1 # -*- coding: utf-8 -*-
2 # This file is part of lyx2lyx
3 # Copyright (C) 2011 The LyX team
5 # This program is free software; you can redistribute it and/or
6 # modify it under the terms of the GNU General Public License
7 # as published by the Free Software Foundation; either version 2
8 # of the License, or (at your option) any later version.
10 # This program is distributed in the hope that it will be useful,
11 # but WITHOUT ANY WARRANTY; without even the implied warranty of
12 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13 # GNU General Public License for more details.
15 # You should have received a copy of the GNU General Public License
16 # along with this program; if not, write to the Free Software
17 # Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
19 """ Convert files to the file format generated by lyx 2.0"""
25 from parser_tools import find_token, find_end_of, find_tokens, \
26 find_token_exact, find_end_of_inset, find_end_of_layout, \
27 find_token_backwards, is_in_inset, get_value, get_quoted_value, \
28 del_token, check_token, get_option_value
30 from lyx2lyx_tools import add_to_preamble, insert_to_preamble, \
31 put_cmd_in_ert, lyx2latex, latex_length, revert_flex_inset, \
32 revert_font_attrs, hex2ratio, str2bool
34 ####################################################################
35 # Private helper functions
37 def remove_option(lines, m, option):
38 ''' removes option from line m. returns whether we did anything '''
39 l = lines[m].find(option)
42 val = lines[m][l:].split('"')[1]
43 lines[m] = lines[m][:l - 1] + lines[m][l+len(option + '="' + val + '"'):]
47 ###############################################################################
49 ### Conversion and reversion routines
51 ###############################################################################
53 def revert_swiss(document):
54 " Set language german-ch to ngerman "
56 if document.language == "german-ch":
57 document.language = "ngerman"
58 i = find_token(document.header, "\\language", 0)
60 document.header[i] = "\\language ngerman"
63 j = find_token(document.body, "\\lang german-ch", j)
66 document.body[j] = document.body[j].replace("\\lang german-ch", "\\lang ngerman")
70 def revert_tabularvalign(document):
71 " Revert the tabular valign option "
74 i = find_token(document.body, "\\begin_inset Tabular", i)
77 end = find_end_of_inset(document.body, i)
79 document.warning("Can't find end of inset at line " + str(i))
82 fline = find_token(document.body, "<features", i, end)
84 document.warning("Can't find features for inset at line " + str(i))
87 p = document.body[fline].find("islongtable")
89 q = document.body[fline].find("tabularvalignment")
91 document.body[fline] = re.sub(r' tabularvalignment=\"[a-z]+\"', "", document.body[fline])
96 tabularvalignment = 'c'
97 # which valignment is specified?
98 m = document.body[fline].find('tabularvalignment="top"')
100 tabularvalignment = 't'
101 m = document.body[fline].find('tabularvalignment="bottom"')
103 tabularvalignment = 'b'
104 # delete tabularvalignment
105 q = document.body[fline].find("tabularvalignment")
107 document.body[fline] = re.sub(r' tabularvalignment=\"[a-z]+\"', "", document.body[fline])
109 # don't add a box when centered
110 if tabularvalignment == 'c':
113 subst = ['\\end_inset', '\\end_layout']
114 document.body[end:end] = subst # just inserts those lines
115 subst = ['\\begin_inset Box Frameless',
116 'position "' + tabularvalignment +'"',
121 # we don't know the width, assume 50%
125 'height_special "totalheight"',
128 '\\begin_layout Plain Layout']
129 document.body[i:i] = subst # this just inserts the array at i
130 # since there could be a tabular inside a tabular, we cannot
135 def revert_phantom_types(document, ptype, cmd):
136 " Reverts phantom to ERT "
139 i = find_token(document.body, "\\begin_inset Phantom " + ptype, i)
142 end = find_end_of_inset(document.body, i)
144 document.warning("Can't find end of inset at line " + str(i))
147 blay = find_token(document.body, "\\begin_layout Plain Layout", i, end)
149 document.warning("Can't find layout for inset at line " + str(i))
152 bend = find_end_of_layout(document.body, blay)
154 document.warning("Malformed LyX document: Could not find end of Phantom inset's layout.")
157 substi = ["\\begin_inset ERT", "status collapsed", "",
158 "\\begin_layout Plain Layout", "", "", "\\backslash",
159 cmd + "{", "\\end_layout", "", "\\end_inset"]
160 substj = ["\\size default", "", "\\begin_inset ERT", "status collapsed", "",
161 "\\begin_layout Plain Layout", "", "}", "\\end_layout", "", "\\end_inset"]
162 # do the later one first so as not to mess up the numbering
163 document.body[bend:end + 1] = substj
164 document.body[i:blay + 1] = substi
165 i = end + len(substi) + len(substj) - (end - bend) - (blay - i) - 2
168 def revert_phantom(document):
169 revert_phantom_types(document, "Phantom", "phantom")
171 def revert_hphantom(document):
172 revert_phantom_types(document, "HPhantom", "hphantom")
174 def revert_vphantom(document):
175 revert_phantom_types(document, "VPhantom", "vphantom")
178 def revert_xetex(document):
179 " Reverts documents that use XeTeX "
181 i = find_token(document.header, '\\use_xetex', 0)
183 document.warning("Malformed LyX document: Missing \\use_xetex.")
185 if not str2bool(get_value(document.header, "\\use_xetex", i)):
186 del document.header[i]
188 del document.header[i]
190 # 1.) set doc encoding to utf8-plain
191 i = find_token(document.header, "\\inputencoding", 0)
193 document.warning("Malformed LyX document: Missing \\inputencoding.")
195 document.header[i] = "\\inputencoding utf8-plain"
197 # 2.) check font settings
199 roman = sans = typew = "default"
201 sf_scale = tt_scale = 100.0
203 i = find_token(document.header, "\\font_roman", 0)
205 document.warning("Malformed LyX document: Missing \\font_roman.")
207 roman = get_value(document.header, "\\font_roman", i)
208 document.header[i] = "\\font_roman default"
210 i = find_token(document.header, "\\font_sans", 0)
212 document.warning("Malformed LyX document: Missing \\font_sans.")
214 sans = get_value(document.header, "\\font_sans", i)
215 document.header[i] = "\\font_sans default"
217 i = find_token(document.header, "\\font_typewriter", 0)
219 document.warning("Malformed LyX document: Missing \\font_typewriter.")
221 typew = get_value(document.header, "\\font_typewriter", i)
222 document.header[i] = "\\font_typewriter default"
224 i = find_token(document.header, "\\font_osf", 0)
226 document.warning("Malformed LyX document: Missing \\font_osf.")
228 osf = str2bool(get_value(document.header, "\\font_osf", i))
229 document.header[i] = "\\font_osf false"
231 i = find_token(document.header, "\\font_sc", 0)
233 document.warning("Malformed LyX document: Missing \\font_sc.")
235 # we do not need this value.
236 document.header[i] = "\\font_sc false"
238 i = find_token(document.header, "\\font_sf_scale", 0)
240 document.warning("Malformed LyX document: Missing \\font_sf_scale.")
242 val = get_value(document.header, '\\font_sf_scale', i)
245 sf_scale = float(val)
247 document.warning("Invalid font_sf_scale value: " + val)
248 document.header[i] = "\\font_sf_scale 100"
250 i = find_token(document.header, "\\font_tt_scale", 0)
252 document.warning("Malformed LyX document: Missing \\font_tt_scale.")
254 val = get_value(document.header, '\\font_tt_scale', i)
257 tt_scale = float(val)
259 document.warning("Invalid font_tt_scale value: " + val)
260 document.header[i] = "\\font_tt_scale 100"
262 # 3.) set preamble stuff
263 pretext = ['%% This document must be processed with xelatex!']
264 pretext.append('\\usepackage{fontspec}')
265 if roman != "default":
266 pretext.append('\\setmainfont[Mapping=tex-text]{' + roman + '}')
267 if sans != "default":
268 sf = '\\setsansfont['
269 if sf_scale != 100.0:
270 sf += 'Scale=' + str(sf_scale / 100.0) + ','
271 sf += 'Mapping=tex-text]{' + sans + '}'
273 if typew != "default":
275 if tt_scale != 100.0:
276 tw += '[Scale=' + str(tt_scale / 100.0) + ']'
277 tw += '{' + typew + '}'
280 pretext.append('\\defaultfontfeatures{Numbers=OldStyle}')
281 pretext.append('\\usepackage{xunicode}')
282 pretext.append('\\usepackage{xltxtra}')
283 insert_to_preamble(document, pretext)
286 def revert_outputformat(document):
287 " Remove default output format param "
289 if not del_token(document.header, '\\default_output_format', 0):
290 document.warning("Malformed LyX document: Missing \\default_output_format.")
