'] and
168 sans_fonts[font_scheme] == fonts['sans'] and
169 typewriter_fonts[font_scheme] == fonts['typewriter']):
170 document.header.insert(insert_line, '\\fontscheme %s' % font_scheme)
171 if font_default_family != 'default':
172 document.preamble.append('\\renewcommand{\\familydefault}{\\%s}' % font_default_family)
173 if font_osf == 'true':
174 document.warning("Ignoring `\\font_osf = true'")
176 font_scheme = 'default'
177 document.header.insert(insert_line, '\\fontscheme %s' % font_scheme)
178 if fonts['roman'] == 'cmr':
179 document.preamble.append('\\renewcommand{\\rmdefault}{cmr}')
180 if font_osf == 'true':
181 document.preamble.append('\\usepackage{eco}')
183 for font in 'lmodern', 'charter', 'utopia', 'beraserif', 'ccfonts', 'chancery':
184 if fonts['roman'] == font:
185 document.preamble.append('\\usepackage{%s}' % font)
186 for font in 'cmss', 'lmss', 'cmbr':
187 if fonts['sans'] == font:
188 document.preamble.append('\\renewcommand{\\sfdefault}{%s}' % font)
189 for font in 'berasans':
190 if fonts['sans'] == font:
191 document.preamble.append('\\usepackage{%s}' % font)
192 for font in 'cmtt', 'lmtt', 'cmtl':
193 if fonts['typewriter'] == font:
194 document.preamble.append('\\renewcommand{\\ttdefault}{%s}' % font)
195 for font in 'courier', 'beramono', 'luximono':
196 if fonts['typewriter'] == font:
197 document.preamble.append('\\usepackage{%s}' % font)
198 if font_default_family != 'default':
199 document.preamble.append('\\renewcommand{\\familydefault}{\\%s}' % font_default_family)
200 if font_osf == 'true':
201 document.warning("Ignoring `\\font_osf = true'")
204 def revert_booktabs(document):
205 " We remove the booktabs flag or everything else will become a mess. "
206 re_row = re.compile(r'^<row.*space="[^"]+".*>$')
207 re_tspace = re.compile(r'\s+topspace="[^"]+"')
208 re_bspace = re.compile(r'\s+bottomspace="[^"]+"')
209 re_ispace = re.compile(r'\s+interlinespace="[^"]+"')
212 i = find_token(document.body, "\\begin_inset Tabular", i)
215 j = find_end_of_inset(document.body, i + 1)
217 document.warning("Malformed LyX document: Could not find end of tabular.")
219 for k in range(i, j):
220 if re.search('^<features.* booktabs="true".*>$', document.body[k]):
221 document.warning("Converting 'booktabs' table to normal table.")
222 document.body[k] = document.body[k].replace(' booktabs="true"', '')
223 if re.search(re_row, document.body[k]):
224 document.warning("Removing extra row space.")
225 document.body[k] = re_tspace.sub('', document.body[k])
226 document.body[k] = re_bspace.sub('', document.body[k])
227 document.body[k] = re_ispace.sub('', document.body[k])
231 def convert_multiencoding(document, forward):
232 """ Fix files with multiple encodings.
233 Files with an inputencoding of "auto" or "default" and multiple languages
234 where at least two languages have different default encodings are encoded
235 in multiple encodings for file formats < 249. These files are incorrectly
236 read and written (as if the whole file was in the encoding of the main
238 This is not true for files written by CJK-LyX, they are always in the locale
242 - converts from fake unicode values to true unicode if forward is true, and
243 - converts from true unicode values to fake unicode if forward is false.
244 document.encoding must be set to the old value (format 248) in both cases.
246 We do this here and not in LyX.py because it is far easier to do the
247 necessary parsing in modern formats than in ancient ones.
249 if document.cjk_encoding != '':
251 encoding_stack = [document.encoding]
252 lang_re = re.compile(r"^\\lang\s(\S+)")
253 if document.inputencoding == "auto" or document.inputencoding == "default":
254 for i in range(len(document.body)):
255 result = lang_re.match(document.body[i])
257 language = result.group(1)
258 if language == "default":
259 document.warning("Resetting encoding from %s to %s." % (encoding_stack[-1], document.encoding), 3)
260 encoding_stack[-1] = document.encoding
262 from lyx2lyx_lang import lang
263 document.warning("Setting encoding from %s to %s." % (encoding_stack[-1], lang[language][3]), 3)
264 encoding_stack[-1] = lang[language][3]
265 elif find_token(document.body, "\\begin_layout", i, i + 1) == i:
266 document.warning("Adding nested encoding %s." % encoding_stack[-1], 3)
267 encoding_stack.append(encoding_stack[-1])
268 elif find_token(document.body, "\\end_layout", i, i + 1) == i:
269 document.warning("Removing nested encoding %s." % encoding_stack[-1], 3)
270 if len(encoding_stack) == 1:
271 # Don't remove the document encoding from the stack
272 document.warning("Malformed LyX document: Unexpected `\\end_layout'.")
274 del encoding_stack[-1]
275 if encoding_stack[-1] != document.encoding:
277 # This line has been incorrectly interpreted as if it was
278 # encoded in 'encoding'.
279 # Convert back to the 8bit string that was in the file.
280 orig = document.body[i].encode(document.encoding)
281 # Convert the 8bit string that was in the file to unicode
282 # with the correct encoding.
283 document.body[i] = orig.decode(encoding_stack[-1])
285 # Convert unicode to the 8bit string that will be written
286 # to the file with the correct encoding.
287 orig = document.body[i].encode(encoding_stack[-1])
288 # Convert the 8bit string that will be written to the
289 # file to fake unicode with the encoding that will later
290 # be used when writing to the file.
291 document.body[i] = orig.decode(document.encoding)
294 def convert_utf8(document):
295 " Set document encoding to UTF-8. "
296 convert_multiencoding(document, True)
297 document.encoding = "utf8"
300 def revert_utf8(document):
301 " Set document encoding to the value corresponding to inputencoding. "
302 i = find_token(document.header, "\\inputencoding", 0)
304 document.header.append("\\inputencoding auto")
305 elif get_value(document.header, "\\inputencoding", i) == "utf8":
306 document.header[i] = "\\inputencoding auto"
307 document.inputencoding = get_value(document.header, "\\inputencoding", 0)
308 document.encoding = get_encoding(document.language, document.inputencoding, 248, document.cjk_encoding)
309 convert_multiencoding(document, False)
312 def revert_cs_label(document):
313 " Remove status flag of charstyle label. "
316 i = find_token(document.body, "\\begin_inset CharStyle", i)
319 # Seach for a line starting 'show_label'
320 # If it is not there, break with a warning message
323 if (document.body[i][:10] == "show_label"):
326 elif (document.body[i][:13] == "\\begin_layout"):
327 document.warning("Malformed LyX document: Missing 'show_label'.")
