# -*- coding: utf-8 -*- # This file is part of lyx2lyx # -*- coding: utf-8 -*- # Copyright (C) 2016 The LyX team # # This program is free software; you can redistribute it and/or # modify it under the terms of the GNU General Public License # as published by the Free Software Foundation; either version 2 # of the License, or (at your option) any later version. # # This program is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the # GNU General Public License for more details. # # You should have received a copy of the GNU General Public License # along with this program; if not, write to the Free Software # Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. """ Convert files to the file format generated by lyx 2.3""" import re, string import unicodedata import sys, os # Uncomment only what you need to import, please. from parser_tools import find_end_of, find_token_backwards, find_end_of_layout, \ find_token, find_end_of_inset, get_value, get_bool_value, \ get_containing_layout, get_quoted_value, del_token # find_tokens, find_token_exact, is_in_inset, \ # check_token, get_option_value from lyx2lyx_tools import add_to_preamble, put_cmd_in_ert # get_ert, lyx2latex, \ # lyx2verbatim, length_in_bp, convert_info_insets # insert_to_preamble, latex_length, revert_flex_inset, \ # revert_font_attrs, hex2ratio, str2bool #################################################################### # Private helper functions ############################################################################### ### ### Conversion and reversion routines ### ############################################################################### def convert_microtype(document): " Add microtype settings. " i = find_token(document.header, "\\font_tt_scale" , 0) if i == -1: document.warning("Malformed LyX document: Can't find \\font_tt_scale.") i = len(document.header) - 1 j = find_token(document.preamble, "\\usepackage{microtype}", 0) if j == -1: document.header.insert(i + 1, "\\use_microtype false") else: document.header.insert(i + 1, "\\use_microtype true") del document.preamble[j] def revert_microtype(document): " Remove microtype settings. " i = find_token(document.header, "\\use_microtype", 0) if i == -1: return use_microtype = get_bool_value(document.header, "\\use_microtype" , i) del document.header[i] if use_microtype: add_to_preamble(document, ["\\usepackage{microtype}"]) def convert_dateinset(document): ' Convert date external inset to ERT ' i = 0 while True: i = find_token(document.body, "\\begin_inset External", i) if i == -1: return j = find_end_of_inset(document.body, i) if j == -1: document.warning("Malformed lyx document: Missing '\\end_inset' in convert_dateinset.") i += 1 continue if get_value(document.body, 'template', i, j) == "Date": document.body[i : j + 1] = put_cmd_in_ert("\\today ") i += 1 continue def convert_inputenc(document): " Replace no longer supported input encoding settings. " i = find_token(document.header, "\\inputenc", 0) if i == -1: return if get_value(document.header, "\\inputencoding", i) == "pt254": document.header[i] = "\\inputencoding pt154" def convert_ibranches(document): ' Add "inverted 0" to branch insets' i = 0 while True: i = find_token(document.body, "\\begin_inset Branch", i) if i == -1: return document.body.insert(i + 1, "inverted 0") i += 1 def revert_ibranches(document): ' Convert inverted branches to explicit anti-branches' # Get list of branches ourbranches = {} i = 0 while True: i = find_token(document.header, "\\branch", i) if i == -1: break branch = document.header[i][8:].strip() if document.header[i+1].startswith("\\selected "): #document.warning(document.header[i+1]) #document.warning(document.header[i+1][10]) selected = int(document.header[i+1][10]) else: document.warning("Malformed LyX document: No selection indicator for branch " + branch) selected = 1 # the value tells us whether the branch is selected ourbranches[document.header[i][8:].strip()] = selected i += 1 # Figure out what inverted branches, if any, have been used # and convert them to "Anti-OldBranch" ibranches = {} i = 0 while True: i = find_token(document.body, "\\begin_inset Branch", i) if i == -1: break if not document.body[i+1].startswith("inverted "): document.warning("Malformed LyX document: Missing 'inverted' tag!") i += 1 continue inverted = document.body[i+1][9] #document.warning(document.body[i+1]) if inverted == "1": branch = document.body[i][20:].strip() #document.warning(branch) if not branch in ibranches: antibranch = "Anti-" + branch while antibranch in ibranches: antibranch = "x" + antibranch ibranches[branch] = antibranch else: antibranch = ibranches[branch] #document.warning(antibranch) document.body[i] = "\\begin_inset Branch " + antibranch # remove "inverted" key del document.body[i+1] i += 1 # now we need to add the new branches to the header for old, new in ibranches.iteritems(): i = find_token(document.header, "\\branch " + old, 0) if i == -1: document.warning("Can't find branch %s even though we found it before!" % (old)) continue j = find_token(document.header, "\\end_branch", i) if j == -1: document.warning("Malformed LyX document! Can't find end of branch " + old) continue # ourbranches[old] - 1 inverts the selection status of the old branch lines = ["\\branch " + new, "\\selected " + str(ourbranches[old] - 1)] # these are the old lines telling us color, etc. lines += document.header[i+2 : j+1] document.header[i:i] = lines def revert_beamer_article_styles(document): " Include (scr)article styles in beamer article " beamer_articles = ["article-beamer", "scrarticle-beamer"] if