293 def revert_backgroundcolor(document):
294 " Reverts background color to preamble code "
295 i = find_token(document.header, "\\backgroundcolor", 0)
298 colorcode = get_value(document.header, '\\backgroundcolor', i)
299 del document.header[i]
300 # don't clutter the preamble if backgroundcolor is not set
301 if colorcode == "#ffffff":
303 red = hex2ratio(colorcode[1:3])
304 green = hex2ratio(colorcode[3:5])
305 blue = hex2ratio(colorcode[5:7])
306 insert_to_preamble(document, \
307 ['% To set the background color',
308 '\\@ifundefined{definecolor}{\\usepackage{color}}{}',
309 '\\definecolor{page_backgroundcolor}{rgb}{' + red + ',' + green + ',' + blue + '}',
310 '\\pagecolor{page_backgroundcolor}'])
313 def add_use_indices(document):
314 " Add \\use_indices if it is missing "
315 i = find_token(document.header, '\\use_indices', 0)
318 i = find_token(document.header, '\\use_bibtopic', 0)
320 i = find_token(document.header, '\\cite_engine', 0)
322 i = find_token(document.header, '\\use_mathdots', 0)
324 i = find_token(document.header, '\\use_mhchem', 0)
326 i = find_token(document.header, '\\use_esint', 0)
328 i = find_token(document.header, '\\use_amsmath', 0)
330 document.warning("Malformed LyX document: Missing \\use_indices.")
332 document.header.insert(i + 1, '\\use_indices 0')
336 def revert_splitindex(document):
337 " Reverts splitindex-aware documents "
338 i = add_use_indices(document)
341 useindices = str2bool(get_value(document.header, "\\use_indices", i))
342 del document.header[i]
345 preamble.append("\\usepackage{splitidx})")
347 # deal with index declarations in the preamble
350 i = find_token(document.header, "\\index", i)
353 k = find_token(document.header, "\\end_index", i)
355 document.warning("Malformed LyX document: Missing \\end_index.")
358 line = document.header[i]
359 l = re.compile(r'\\index (.*)$')
362 ishortcut = get_value(document.header, '\\shortcut', i, k)
364 preamble.append("\\newindex[" + iname + "]{" + ishortcut + "}")
365 del document.header[i:k + 1]
367 insert_to_preamble(document, preamble)
369 # deal with index insets
370 # these need to have the argument removed
373 i = find_token(document.body, "\\begin_inset Index", i)
376 line = document.body[i]
377 l = re.compile(r'\\begin_inset Index (.*)$')
380 if itype == "idx" or indices == "false":
381 document.body[i] = "\\begin_inset Index"
383 k = find_end_of_inset(document.body, i)
385 document.warning("Can't find end of index inset!")
388 content = lyx2latex(document, document.body[i:k])
390 content = content.replace('"', r'\"')
391 subst = put_cmd_in_ert("\\sindex[" + itype + "]{" + content + "}")
392 document.body[i:k + 1] = subst
395 # deal with index_print insets
398 i = find_token(document.body, "\\begin_inset CommandInset index_print", i)
401 k = find_end_of_inset(document.body, i)
402 ptype = get_quoted_value(document.body, 'type', i, k)
404 j = find_token(document.body, "type", i, k)
407 del document.body[i:k + 1]
409 subst = put_cmd_in_ert("\\printindex[" + ptype + "]{}")
410 document.body[i:k + 1] = subst
414 def convert_splitindex(document):
415 " Converts index and printindex insets to splitindex-aware format "
416 add_use_indices(document)
419 i = find_token(document.body, "\\begin_inset Index", i)
422 document.body[i] = document.body[i].replace("\\begin_inset Index",
423 "\\begin_inset Index idx")
427 i = find_token(document.body, "\\begin_inset CommandInset index_print", i)
430 if document.body[i + 1].find('LatexCommand printindex') == -1:
431 document.warning("Malformed LyX document: Incomplete printindex inset.")
433 subst = ["LatexCommand printindex",
435 document.body[i + 1:i + 2] = subst
439 def revert_subindex(document):
440 " Reverts \\printsubindex CommandInset types "
441 i = add_use_indices(document)
444 useindices = str2bool(get_value(document.header, "\\use_indices", i))
447 i = find_token(document.body, "\\begin_inset CommandInset index_print", i)
450 k = find_end_of_inset(document.body, i)
451 ctype = get_value(document.body, 'LatexCommand', i, k)
452 if ctype != "printsubindex":
455 ptype = get_quoted_value(document.body, 'type', i, k)
457 del document.body[i:k + 1]
459 subst = put_cmd_in_ert("\\printsubindex[" + ptype + "]{}")
460 document.body[i:k + 1] = subst
464 def revert_printindexall(document):
465 " Reverts \\print[sub]index* CommandInset types "
466 i = add_use_indices(document)
469 useindices = str2bool(get_value(document.header, "\\use_indices", i))
472 i = find_token(document.body, "\\begin_inset CommandInset index_print", i)
475 k = find_end_of_inset(document.body, i)
476 ctype = get_value(document.body, 'LatexCommand', i, k)
477 if ctype != "printindex*" and ctype != "printsubindex*":
481 del document.body[i:k + 1]
483 subst = put_cmd_in_ert("\\" + ctype + "{}")
484 document.body[i:k + 1] = subst
488 def revert_strikeout(document):
489 " Reverts \\strikeout font attribute "
490 changed = revert_font_attrs(document.body, "\\uuline", "\\uuline")
491 changed = revert_font_attrs(document.body, "\\uwave", "\\uwave") or changed
492 changed = revert_font_attrs(document.body, "\\strikeout", "\\sout") or changed
494 insert_to_preamble(document, \
495 ['% for proper underlining',
496 '\\PassOptionsToPackage{normalem}{ulem}',
497 '\\usepackage{ulem}'])
500 def revert_ulinelatex(document):
501 " Reverts \\uline font attribute "
502 i = find_token(document.body, '\\bar under', 0)
505 insert_to_preamble(document,\
506 ['% for proper underlining',
507 '\\PassOptionsToPackage{normalem}{ulem}',
508 '\\usepackage{ulem}',
509 '\\let\\cite@rig\\cite',
510 '\\newcommand{\\b@xcite}[2][\\%]{\\def\\def@pt{\\%}\\def\\pas@pt{#1}',
511 ' \\mbox{\\ifx\\def@pt\\pas@pt\\cite@rig{#2}\\else\\cite@rig[#1]{#2}\\fi}}',
512 '\\renewcommand{\\underbar}[1]{{\\let\\cite\\b@xcite\\uline{#1}}}'])
515 def revert_custom_processors(document):
516 " Remove bibtex_command and index_command params "
518 if not del_token(document.header, '\\bibtex_command', 0):
519 document.warning("Malformed LyX document: Missing \\bibtex_command.")
521 if not del_token(document.header, '\\index_command', 0):
522 document.warning("Malformed LyX document: Missing \\index_command.")
525 def convert_nomencl_width(document):
526 " Add set_width param to nomencl_print "
529 i = find_token(document.body, "\\begin_inset CommandInset nomencl_print", i)
532 document.body.insert(i + 2, "set_width \"none\"")
536 def revert_nomencl_width(document):
537 " Remove set_width param from nomencl_print "
540 i = find_token(document.body, "\\begin_inset CommandInset nomencl_print", i)
543 j = find_end_of_inset(document.body, i)
544 if not del_token(document.body, "set_width", i, j):
545 document.warning("Can't find set_width option for nomencl_print!")
549 def revert_nomencl_cwidth(document):
550 " Remove width param from nomencl_print "
553 i = find_token(document.body, "\\begin_inset CommandInset nomencl_print", i)
556 j = find_end_of_inset(document.body, i)
557 l = find_token(document.body, "width", i, j)
561 width = get_quoted_value(document.body, "width", i, j)
563 insert_to_preamble(document, ["\\setlength{\\nomlabelwidth}{" + width + "}"])
567 def revert_applemac(document):
568 " Revert applemac encoding to auto "
569 if document.encoding != "applemac":
571 document.encoding = "auto"
572 i = find_token(document.header, "\\encoding", 0)
574 document.header[i] = "\\encoding auto"
577 def revert_longtable_align(document):
578 " Remove longtable alignment setting "
581 i = find_token(document.body, "\\begin_inset Tabular", i)
584 end = find_end_of_inset(document.body, i)
586 document.warning("Can't find end of inset at line " + str(i))
589 fline = find_token(document.body, "<features", i, end)
591 document.warning("Can't find features for inset at line " + str(i))
594 j = document.body[fline].find("longtabularalignment")
598 # FIXME Is this correct? It wipes out everything after the
600 document.body[fline] = document.body[fline][:j - 1] + '>'
601 # since there could be a tabular inside this one, we
602 # cannot jump to end.
606 def revert_branch_filename(document):
607 " Remove \\filename_suffix parameter from branches "
610 i = find_token(document.header, "\\filename_suffix", i)
613 del document.header[i]
616 def revert_paragraph_indentation(document):
617 " Revert custom paragraph indentation to preamble code "
618 i = find_token(document.header, "\\paragraph_indentation", 0)
621 length = get_value(document.header, "\\paragraph_indentation", i)
622 # we need only remove the line if indentation is default
623 if length != "default":
624 # handle percent lengths
625 length = latex_length(length)[1]
626 insert_to_preamble(document, ["\\setlength{\\parindent}{" + length + "}"])
627 del document.header[i]
630 def revert_percent_skip_lengths(document):
631 " Revert relative lengths for paragraph skip separation to preamble code "
632 i = find_token(document.header, "\\defskip", 0)
635 length = get_value(document.header, "\\defskip", i)
636 # only revert when a custom length was set and when
637 # it used a percent length
638 if length in ('smallskip', 'medskip', 'bigskip'):
640 # handle percent lengths
641 percent, length = latex_length(length)
643 insert_to_preamble(document, ["\\setlength{\\parskip}{" + length + "}"])
644 # set defskip to medskip as default
645 document.header[i] = "\\defskip medskip"
648 def revert_percent_vspace_lengths(document):
649 " Revert relative VSpace lengths to ERT "
652 i = find_token(document.body, "\\begin_inset VSpace", i)
655 # only revert if a custom length was set and if
656 # it used a percent length
657 r = re.compile(r'\\begin_inset VSpace (.*)$')
658 m = r.match(document.body[i])
660 if length in ('defskip', 'smallskip', 'medskip', 'bigskip', 'vfill'):
663 # check if the space has a star (protected space)
664 protected = (document.body[i].rfind("*") != -1)
666 length = length.rstrip('*')
667 # handle percent lengths
668 percent, length = latex_length(length)
669 # revert the VSpace inset to ERT
672 subst = put_cmd_in_ert("\\vspace*{" + length + "}")
674 subst = put_cmd_in_ert("\\vspace{" + length + "}")
675 document.body[i:i + 2] = subst
679 def revert_percent_hspace_lengths(document):
680 " Revert relative HSpace lengths to ERT "
683 i = find_token_exact(document.body, "\\begin_inset space \\hspace", i)
686 j = find_end_of_inset(document.body, i)
688 document.warning("Can't find end of inset at line " + str(i))
691 # only revert if a custom length was set...