334 def convert_bibitem(document):
336 \bibitem [option]{argument}
340 \begin_inset LatexCommand bibitem
346 This must be called after convert_commandparams.
350 i = find_token(document.body, "\\bibitem", i)
353 j = document.body[i].find('[') + 1
354 k = document.body[i].rfind(']')
355 if j == 0: # No optional argument found
358 option = document.body[i][j:k]
359 j = document.body[i].rfind('{') + 1
360 k = document.body[i].rfind('}')
361 argument = document.body[i][j:k]
362 lines = ['\\begin_inset LatexCommand bibitem']
364 lines.append('label "%s"' % option.replace('"', '\\"'))
365 lines.append('key "%s"' % argument.replace('"', '\\"'))
367 lines.append('\\end_inset')
368 document.body[i:i+1] = lines
372 commandparams_info = {
373 # command : [option1, option2, argument]
374 "bibitem" : ["label", "", "key"],
375 "bibtex" : ["options", "btprint", "bibfiles"],
376 "cite" : ["after", "before", "key"],
377 "citet" : ["after", "before", "key"],
378 "citep" : ["after", "before", "key"],
379 "citealt" : ["after", "before", "key"],
380 "citealp" : ["after", "before", "key"],
381 "citeauthor" : ["after", "before", "key"],
382 "citeyear" : ["after", "before", "key"],
383 "citeyearpar" : ["after", "before", "key"],
384 "citet*" : ["after", "before", "key"],
385 "citep*" : ["after", "before", "key"],
386 "citealt*" : ["after", "before", "key"],
387 "citealp*" : ["after", "before", "key"],
388 "citeauthor*" : ["after", "before", "key"],
389 "Citet" : ["after", "before", "key"],
390 "Citep" : ["after", "before", "key"],
391 "Citealt" : ["after", "before", "key"],
392 "Citealp" : ["after", "before", "key"],
393 "Citeauthor" : ["after", "before", "key"],
394 "Citet*" : ["after", "before", "key"],
395 "Citep*" : ["after", "before", "key"],
396 "Citealt*" : ["after", "before", "key"],
397 "Citealp*" : ["after", "before", "key"],
398 "Citeauthor*" : ["after", "before", "key"],
399 "citefield" : ["after", "before", "key"],
400 "citetitle" : ["after", "before", "key"],
401 "cite*" : ["after", "before", "key"],
402 "hfill" : ["", "", ""],
403 "index" : ["", "", "name"],
404 "printindex" : ["", "", "name"],
405 "label" : ["", "", "name"],
406 "eqref" : ["name", "", "reference"],
407 "pageref" : ["name", "", "reference"],
408 "prettyref" : ["name", "", "reference"],
409 "ref" : ["name", "", "reference"],
410 "vpageref" : ["name", "", "reference"],
411 "vref" : ["name", "", "reference"],
412 "tableofcontents" : ["", "", "type"],
413 "htmlurl" : ["name", "", "target"],
414 "url" : ["name", "", "target"]}
417 def convert_commandparams(document):
420 \begin_inset LatexCommand \cmdname[opt1][opt2]{arg}
425 \begin_inset LatexCommand cmdname
431 name1, name2 and name3 can be different for each command.
433 # \begin_inset LatexCommand bibitem was not the official version (see
434 # convert_bibitem()), but could be read in, so we convert it here, too.
438 i = find_token(document.body, "\\begin_inset LatexCommand", i)
441 command = document.body[i][26:].strip()
443 document.warning("Malformed LyX document: Missing LatexCommand name.")
447 j = find_token(document.body, "\\end_inset", i + 1)
449 document.warning("Malformed document")
451 command += "".join(document.body[i+1:j])
452 document.body[i+1:j] = []
454 # The following parser is taken from the original InsetCommandParams::scanCommand
460 # Used to handle things like \command[foo[bar]]{foo{bar}}
464 if ((state == "CMDNAME" and c == ' ') or
465 (state == "CMDNAME" and c == '[') or
466 (state == "CMDNAME" and c == '{')):
468 if ((state == "OPTION" and c == ']') or
469 (state == "SECOPTION" and c == ']') or
470 (state == "CONTENT" and c == '}')):
474 nestdepth = nestdepth - 1
475 if ((state == "OPTION" and c == '[') or
476 (state == "SECOPTION" and c == '[') or
477 (state == "CONTENT" and c == '{')):
478 nestdepth = nestdepth + 1
479 if state == "CMDNAME":
481 elif state == "OPTION":
483 elif state == "SECOPTION":
485 elif state == "CONTENT":
490 elif c == '[' and b != ']':
492 nestdepth = 0 # Just to be sure
493 elif c == '[' and b == ']':
495 nestdepth = 0 # Just to be sure
498 nestdepth = 0 # Just to be sure
501 # Now we have parsed the command, output the parameters
502 lines = ["\\begin_inset LatexCommand %s" % name]
504 if commandparams_info[name][0] == "":
505 document.warning("Ignoring invalid option `%s' of command `%s'." % (option1, name))
507 lines.append('%s "%s"' % (commandparams_info[name][0], option1.replace('"', '\\"')))
509 if commandparams_info[name][1] == "":
510 document.warning("Ignoring invalid second option `%s' of command `%s'." % (option2, name))
512 lines.append('%s "%s"' % (commandparams_info[name][1], option2.replace('"', '\\"')))
514 if commandparams_info[name][2] == "":
515 document.warning("Ignoring invalid argument `%s' of command `%s'." % (argument, name))
517 lines.append('%s "%s"' % (commandparams_info[name][2], argument.replace('"', '\\"')))
518 document.body[i:i+1] = lines
522 def revert_commandparams(document):
523 regex = re.compile(r'(\S+)\s+(.+)')
526 i = find_token(document.body, "\\begin_inset LatexCommand", i)
529 name = document.body[i].split()[2]
530 j = find_end_of_inset(document.body, i + 1)
535 for k in range(i + 1, j):
536 match = re.match(regex, document.body[k])
538 pname = match.group(1)
539 pvalue = match.group(2)
540 if pname == "preview":
541 preview_line = document.body[k]
542 elif (commandparams_info[name][0] != "" and
543 pname == commandparams_info[name][0]):
544 option1 = pvalue.strip('"').replace('\\"', '"')
545 elif (commandparams_info[name][1] != "" and
546 pname == commandparams_info[name][1]):
547 option2 = pvalue.strip('"').replace('\\"', '"')
548 elif (commandparams_info[name][2] != "" and
549 pname == commandparams_info[name][2]):
550 argument = pvalue.strip('"').replace('\\"', '"')
551 elif document.body[k].strip() != "":
552 document.warning("Ignoring unknown contents `%s' in command inset %s." % (document.body[k], name))
553 if name == "bibitem":
555 lines = ["\\bibitem {%s}" % argument]
557 lines = ["\\bibitem [%s]{%s}" % (option1, argument)]
561 lines = ["\\begin_inset LatexCommand \\%s{%s}" % (name, argument)]
563 lines = ["\\begin_inset LatexCommand \\%s[][%s]{%s}" % (name, option2, argument)]
566 lines = ["\\begin_inset LatexCommand \\%s[%s]{%s}" % (name, option1, argument)]
568 lines = ["\\begin_inset LatexCommand \\%s[%s][%s]{%s}" % (name, option1, option2, argument)]
569 if name != "bibitem":
570 if preview_line != "":
571 lines.append(preview_line)
573 lines.append('\\end_inset')
574 document.body[i:j+1] = lines
578 def revert_nomenclature(document):
579 " Convert nomenclature entry to ERT. "
580 regex = re.compile(r'(\S+)\s+(.+)')
584 i = find_token(document.body, "\\begin_inset LatexCommand nomenclature", i)
588 j = find_end_of_inset(document.body, i + 1)
593 for k in range(i + 1, j):
594 match = re.match(regex, document.body[k])
596 name = match.group(1)