document.textclass not in beamer_articles: return inclusion = "article.layout" if document.textclass == "scrarticle-beamer": inclusion = "scrartcl.layout" i = find_token(document.header, "\\begin_local_layout", 0) if i == -1: k = find_token(document.header, "\\language", 0) if k == -1: # this should not happen document.warning("Malformed LyX document! No \\language header found!") return document.header[k-1 : k-1] = ["\\begin_local_layout", "\\end_local_layout"] i = k - 1 j = find_end_of(document.header, i, "\\begin_local_layout", "\\end_local_layout") if j == -1: # this should not happen document.warning("Malformed LyX document: Can't find end of local layout!") return document.header[i+1 : i+1] = [ "### Inserted by lyx2lyx (more [scr]article styles) ###", "Input " + inclusion, "Input beamer.layout", "Provides geometry 0", "Provides hyperref 0", "DefaultFont", " Family Roman", " Series Medium", " Shape Up", " Size Normal", " Color None", "EndFont", "Preamble", " \\usepackage{beamerarticle,pgf}", " % this default might be overridden by plain title style", " \\newcommand\makebeamertitle{\\frame{\\maketitle}}%", " \\AtBeginDocument{", " \\let\\origtableofcontents=\\tableofcontents", " \\def\\tableofcontents{\\@ifnextchar[{\\origtableofcontents}{\\gobbletableofcontents}}", " \\def\\gobbletableofcontents#1{\\origtableofcontents}", " }", "EndPreamble", "### End of insertion by lyx2lyx (more [scr]article styles) ###" ] def convert_beamer_article_styles(document): " Remove included (scr)article styles in beamer article " beamer_articles = ["article-beamer", "scrarticle-beamer"] if document.textclass not in beamer_articles: return i = find_token(document.header, "\\begin_local_layout", 0) if i == -1: return j = find_end_of(document.header, i, "\\begin_local_layout", "\\end_local_layout") if j == -1: # this should not happen document.warning("Malformed LyX document: Can't find end of local layout!") return k = find_token(document.header, "### Inserted by lyx2lyx (more [scr]article styles) ###", i, j) if k != -1: l = find_token(document.header, "### End of insertion by lyx2lyx (more [scr]article styles) ###", i, j) if l == -1: # this should not happen document.warning("End of lyx2lyx local layout insertion not found!") return if k == i + 1 and l == j - 1: # that was all the local layout there was document.header[i : j + 1] = [] else: document.header[k : l + 1] = [] def revert_bosnian(document): "Set the document language to English but assure Bosnian output" if document.language == "bosnian": document.language = "english" i = find_token(document.header, "\\language bosnian", 0) if i != -1: document.header[i] = "\\language english" j = find_token(document.header, "\\language_package default", 0) if j != -1: document.header[j] = "\\language_package babel" k = find_token(document.header, "\\options", 0) if k != -1: document.header[k] = document.header[k].replace("\\options", "\\options bosnian,") else: l = find_token(document.header, "\\use_default_options", 0) document.header.insert(l + 1, "\\options bosnian") def revert_friulan(document): "Set the document language to English but assure Friulan output" if document.language == "friulan": document.language = "english" i = find_token(document.header, "\\language friulan", 0) if i != -1: document.header[i] = "\\language english" j = find_token(document.header, "\\language_package default", 0) if j != -1: document.header[j] = "\\language_package babel" k = find_token(document.header, "\\options", 0) if k != -1: document.header[k] = document.header[k].replace("\\options", "\\options friulan,") else: l = find_token(document.header, "\\use_default_options", 0) document.header.insert(l + 1, "\\options friulan") def revert_macedonian(document): "Set the document language to English but assure Macedonian output" if document.language == "macedonian": document.language = "english" i = find_token(document.header, "\\language macedonian", 0) if i != -1: document.header[i] = "\\language english" j = find_token(document.header, "\\language_package default", 0) if j != -1: document.header[j] = "\\language_package babel" k = find_token(document.header, "\\options", 0) if k != -1: document.header[k] = document.header[k].replace("\\options", "\\options macedonian,") else: l = find_token(document.header, "\\use_default_options", 0) document.header.insert(l + 1, "\\options macedonian") def revert_piedmontese(document): "Set the document language to English but assure Piedmontese output" if document.language == "piedmontese": document.language = "english" i = find_token(document.header, "\\language piedmontese", 0) if i != -1: document.header[i] = "\\language english" j = find_token(document.header, "\\language_package default", 0) if j != -1: document.header[j] = "\\language_package babel" k = find_token(document.header, "\\options", 0) if k != -1: document.header[k] = document.header[k].replace("\\options", "\\options piedmontese,") else: l = find_token(document.header, "\\use_default_options", 0) document.header.insert(l + 1, "\\options piedmontese") def revert_romansh(document): "Set the document language to English but assure Romansh output" if document.language == "romansh": document.language = "english" i = find_token(document.header, "\\language romansh", 0) if i != -1: document.header[i] = "\\language english" j = find_token(document.header, "\\language_package default", 0) if j != -1: document.header[j] = "\\language_package babel" k = find_token(document.header, "\\options", 0) if k != -1: document.header[k] = document.header[k].replace("\\options", "\\options