692 length = get_value(document.body, '\\length', i + 1, j)
694 document.warning("Malformed lyx document: Missing '\\length' in Space inset.")
698 if document.body[i].find("\\hspace*{}") != -1:
700 # ...and if it used a percent length
701 percent, length = latex_length(length)
702 # revert the HSpace inset to ERT
704 subst = put_cmd_in_ert("\\hspace" + protected + "{" + length + "}")
705 document.body[i:j + 1] = subst
706 # if we did a substitution, this will still be ok
710 def revert_hspace_glue_lengths(document):
711 " Revert HSpace glue lengths to ERT "
714 i = find_token_exact(document.body, "\\begin_inset space \\hspace", i)
717 j = find_end_of_inset(document.body, i)
719 document.warning("Can't find end of inset at line " + str(i))
722 length = get_value(document.body, '\\length', i + 1, j)
724 document.warning("Malformed lyx document: Missing '\\length' in Space inset.")
728 if document.body[i].find("\\hspace*{}") != -1:
730 # only revert if the length contains a plus or minus at pos != 0
731 if length.find('-',1) != -1 or length.find('+',1) != -1:
732 # handle percent lengths
733 length = latex_length(length)[1]
734 # revert the HSpace inset to ERT
735 subst = put_cmd_in_ert("\\hspace" + protected + "{" + length + "}")
736 document.body[i:j+1] = subst
740 def convert_author_id(document):
741 " Add the author_id to the \\author definition and make sure 0 is not used"
744 re_author = re.compile(r'(\\author) (\".*\")\s*(.*)$')
747 i = find_token(document.header, "\\author", i)
750 m = re_author.match(document.header[i])
754 document.header[i] = "\\author %i %s %s" % (anum, name, email)
760 i = find_token(document.body, "\\change_", i)
763 change = document.body[i].split(' ');
766 author_id = int(change[1])
768 document.body[i] = "%s %i %s" % (type, author_id + 1, time)
772 def revert_author_id(document):
773 " Remove the author_id from the \\author definition "
776 rx = re.compile(r'(\\author)\s+(-?\d+)\s+(\".*\")\s*(.*)$')
780 i = find_token(document.header, "\\author", i)
783 m = rx.match(document.header[i])
785 author_id = int(m.group(2))
786 idmap[author_id] = anum
789 document.header[i] = "\\author %s %s" % (name, email)
791 # FIXME Should this be incremented if we didn't match?
796 i = find_token(document.body, "\\change_", i)
799 change = document.body[i].split(' ');
802 author_id = int(change[1])
804 document.body[i] = "%s %i %s" % (type, idmap[author_id], time)
808 def revert_suppress_date(document):
809 " Revert suppressing of default document date to preamble code "
810 i = find_token(document.header, "\\suppress_date", 0)
813 # remove the preamble line and write to the preamble
814 # when suppress_date was true
815 date = str2bool(get_value(document.header, "\\suppress_date", i))
817 add_to_preamble(document, ["\\date{}"])
818 del document.header[i]
821 def convert_mhchem(document):
822 "Set mhchem to off for versions older than 1.6.x"
823 if document.initial_format < 277:
824 # LyX 1.5.x and older did never load mhchem.
825 # Therefore we must switch it off: Documents that use mhchem have
826 # a manual \usepackage anyway, and documents not using mhchem but
827 # custom macros with the same names as mhchem commands might get
828 # corrupted if mhchem is automatically loaded.
831 # LyX 1.6.x did always load mhchem automatically.
833 i = find_token(document.header, "\\use_esint", 0)
836 i = find_token(document.header, "\\use_amsmath", 0)
838 document.warning("Malformed LyX document: Could not find amsmath os esint setting.")
840 document.header.insert(i + 1, "\\use_mhchem %d" % mhchem)
843 def revert_mhchem(document):
844 "Revert mhchem loading to preamble code"
847 i = find_token(document.header, "\\use_mhchem", 0)
849 document.warning("Malformed LyX document: Could not find mhchem setting.")
852 val = get_value(document.header, "\\use_mhchem", i)
857 del document.header[i]
866 i = find_token(document.body, "\\begin_inset Formula", i)
869 line = document.body[i]
870 if line.find("\\ce{") != -1 or line.find("\\cf{") != -1:
876 pre = ["\\PassOptionsToPackage{version=3}{mhchem}",
877 "\\usepackage{mhchem}"]
878 insert_to_preamble(document, pre)
881 def revert_fontenc(document):
882 " Remove fontencoding param "
883 if not del_token(document.header, '\\fontencoding', 0):
884 document.warning("Malformed LyX document: Missing \\fontencoding.")
887 def merge_gbrief(document):
888 " Merge g-brief-en and g-brief-de to one class "
890 if document.textclass != "g-brief-de":
891 if document.textclass == "g-brief-en":
892 document.textclass = "g-brief"
893 document.set_textclass()
896 obsoletedby = { "Brieftext": "Letter",
897 "Unterschrift": "Signature",
899 "Zusatz": "Addition",
902 "RetourAdresse": "ReturnAddress",
903 "MeinZeichen": "MyRef",
904 "IhrZeichen": "YourRef",
905 "IhrSchreiben": "YourMail",
908 "Konto": "BankAccount",
909 "Postvermerk": "PostalComment",
910 "Adresse": "Address",
912 "Betreff": "Reference",
919 i = find_token(document.body, "\\begin_layout", i)
923 layout = document.body[i][14:]
924 if layout in obsoletedby:
925 document.body[i] = "\\begin_layout " + obsoletedby[layout]
929 document.textclass = "g-brief"
930 document.set_textclass()
933 def revert_gbrief(document):
934 " Revert g-brief to g-brief-en "
935 if document.textclass == "g-brief":
936 document.textclass = "g-brief-en"
937 document.set_textclass()
940 def revert_html_options(document):
941 " Remove html options "
942 del_token(document.header, '\\html_use_mathml', 0)
943 del_token(document.header, '\\html_be_strict', 0)
946 def revert_includeonly(document):
949 i = find_token(document.header, "\\begin_includeonly", i)
952 j = find_end_of(document.header, i, "\\begin_includeonly", "\\end_includeonly")
954 document.warning("Unable to find end of includeonly section!!")
956 document.header[i : j + 1] = []
959 def revert_includeall(document):
960 " Remove maintain_unincluded_children param "
961 del_token(document.header, '\\maintain_unincluded_children', 0)
964 def revert_multirow(document):
965 " Revert multirow cells in tables to TeX-code"
967 # first, let's find out if we need to do anything
968 # cell type 3 is multirow begin cell
969 i = find_token(document.body, '<cell multirow="3"', 0)
973 add_to_preamble(document, ["\\usepackage{multirow}"])
977 # find begin/end of table
978 begin_table = find_token(document.body, '<lyxtabular version=', begin_table)
979 if begin_table == -1:
981 end_table = find_end_of(document.body, begin_table, '<lyxtabular', '</lyxtabular>')
983 document.warning("Malformed LyX document: Could not find end of table.")
986 # does this table have multirow?
987 i = find_token(document.body, '<cell multirow="3"', begin_table, end_table)
989 begin_table = end_table
992 # store the number of rows and columns
993 numrows = get_option_value(document.body[begin_table], "rows")
994 numcols = get_option_value(document.body[begin_table], "columns")
996 numrows = int(numrows)
997 numcols = int(numcols)
999 document.warning("Unable to determine rows and columns!")
1000 begin_table = end_table
1005 # collect info on rows and columns of this table.
1006 begin_row = begin_table
1007 for row in range(numrows):
1008 begin_row = find_token(document.body, '<row>', begin_row, end_table)
1010 document.warning("Can't find row " + str(row + 1))
1012 end_row = find_end_of(document.body, begin_row, '<row>', '</row>')
1014 document.warning("Can't find end of row " + str(row + 1))
1016 begin_cell = begin_row
1017 multirows.append([])
1018 for column in range(numcols):
1019 begin_cell = find_token(document.body, '<cell ', begin_cell, end_row)
1020 if begin_cell == -1:
1021 document.warning("Can't find column " + str(column + 1) + \
1022 "in row " + str(row + 1))
1025 # this will fail if someone puts "</cell>" in a cell, but
1026 # that seems fairly unlikely.