597 value = match.group(2)
598 if name == "preview":
599 preview_line = document.body[k]
600 elif name == "symbol":
601 symbol = value.strip('"').replace('\\"', '"')
602 elif name == "description":
603 description = value.strip('"').replace('\\"', '"')
604 elif name == "prefix":
605 prefix = value.strip('"').replace('\\"', '"')
606 elif document.body[k].strip() != "":
607 document.warning("Ignoring unknown contents `%s' in nomenclature inset." % document.body[k])
609 command = 'nomenclature{%s}{%s}' % (symbol, description)
611 command = 'nomenclature[%s]{%s}{%s}' % (prefix, symbol, description)
612 document.body[i:j+1] = ['\\begin_inset ERT',
615 '\\begin_layout %s' % document.default_layout,
624 if use_nomencl and find_token(document.preamble, '\\usepackage{nomencl}[2005/09/22]', 0) == -1:
625 document.preamble.append('\\usepackage{nomencl}[2005/09/22]')
626 document.preamble.append('\\makenomenclature')
629 def revert_printnomenclature(document):
630 " Convert printnomenclature to ERT. "
631 regex = re.compile(r'(\S+)\s+(.+)')
635 i = find_token(document.body, "\\begin_inset LatexCommand printnomenclature", i)
639 j = find_end_of_inset(document.body, i + 1)
642 for k in range(i + 1, j):
643 match = re.match(regex, document.body[k])
645 name = match.group(1)
646 value = match.group(2)
647 if name == "preview":
648 preview_line = document.body[k]
649 elif name == "labelwidth":
650 labelwidth = value.strip('"').replace('\\"', '"')
651 elif document.body[k].strip() != "":
652 document.warning("Ignoring unknown contents `%s' in printnomenclature inset." % document.body[k])
654 command = 'nomenclature{}'
656 command = 'nomenclature[%s]' % labelwidth
657 document.body[i:j+1] = ['\\begin_inset ERT',
660 '\\begin_layout %s' % document.default_layout,
669 if use_nomencl and find_token(document.preamble, '\\usepackage{nomencl}[2005/09/22]', 0) == -1:
670 document.preamble.append('\\usepackage{nomencl}[2005/09/22]')
671 document.preamble.append('\\makenomenclature')
674 def convert_esint(document):
675 " Add \\use_esint setting to header. "
676 i = find_token(document.header, "\\cite_engine", 0)
678 document.warning("Malformed LyX document: Missing `\\cite_engine'.")
680 # 0 is off, 1 is auto, 2 is on.
681 document.header.insert(i, '\\use_esint 0')
684 def revert_esint(document):
685 " Remove \\use_esint setting from header. "
686 i = find_token(document.header, "\\use_esint", 0)
688 document.warning("Malformed LyX document: Missing `\\use_esint'.")
690 use_esint = document.header[i].split()[1]
691 del document.header[i]
692 # 0 is off, 1 is auto, 2 is on.
694 document.preamble.append('\\usepackage{esint}')
697 def revert_clearpage(document):
701 i = find_token(document.body, "\\clearpage", i)
704 document.body[i:i+1] = ['\\begin_inset ERT',
707 '\\begin_layout %s' % document.default_layout,
718 def revert_cleardoublepage(document):
719 " cleardoublepage -> ERT "
722 i = find_token(document.body, "\\cleardoublepage", i)
725 document.body[i:i+1] = ['\\begin_inset ERT',
728 '\\begin_layout %s' % document.default_layout,
739 def convert_lyxline(document):
740 " remove fontsize commands for \lyxline "
741 # The problematic is: The old \lyxline definition doesn't handle the fontsize
742 # to change the line thickness. The new definiton does this so that imported
743 # \lyxlines would have a different line thickness. The eventual fontsize command
744 # before \lyxline is therefore removed to get the same output.
745 fontsizes = ["tiny", "scriptsize", "footnotesize", "small", "normalsize",
746 "large", "Large", "LARGE", "huge", "Huge"]
747 for n in range(0, len(fontsizes)):
750 while i < len(document.body):
751 i = find_token(document.body, "\\size " + fontsizes[n], i)
752 k = find_token(document.body, "\\lyxline", i)
753 # the corresponding fontsize command is always 2 lines before the \lyxline
754 if (i != -1 and k == i+2):
755 document.body[i:i+1] = []
761 def revert_encodings(document):
762 " Set new encodings to auto. "
763 encodings = ["8859-6", "8859-8", "cp437", "cp437de", "cp850", "cp852",
764 "cp855", "cp858", "cp862", "cp865", "cp866", "cp1250",
765 "cp1252", "cp1256", "cp1257", "latin10", "pt254", "tis620-0"]
766 i = find_token(document.header, "\\inputencoding", 0)
768 document.header.append("\\inputencoding auto")
770 inputenc = get_value(document.header, "\\inputencoding", i)
771 if inputenc in encodings:
772 document.header[i] = "\\inputencoding auto"
773 document.inputencoding = get_value(document.header, "\\inputencoding", 0)
776 def convert_caption(document):
777 " Convert caption layouts to caption insets. "
780 i = find_token(document.body, "\\begin_layout Caption", i)
783 j = find_end_of_layout(document.body, i)
785 document.warning("Malformed LyX document: Missing `\\end_layout'.")
788 document.body[j:j] = ["\\end_layout", "", "\\end_inset", "", ""]
789 document.body[i:i+1] = ["\\begin_layout %s" % document.default_layout,
790 "\\begin_inset Caption", "",
791 "\\begin_layout %s" % document.default_layout]
795 def revert_caption(document):
796 " Convert caption insets to caption layouts. "
797 " This assumes that the text class has a caption style. "
800 i = find_token(document.body, "\\begin_inset Caption", i)
804 # We either need to delete the previous \begin_layout line, or we
805 # need to end the previous layout if this inset is not in the first
806 # position of the paragraph.
807 layout_before = find_token_backwards(document.body, "\\begin_layout", i)
808 if layout_before == -1:
809 document.warning("Malformed LyX document: Missing `\\begin_layout'.")
811 layout_line = document.body[layout_before]
812 del_layout_before = True
813 l = layout_before + 1
815 if document.body[l] != "":
816 del_layout_before = False
819 if del_layout_before:
820 del document.body[layout_before:i]
823 document.body[i:i] = ["\\end_layout", ""]
826 # Find start of layout in the inset and end of inset
827 j = find_token(document.body, "\\begin_layout", i)
829 document.warning("Malformed LyX document: Missing `\\begin_layout'.")
831 k = find_end_of_inset(document.body, i)
833 document.warning("Malformed LyX document: Missing `\\end_inset'.")
836 # We either need to delete the following \end_layout line, or we need
837 # to restart the old layout if this inset is not at the paragraph end.
838 layout_after = find_token(document.body, "\\end_layout", k)
839 if layout_after == -1:
840 document.warning("Malformed LyX document: Missing `\\end_layout'.")
842 del_layout_after = True
844 while l < layout_after:
845 if document.body[l] != "":
846 del_layout_after = False
850 del document.body[k+1:layout_after+1]
852 document.body[k+1:k+1] = [layout_line, ""]
854 # delete \begin_layout and \end_inset and replace \begin_inset with
855 # "\begin_layout Caption". This works because we can only have one
856 # paragraph in the caption inset: The old \end_layout will be recycled.