romansh,") else: l = find_token(document.header, "\\use_default_options", 0) document.header.insert(l + 1, "\\options romansh") def revert_amharic(document): "Set the document language to English but assure Amharic output" if document.language == "amharic": document.language = "english" i = find_token(document.header, "\\language amharic", 0) if i != -1: document.header[i] = "\\language english" j = find_token(document.header, "\\language_package default", 0) if j != -1: document.header[j] = "\\language_package default" add_to_preamble(document, ["\\AtBeginDocument{\setotherlanguage{amharic}}"]) document.body[2 : 2] = ["\\begin_layout Standard", "\\begin_inset ERT", "status open", "", "\\begin_layout Plain Layout", "", "", "\\backslash", "resetdefaultlanguage{amharic}", "\\end_layout", "", "\\end_inset", "", "", "\\end_layout", ""] def revert_asturian(document): "Set the document language to English but assure Asturian output" if document.language == "asturian": document.language = "english" i = find_token(document.header, "\\language asturian", 0) if i != -1: document.header[i] = "\\language english" j = find_token(document.header, "\\language_package default", 0) if j != -1: document.header[j] = "\\language_package default" add_to_preamble(document, ["\\AtBeginDocument{\setotherlanguage{asturian}}"]) document.body[2 : 2] = ["\\begin_layout Standard", "\\begin_inset ERT", "status open", "", "\\begin_layout Plain Layout", "", "", "\\backslash", "resetdefaultlanguage{asturian}", "\\end_layout", "", "\\end_inset", "", "", "\\end_layout", ""] def revert_kannada(document): "Set the document language to English but assure Kannada output" if document.language == "kannada": document.language = "english" i = find_token(document.header, "\\language kannada", 0) if i != -1: document.header[i] = "\\language english" j = find_token(document.header, "\\language_package default", 0) if j != -1: document.header[j] = "\\language_package default" add_to_preamble(document, ["\\AtBeginDocument{\setotherlanguage{kannada}}"]) document.body[2 : 2] = ["\\begin_layout Standard", "\\begin_inset ERT", "status open", "", "\\begin_layout Plain Layout", "", "", "\\backslash", "resetdefaultlanguage{kannada}", "\\end_layout", "", "\\end_inset", "", "", "\\end_layout", ""] def revert_khmer(document): "Set the document language to English but assure Khmer output" if document.language == "khmer": document.language = "english" i = find_token(document.header, "\\language khmer", 0) if i != -1: document.header[i] = "\\language english" j = find_token(document.header, "\\language_package default", 0) if j != -1: document.header[j] = "\\language_package default" add_to_preamble(document, ["\\AtBeginDocument{\setotherlanguage{khmer}}"]) document.body[2 : 2] = ["\\begin_layout Standard", "\\begin_inset ERT", "status open", "", "\\begin_layout Plain Layout", "", "", "\\backslash", "resetdefaultlanguage{khmer}", "\\end_layout", "", "\\end_inset", "", "", "\\end_layout", ""] def revert_urdu(document): "Set the document language to English but assure Urdu output" if document.language == "urdu": document.language = "english" i = find_token(document.header, "\\language urdu", 0) if i != -1: document.header[i] = "\\language english" j = find_token(document.header, "\\language_package default", 0) if j != -1: document.header[j] = "\\language_package default" add_to_preamble(document, ["\\AtBeginDocument{\setotherlanguage{urdu}}"]) document.body[2 : 2] = ["\\begin_layout Standard", "\\begin_inset ERT", "status open", "", "\\begin_layout Plain Layout", "", "", "\\backslash", "resetdefaultlanguage{urdu}", "\\end_layout", "", "\\end_inset", "", "", "\\end_layout", ""] def revert_syriac(document): "Set the document language to English but assure Syriac output" if document.language == "syriac": document.language = "english" i = find_token(document.header, "\\language syriac", 0) if i != -1: document.header[i] = "\\language english" j = find_token(document.header, "\\language_package default", 0) if j != -1: document.header[j] = "\\language_package default" add_to_preamble(document, ["\\AtBeginDocument{\setotherlanguage{syriac}}"]) document.body[2 : 2] = ["\\begin_layout Standard", "\\begin_inset ERT", "status open", "", "\\begin_layout Plain Layout", "", "", "\\backslash", "resetdefaultlanguage{syriac}", "\\end_layout", "", "\\end_inset", "", "", "\\end_layout", ""] def revert_quotes(document): " Revert Quote Insets in verbatim or Hebrew context to plain quotes " # First handle verbatim insets i = 0 j = 0 while i < len(document.body): words = document.body[i].split() if len(words) > 1 and words[0] == "\\begin_inset" and \ ( words[1] in ["ERT", "listings"] or ( len(words) > 2 and words[2] in ["URL", "Chunk", "Sweave", "S/R"]) ): j = find_end_of_inset(document.body, i) if j == -1: document.warning("Malformed LyX document: Can't find end of " + words[1] + " inset at line " + str(i)) i += 1 continue while True: k = find_token(document.body, '\\begin_inset Quotes', i, j) if k == -1: i += 1 break l = find_end_of_inset(document.body, k) if l == -1: document.warning("Malformed LyX document: Can't find end of Quote inset at line " + str(k)) i = k continue replace = "\"" if document.body[k].endswith("s"): replace = "'" document.body[k:l+1] = [replace] else: i += 1 continue # Now verbatim layouts i = 0 j = 0 while i < len(document.body): words = document.body[i].split() if len(words) > 1 and words[0] == "\\begin_layout" and \ words[1] in ["Verbatim", "Verbatim*", "Code", "Author_Email", "Author_URL"]: j = find_end_of_layout(document.body, i) if j == -1: document.warning("Malformed LyX document: Can't find end of " + words[1] + " layout at line " + str(i)) i += 1 continue while True: k = find_token(document.body, '\\begin_inset Quotes', i, j) if k == -1: i += 1 break l = find_end_of_inset(document.body, k) if l == -1: document.warning("Malformed LyX document: Can't find end of Quote inset at line " + str(k)) i = k continue replace = "\"" if document.body[k].endswith("s"): replace = "'" document.body[k:l+1] = [replace] else: i += 1 continue # Now handle Hebrew if not document.language == "hebrew" and find_token(document.body, '\\lang hebrew', 0) == -1: return i = 0 j = 0 while True: k = find_token(document.body, '\\begin_inset Quotes', i) if k == -1: return l = find_end_of_inset(document.body, k) if l == -1: document.warning("Malformed LyX document: Can't find end of Quote inset at line " + str(k)) i = k continue hebrew = False parent = get_containing_layout(document.body, k) ql = find_token_backwards(document.body, "\\lang", k) if ql == -1 or ql < parent[1]: hebrew = document.language == "hebrew" elif document.body[ql] == "\\lang hebrew": hebrew = True if hebrew: replace = "\"" if document.body[k].endswith("s"): replace = "'" document.body[k:l+1] = [replace] i = l def revert_iopart(document): " Input new styles via local layout " if document.textclass != "iopart": return i = find_token(document.header, "\\begin_local_layout", 0) if i == -1: k = find_token(document.header, "\\language", 0) if k == -1: # this should not happen document.warning("Malformed LyX document! No \\language header found!") return document.header[k-1 : k-1] = ["\\begin_local_layout", "\\end_local_layout"] i = k-1 j = find_end_of(document.header, i, "\\begin_local_layout", "\\end_local_layout") if j == -1: # this should not happen document.warning("Malformed LyX document! Can't find end of local layout!") return document.header[i+1 : i+1] = [ "### Inserted by lyx2lyx (stdlayouts) ###", "Input stdlayouts.inc", "### End of insertion by lyx2lyx (stdlayouts) ###" ] def convert_iopart(document): " Remove local layout we added, if it is there " if document.textclass != "iopart": return i = find_token(document.header, "\\begin_local_layout", 0) if i == -1: return j = find_end_of(document.header, i, "\\begin_local_layout", "\\end_local_layout") if j == -1: # this should not happen document.warning("Malformed LyX document! Can't find end of local layout!") return k = find_token(document.header, "### Inserted by lyx2lyx (stdlayouts) ###", i, j) if k != -1: l = find_token(document.header, "### End of insertion by lyx2lyx (stdlayouts) ###", i, j) if l == -1: # this should not happen document.warning("End of lyx2lyx local layout insertion not found!") return if k == i + 1 and l == j - 1: # that was all the local layout there was document.header[i : j + 1] = [] else: document.header[k : l + 1] = [] def convert_quotestyle(document): " Convert \\quotes_language to \\quotes_style " i = find_token(document.header, "\\quotes_language", 0) if i == -1: document.warning("Malformed LyX document! Can't find \\quotes_language!") return val = get_value(document.header, "\\quotes_language", i) document.header[i] = "\\quotes_style " + val def revert_quotestyle(document): " Revert \\quotes_style to \\quotes_language " i = find_token(document.header, "\\quotes_style", 0) if i == -1: document.warning("Malformed LyX document! Can't find \\quotes_style!") return val = get_value(document.header, "\\quotes_style", i) document.header[i] = "\\quotes_language " + val def revert_plainquote(document): " Revert plain quote insets " # First, revert style setting i = find_token(document.header, "\\quotes_style plain", 0) if i != -1: document.header[i] = "\\quotes_style english" # now the insets i = 0 j = 0 while True: k = find_token(document.body, '\\begin_inset Quotes q', i) if k == -1: return l = find_end_of_inset(document.body, k) if l == -1: document.warning("Malformed LyX document: Can't find end of Quote inset at line " + str(k)) i = k continue replace = "\"" if document.body[k].endswith("s"): replace = "'" document.body[k:l+1] = [replace] i = l def convert_frenchquotes(document): " Convert french quote insets to swiss " # First, revert style setting i = find_token(document.header, "\\quotes_style french", 0) if i != -1: document.header[i] = "\\quotes_style swiss" # now the insets i = 0 while True: i = find_token(document.body, '\\begin_inset Quotes f', i) if i == -1: return val = get_value(document.body, "\\begin_inset Quotes", i)[7:] newval = val.replace("f", "c", 1) document.body[i] = document.body[i].replace(val, newval) i += 1 def revert_swissquotes(document): " Revert swiss quote insets to french " # First, revert style setting i = find_token(document.header, "\\quotes_style swiss", 0) if i != -1: document.header[i] = "\\quotes_style french" # now the insets i = 0 while True: i = find_token(document.body, '\\begin_inset Quotes c', i) if i == -1: return val = get_value(document.body, "\\begin_inset Quotes", i)[7:] newval = val.replace("c", "f", 1) document.body[i] = document.body[i].replace(val, newval) i += 1 def revert_britishquotes(document): " Revert british quote insets to english " # First, revert style setting i = find_token(document.header, "\\quotes_style british", 0) if i != -1: document.header[i] = "\\quotes_style english" # now the insets i = 0 while True: i = find_token(document.body, '\\begin_inset Quotes b', i) if i == -1: return val = get_value(document.body, "\\begin_inset Quotes", i)[7:] newval = val.replace("b", "e", 1) if val[2] == "d": # opening mark newval = newval.replace("d", "s") else: # closing mark newval = newval.replace("s", "d") document.body[i] = document.body[i].replace(val, newval) i += 1 def revert_swedishgquotes(document): " Revert swedish quote insets " # First, revert style setting i = find_token(document.header, "\\quotes_style swedishg", 0) if i != -1: document.header[i] = "\\quotes_style danish" # now the insets i = 0 while True: i = find_token(document.body, '\\begin_inset Quotes w', i) if i == -1: return val = get_value(document.body, "\\begin_inset Quotes", i)[7:] if val[2] == "d": # outer marks newval = val.replace("w", "a", 1).replace("r", "l") else: # inner marks newval = val.replace("w", "s", 1) document.body[i] = document.body[i].replace(val, newval) i += 1 def revert_frenchquotes(document): " Revert french inner quote insets " i = 0 while True: i = find_token(document.body, '\\begin_inset Quotes f', i) if i == -1: return val = get_value(document.body, "\\begin_inset Quotes", i)[7:] if val[2] == "s": # inner marks newval = val.replace("f", "e", 1).replace("s", "d") document.body[i] = document.body[i].replace(val, newval) i += 1 def revert_frenchinquotes(document): " Revert inner frenchin quote insets " # First, revert style setting i = find_token(document.header, "\\quotes_style frenchin", 0) if i != -1: document.header[i] = "\\quotes_style french" # now the insets i = 0 while True: i = find_token(document.body, '\\begin_inset Quotes i', i) if i == -1: return val = get_value(document.body, "\\begin_inset Quotes", i)[7:] newval = val.replace("i", "f", 1) if val[2] == "s": # inner marks newval = newval.replace("s", "d") document.body[i] = document.body[i].replace(val, newval) i += 1 def revert_russianquotes(document): " Revert russian quote insets " # First, revert style setting i = find_token(document.header, "\\quotes_style russian", 0) if i != -1: document.header[i] = "\\quotes_style french" # now the insets i = 0 while True: i = find_token(document.body, '\\begin_inset Quotes r', i) if i == -1: return val = get_value(document.body, "\\begin_inset Quotes", i)[7:] newval = val if val[2] == "s": # inner marks newval = val.replace("r", "g", 1).replace("s", "d") else: # outer marks newval = val.replace("r", "f", 1) document.body[i] = document.body[i].replace(val, newval) i += 1 def revert_dynamicquotes(document): " Revert dynamic quote insets " # First, revert header i = find_token(document.header, "\\dynamic_quotes", 0) if i != -1: del document.header[i] # Get global style style = "english" i = find_token(document.header, "\\quotes_style", 0) if i == -1: document.warning("Malformed document! Missing \\quotes_style") else: style = get_value(document.header, "\\quotes_style", i) s = "e" if style == "english": s = "e" elif style == "swedish": s = "s" elif style == "german": s = "g" elif style == "polish": s = "p" elif style == "swiss": s = "c" elif style == "danish": s = "a" elif style == "plain": s = "q" elif style == "british": s = "b" elif style == "swedishg": s = "w" elif style == "french": s = "f" elif style == "frenchin": s = "i" elif style == "russian": s = "r" # now transform the insets i = 0 while True: i = find_token(document.body, '\\begin_inset Quotes x', i) if i == -1: return document.body[i] = document.body[i].replace("x", s) i += 1 def revert_cjkquotes(document): " Revert cjk quote insets " # Get global style style = "english" i = find_token(document.header, "\\quotes_style", 0) if i == -1: document.warning("Malformed document! Missing \\quotes_style") else: style = get_value(document.header, "\\quotes_style", i) global_cjk = style.find("cjk") != -1 if global_cjk: document.header[i] = "\\quotes_style english" # transform dynamic insets s = "j" if style == "cjkangle": s = "k" i = 0 while True: i = find_token(document.body, '\\begin_inset Quotes x', i) if i == -1: break document.body[i] = document.body[i].replace("x", s) i += 1 cjk_langs = ["chinese-simplified", "chinese-traditional", "japanese", "japanese-cjk", "korean"] i = 0 j = 0 while True: k = find_token(document.body, '\\begin_inset Quotes j', i) if k == -1: break l = find_end_of_inset(document.body, k) if l == -1: document.warning("Malformed LyX document: Can't find end of Quote inset at line " + str(k)) i = k continue cjk = False parent = get_containing_layout(document.body, k) ql = find_token_backwards(document.body, "\\lang", k) if ql == -1 or ql < parent[1]: cjk = document.language in cjk_langs elif document.body[ql].split()[1] in cjk_langs: cjk = True val = get_value(document.body, "\\begin_inset Quotes", i)[7:] replace = [] if val[2] == "s": # inner marks if val[1] == "l": # inner opening mark if cjk: replace = [u"\u300E"] else: replace = ["\\begin_inset Formula $\\llceil$", "\\end_inset"] else: # inner closing mark if cjk: replace = [u"\u300F"] else: replace = ["\\begin_inset Formula $\\rrfloor$", "\\end_inset"] else: # outer marks if val[1] == "l": # outer opening mark if cjk: replace = [u"\u300C"] else: replace = ["\\begin_inset Formula $\\lceil$", "\\end_inset"] else: # outer closing mark if cjk: replace = [u"\u300D"] else: replace = ["\\begin_inset Formula $\\rfloor$", "\\end_inset"] document.body[k:l+1] = replace i = l i = 0 j = 0 while True: k = find_token(document.body, '\\begin_inset Quotes k', i) if k == -1: return l = find_end_of_inset(document.body, k) if l == -1: document.warning("Malformed LyX document: Can't find end of Quote inset at line " + str(k)) i = k continue cjk = False parent = get_containing_layout(document.body, k) ql = find_token_backwards(document.body, "\\lang", k) if ql == -1 or ql < parent[1]: cjk = document.language in cjk_langs