1027 end_cell = find_end_of(document.body, begin_cell, '<cell', '</cell>')
1029 document.warning("Can't find end of column " + str(column + 1) + \
1030 "in row " + str(row + 1))
1032 multirows[row].append([begin_cell, end_cell, 0])
1033 if document.body[begin_cell].find('multirow="3"') != -1:
1034 multirows[row][column][2] = 3 # begin multirow
1035 mrstarts.append([row, column])
1036 elif document.body[begin_cell].find('multirow="4"') != -1:
1037 multirows[row][column][2] = 4 # in multirow
1038 begin_cell = end_cell
1040 # end of table info collection
1042 # work from the back to avoid messing up numbering
1048 col_width = get_option_value(document.body[begin_table + 2 + col], "width")
1049 # "0pt" means that no width is specified
1050 if not col_width or col_width == "0pt":
1052 # determine the number of cells that are part of the multirow
1054 for r in range(row + 1, numrows):
1055 if multirows[r][col][2] != 4:
1058 # take the opportunity to revert this line
1059 lineno = multirows[r][col][0]
1060 document.body[lineno] = document.body[lineno].\
1061 replace(' multirow="4" ', ' ').\
1062 replace('valignment="middle"', 'valignment="top"').\
1063 replace(' topline="true" ', ' ')
1064 # remove bottom line of previous multirow-part cell
1065 lineno = multirows[r-1][col][0]
1066 document.body[lineno] = document.body[lineno].replace(' bottomline="true" ', ' ')
1067 # revert beginning cell
1068 bcell = multirows[row][col][0]
1069 ecell = multirows[row][col][1]
1070 document.body[bcell] = document.body[bcell].\
1071 replace(' multirow="3" ', ' ').\
1072 replace('valignment="middle"', 'valignment="top"')
1073 blay = find_token(document.body, "\\begin_layout", bcell, ecell)
1075 document.warning("Can't find layout for cell!")
1077 bend = find_end_of_layout(document.body, blay)
1079 document.warning("Can't find end of layout for cell!")
1081 # do the later one first, so as not to mess up the numbering
1082 # we are wrapping the whole cell in this ert
1083 # so before the end of the layout...
1084 document.body[bend:bend] = put_cmd_in_ert("}")
1085 # ...and after the beginning
1086 document.body[blay + 1:blay + 1] = \
1087 put_cmd_in_ert("\\multirow{" + str(nummrs) + "}{" + col_width + "}{")
1089 begin_table = end_table
1092 def convert_math_output(document):
1093 " Convert \html_use_mathml to \html_math_output "
1094 i = find_token(document.header, "\\html_use_mathml", 0)
1097 rgx = re.compile(r'\\html_use_mathml\s+(\w+)')
1098 m = rgx.match(document.header[i])
1099 newval = "0" # MathML
1101 val = str2bool(m.group(1))
1103 newval = "2" # Images
1105 document.warning("Can't match " + document.header[i])
1106 document.header[i] = "\\html_math_output " + newval
1109 def revert_math_output(document):
1110 " Revert \html_math_output to \html_use_mathml "
1111 i = find_token(document.header, "\\html_math_output", 0)
1114 rgx = re.compile(r'\\html_math_output\s+(\d)')
1115 m = rgx.match(document.header[i])
1119 if val == "1" or val == "2":
1122 document.warning("Unable to match " + document.header[i])
1123 document.header[i] = "\\html_use_mathml " + newval
1127 def revert_inset_preview(document):
1128 " Dissolves the preview inset "
1131 i = find_token(document.body, "\\begin_inset Preview", i)
1134 iend = find_end_of_inset(document.body, i)
1136 document.warning("Malformed LyX document: Could not find end of Preview inset.")
1140 # This has several issues.
1141 # We need to do something about the layouts inside InsetPreview.
1142 # If we just leave the first one, then we have something like:
1143 # \begin_layout Standard
1145 # \begin_layout Standard
1146 # and we get a "no \end_layout" error. So something has to be done.
1147 # Ideally, we would check if it is the same as the layout we are in.
1148 # If so, we just remove it; if not, we end the active one. But it is
1149 # not easy to know what layout we are in, due to depth changes, etc,
1150 # and it is not clear to me how much work it is worth doing. In most
1151 # cases, the layout will probably be the same.
1153 # For the same reason, we have to remove the \end_layout tag at the
1154 # end of the last layout in the inset. Again, that will sometimes be
1155 # wrong, but it will usually be right. To know what to do, we would
1156 # again have to know what layout the inset is in.
1158 blay = find_token(document.body, "\\begin_layout", i, iend)
1160 document.warning("Can't find layout for preview inset!")
1161 # always do the later one first...
1162 del document.body[iend]
1163 del document.body[i]
1164 # deletions mean we do not need to reset i
1167 # This is where we would check what layout we are in.
1168 # The check for Standard is definitely wrong.
1170 # lay = document.body[blay].split(None, 1)[1]
1171 # if lay != oldlayout:
1172 # # record a boolean to tell us what to do later....
1173 # # better to do it later, since (a) it won't mess up
1174 # # the numbering and (b) we only modify at the end.
1176 # we want to delete the last \\end_layout in this inset, too.
1177 # note that this may not be the \\end_layout that goes with blay!!
1178 bend = find_end_of_layout(document.body, blay)
1180 tmp = find_token(document.body, "\\end_layout", bend + 1, iend)
1185 document.warning("Unable to find last layout in preview inset!")
1186 del document.body[iend]
1187 del document.body[i]
1188 # deletions mean we do not need to reset i
1190 # always do the later one first...
1191 del document.body[iend]
1192 del document.body[bend]
1193 del document.body[i:blay + 1]
1194 # we do not need to reset i
1197 def revert_equalspacing_xymatrix(document):
1198 " Revert a Formula with xymatrix@! to an ERT inset "
1200 has_preamble = False
1201 has_equal_spacing = False
1204 i = find_token(document.body, "\\begin_inset Formula", i)
1207 j = find_end_of_inset(document.body, i)
1209 document.warning("Malformed LyX document: Could not find end of Formula inset.")
1213 for curline in range(i,j):
1214 found = document.body[curline].find("\\xymatrix@!")
1219 has_equal_spacing = True
1220 content = [document.body[i][21:]]
1221 content += document.body[i + 1:j]
1222 subst = put_cmd_in_ert(content)
1223 document.body[i:j + 1] = subst
1224 i += len(subst) - (j - i) + 1
1226 for curline in range(i,j):
1227 l = document.body[curline].find("\\xymatrix")
1229 has_preamble = True;
1233 if has_equal_spacing and not has_preamble:
1234 add_to_preamble(document, ['\\usepackage[all]{xy}'])
1237 def revert_notefontcolor(document):
1238 " Reverts greyed-out note font color to preamble code "
1240 i = find_token(document.header, "\\notefontcolor", 0)
1244 colorcode = get_value(document.header, '\\notefontcolor', i)
1245 del document.header[i]
1247 # are there any grey notes?
1248 if find_token(document.body, "\\begin_inset Note Greyedout", 0) == -1:
1249 # no need to do anything else, and \renewcommand will throw
1250 # an error since lyxgreyedout will not exist.
1253 # the color code is in the form #rrggbb where every character denotes a hex number
1254 red = hex2ratio(colorcode[1:3])
1255 green = hex2ratio(colorcode[3:5])
1256 blue = hex2ratio(colorcode[5:7])
1257 # write the preamble
1258 insert_to_preamble(document,
1259 [ '% for greyed-out notes',
1260 '\\@ifundefined{definecolor}{\\usepackage{color}}{}'
1261 '\\definecolor{note_fontcolor}{rgb}{%s,%s,%s}' % (red, green, blue),
1262 '\\renewenvironment{lyxgreyedout}',
1263 ' {\\textcolor{note_fontcolor}\\bgroup}{\\egroup}'])
1266 def revert_turkmen(document):
1267 "Set language Turkmen to English"
1269 if document.language == "turkmen":
1270 document.language = "english"
1271 i = find_token(document.header, "\\language", 0)
1273 document.header[i] = "\\language english"
1277 j = find_token(document.body, "\\lang turkmen", j)
1280 document.body[j] = document.body[j].replace("\\lang turkmen", "\\lang english")
1284 def revert_fontcolor(document):
1285 " Reverts font color to preamble code "
1286 i = find_token(document.header, "\\fontcolor", 0)
1289 colorcode = get_value(document.header, '\\fontcolor', i)
1290 del document.header[i]
1291 # don't clutter the preamble if font color is not set
1292 if colorcode == "#000000":
1294 # the color code is in the form #rrggbb where every character denotes a hex number
1295 red = hex2ratio(colorcode[1:3])
1296 green = hex2ratio(colorcode[3:5])
1297 blue = hex2ratio(colorcode[5:7])
1298 # write the preamble
1299 insert_to_preamble(document,
1300 ['% Set the font color',
1301 '\\@ifundefined{definecolor}{\\usepackage{color}}{}',
1302 '\\definecolor{document_fontcolor}{rgb}{%s,%s,%s}' % (red, green, blue),
1303 '\\color{document_fontcolor}'])
1306 def revert_shadedboxcolor(document):
1307 " Reverts shaded box color to preamble code "
1308 i = find_token(document.header, "\\boxbgcolor", 0)
1311 colorcode = get_value(document.header, '\\boxbgcolor', i)
1312 del document.header[i]
1313 # the color code is in the form #rrggbb
1314 red = hex2ratio(colorcode[1:3])
1315 green = hex2ratio(colorcode[3:5])
1316 blue = hex2ratio(colorcode[5:7])
1317 # write the preamble
1318 insert_to_preamble(document,
1319 ['% Set the color of boxes with shaded background',
1320 '\\@ifundefined{definecolor}{\\usepackage{color}}{}',
1321 "\\definecolor{shadecolor}{rgb}{%s,%s,%s}" % (red, green, blue)])
1324 def revert_lyx_version(document):
1325 " Reverts LyX Version information from Inset Info "
1326 version = "LyX version"
1328 import lyx2lyx_version
1329 version = lyx2lyx_version.version
1335 i = find_token(document.body, '\\begin_inset Info', i)
1338 j = find_end_of_inset(document.body, i + 1)
1340 document.warning("Malformed LyX document: Could not find end of Info inset.")