858 if document.body[k] == "":
861 if document.body[j] == "":
863 document.body[i] = "\\begin_layout Caption"
864 if document.body[i+1] == "":
865 del document.body[i+1]
869 # Accents of InsetLaTeXAccent
871 "`" : u'\u0300', # grave
872 "'" : u'\u0301', # acute
873 "^" : u'\u0302', # circumflex
874 "~" : u'\u0303', # tilde
875 "=" : u'\u0304', # macron
876 "u" : u'\u0306', # breve
877 "." : u'\u0307', # dot above
878 "\"": u'\u0308', # diaeresis
879 "r" : u'\u030a', # ring above
880 "H" : u'\u030b', # double acute
881 "v" : u'\u030c', # caron
882 "b" : u'\u0320', # minus sign below
883 "d" : u'\u0323', # dot below
884 "c" : u'\u0327', # cedilla
885 "k" : u'\u0328', # ogonek
886 "t" : u'\u0361' # tie. This is special: It spans two characters, but
887 # only one is given as argument, so we don't need to
888 # treat it differently.
892 # special accents of InsetLaTeXAccent without argument
893 special_accent_map = {
894 'i' : u'\u0131', # dotless i
895 'j' : u'\u0237', # dotless j
896 'l' : u'\u0142', # l with stroke
897 'L' : u'\u0141' # L with stroke
901 # special accent arguments of InsetLaTeXAccent
903 '\\i' : u'\u0131', # dotless i
904 '\\j' : u'\u0237' # dotless j
908 def _convert_accent(accent, accented_char):
912 if type in special_accent_map:
913 return special_accent_map[type]
914 # a missing char is treated as space by LyX
916 elif type == 'q' and char in ['t', 'd', 'l', 'L']:
917 # Special caron, only used with t, d, l and L.
918 # It is not in the map because we convert it to the same unicode
919 # character as the normal caron: \q{} is only defined if babel with
920 # the czech or slovak language is used, and the normal caron
921 # produces the correct output if the T1 font encoding is used.
922 # For the same reason we never convert to \q{} in the other direction.
924 elif char in accented_map:
925 char = accented_map[char]
926 elif (len(char) > 1):
927 # We can only convert accents on a single char
929 a = accent_map.get(type)
931 return unicodedata.normalize("NFC", "%s%s" % (char, a))
935 def convert_ertbackslash(body, i, ert, default_layout):
936 r""" -------------------------------------------------------------------------------------------
937 Convert backslashes and '\n' into valid ERT code, append the converted
938 text to body[i] and return the (maybe incremented) line index i"""
942 body[i] = body[i] + '\\backslash '
946 body[i+1:i+1] = ['\\end_layout', '', '\\begin_layout %s' % default_layout, '']
949 body[i] = body[i] + c
953 def convert_accent(document):
954 # The following forms are supported by LyX:
955 # '\i \"{a}' (standard form, as written by LyX)
956 # '\i \"{}' (standard form, as written by LyX if the accented char is a space)
957 # '\i \"{ }' (also accepted if the accented char is a space)
958 # '\i \" a' (also accepted)
959 # '\i \"' (also accepted)
960 re_wholeinset = re.compile(r'^(.*)(\\i\s+)(.*)$')
961 re_contents = re.compile(r'^([^\s{]+)(.*)$')
962 re_accentedcontents = re.compile(r'^\s*{?([^{}]*)}?\s*$')
965 i = find_re(document.body, re_wholeinset, i)
968 match = re_wholeinset.match(document.body[i])
969 prefix = match.group(1)
970 contents = match.group(3).strip()
971 match = re_contents.match(contents)
973 # Strip first char (always \)
974 accent = match.group(1)[1:]
975 accented_contents = match.group(2).strip()
976 match = re_accentedcontents.match(accented_contents)
977 accented_char = match.group(1)
978 converted = _convert_accent(accent, accented_char)
981 contents = '%s{%s}' % (accent, accented_char),
983 document.body[i] = '%s%s' % (prefix, converted)
986 document.warning("Converting unknown InsetLaTeXAccent `\\i %s' to ERT." % contents)
987 document.body[i] = prefix
988 document.body[i+1:i+1] = ['\\begin_inset ERT',
991 '\\begin_layout %s' % document.default_layout,
995 i = convert_ertbackslash(document.body, i + 7,
997 document.default_layout)
998 document.body[i+1:i+1] = ['\\end_layout',
1004 def revert_accent(document):
1005 inverse_accent_map = {}
1006 for k in accent_map:
1007 inverse_accent_map[accent_map[k]] = k
1008 inverse_special_accent_map = {}
1009 for k in special_accent_map:
1010 inverse_special_accent_map[special_accent_map[k]] = k
1011 inverse_accented_map = {}
1012 for k in accented_map:
1013 inverse_accented_map[accented_map[k]] = k
1015 # Since LyX may insert a line break within a word we must combine all
1016 # words before unicode normalization.
1017 # We do this only if the next line starts with an accent, otherwise we
1018 # would create things like '\begin_inset ERTstatus'.
1019 numberoflines = len(document.body)
1020 for i in range(numberoflines-1):
1021 if document.body[i] == '' or document.body[i+1] == '' or document.body[i][-1] == ' ':
1023 if (document.body[i+1][0] in inverse_accent_map):
1024 # the last character of this line and the first of the next line
1025 # form probably a surrogate pair.
1026 while (len(document.body[i+1]) > 0 and document.body[i+1][0] != ' '):
1027 document.body[i] += document.body[i+1][0]
1028 document.body[i+1] = document.body[i+1][1:]
1030 # Normalize to "Normal form D" (NFD, also known as canonical decomposition).
1031 # This is needed to catch all accented characters.
1032 for i in range(numberoflines):
1033 # Unfortunately we have a mixture of unicode strings and plain strings,
1034 # because we never use u'xxx' for string literals, but 'xxx'.
1035 # Therefore we may have to try two times to normalize the data.