elif document.body[ql].split()[1] in cjk_langs: cjk = True val = get_value(document.body, "\\begin_inset Quotes", i)[7:] replace = [] if val[2] == "s": # inner marks if val[1] == "l": # inner opening mark if cjk: replace = [u"\u3008"] else: replace = ["\\begin_inset Formula $\\langle$", "\\end_inset"] else: # inner closing mark if cjk: replace = [u"\u3009"] else: replace = ["\\begin_inset Formula $\\rangle$", "\\end_inset"] else: # outer marks if val[1] == "l": # outer opening mark if cjk: replace = [u"\u300A"] else: replace = ["\\begin_inset Formula $\\langle\\kern -2.5pt\\langle$", "\\end_inset"] else: # outer closing mark if cjk: replace = [u"\u300B"] else: replace = ["\\begin_inset Formula $\\rangle\\kern -2.5pt\\rangle$", "\\end_inset"] document.body[k:l+1] = replace i = l def revert_crimson(document): " Revert native Cochineal/Crimson font definition to LaTeX " if find_token(document.header, "\\use_non_tex_fonts false", 0) != -1: preamble = "" i = find_token(document.header, "\\font_roman \"cochineal\"", 0) if i != -1: osf = False j = find_token(document.header, "\\font_osf true", 0) if j != -1: osf = True preamble = "\\usepackage" if osf: document.header[j] = "\\font_osf false" preamble += "[proportional,osf]" preamble += "{cochineal}" add_to_preamble(document, [preamble]) document.header[i] = document.header[i].replace("cochineal", "default") def revert_cochinealmath(document): " Revert cochineal newtxmath definitions to LaTeX " if find_token(document.header, "\\use_non_tex_fonts false", 0) != -1: i = find_token(document.header, "\\font_math \"cochineal-ntxm\"", 0) if i != -1: add_to_preamble(document, "\\usepackage[cochineal]{newtxmath}") document.header[i] = document.header[i].replace("cochineal-ntxm", "auto") def revert_labelonly(document): " Revert labelonly tag for InsetRef " i = 0 while (True): i = find_token(document.body, "\\begin_inset CommandInset ref", i) if i == -1: return j = find_end_of_inset(document.body, i) if j == -1: document.warning("Can't find end of reference inset at line %d!!" %(i)) i += 1 continue k = find_token(document.body, "LatexCommand labelonly", i, j) if k == -1: i = j continue label = get_quoted_value(document.body, "reference", i, j) if not label: document.warning("Can't find label for reference at line %d!" %(i)) i = j + 1 continue document.body[i:j+1] = put_cmd_in_ert([label]) i += 1 def revert_plural_refs(document): " Revert plural and capitalized references " i = find_token(document.header, "\\use_refstyle 1", 0) use_refstyle = (i != 0) i = 0 while (True): i = find_token(document.body, "\\begin_inset CommandInset ref", i) if i == -1: return j = find_end_of_inset(document.body, i) if j == -1: document.warning("Can't find end of reference inset at line %d!!" %(i)) i += 1 continue plural = caps = suffix = False k = find_token(document.body, "LaTeXCommand formatted", i, j) if k != -1 and use_refstyle: plural = get_bool_value(document.body, "plural", i, j, False) caps = get_bool_value(document.body, "caps", i, j, False) label = get_quoted_value(document.body, "reference", i, j) if label: try: (prefix, suffix) = label.split(":", 1) except: document.warning("No `:' separator in formatted reference at line %d!" % (i)) else: document.warning("Can't find label for reference at line %d!" % (i)) # this effectively tests also for use_refstyle and a formatted reference # we do this complicated test because we would otherwise do this erasure # over and over and over if not ((plural or caps) and suffix): del_token(document.body, "plural", i, j) del_token(document.body, "caps", i, j - 1) # since we deleted a line i = j - 1 continue if caps: prefix = prefix[0].title() + prefix[1:] cmd = "\\" + prefix + "ref" if plural: cmd += "[s]" cmd += "{" + suffix + "}" document.body[i:j+1] = put_cmd_in_ert([cmd]) i += 1 def revert_noprefix(document): " Revert labelonly tags with 'noprefix' set " i = 0 while (True): i = find_token(document.body, "\\begin_inset CommandInset ref", i) if i == -1: return j = find_end_of_inset(document.body, i) if j == -1: document.warning("Can't find end of reference inset at line %d!!" %(i)) i += 1 continue k = find_token(document.body, "LatexCommand labelonly", i, j) if k == -1: i = j continue noprefix = get_bool_value(document.body, "noprefix", i, j) if not noprefix: del_token(document.body, "noprefix", i, j) i = j continue label = get_quoted_value(document.body, "reference", i, j) if not label: document.warning("Can't find label for reference at line %d!" %(i)) i = j + 1 continue try: (prefix, suffix) = label.split(":", 1) except: document.warning("No `:' separator in formatted reference at line %d!" % (i)) # we'll leave this as an ordinary labelonly reference del_token(document.body, "noprefix", i, j) i = j continue document.body[i:j+1] = put_cmd_in_ert([suffix]) i += 1 def revert_biblatex(document): " Revert biblatex support " # # Header # # 1. Get cite engine engine = "basic" i = find_token(document.header, "\\cite_engine", 0) if i == -1: document.warning("Malformed document! Missing \\cite_engine") else: engine = get_value(document.header, "\\cite_engine", i) # 2. Store biblatex state and revert to natbib biblatex = False if engine in ["biblatex", "biblatex-natbib"]: biblatex = True document.header[i] = "\cite_engine natbib" # 3. Store and remove new document headers bibstyle = "" i = find_token(document.header, "\\biblatex_bibstyle", 0) if i != -1: bibstyle = get_value(document.header, "\\biblatex_bibstyle", i) del document.header[i] citestyle = "" i = find_token(document.header, "\\biblatex_citestyle", 0) if i != -1: citestyle = get_value(document.header, "\\biblatex_citestyle", i) del document.header[i] biblio_options = "" i = find_token(document.header, "\\biblio_options", 0) if i != -1: biblio_options = get_value(document.header, "\\biblio_options", i) del document.header[i] if biblatex: bbxopts = "[natbib=true" if bibstyle != "": bbxopts += ",bibstyle=" + bibstyle if citestyle != "": bbxopts += ",citestyle=" + citestyle if biblio_options != "": bbxopts += "," + biblio_options bbxopts += "]" add_to_preamble(document, "\\usepackage" + bbxopts + "{biblatex}") # # Body # # 1. Bibtex insets i = 0 bibresources = [] while (True): i = find_token(document.body, "\\begin_inset CommandInset bibtex", i) if i == -1: break j = find_end_of_inset(document.body, i) if j == -1: document.warning("Can't find end of bibtex inset at line %d!!" %(i)) i += 1 continue bibs = get_quoted_value(document.body, "bibfiles", i, j) opts = get_quoted_value(document.body, "biblatexopts", i, j) # store resources if bibs: bibresources += bibs.split(",") else: document.warning("Can't find bibfiles for bibtex inset at line %d!" %(i)) # remove biblatexopts line k = find_token(document.body, "biblatexopts", i, j) if k != -1: del document.body[k] # Re-find inset end line j = find_end_of_inset(document.body, i) # Insert ERT \\printbibliography and wrap bibtex inset to a Note if biblatex: pcmd = "printbibliography" if opts: pcmd += "[" + opts + "]" repl = ["\\begin_inset ERT", "status open", "", "\\begin_layout Plain Layout",\ "", "", "\\backslash", pcmd, "\\end_layout", "", "\\end_inset", "", "",\ "\\end_layout", "", "\\begin_layout Standard", "\\begin_inset Note Note",\ "status open", "", "\\begin_layout Plain Layout" ] repl += document.body[i:j+1] repl += ["", "\\end_layout", "", "\\end_inset", "", ""] document.body[i:j+1] = repl j += 27 i = j + 1 if biblatex: for b in bibresources: add_to_preamble(document, "\\addbibresource{" + b + ".bib}") # 2. Citation insets # Specific citation insets used in biblatex that need to be reverted to ERT new_citations = { "Cite" : "Cite", "citebyear" : "citeyear", "citeyear" : "cite*", "Footcite" : "Smartcite", "footcite" : "smartcite", "Autocite" : "Autocite", "autocite" : "autocite", "citetitle" : "citetitle", "citetitle*" : "citetitle*", "fullcite" : "fullcite", "footfullcite" : "footfullcite", "supercite" : "supercite", "citeauthor" : "citeauthor", "citeauthor*" : "citeauthor*", "Citeauthor" : "Citeauthor", "Citeauthor*" : "Citeauthor*" } # All commands accepted by LyX < 2.3. Everything else throws an error. old_citations = [ "cite", "nocite", "citet", "citep", "citealt", "citealp",\ "citeauthor", "citeyear", "citeyearpar", "citet*", "citep*",\ "citealt*", "citealp*", "citeauthor*", "Citet", "Citep",\ "Citealt", "Citealp", "Citeauthor", "Citet*", "Citep*",\ "Citealt*", "Citealp*", "Citeauthor*", "fullcite", "footcite",\ "footcitet", "footcitep", "footcitealt", "footcitealp",\ "footciteauthor", "footciteyear", "footciteyearpar",\ "citefield", "citetitle", "cite*" ] i = 0 while (True): i = find_token(document.body, "\\begin_inset CommandInset citation", i) if i == -1: break j = find_end_of_inset(document.body, i) if j == -1: document.warning("Can't find end of citation inset at line %d!!" %(i)) i += 1 continue k = find_token(document.body, "LatexCommand", i, j) if k == -1: document.warning("Can't find LatexCommand for citation inset at line %d!" %(i)) i = j + 1 continue cmd = get_value(document.body, "LatexCommand", k) if biblatex and cmd in list(new_citations.keys()): pre = get_quoted_value(document.body, "before", i, j) post = get_quoted_value(document.body, "after", i, j) key = get_quoted_value(document.body, "key", i, j) if not key: document.warning("Citation inset at line %d does not have a key!" %(i)) key = "???" # Replace known new commands with ERT res = "\\" + new_citations[cmd] if pre: res += "[" + pre + "]" if post: res += "[" + post + "]" elif pre: res += "[]" res += "{" + key + "}" document.body[i:j+1] = put_cmd_in_ert([res]) elif cmd not in old_citations: # Reset unknown commands to cite. This is what LyX does as well # (but LyX 2.2 would break on unknown commands) document.body[k] = "LatexCommand cite" document.warning("Reset unknown cite command '%s' with cite" % cmd) i = j + 1 # Emulate the old biblatex-workaround (pretend natbib in order to use the styles) if biblatex: i = find_token(document.header, "\\begin_local_layout", 0) if i == -1: k = find_token(document.header, "\\language", 0) if k == -1: # this should not happen document.warning("Malformed LyX document! No \\language header found!") return document.header[k-1 : k-1] = ["\\begin_local_layout", "\\end_local_layout"] i = k-1 j = find_end_of(document.header, i, "\\begin_local_layout", "\\end_local_layout") if j == -1: # this should not happen document.warning("Malformed LyX document! Can't find end of local layout!") return document.header[i+1 : i+1] = [ "### Inserted by lyx2lyx (biblatex emulation) ###", "Provides natbib 1", "### End of insertion by lyx2lyx (biblatex emulation) ###" ] def revert_citekeyonly(document): " Revert keyonly cite command to ERT " i = 0 while (True): i = find_token(document.body, "\\begin_inset CommandInset citation", i) if i == -1: break j = find_end_of_inset(document.body, i) if j == -1: document.warning("Can't find end of citation inset at line %d!!" %(i)) i += 1 continue k = find_token(document.body, "LatexCommand", i, j) if k == -1: document.warning("Can't find LatexCommand for citation inset at line %d!" %(i)) i = j + 1 continue cmd = get_value(document.body, "LatexCommand", k) if cmd != "keyonly": i = j + 1 