1349 typ = get_quoted_value(document.body, "type", i, j)
1350 arg = get_quoted_value(document.body, "arg", i, j)
1351 if arg != "version" or typ != "lyxinfo":
1355 # We do not actually know the version of LyX used to produce the document.
1356 # But we can use our version, since we are reverting.
1358 # Now we want to check if the line after "\end_inset" is empty. It normally
1359 # is, so we want to remove it, too.
1361 if document.body[j + 1].strip() == "":
1363 document.body[i: lastline] = s
1367 def revert_math_scale(document):
1368 " Remove math scaling and LaTeX options "
1369 del_token(document.header, '\\html_math_img_scale', 0)
1370 del_token(document.header, '\\html_latex_start', 0)
1371 del_token(document.header, '\\html_latex_end', 0)
1374 def revert_pagesizes(document):
1375 " Revert page sizes to default "
1376 i = find_token(document.header, '\\papersize', 0)
1378 size = document.header[i][11:]
1379 if size == "a0paper" or size == "a1paper" or size == "a2paper" \
1380 or size == "a6paper" or size == "b0paper" or size == "b1paper" \
1381 or size == "b2paper" or size == "b6paper" or size == "b0j" \
1382 or size == "b1j" or size == "b2j" or size == "b3j" or size == "b4j" \
1383 or size == "b5j" or size == "b6j":
1384 del document.header[i]
1387 def revert_DIN_C_pagesizes(document):
1388 " Revert DIN C page sizes to default "
1389 i = find_token(document.header, '\\papersize', 0)
1391 size = document.header[i][11:]
1392 if size == "c0paper" or size == "c1paper" or size == "c2paper" \
1393 or size == "c3paper" or size == "c4paper" or size == "c5paper" \
1394 or size == "c6paper":
1395 del document.header[i]
1398 def convert_html_quotes(document):
1399 " Remove quotes around html_latex_start and html_latex_end "
1401 i = find_token(document.header, '\\html_latex_start', 0)
1403 line = document.header[i]
1404 l = re.compile(r'\\html_latex_start\s+"(.*)"')
1407 document.header[i] = "\\html_latex_start " + m.group(1)
1409 i = find_token(document.header, '\\html_latex_end', 0)
1411 line = document.header[i]
1412 l = re.compile(r'\\html_latex_end\s+"(.*)"')
1415 document.header[i] = "\\html_latex_end " + m.group(1)
1418 def revert_html_quotes(document):
1419 " Remove quotes around html_latex_start and html_latex_end "
1421 i = find_token(document.header, '\\html_latex_start', 0)
1423 line = document.header[i]
1424 l = re.compile(r'\\html_latex_start\s+(.*)')
1427 document.warning("Weird html_latex_start line: " + line)
1428 del document.header[i]
1430 document.header[i] = "\\html_latex_start \"" + m.group(1) + "\""
1432 i = find_token(document.header, '\\html_latex_end', 0)
1434 line = document.header[i]
1435 l = re.compile(r'\\html_latex_end\s+(.*)')
1438 document.warning("Weird html_latex_end line: " + line)
1439 del document.header[i]
1441 document.header[i] = "\\html_latex_end \"" + m.group(1) + "\""
1444 def revert_output_sync(document):
1445 " Remove forward search options "
1446 del_token(document.header, '\\output_sync_macro', 0)
1447 del_token(document.header, '\\output_sync', 0)
1450 def revert_align_decimal(document):
1453 i = find_token(document.body, "\\begin_inset Tabular", i)
1456 j = find_end_of_inset(document.body, i)
1458 document.warning("Unable to find end of Tabular inset at line " + str(i))
1461 cell = find_token(document.body, "<cell", i, j)
1463 document.warning("Can't find any cells in Tabular inset at line " + str(i))
1468 k = find_token(document.body, "<column", k, cell)
1471 if document.body[k].find('alignment="decimal"') == -1:
1474 remove_option(document.body, k, 'decimal_point')
1475 document.body[k] = \
1476 document.body[k].replace('alignment="decimal"', 'alignment="center"')
1480 def convert_optarg(document):
1481 " Convert \\begin_inset OptArg to \\begin_inset Argument "
1484 i = find_token(document.body, '\\begin_inset OptArg', i)
1487 document.body[i] = "\\begin_inset Argument"
1491 def revert_argument(document):
1492 " Convert \\begin_inset Argument to \\begin_inset OptArg "
1495 i = find_token(document.body, '\\begin_inset Argument', i)
1498 document.body[i] = "\\begin_inset OptArg"
1502 def revert_makebox(document):
1503 " Convert \\makebox to TeX code "
1506 i = find_token(document.body, '\\begin_inset Box', i)
1509 z = find_end_of_inset(document.body, i)
1511 document.warning("Malformed LyX document: Can't find end of box inset.")
1514 blay = find_token(document.body, "\\begin_layout", i, z)
1516 document.warning("Malformed LyX document: Can't find layout in box.")
1519 j = find_token(document.body, 'use_makebox', i)
1520 if j == -1 or j != i +6:
1521 document.warning("Malformed LyX document: Can't find use_makebox statement in box.")
1524 # delete use_makebox
1525 if not check_token(document.body[i], "\\begin_inset Box Frameless") \
1526 or get_value(document.body, 'use_makebox', j) != 1:
1527 del document.body[j]
1530 bend = find_end_of_layout(document.body, blay)
1531 if bend == -1 or bend > z:
1532 document.warning("Malformed LyX document: Can't find end of layout in box.")
1535 # determine the alignment
1536 align = get_quoted_value(document.body, 'hor_pos', i, blay, "c")
1537 # determine the width
1538 length = get_quoted_value(document.body, 'width', i, blay, "50col%")
1539 length = latex_length(length)[1]
1540 # remove the \end_layout \end_inset pair
1541 document.body[bend:z + 1] = put_cmd_in_ert("}")
1542 subst = "\\makebox[" + length + "][" \
1544 document.body[i:blay + 1] = put_cmd_in_ert(subst)
1548 def convert_use_makebox(document):
1549 " Adds use_makebox option for boxes "
1552 i = find_token(document.body, '\\begin_inset Box', i)
1555 k = find_token(document.body, 'use_parbox', i)
1556 if k == -1 or k != i + 5:
1557 document.warning("Malformed LyX document: Can't find use_parbox statement in box.")
1561 document.body.insert(k + 1, "use_makebox 0")
1565 def revert_IEEEtran(document):
1566 " Convert IEEEtran layouts and styles to TeX code "
1568 if document.textclass != "IEEEtran":
1571 revert_flex_inset(document.body, "IEEE membership", "\\IEEEmembership")
1572 revert_flex_inset(document.body, "Lowercase", "\\MakeLowercase")
1574 layouts = ("Special Paper Notice", "After Title Text", "Publication ID",
1575 "Page headings", "Biography without photo")
1576 latexcmd = {"Special Paper Notice": "\\IEEEspecialpapernotice",
1577 "After Title Text": "\\IEEEaftertitletext",
1578 "Publication ID": "\\IEEEpubid"}
1579 obsoletedby = {"Page headings": "MarkBoth",
1580 "Biography without photo": "BiographyNoPhoto"}
1582 for layout in layouts:
1585 i = find_token(document.body, '\\begin_layout ' + layout, i)
1588 j = find_end_of_layout(document.body, i)
1590 document.warning("Malformed LyX document: Can't find end of " + layout + " layout.")
1593 if layout in list(obsoletedby.keys()):
1594 document.body[i] = "\\begin_layout " + obsoletedby[layout]
1597 content = lyx2latex(document, document.body[i:j + 1])
1598 add_to_preamble(document, [latexcmd[layout] + "{" + content + "}"])
1599 del document.body[i:j + 1]
1600 # no need to reset i
1603 def convert_prettyref(document):
1604 " Converts prettyref references to neutral formatted refs "
1605 re_ref = re.compile("^\s*reference\s+\"(\w+):(\S+)\"")
1606 nm_ref = re.compile("^\s*name\s+\"(\w+):(\S+)\"")
1610 i = find_token(document.body, "\\begin_inset CommandInset ref", i)
1613 j = find_end_of_inset(document.body, i)
1615 document.warning("Malformed LyX document: No end of InsetRef!")