1037 document.body[i] = unicodedata.normalize("NFD", document.body[i])
1039 document.body[i] = unicodedata.normalize("NFD", unicode(document.body[i], 'utf-8'))
1041 # Replace accented characters with InsetLaTeXAccent
1042 # Do not convert characters that can be represented in the chosen
1044 encoding_stack = [get_encoding(document.language, document.inputencoding, 248, document.cjk_encoding)]
1045 lang_re = re.compile(r"^\\lang\s(\S+)")
1047 while i < len(document.body):
1049 if (document.inputencoding == "auto" or document.inputencoding == "default") and document.cjk_encoding != '':
1050 # Track the encoding of the current line
1051 result = lang_re.match(document.body[i])
1053 language = result.group(1)
1054 if language == "default":
1055 encoding_stack[-1] = document.encoding
1057 from lyx2lyx_lang import lang
1058 encoding_stack[-1] = lang[language][3]
1060 elif find_token(document.body, "\\begin_layout", i, i + 1) == i:
1061 encoding_stack.append(encoding_stack[-1])
1063 elif find_token(document.body, "\\end_layout", i, i + 1) == i:
1064 del encoding_stack[-1]
1067 for j in range(len(document.body[i])):
1068 # dotless i and dotless j are both in special_accent_map and can
1069 # occur as an accented character, so we need to test that the
1070 # following character is no accent
1071 if (document.body[i][j] in inverse_special_accent_map and
1072 (j == len(document.body[i]) - 1 or document.body[i][j+1] not in inverse_accent_map)):
1073 accent = document.body[i][j]
1075 dummy = accent.encode(encoding_stack[-1])
1076 except UnicodeEncodeError:
1077 # Insert the rest of the line as new line
1078 if j < len(document.body[i]) - 1:
1079 document.body.insert(i+1, document.body[i][j+1:])
1080 # Delete the accented character
1082 document.body[i] = document.body[i][:j-1]
1084 document.body[i] = u''
1085 # Finally add the InsetLaTeXAccent
1086 document.body[i] += "\\i \\%s{}" % inverse_special_accent_map[accent]
1088 elif j > 0 and document.body[i][j] in inverse_accent_map:
1089 accented_char = document.body[i][j-1]
1090 if accented_char == ' ':
1091 # Conform to LyX output
1093 elif accented_char in inverse_accented_map:
1094 accented_char = inverse_accented_map[accented_char]
1095 accent = document.body[i][j]
1097 dummy = unicodedata.normalize("NFC", accented_char + accent).encode(encoding_stack[-1])
1098 except UnicodeEncodeError:
1099 # Insert the rest of the line as new line
1100 if j < len(document.body[i]) - 1:
1101 document.body.insert(i+1, document.body[i][j+1:])
1102 # Delete the accented characters
1104 document.body[i] = document.body[i][:j-2]
1106 document.body[i] = u''
1107 # Finally add the InsetLaTeXAccent
1108 document.body[i] += "\\i \\%s{%s}" % (inverse_accent_map[accent], accented_char)
1112 # Normalize to "Normal form C" (NFC, pre-composed characters) again
1113 for i in range(numberoflines):
1114 document.body[i] = unicodedata.normalize("NFC", document.body[i])
1117 def normalize_font_whitespace_259(document):
1118 """ Before format 259 the font changes were ignored if a
1119 whitespace was the first or last character in the sequence, this function
1120 transfers the whitespace outside."""
1122 char_properties = {"\\series": "default",
1123 "\\emph": "default",
1125 "\\shape": "default",
1127 "\\family": "default"}
1128 return normalize_font_whitespace(document, char_properties)
1130 def normalize_font_whitespace_274(document):
1131 """ Before format 259 (sic) the font changes were ignored if a
1132 whitespace was the first or last character in the sequence. This was
1133 corrected for most font properties in format 259, but the language
1134 was forgotten then. This function applies the same conversion done
1135 there (namely, transfers the whitespace outside) for font language
1136 changes, as well."""
1138 char_properties = {"\\lang": "default"}
1139 return normalize_font_whitespace(document, char_properties)
1141 def get_paragraph_language(document, i):
1142 """ Return the language of the paragraph in which line i of the document
1143 body is. If the first thing in the paragraph is a \\lang command, that
1144 is the paragraph's langauge; otherwise, the paragraph's language is the
1145 document's language."""
1147 lines = document.body
1149 first_nonempty_line = \
1150 find_nonempty_line(lines, find_beginning_of_layout(lines, i) + 1)
1152 words = lines[first_nonempty_line].split()
1154 if len(words) > 1 and words[0] == "\\lang":
1157 return document.language
1159 def normalize_font_whitespace(document, char_properties):
1160 """ Before format 259 the font changes were ignored if a
1161 whitespace was the first or last character in the sequence, this function
1162 transfers the whitespace outside. Only a change in one of the properties
1163 in the provided char_properties is handled by this function."""
1165 if document.backend != "latex":
1168 lines = document.body
1173 while i < len(lines):
1174 words = lines[i].split()
1176 if len(words) > 0 and words[0] == "\\begin_layout":
1177 # a new paragraph resets all font changes
1179 # also reset the default language to be the paragraph's language
1180 if "\\lang" in char_properties.keys():
1181 char_properties["\\lang"] = \
1182 get_paragraph_language(document, i + 1)
1184 elif len(words) > 1 and words[0] in char_properties.keys():
1185 # we have a font change
1186 if char_properties[words[0]] == words[1]:
1187 # property gets reset
1188 if words[0] in changes.keys():
1189 del changes[words[0]]
1190 defaultproperty = True
1193 changes[words[0]] = words[1]
1194 defaultproperty = False
1196 # We need to explicitly reset all changed properties if we find
1197 # a space below, because LyX 1.4 would output the space after
1198 # closing the previous change and before starting the new one,
1199 # and closing a font change means to close all properties, not
1200 # just the changed one.
1202 if lines[i-1] and lines[i-1][-1] == " ":
1203 lines[i-1] = lines[i-1][:-1]
1204 # a space before the font change
1206 for k in changes.keys():
1207 # exclude property k because that is already in lines[i]
1209 added_lines[1:1] = ["%s %s" % (k, changes[k])]
1210 for k in changes.keys():
1211 # exclude property k because that must be added below anyway
1213 added_lines[0:0] = ["%s %s" % (k, char_properties[k])]
1215 # Property is reset in lines[i], so add the new stuff afterwards
1216 lines[i+1:i+1] = added_lines
1218 # Reset property for the space
1219 added_lines[0:0] = ["%s %s" % (words[0], char_properties[words[0]])]
1220 lines[i:i] = added_lines
1221 i = i + len(added_lines)
1223 elif lines[i+1] and lines[i+1][0] == " " and (len(changes) > 0 or not defaultproperty):
1224 # a space after the font change
1225 if (lines[i+1] == " " and lines[i+2]):
1226 next_words = lines[i+2].split()
1227 if len(next_words) > 0 and next_words[0] == words[0]:
1228 # a single blank with a property different from the
1229 # previous and the next line must not be changed
1232 lines[i+1] = lines[i+1][1:]
1234 for k in changes.keys():
1235 # exclude property k because that is already in lines[i]
1237 added_lines[1:1] = ["%s %s" % (k, changes[k])]
1238 for k in changes.keys():
1239 # exclude property k because that must be added below anyway
1241 added_lines[0:0] = ["%s %s" % (k, char_properties[k])]
1242 # Reset property for the space
1243 added_lines[0:0] = ["%s %s" % (words[0], char_properties[words[0]])]
1244 lines[i:i] = added_lines
1245 i = i + len(added_lines)
1250 def revert_utf8x(document):
1251 " Set utf8x encoding to utf8. "
1252 i = find_token(document.header, "\\inputencoding", 0)
1254 document.header.append("\\inputencoding auto")
1256 inputenc = get_value(document.header, "\\inputencoding", i)
1257 if inputenc == "utf8x":
1258 document.header[i] = "\\inputencoding utf8"
1259 document.inputencoding = get_value(document.header, "\\inputencoding", 0)
1262 def revert_utf8plain(document):
1263 " Set utf8plain encoding to utf8. "
1264 i = find_token(document.header, "\\inputencoding", 0)
1266 document.header.append("\\inputencoding auto")
1268 inputenc = get_value(document.header, "\\inputencoding", i)
1269 if inputenc == "utf8-plain":
1270 document.header[i] = "\\inputencoding utf8"
1271 document.inputencoding = get_value(document.header, "\\inputencoding", 0)
1274 def revert_beamer_alert(document):
1275 " Revert beamer's \\alert inset back to ERT. "
1278 i = find_token(document.body, "\\begin_inset CharStyle Alert", i)
1281 document.body[i] = "\\begin_inset ERT"
1284 if (document.body[i][:13] == "\\begin_layout"):
1285 # Insert the \alert command
1286 document.body[i + 1] = "\\alert{" + document.body[i + 1] + '}'
1293 def revert_beamer_structure(document):
1294 " Revert beamer's \\structure inset back to ERT. "
1297 i = find_token(document.body, "\\begin_inset CharStyle Structure", i)
1300 document.body[i] = "\\begin_inset ERT"
1303 if (document.body[i][:13] == "\\begin_layout"):
1304 document.body[i + 1] = "\\structure{" + document.body[i + 1] + '}'
1311 def convert_changes(document):
1312 " Switch output_changes off if tracking_changes is off. "
1313 i = find_token(document.header, '\\tracking_changes', 0)
1315 document.warning("Malformed lyx document: Missing '\\tracking_changes'.")