continue key = get_quoted_value(document.body, "key", i, j) if not key: document.warning("Citation inset at line %d does not have a key!" %(i)) # Replace known new commands with ERT document.body[i:j+1] = put_cmd_in_ert([key]) i = j + 1 def revert_bibpackopts(document): " Revert support for natbib/jurabib package options " engine = "basic" i = find_token(document.header, "\\cite_engine", 0) if i == -1: document.warning("Malformed document! Missing \\cite_engine") else: engine = get_value(document.header, "\\cite_engine", i) biblatex = False if engine not in ["natbib", "jurabib"]: return i = find_token(document.header, "\\biblio_options", 0) if i == -1: # Nothing to do if we have no options return biblio_options = get_value(document.header, "\\biblio_options", i) del document.header[i] if not biblio_options: # Nothing to do for empty options return i = find_token(document.header, "\\begin_local_layout", 0) if i == -1: k = find_token(document.header, "\\language", 0) if k == -1: # this should not happen document.warning("Malformed LyX document! No \\language header found!") return document.header[k-1 : k-1] = ["\\begin_local_layout", "\\end_local_layout"] i = k - 1 j = find_end_of(document.header, i, "\\begin_local_layout", "\\end_local_layout") if j == -1: # this should not happen document.warning("Malformed LyX document! Can't find end of local layout!") return document.header[i+1 : i+1] = [ "### Inserted by lyx2lyx (bibliography package options) ###", "PackageOptions " + engine + " " + biblio_options, "### End of insertion by lyx2lyx (bibliography package options) ###" ] def revert_qualicites(document): " Revert qualified citation list commands to ERT " # Citation insets that support qualified lists, with their LaTeX code ql_citations = { "cite" : "cites", "Cite" : "Cites", "citet" : "textcites", "Citet" : "Textcites", "citep" : "parencites", "Citep" : "Parencites", "Footcite" : "Smartcites", "footcite" : "smartcites", "Autocite" : "Autocites", "autocite" : "autocites", } # Get cite engine engine = "basic" i = find_token(document.header, "\\cite_engine", 0) if i == -1: document.warning("Malformed document! Missing \\cite_engine") else: engine = get_value(document.header, "\\cite_engine", i) biblatex = engine in ["biblatex", "biblatex-natbib"] i = 0 while (True): i = find_token(document.body, "\\begin_inset CommandInset citation", i) if i == -1: break j = find_end_of_inset(document.body, i) if j == -1: document.warning("Can't find end of citation inset at line %d!!" %(i)) i += 1 continue pres = find_token(document.body, "pretextlist", i, j) posts = find_token(document.body, "posttextlist", i, j) if pres == -1 and posts == -1: # nothing to do. i = j + 1 continue pretexts = get_quoted_value(document.body, "pretextlist", pres) posttexts = get_quoted_value(document.body, "posttextlist", posts) k = find_token(document.body, "LatexCommand", i, j) if k == -1: document.warning("Can't find LatexCommand for citation inset at line %d!" %(i)) i = j + 1 continue cmd = get_value(document.body, "LatexCommand", k) if biblatex and cmd in list(ql_citations.keys()): pre = get_quoted_value(document.body, "before", i, j) post = get_quoted_value(document.body, "after", i, j) key = get_quoted_value(document.body, "key", i, j) if not key: document.warning("Citation inset at line %d does not have a key!" %(i)) key = "???" keys = key.split(",") prelist = pretexts.split("\t") premap = dict() for pp in prelist: ppp = pp.split(" ", 1) premap[ppp[0]] = ppp[1] postlist = posttexts.split("\t") postmap = dict() for pp in postlist: ppp = pp.split(" ", 1) postmap[ppp[0]] = ppp[1] # Replace known new commands with ERT if "(" in pre or ")" in pre: pre = "{" + pre + "}" if "(" in post or ")" in post: post = "{" + post + "}" res = "\\" + ql_citations[cmd] if pre: res += "(" + pre + ")" if post: res += "(" + post + ")" elif pre: res += "()" for kk in keys: if premap.get(kk, "") != "": res += "[" + premap[kk] + "]" if postmap.get(kk, "") != "": res += "[" + postmap[kk] + "]" elif premap.get(kk, "") != "": res += "[]" res += "{" + kk + "}" document.body[i:j+1] = put_cmd_in_ert([res]) else: # just remove the params del document.body[posttexts] del document.body[pretexts] i += 1 ## # Conversion hub # supported_versions = ["2.3.0", "2.3"] convert = [ [509, [convert_microtype]], [510, [convert_dateinset]], [511, [convert_ibranches]], [512, [convert_beamer_article_styles]], [513, []], [514, []], [515, []], [516, [convert_inputenc]], [517, []], [518, [convert_iopart]], [519, [convert_quotestyle]], [520, []], [521, [convert_frenchquotes]], [522, []], [523, []], [524, []], [525, []], [526, []], [527, []], [528, []], [529, []], [530, []], [531, []] ] revert = [ [530, [revert_qualicites]], [529, [revert_bibpackopts]], [528, [revert_citekeyonly]], [527, [revert_biblatex]], [526, [revert_noprefix]], [525, [revert_plural_refs]], [524, [revert_labelonly]], [523, [revert_crimson, revert_cochinealmath]], [522, [revert_cjkquotes]], [521, [revert_dynamicquotes]], [520, [revert_britishquotes, revert_swedishgquotes, revert_frenchquotes, revert_frenchinquotes, revert_russianquotes, revert_swissquotes]], [519, [revert_plainquote]], [518, [revert_quotestyle]], [517, [revert_iopart]], [516, [revert_quotes]], [515, []], [514, [revert_urdu, revert_syriac]], [513, [revert_amharic, revert_asturian, revert_kannada, revert_khmer]], [512, [revert_bosnian, revert_friulan, revert_macedonian, revert_piedmontese, revert_romansh]], [511, [revert_beamer_article_styles]], [510, [revert_ibranches]], [509, []], [508, [revert_microtype]] ] if __name__ == "__main__": pass