1618 k = find_token(document.body, "LatexCommand prettyref", i, j)
1620 document.body[k] = "LatexCommand formatted"
1622 document.header.insert(-1, "\\use_refstyle 0")
1625 def revert_refstyle(document):
1626 " Reverts neutral formatted refs to prettyref "
1627 re_ref = re.compile("^reference\s+\"(\w+):(\S+)\"")
1628 nm_ref = re.compile("^\s*name\s+\"(\w+):(\S+)\"")
1632 i = find_token(document.body, "\\begin_inset CommandInset ref", i)
1635 j = find_end_of_inset(document.body, i)
1637 document.warning("Malformed LyX document: No end of InsetRef")
1640 k = find_token(document.body, "LatexCommand formatted", i, j)
1642 document.body[k] = "LatexCommand prettyref"
1644 i = find_token(document.header, "\\use_refstyle", 0)
1646 document.header.pop(i)
1649 def revert_nameref(document):
1650 " Convert namerefs to regular references "
1651 cmds = ["Nameref", "nameref"]
1653 rx = re.compile(r'reference "(.*)"')
1656 oldcmd = "LatexCommand " + cmd
1658 # It seems better to look for this, as most of the reference
1659 # insets won't be ones we care about.
1660 i = find_token(document.body, oldcmd, i)
1665 # Make sure it is actually in an inset!
1666 # A normal line could begin with "LatexCommand nameref"!
1667 val = is_in_inset(document.body, cmdloc, \
1668 "\\begin_inset CommandInset ref")
1673 # ok, so it is in an InsetRef
1674 refline = find_token(document.body, "reference", stins, endins)
1676 document.warning("Can't find reference for inset at line " + stinst + "!!")
1678 m = rx.match(document.body[refline])
1680 document.warning("Can't match reference line: " + document.body[ref])
1684 newcontent = put_cmd_in_ert('\\' + cmd + '{' + ref + '}')
1685 document.body[stins:endins + 1] = newcontent
1688 add_to_preamble(document, ["\\usepackage{nameref}"])
1691 def remove_Nameref(document):
1692 " Convert Nameref commands to nameref commands "
1695 # It seems better to look for this, as most of the reference
1696 # insets won't be ones we care about.
1697 i = find_token(document.body, "LatexCommand Nameref" , i)
1703 # Make sure it is actually in an inset!
1704 val = is_in_inset(document.body, cmdloc, \
1705 "\\begin_inset CommandInset ref")
1708 document.body[cmdloc] = "LatexCommand nameref"
1711 def revert_mathrsfs(document):
1712 " Load mathrsfs if \mathrsfs us use in the document "
1714 for line in document.body:
1715 if line.find("\\mathscr{") != -1:
1716 add_to_preamble(document, ["\\usepackage{mathrsfs}"])
1720 def convert_flexnames(document):
1721 "Convert \\begin_inset Flex Custom:Style to \\begin_inset Flex Style and similarly for CharStyle and Element."
1724 rx = re.compile(r'^\\begin_inset Flex (?:Custom|CharStyle|Element):(.+)$')
1726 i = find_token(document.body, "\\begin_inset Flex", i)
1729 m = rx.match(document.body[i])
1731 document.body[i] = "\\begin_inset Flex " + m.group(1)
1736 "Alert" : "CharStyle:Alert",
1737 "Code" : "CharStyle:Code",
1738 "Concepts" : "CharStyle:Concepts",
1739 "E-Mail" : "CharStyle:E-Mail",
1740 "Emph" : "CharStyle:Emph",
1741 "Expression" : "CharStyle:Expression",
1742 "Initial" : "CharStyle:Initial",
1743 "Institute" : "CharStyle:Institute",
1744 "Meaning" : "CharStyle:Meaning",
1745 "Noun" : "CharStyle:Noun",
1746 "Strong" : "CharStyle:Strong",
1747 "Structure" : "CharStyle:Structure",
1748 "ArticleMode" : "Custom:ArticleMode",
1749 "Endnote" : "Custom:Endnote",
1750 "Glosse" : "Custom:Glosse",
1751 "PresentationMode" : "Custom:PresentationMode",
1752 "Tri-Glosse" : "Custom:Tri-Glosse"
1756 "Abbrev" : "Element:Abbrev",
1757 "CCC-Code" : "Element:CCC-Code",
1758 "Citation-number" : "Element:Citation-number",
1759 "City" : "Element:City",
1760 "Code" : "Element:Code",
1761 "CODEN" : "Element:CODEN",
1762 "Country" : "Element:Country",
1763 "Day" : "Element:Day",
1764 "Directory" : "Element:Directory",
1765 "Dscr" : "Element:Dscr",
1766 "Email" : "Element:Email",
1767 "Emph" : "Element:Emph",
1768 "Filename" : "Element:Filename",
1769 "Firstname" : "Element:Firstname",
1770 "Fname" : "Element:Fname",
1771 "GuiButton" : "Element:GuiButton",
1772 "GuiMenu" : "Element:GuiMenu",
1773 "GuiMenuItem" : "Element:GuiMenuItem",
1774 "ISSN" : "Element:ISSN",
1775 "Issue-day" : "Element:Issue-day",
1776 "Issue-months" : "Element:Issue-months",
1777 "Issue-number" : "Element:Issue-number",
1778 "KeyCap" : "Element:KeyCap",
1779 "KeyCombo" : "Element:KeyCombo",
1780 "Keyword" : "Element:Keyword",
1781 "Literal" : "Element:Literal",
1782 "MenuChoice" : "Element:MenuChoice",
1783 "Month" : "Element:Month",
1784 "Orgdiv" : "Element:Orgdiv",
1785 "Orgname" : "Element:Orgname",
1786 "Postcode" : "Element:Postcode",
1787 "SS-Code" : "Element:SS-Code",
1788 "SS-Title" : "Element:SS-Title",
1789 "State" : "Element:State",
1790 "Street" : "Element:Street",
1791 "Surname" : "Element:Surname",
1792 "Volume" : "Element:Volume",
1793 "Year" : "Element:Year"
1797 def revert_flexnames(document):
1798 if document.backend == "latex":
1799 flexlist = flex_insets
1801 flexlist = flex_elements
1803 rx = re.compile(r'^\\begin_inset Flex\s+(.+)$')
1806 i = find_token(document.body, "\\begin_inset Flex", i)
1809 m = rx.match(document.body[i])
1811 document.warning("Illegal flex inset: " + document.body[i])
1815 if style in flexlist:
1816 document.body[i] = "\\begin_inset Flex " + flexlist[style]
1820 def convert_mathdots(document):
1821 " Load mathdots automatically "
1822 i = find_token(document.header, "\\use_mhchem" , 0)
1824 i = find_token(document.header, "\\use_esint" , 0)
1826 document.warning("Malformed LyX document: Can't find \\use_mhchem.")
1828 j = find_token(document.preamble, "\\usepackage{mathdots}", 0)
1830 document.header.insert(i + 1, "\\use_mathdots 0")
1832 document.header.insert(i + 1, "\\use_mathdots 2")
1833 del document.preamble[j]
1836 def revert_mathdots(document):
1837 " Load mathdots if used in the document "
1839 mathdots = find_token(document.header, "\\use_mathdots" , 0)
1841 document.warning("No \\use_mathdots line. Assuming auto.")
1843 val = get_value(document.header, "\\use_mathdots", mathdots)
1844 del document.header[mathdots]
1848 document.warning("Invalid \\use_mathdots value: " + val + ". Assuming auto.")
1849 # probably usedots has not been changed, but be safe.
1857 add_to_preamble(document, ["\\usepackage{mathdots}"])
1860 # so we are in the auto case. we want to load mathdots if \iddots is used.
1863 i = find_token(document.body, '\\begin_inset Formula', i)
1866 j = find_end_of_inset(document.body, i)
1868 document.warning("Malformed LyX document: Can't find end of Formula inset at line " + str(i))
1871 code = "\n".join(document.body[i:j])
1872 if code.find("\\iddots") != -1:
1873 add_to_preamble(document, ["\\@ifundefined{iddots}{\\usepackage{mathdots}}"])
1878 def convert_rule(document):
1879 " Convert \\lyxline to CommandInset line. "
1882 inset = ['\\begin_inset CommandInset line',
1883 'LatexCommand rule',
1887 '\\end_inset', '', '']
1889 # if paragraphs are indented, we may have to unindent to get the
1890 # line to be full-width.
1891 indent = get_value(document.header, "\\paragraph_separation", 0)
1892 have_indent = (indent == "indent")
1895 i = find_token(document.body, "\\lyxline" , i)
1899 # we need to find out if this line follows other content
1900 # in its paragraph. find its layout....
1901 lastlay = find_token_backwards(document.body, "\\begin_layout", i)
1903 document.warning("Can't find layout for line at " + str(i))
1904 # do the best we can.
1905 document.body[i:i+1] = inset
1909 # ...and look for other content before it.
1911 for line in document.body[lastlay + 1:i]:
1912 # is it empty or a paragraph option?
1913 if not line or line[0] == '\\':
1919 document.body[i:i+1] = inset
1921 # we need to unindent, lest the line be too long
1922 document.body.insert(lastlay + 1, "\\noindent")
1925 # so our line is in the middle of a paragraph
1926 # we need to add a new line, lest this line follow the
1927 # other content on that line and run off the side of the page
1928 document.body[i:i+1] = inset
1929 document.body[i:i] = ["\\begin_inset Newline newline", "\\end_inset", ""]
1933 def revert_rule(document):
1934 " Revert line insets to Tex code "
1937 i = find_token(document.body, "\\begin_inset CommandInset line" , i)
1941 j = find_token(document.body, "\\end_inset" , i)
1943 document.warning("Malformed LyX document: Can't find end of line inset.")