1317 j = find_token(document.header, '\\output_changes', 0)
1319 document.warning("Malformed lyx document: Missing '\\output_changes'.")
1321 tracking_changes = get_value(document.header, "\\tracking_changes", i)
1322 output_changes = get_value(document.header, "\\output_changes", j)
1323 if tracking_changes == "false" and output_changes == "true":
1324 document.header[j] = "\\output_changes false"
1327 def revert_ascii(document):
1328 " Set ascii encoding to auto. "
1329 i = find_token(document.header, "\\inputencoding", 0)
1331 document.header.append("\\inputencoding auto")
1333 inputenc = get_value(document.header, "\\inputencoding", i)
1334 if inputenc == "ascii":
1335 document.header[i] = "\\inputencoding auto"
1336 document.inputencoding = get_value(document.header, "\\inputencoding", 0)
1339 def normalize_language_name(document):
1340 lang = { "brazil": "brazilian",
1341 "portuges": "portuguese"}
1343 if document.language in lang:
1344 document.language = lang[document.language]
1345 i = find_token(document.header, "\\language", 0)
1346 document.header[i] = "\\language %s" % document.language
1349 def revert_language_name(document):
1350 lang = { "brazilian": "brazil",
1351 "portuguese": "portuges"}
1353 if document.language in lang:
1354 document.language = lang[document.language]
1355 i = find_token(document.header, "\\language", 0)
1356 document.header[i] = "\\language %s" % document.language
1359 # \textclass cv -> \textclass simplecv
1360 def convert_cv_textclass(document):
1361 if document.textclass == "cv":
1362 document.textclass = "simplecv"
1365 def revert_cv_textclass(document):
1366 if document.textclass == "simplecv":
1367 document.textclass = "cv"
1371 # add scaleBeforeRotation graphics param
1372 def convert_graphics_rotation(document):
1373 " add scaleBeforeRotation graphics parameter. "
1376 i = find_token(document.body, "\\begin_inset Graphics", i)
1379 j = find_end_of_inset(document.body, i+1)
1382 document.warning("Malformed LyX document: Could not find end of graphics inset.")
1383 # Seach for rotateAngle and width or height or scale
1384 # If these params are not there, nothing needs to be done.
1385 k = find_token(document.body, "\trotateAngle", i + 1, j)
1386 l = find_tokens(document.body, ["\twidth", "\theight", "\tscale"], i + 1, j)
1387 if (k != -1 and l != -1):
1388 document.body.insert(j, 'scaleBeforeRotation')
1393 # remove scaleBeforeRotation graphics param
1394 def revert_graphics_rotation(document):
1395 " remove scaleBeforeRotation graphics parameter. "
1398 i = find_token(document.body, "\\begin_inset Graphics", i)
1401 j = find_end_of_inset(document.body, i + 1)
1404 document.warning("Malformed LyX document: Could not find end of graphics inset.")
1405 # If there's a scaleBeforeRotation param, just remove that
1406 k = find_token(document.body, "\tscaleBeforeRotation", i + 1, j)
1408 del document.body[k]
1410 # if not, and if we have rotateAngle and width or height or scale,
1411 # we have to put the rotateAngle value to special
1412 rotateAngle = get_value(document.body, 'rotateAngle', i + 1, j)
1413 special = get_value(document.body, 'special', i + 1, j)
1414 if rotateAngle != "":
1415 k = find_tokens(document.body, ["\twidth", "\theight", "\tscale"], i + 1, j)
1419 document.body.insert(j-1, '\tspecial angle=%s' % rotateAngle)
1421 l = find_token(document.body, "\tspecial", i + 1, j)
1422 document.body[l] = document.body[l].replace(special, 'angle=%s,%s' % (rotateAngle, special))
1423 k = find_token(document.body, "\trotateAngle", i + 1, j)
1425 del document.body[k]
1430 def convert_tableborder(document):
1431 # The problematic is: LyX double the table cell border as it ignores the "|" character in
1432 # the cell arguments. A fix takes care of this and therefore the "|" has to be removed
1434 while i < len(document.body):
1435 h = document.body[i].find("leftline=\"true\"", 0, len(document.body[i]))
1436 k = document.body[i].find("|>{", 0, len(document.body[i]))
1437 # the two tokens have to be in one line
1438 if (h != -1 and k != -1):
1440 document.body[i] = document.body[i][:k] + document.body[i][k+1:len(document.body[i])-1]
1444 def revert_tableborder(document):
1446 while i < len(document.body):
1447 h = document.body[i].find("leftline=\"true\"", 0, len(document.body[i]))
1448 k = document.body[i].find(">{", 0, len(document.body[i]))
1449 # the two tokens have to be in one line
1450 if (h != -1 and k != -1):
1452 document.body[i] = document.body[i][:k] + '|' + document.body[i][k:]
1456 def revert_armenian(document):
1458 # set inputencoding from armscii8 to auto
1459 if document.inputencoding == "armscii8":
1460 i = find_token(document.header, "\\inputencoding", 0)
1462 document.header[i] = "\\inputencoding auto"
1463 # check if preamble exists, if not k is set to -1
1466 while i < len(document.preamble):
1468 k = document.preamble[i].find("\\", 0, len(document.preamble[i]))
1470 k = document.preamble[i].find("%", 0, len(document.preamble[i]))
1472 # add the entry \usepackage{armtex} to the document preamble
1473 if document.language == "armenian":
1474 # set the armtex entry as the first preamble line
1476 document.preamble[0:0] = ["\\usepackage{armtex}"]
1477 # create the preamble when it doesn't exist
1479 document.preamble.append('\\usepackage{armtex}')
1480 # Set document language from armenian to english
1481 if document.language == "armenian":
1482 document.language = "english"
1483 i = find_token(document.header, "\\language", 0)
1485 document.header[i] = "\\language english"
1488 def revert_CJK(document):
1489 " Set CJK encodings to default and languages chinese, japanese and korean to english. "
1490 encodings = ["Bg5", "Bg5+", "GB", "GBt", "GBK", "JIS",
1491 "KS", "SJIS", "UTF8", "EUC-TW", "EUC-JP"]
1492 i = find_token(document.header, "\\inputencoding", 0)
1494 document.header.append("\\inputencoding auto")
1496 inputenc = get_value(document.header, "\\inputencoding", i)
1497 if inputenc in encodings:
1498 document.header[i] = "\\inputencoding default"
1499 document.inputencoding = get_value(document.header, "\\inputencoding", 0)
1501 if document.language == "chinese-simplified" or \
1502 document.language == "chinese-traditional" or \
1503 document.language == "japanese" or document.language == "korean":
1504 document.language = "english"
1505 i = find_token(document.header, "\\language", 0)
1507 document.header[i] = "\\language english"
1510 def revert_preamble_listings_params(document):
1511 " Revert preamble option \listings_params "
1512 i = find_token(document.header, "\\listings_params", 0)
1514 document.preamble.append('\\usepackage{listings}')