1945 # determine the optional offset
1946 offset = get_quoted_value(document.body, 'offset', i, j)
1948 offset = '[' + offset + ']'
1949 # determine the width
1950 width = get_quoted_value(document.body, 'width', i, j, "100col%")
1951 width = latex_length(width)[1]
1952 # determine the height
1953 height = get_quoted_value(document.body, 'height', i, j, "1pt")
1954 height = latex_length(height)[1]
1955 # output the \rule command
1956 subst = "\\rule[" + offset + "]{" + width + "}{" + height + "}"
1957 document.body[i:j + 1] = put_cmd_in_ert(subst)
1958 i += len(subst) - (j - i)
1961 def revert_diagram(document):
1962 " Add the feyn package if \\Diagram is used in math "
1965 i = find_token(document.body, '\\begin_inset Formula', i)
1968 j = find_end_of_inset(document.body, i)
1970 document.warning("Malformed LyX document: Can't find end of Formula inset.")
1972 lines = "\n".join(document.body[i:j])
1973 if lines.find("\\Diagram") == -1:
1976 add_to_preamble(document, ["\\usepackage{feyn}"])
1977 # only need to do it once!
1980 chapters = ("amsbook", "book", "docbook-book", "elsart", "extbook", "extreport",
1981 "jbook", "jreport", "jsbook", "literate-book", "literate-report", "memoir",
1982 "mwbk", "mwrep", "recipebook", "report", "scrbook", "scrreprt", "svmono",
1983 "svmult", "tbook", "treport", "tufte-book")
1985 def convert_bibtex_clearpage(document):
1986 " insert a clear(double)page bibliographystyle if bibtotoc option is used "
1988 if document.textclass not in chapters:
1991 i = find_token(document.header, '\\papersides', 0)
1994 document.warning("Malformed LyX document: Can't find papersides definition.")
1995 document.warning("Assuming single sided.")
1998 val = get_value(document.header, "\\papersides", i)
2003 if sides != 1 and sides != 2:
2004 document.warning("Invalid papersides value: " + val)
2005 document.warning("Assuming single sided.")
2010 j = find_token(document.body, "\\begin_inset CommandInset bibtex", j)
2014 k = find_end_of_inset(document.body, j)
2016 document.warning("Can't find end of Bibliography inset at line " + str(j))
2020 # only act if there is the option "bibtotoc"
2021 val = get_value(document.body, 'options', j, k)
2023 document.warning("Can't find options for bibliography inset at line " + str(j))
2027 if val.find("bibtotoc") == -1:
2031 # so we want to insert a new page right before the paragraph that
2032 # this bibliography thing is in.
2033 lay = find_token_backwards(document.body, "\\begin_layout", j)
2035 document.warning("Can't find layout containing bibliography inset at line " + str(j))
2042 cmd = "cleardoublepage"
2043 subst = ['\\begin_layout Standard',
2044 '\\begin_inset Newpage ' + cmd,
2045 '\\end_inset', '', '',
2047 document.body[lay:lay] = subst
2051 def check_passthru(document):
2052 tc = document.textclass
2053 ok = (tc == "literate-article" or tc == "literate-book" or tc == "literate-report")
2055 mods = document.get_module_list()
2057 if mod == "sweave" or mod == "noweb":
2063 def convert_passthru(document):
2064 " http://www.mail-archive.com/lyx-devel@lists.lyx.org/msg161298.html "
2065 if not check_passthru:
2068 rx = re.compile("\\\\begin_layout \s*(\w+)")
2070 for lay in ["Chunk", "Scrap"]:
2072 beg = find_token(document.body, "\\begin_layout " + lay, beg)
2075 end = find_end_of_layout(document.body, beg)
2077 document.warning("Can't find end of layout at line " + str(beg))
2081 # we are now going to replace newline insets within this layout
2082 # by new instances of this layout. so we have repeated layouts
2083 # instead of newlines.
2085 # if the paragraph has any customization, however, we do not want to
2086 # do the replacement.
2087 if document.body[beg + 1].startswith("\\"):
2093 ns = find_token(document.body, "\\begin_inset Newline newline", ns, end)
2096 ne = find_end_of_inset(document.body, ns)
2097 if ne == -1 or ne > end:
2098 document.warning("Can't find end of inset at line " + str(nb))
2101 if document.body[ne + 1] == "":
2103 subst = ["\\end_layout", "", "\\begin_layout " + lay]
2104 document.body[ns:ne + 1] = subst
2105 # now we need to adjust end, in particular, but might as well
2106 # do ns properly, too
2107 newlines = (ne - ns) - len(subst)
2111 # ok, we now want to find out if the next layout is the
2112 # same as this one. if so, we will insert an extra copy of it
2114 next = find_token(document.body, "\\begin_layout", end)
2116 m = rx.match(document.body[next])
2118 nextlay = m.group(1)
2120 subst = ["\\begin_layout " + lay, "", "\\end_layout", ""]
2121 document.body[next:next] = subst
2125 beg += 4 # for the extra layout
2128 def revert_passthru(document):
2129 " http://www.mail-archive.com/lyx-devel@lists.lyx.org/msg161298.html "
2130 if not check_passthru:
2132 rx = re.compile("\\\\begin_layout \s*(\w+)")
2134 for lay in ["Chunk", "Scrap"]:
2136 beg = find_token(document.body, "\\begin_layout " + lay, beg)
2139 end = find_end_of_layout(document.body, beg)
2141 document.warning("Can't find end of layout at line " + str(beg))
2145 # we now want to find out if the next layout is the
2146 # same as this one. but we will need to do this over and
2149 next = find_token(document.body, "\\begin_layout", end)
2152 m = rx.match(document.body[next])
2155 nextlay = m.group(1)
2158 # so it is the same layout again. we now want to know if it is empty.
2159 # but first let's check and make sure there is no content between the
2160 # two layouts. i'm not sure if that can happen or not.
2161 for l in range(end + 1, next):
2162 if document.body[l] != "":
2163 document.warning("Found content between adjacent " + lay + " layouts!")
2165 nextend = find_end_of_layout(document.body, next)
2167 document.warning("Can't find end of layout at line " + str(next))
2170 for l in range(next + 1, nextend):
2171 if document.body[l] != "":
2175 # empty layouts just get removed
2176 # should we check if it's before yet another such layout?
2177 del document.body[next : nextend + 1]
2178 # and we do not want to check again. we know the next layout
2179 # should be another Chunk and should be left as is.
2182 # if it's not empty, then we want to insert a newline in place
2183 # of the layout switch
2184 subst = ["\\begin_inset Newline newline", "\\end_inset", ""]
2185 document.body[end : next + 1] = subst
2186 # and now we have to find the end of the new, larger layout
2187 newend = find_end_of_layout(document.body, beg)
2189 document.warning("Can't find end of new layout at line " + str(beg))
2195 def revert_multirowOffset(document):
2196 " Revert multirow cells with offset in tables to TeX-code"
2197 # this routine is the same as the revert_multirow routine except that
2198 # it checks additionally for the offset
2200 # first, let's find out if we need to do anything
2201 i = find_token(document.body, '<cell multirow="3" mroffset=', 0)
2205 add_to_preamble(document, ["\\usepackage{multirow}"])
2207 rgx = re.compile(r'mroffset="[^"]+?"')
2211 # find begin/end of table
2212 begin_table = find_token(document.body, '<lyxtabular version=', begin_table)
2213 if begin_table == -1:
2215 end_table = find_end_of(document.body, begin_table, '<lyxtabular', '</lyxtabular>')
2217 document.warning("Malformed LyX document: Could not find end of table.")
2220 # does this table have multirow?
2221 i = find_token(document.body, '<cell multirow="3"', begin_table, end_table)
2223 begin_table = end_table
2226 # store the number of rows and columns
2227 numrows = get_option_value(document.body[begin_table], "rows")
2228 numcols = get_option_value(document.body[begin_table], "columns")
2230 numrows = int(numrows)
2231 numcols = int(numcols)
2233 document.warning("Unable to determine rows and columns!")
2234 begin_table = end_table
2239 # collect info on rows and columns of this table.
2240 begin_row = begin_table
2241 for row in range(numrows):
2242 begin_row = find_token(document.body, '<row>', begin_row, end_table)
2244 document.warning("Can't find row " + str(row + 1))
2246 end_row = find_end_of(document.body, begin_row, '<row>', '</row>')
2248 document.warning("Can't find end of row " + str(row + 1))
2250 begin_cell = begin_row
2251 multirows.append([])
2252 for column in range(numcols):
2253 begin_cell = find_token(document.body, '<cell ', begin_cell, end_row)
2254 if begin_cell == -1:
2255 document.warning("Can't find column " + str(column + 1) + \
2256 "in row " + str(row + 1))
2259 # this will fail if someone puts "</cell>" in a cell, but
2260 # that seems fairly unlikely.