1515 document.preamble.append('\\lstset{%s}' % document.header[i].split()[1].strip('"'))
1516 document.header.pop(i);
1519 def revert_listings_inset(document):
1520 r''' Revert listings inset to \lstinline or \begin, \end lstlisting, translate
1524 lstparams "language=Delphi"
1528 \begin_layout Standard
1538 \begin_layout Standard
1542 lstinline[language=Delphi]{var i = 10;}
1547 There can be an caption inset in this inset
1549 \begin_layout Standard
1550 \begin_inset Caption
1552 \begin_layout Standard
1554 \begin_inset LatexCommand label
1570 i = find_token(document.body, '\\begin_inset listings', i)
1574 if not '\\usepackage{listings}' in document.preamble:
1575 document.preamble.append('\\usepackage{listings}')
1576 j = find_end_of_inset(document.body, i + 1)
1578 # this should not happen
1584 for line in range(i + 1, i + 4):
1585 if document.body[line].startswith('inline'):
1586 inline = document.body[line].split()[1]
1587 if document.body[line].startswith('lstparams'):
1588 params = document.body[line].split()[1].strip('"')
1589 if document.body[line].startswith('status'):
1590 status = document.body[line].split()[1].strip()
1595 cap = find_token(document.body, '\\begin_inset Caption', i)
1597 cap_end = find_end_of_inset(document.body, cap + 1)
1599 # this should not happen
1602 lbl = find_token(document.body, '\\begin_inset LatexCommand label', cap + 1)
1604 lbl_end = find_end_of_inset(document.body, lbl + 1)
1606 # this should not happen
1611 for line in document.body[lbl : lbl_end + 1]:
1612 if line.startswith('name '):
1613 label = line.split()[1].strip('"')
1615 for line in document.body[cap : lbl ] + document.body[lbl_end + 1 : cap_end + 1]:
1616 if not line.startswith('\\'):
1617 caption += line.strip()
1620 # looking for the oneline code for lstinline
1621 inlinecode = document.body[find_end_of_layout(document.body,
1622 find_token(document.body, '\\begin_layout %s' % document.default_layout, i + 1) +1 ) - 1]
1623 if len(caption) > 0:
1624 if len(params) == 0:
1625 params = 'caption={%s}' % caption
1627 params += ',caption={%s}' % caption
1629 if len(params) == 0:
1630 params = 'label={%s}' % label
1632 params += ',label={%s}' % label
1634 params = '[%s]' % params
1635 params = params.replace('\\', '\\backslash\n')
1636 if inline == 'true':
1637 document.body[i:(j+1)] = [r'\begin_inset ERT',
1638 'status %s' % status,
1639 r'\begin_layout %s' % document.default_layout,
1643 'lstinline%s{%s}' % (params, inlinecode),
1648 document.body[i: j+1] = [r'\begin_inset ERT',
1649 'status %s' % status,
1651 r'\begin_layout %s' % document.default_layout,
1655 r'begin{lstlisting}%s' % params,
1657 ] + document.body[k : j - 1] + \
1659 r'\begin_layout %s' % document.default_layout,
1668 def revert_include_listings(document):
1669 r''' Revert lstinputlisting Include option , translate
1670 \begin_inset Include \lstinputlisting{file}[opt]
1680 \begin_layout Standard
1684 lstinputlisting{file}[opt]
1692 i = find_token(document.body, r'\begin_inset Include \lstinputlisting', i)
1696 if not '\\usepackage{listings}' in document.preamble:
1697 document.preamble.append('\\usepackage{listings}')
1698 j = find_end_of_inset(document.body, i + 1)
1700 # this should not happen
1702 # find command line lstinputlisting{file}[options]
1703 cmd, file, option = '', '', ''
1704 if re.match(r'\\(lstinputlisting){([.\w]*)}(.*)', document.body[i].split()[2]):
1705 cmd, file, option = re.match(r'\\(lstinputlisting){([.\w]*)}(.*)', document.body[i].split()[2]).groups()
1706 option = option.replace('\\', '\\backslash\n')
1707 document.body[i : j + 1] = [r'\begin_inset ERT',
1710 r'\begin_layout %s' % document.default_layout,
1714 '%s%s{%s}' % (cmd, option, file),
1720 def revert_ext_font_sizes(document):
1721 if document.backend != "latex": return
1722 if not document.textclass.startswith("ext"): return
1724 fontsize = get_value(document.header, '\\paperfontsize', 0)
1725 if fontsize not in ('10', '11', '12'): return
1728 i = find_token(document.header, '\\paperfontsize', 0)
1729 document.header[i] = '\\paperfontsize default'
1731 i = find_token(document.header, '\\options', 0)
1733 i = find_token(document.header, '\\textclass', 0) + 1
1734 document.header[i:i] = ['\\options %s' % fontsize]
1736 document.header[i] += ',%s' % fontsize
1739 def convert_ext_font_sizes(document):
1740 if document.backend != "latex": return
1741 if not document.textclass.startswith("ext"): return
1743 fontsize = get_value(document.header, '\\paperfontsize', 0)
1744 if fontsize != 'default': return
1746 i = find_token(document.header, '\\options', 0)
1749 options = get_value(document.header, '\\options', i)
1751 fontsizes = '10pt', '11pt', '12pt'
1752 for fs in fontsizes:
1753 if options.find(fs) != -1:
1755 else: # this else will only be attained if the for cycle had no match
1758 options = options.split(',')
1759 for j, opt in enumerate(options):
1760 if opt in fontsizes:
1767 k = find_token(document.header, '\\paperfontsize', 0)
1768 document.header[k] = '\\paperfontsize %s' % fontsize
1771 document.header[i] = '\\options %s' % ','.join(options)
1773 del document.header[i]
1776 def revert_separator_layout(document):
1777 r'''Revert --Separator-- to a lyx note
1780 \begin_layout --Separator--
1786 \begin_layout Standard
1787 \begin_inset Note Note
1790 \begin_layout Standard
1803 i = find_token(document.body, r'\begin_layout --Separator--', i)
1806 j = find_end_of_layout(document.body, i + 1)
1808 # this should not happen
1810 document.body[i : j + 1] = [r'\begin_layout %s' % document.default_layout,
1811 r'\begin_inset Note Note',
1814 r'\begin_layout %s' % document.default_layout,
1815 'Separate Environment',
1819 document.body[ i + 1 : j] + \
1825 def convert_arabic (document):
1826 if document.language == "arabic":
1827 document.language = "arabic_arabtex"
1828 i = find_token(document.header, "\\language", 0)
1830 document.header[i] = "\\language arabic_arabtex"
1832 while i < len(document.body):
1833 h = document.body[i].find("\lang arabic", 0, len(document.body[i]))
1835 # change the language name
1836 document.body[i] = '\lang arabic_arabtex'
1840 def revert_arabic (document):
1841 if document.language == "arabic_arabtex":
1842 document.language = "arabic"
1843 i = find_token(document.header, "\\language", 0)
1845 document.header[i] = "\\language arabic"
1847 while i < len(document.body):
1848 h = document.body[i].find("\lang arabic_arabtex", 0, len(document.body[i]))
1850 # change the language name
1851 document.body[i] = '\lang arabic'
1855 def read_unicodesymbols():
1856 " Read the unicodesymbols list of unicode characters and corresponding commands."
1857 pathname = os.path.abspath(os.path.dirname(sys.argv[0]))
1858 fp = open(os.path.join(pathname.strip('lyx2lyx'), 'unicodesymbols'))
1860 for line in fp.readlines():
1862 line=line.replace(' "',' ') # remove all quotation marks with spaces before
1863 line=line.replace('" ',' ') # remove all quotation marks with spaces after
1864 line=line.replace(r'\"','"') # replace \" by " (for characters with diaeresis)
1866 # flag1 and flag2 are preamble and other flags
1867 [ucs4,command,flag1,flag2] =line.split(None,3)
1868 spec_chars[unichr(eval(ucs4))] = [command, flag1, flag2]
1876 def revert_unicode(document):
1877 '''Transform unicode characters that can not be written using the
1878 document encoding to commands according to the unicodesymbols
1879 file. Characters that can not be replaced by commands are replaced by
1880 an replacement string. Flags other than 'combined' are currently not
1883 replacement_character = '???'
1884 spec_chars = read_unicodesymbols()
1886 # Define strings to start and end ERT and math insets
1887 ert_intro='\n\n\\begin_inset ERT\nstatus collapsed\n\\begin_layout %s\n\\backslash' % document.default_layout
1888 ert_outro='\n\\end_layout\n\n\\end_inset\n'
1889 math_intro='\n\\begin_inset Formula $'
1890 math_outro='$\n\\end_inset'
1891 # Find unicode characters and replace them
1892 in_ert = False # flag set to 1 if in ERT inset
1893 in_math = False # flag set to 1 if in math inset
1894 insets = [] # list of active insets
1896 # Go through the file to capture all combining characters
1897 last_char = '' # to store the previous character
1900 while i < len(document.body):
1901 line = document.body[i]
1903 if line.find('\\begin_inset') > -1:
1904 # check which inset to start
1905 if line.find('\\begin_inset ERT') > -1:
1907 insets.append('ert')
1908 elif line.find('\\begin_inset Formula') > -1:
1910 insets.append('math')
1912 insets.append('other')
1913 if line.find('\\end_inset') > -1:
1914 # check which inset to end
1916 cur_inset = insets.pop()
1917 if cur_inset == 'ert':
1919 elif cur_inset == 'math':
1922 pass # end of other inset
1924 pass # inset list was empty (for some reason)
1926 # Try to write the line
1928 # If all goes well the line is written here
1929 dummy = line.encode(document.encoding)
1930 last_char = line[-1]
1933 # Error, some character(s) in the line need to be replaced
1935 for character in line:
1937 # Try to write the character
1938 dummy = character.encode(document.encoding)
1939 mod_line += character
1940 last_char = character
1942 # Try to replace with ERT/math inset
1943 if spec_chars.has_key(character):
1944 command = spec_chars[character][0] # the command to replace unicode
1945 flag1 = spec_chars[character][1]
1946 flag2 = spec_chars[character][2]
1947 if flag1.find('combining') > -1 or flag2.find('combining') > -1:
1948 # We have a character that should be combined with the previous
1949 command += '{' + last_char + '}'
1950 # Remove the last character. Ignore if it is whitespace
1951 if len(last_char.rstrip()):
1952 # last_char was found and is not whitespace
1954 mod_line = mod_line[:-1]
1955 else: # last_char belongs to the last line
1956 document.body[i-1] = document.body[i-1][:-1]
1958 # The last character was replaced by a command. For now it is
1959 # ignored. This could be handled better.
1961 if command[0:2] == '\\\\':
1962 if command[2:12]=='ensuremath':
1965 command = command.replace('\\\\ensuremath{\\\\', '$\n\\backslash')
1966 command = command.replace('}', '$\n')
1968 # add a math inset with the replacement character
1969 command = command.replace('\\\\ensuremath{\\', math_intro)
1970 command = command.replace('}', math_outro)
1972 # we are already in a math inset
1973 command = command.replace('\\\\ensuremath{\\', '')
1974 command = command.replace('}', '')
1977 # avoid putting an ERT in a math; instead put command as text
1978 command = command.replace('\\\\', '\mathrm{')
1979 command = command + '}'
1981 # add an ERT inset with the replacement character
1982 command = command.replace('\\\\', ert_intro)
1983 command = command + ert_outro
1985 command = command.replace('\\\\', '\n\\backslash')
1986 last_char = '' # indicate that the character should not be removed
1989 # Replace with replacement string
1990 mod_line += replacement_character
1991 document.body[i:i+1] = mod_line.split('\n')
1992 i += len(mod_line.split('\n'))
1999 supported_versions = ["1.5.0","1.5"]
2000 convert = [[246, []],
2001 [247, [convert_font_settings]],
2003 [249, [convert_utf8]],
2006 [252, [convert_commandparams, convert_bibitem]],
2008 [254, [convert_esint]],
2011 [257, [convert_caption]],
2012 [258, [convert_lyxline]],
2013 [259, [convert_accent, normalize_font_whitespace_259]],
2015 [261, [convert_changes]],
2017 [263, [normalize_language_name]],
2018 [264, [convert_cv_textclass]],
2019 [265, [convert_tableborder]],
2025 [271, [convert_ext_font_sizes]],
2028 [274, [normalize_font_whitespace_274]],
2029 [275, [convert_graphics_rotation]],
2030 [276, [convert_arabic]]
2034 [275, [revert_arabic]],
2035 [274, [revert_graphics_rotation]],
2037 [272, [revert_separator_layout]],
2038 [271, [revert_preamble_listings_params, revert_listings_inset, revert_include_listings]],
2039 [270, [revert_ext_font_sizes]],
2040 [269, [revert_beamer_alert, revert_beamer_structure]],
2041 [268, [revert_preamble_listings_params, revert_listings_inset, revert_include_listings]],
2042 [267, [revert_CJK]],
2043 [266, [revert_utf8plain]],
2044 [265, [revert_armenian]],
2045 [264, [revert_tableborder]],
2046 [263, [revert_cv_textclass]],
2047 [262, [revert_language_name]],
2048 [261, [revert_ascii]],
2050 [259, [revert_utf8x]],
2053 [256, [revert_caption]],
2054 [255, [revert_encodings]],
2055 [254, [revert_clearpage, revert_cleardoublepage]],
2056 [253, [revert_esint]],
2057 [252, [revert_nomenclature, revert_printnomenclature]],
2058 [251, [revert_commandparams]],
2059 [250, [revert_cs_label]],
2061 [248, [revert_accent, revert_utf8, revert_unicode]],
2062 [247, [revert_booktabs]],
2063 [246, [revert_font_settings]],
2064 [245, [revert_framed]]]
2067 if __name__ == "__main__":