2261 end_cell = find_end_of(document.body, begin_cell, '<cell', '</cell>')
2263 document.warning("Can't find end of column " + str(column + 1) + \
2264 "in row " + str(row + 1))
2266 multirows[row].append([begin_cell, end_cell, 0])
2267 if document.body[begin_cell].find('multirow="3" mroffset=') != -1:
2268 multirows[row][column][2] = 3 # begin multirow
2269 mrstarts.append([row, column])
2270 elif document.body[begin_cell].find('multirow="4"') != -1:
2271 multirows[row][column][2] = 4 # in multirow
2272 begin_cell = end_cell
2274 # end of table info collection
2276 # work from the back to avoid messing up numbering
2282 col_width = get_option_value(document.body[begin_table + 2 + col], "width")
2283 # "0pt" means that no width is specified
2284 if not col_width or col_width == "0pt":
2286 # determine the number of cells that are part of the multirow
2288 for r in range(row + 1, numrows):
2289 if multirows[r][col][2] != 4:
2292 # take the opportunity to revert this line
2293 lineno = multirows[r][col][0]
2294 document.body[lineno] = document.body[lineno].\
2295 replace(' multirow="4" ', ' ').\
2296 replace('valignment="middle"', 'valignment="top"').\
2297 replace(' topline="true" ', ' ')
2298 # remove bottom line of previous multirow-part cell
2299 lineno = multirows[r-1][col][0]
2300 document.body[lineno] = document.body[lineno].replace(' bottomline="true" ', ' ')
2301 # revert beginning cell
2302 bcell = multirows[row][col][0]
2303 ecell = multirows[row][col][1]
2304 offset = get_option_value(document.body[bcell], "mroffset")
2305 document.body[bcell] = document.body[bcell].\
2306 replace(' multirow="3" ', ' ').\
2307 replace('valignment="middle"', 'valignment="top"')
2308 # remove mroffset option
2309 document.body[bcell] = rgx.sub('', document.body[bcell])
2311 blay = find_token(document.body, "\\begin_layout", bcell, ecell)
2313 document.warning("Can't find layout for cell!")
2315 bend = find_end_of_layout(document.body, blay)
2317 document.warning("Can't find end of layout for cell!")
2319 # do the later one first, so as not to mess up the numbering
2320 # we are wrapping the whole cell in this ert
2321 # so before the end of the layout...
2322 document.body[bend:bend] = put_cmd_in_ert("}")
2323 # ...and after the beginning
2324 document.body[blay + 1:blay + 1] = \
2325 put_cmd_in_ert("\\multirow{" + str(nummrs) + "}{" + col_width + "}[" \
2328 # on to the next table
2329 begin_table = end_table
2332 def revert_script(document):
2333 " Convert subscript/superscript inset to TeX code "
2335 foundsubscript = False
2337 i = find_token(document.body, '\\begin_inset script', i)
2340 z = find_end_of_inset(document.body, i)
2342 document.warning("Malformed LyX document: Can't find end of script inset.")
2345 blay = find_token(document.body, "\\begin_layout", i, z)
2347 document.warning("Malformed LyX document: Can't find layout in script inset.")
2351 if check_token(document.body[i], "\\begin_inset script subscript"):
2352 subst = '\\textsubscript{'
2353 foundsubscript = True
2354 elif check_token(document.body[i], "\\begin_inset script superscript"):
2355 subst = '\\textsuperscript{'
2357 document.warning("Malformed LyX document: Unknown type of script inset.")
2360 bend = find_end_of_layout(document.body, blay)
2361 if bend == -1 or bend > z:
2362 document.warning("Malformed LyX document: Can't find end of layout in script inset.")
2365 # remove the \end_layout \end_inset pair
2366 document.body[bend:z + 1] = put_cmd_in_ert("}")
2367 document.body[i:blay + 1] = put_cmd_in_ert(subst)
2369 # these classes provide a \textsubscript command:
2370 # FIXME: Would be nice if we could use the information of the .layout file here
2371 classes = ["memoir", "scrartcl", "scrbook", "scrlttr2", "scrreprt"]
2372 if foundsubscript and find_token_exact(classes, document.textclass, 0) == -1:
2373 add_to_preamble(document, ['\\usepackage{subscript}'])
2376 def convert_use_xetex(document):
2377 " convert \\use_xetex to \\use_non_tex_fonts "
2378 i = find_token(document.header, "\\use_xetex", 0)
2380 document.header.insert(-1, "\\use_non_tex_fonts 0")
2382 val = get_value(document.header, "\\use_xetex", 0)
2383 document.header[i] = "\\use_non_tex_fonts " + val
2386 def revert_use_xetex(document):
2387 " revert \\use_non_tex_fonts to \\use_xetex "
2389 i = find_token(document.header, "\\use_non_tex_fonts", 0)
2391 document.warning("Malformed document. No \\use_non_tex_fonts param!")
2394 val = get_value(document.header, "\\use_non_tex_fonts", 0)
2395 document.header[i] = "\\use_xetex " + val
2398 def revert_labeling(document):
2399 koma = ("scrartcl", "scrarticle-beamer", "scrbook", "scrlettr",
2400 "scrlttr2", "scrreprt")
2401 if document.textclass in koma:
2405 i = find_token_exact(document.body, "\\begin_layout Labeling", i)
2408 document.body[i] = "\\begin_layout List"
2411 def revert_langpack(document):
2412 " revert \\language_package parameter "
2414 i = find_token(document.header, "\\language_package", 0)
2416 document.warning("Malformed document. No \\language_package param!")
2419 del document.header[i]
2422 def convert_langpack(document):
2423 " Add \\language_package parameter "
2424 i = find_token(document.header, "\language" , 0)
2426 document.warning("Malformed document. No \\language defined!")
2429 document.header.insert(i + 1, "\\language_package default")
2432 def revert_tabularwidth(document):
2435 i = find_token(document.body, "\\begin_inset Tabular", i)
2438 j = find_end_of_inset(document.body, i)
2440 document.warning("Unable to find end of Tabular inset at line " + str(i))
2444 features = find_token(document.body, "<features", i, j)
2446 document.warning("Can't find any features in Tabular inset at line " + str(i))
2449 if document.body[features].find('alignment="tabularwidth"') != -1:
2450 remove_option(document.body, features, 'tabularwidth')
2452 def revert_html_css_as_file(document):
2453 if not del_token(document.header, '\\html_css_as_file', 0):
2454 document.warning("Malformed LyX document: Missing \\html_css_as_file.")
2461 supported_versions = ["2.0.0","2.0"]
2462 convert = [[346, []],
2468 [352, [convert_splitindex]],
2475 [359, [convert_nomencl_width]],
2485 [369, [convert_author_id]],
2487 [371, [convert_mhchem]],
2489 [373, [merge_gbrief]],
2495 [379, [convert_math_output]],
2505 [389, [convert_html_quotes]],
2509 [393, [convert_optarg]],
2510 [394, [convert_use_makebox]],
2513 [397, [remove_Nameref]],
2515 [399, [convert_mathdots]],
2516 [400, [convert_rule]],
2518 [402, [convert_bibtex_clearpage]],
2519 [403, [convert_flexnames]],
2520 [404, [convert_prettyref]],
2522 [406, [convert_passthru]],
2525 [409, [convert_use_xetex]],
2527 [411, [convert_langpack]],
2532 revert = [[412, [revert_html_css_as_file]],
2533 [411, [revert_tabularwidth]],
2534 [410, [revert_langpack]],
2535 [409, [revert_labeling]],
2536 [408, [revert_use_xetex]],
2537 [407, [revert_script]],
2538 [406, [revert_multirowOffset]],
2539 [405, [revert_passthru]],
2541 [403, [revert_refstyle]],
2542 [402, [revert_flexnames]],
2544 [400, [revert_diagram]],
2545 [399, [revert_rule]],
2546 [398, [revert_mathdots]],
2547 [397, [revert_mathrsfs]],
2549 [395, [revert_nameref]],
2550 [394, [revert_DIN_C_pagesizes]],
2551 [393, [revert_makebox]],
2552 [392, [revert_argument]],
2554 [390, [revert_align_decimal, revert_IEEEtran]],
2555 [389, [revert_output_sync]],
2556 [388, [revert_html_quotes]],
2557 [387, [revert_pagesizes]],
2558 [386, [revert_math_scale]],
2559 [385, [revert_lyx_version]],
2560 [384, [revert_shadedboxcolor]],
2561 [383, [revert_fontcolor]],
2562 [382, [revert_turkmen]],
2563 [381, [revert_notefontcolor]],
2564 [380, [revert_equalspacing_xymatrix]],
2565 [379, [revert_inset_preview]],
2566 [378, [revert_math_output]],
2568 [376, [revert_multirow]],
2569 [375, [revert_includeall]],
2570 [374, [revert_includeonly]],
2571 [373, [revert_html_options]],
2572 [372, [revert_gbrief]],
2573 [371, [revert_fontenc]],
2574 [370, [revert_mhchem]],
2575 [369, [revert_suppress_date]],
2576 [368, [revert_author_id]],
2577 [367, [revert_hspace_glue_lengths]],
2578 [366, [revert_percent_vspace_lengths, revert_percent_hspace_lengths]],
2579 [365, [revert_percent_skip_lengths]],
2580 [364, [revert_paragraph_indentation]],
2581 [363, [revert_branch_filename]],
2582 [362, [revert_longtable_align]],
2583 [361, [revert_applemac]],
2585 [359, [revert_nomencl_cwidth]],
2586 [358, [revert_nomencl_width]],
2587 [357, [revert_custom_processors]],
2588 [356, [revert_ulinelatex]],
2590 [354, [revert_strikeout]],
2591 [353, [revert_printindexall]],
2592 [352, [revert_subindex]],
2593 [351, [revert_splitindex]],
2594 [350, [revert_backgroundcolor]],
2595 [349, [revert_outputformat]],
2596 [348, [revert_xetex]],
2597 [347, [revert_phantom, revert_hphantom, revert_vphantom]],
2598 [346, [revert_tabularvalign]],
2599 [345, [revert_swiss]]
2603 if __name__ == "__main__":