X-Git-Url: https://git.lyx.org/gitweb/?a=blobdiff_plain;f=lib%2Flyx2lyx%2Flyx_2_0.py;h=ad5636421bb23d16acfd4e9772b17a748bb62243;hb=b3b7675f544128ea4bbff564262774533af3598f;hp=681e41811818b6c1fca21d08d8a8038cce60a7d6;hpb=ac0ee09a237d4f8158840deada1b3f32d9c32d17;p=lyx.git diff --git a/lib/lyx2lyx/lyx_2_0.py b/lib/lyx2lyx/lyx_2_0.py index 681e418118..ad5636421b 100644 --- a/lib/lyx2lyx/lyx_2_0.py +++ b/lib/lyx2lyx/lyx_2_0.py @@ -1,7 +1,6 @@ # -*- coding: utf-8 -*- # This file is part of lyx2lyx -# -*- coding: utf-8 -*- -# Copyright (C) 2010 The LyX team +# Copyright (C) 2011 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 @@ -15,7 +14,7 @@ # # 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. +# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA """ Convert files to the file format generated by lyx 2.0""" @@ -24,36 +23,27 @@ import unicodedata import sys, os from parser_tools import find_token, find_end_of, find_tokens, \ - find_end_of_inset, find_end_of_layout, get_value, get_value_string - + find_token_exact, find_end_of_inset, find_end_of_layout, \ + find_token_backwards, is_in_inset, get_value, get_quoted_value, \ + del_token, check_token, get_option_value + from lyx2lyx_tools import add_to_preamble, insert_to_preamble, \ put_cmd_in_ert, lyx2latex, latex_length, revert_flex_inset, \ - revert_font_attrs, revert_layout_command, hex2ratio + revert_font_attrs, hex2ratio, str2bool #################################################################### # Private helper functions -def remove_option(document, m, option): +def remove_option(lines, m, option): ''' removes option from line m. returns whether we did anything ''' - l = document.body[m].find(option) + l = lines[m].find(option) if l == -1: return False - val = document.body[m][l:].split('"')[1] - document.body[m] = document.body[m][:l - 1] + document.body[m][l+len(option + '="' + val + '"'):] + val = lines[m][l:].split('"')[1] + lines[m] = lines[m][:l - 1] + lines[m][l+len(option + '="' + val + '"'):] return True -# DO NOT USE THIS ROUTINE ANY MORE. Better yet, replace the uses that -# have been made of it with uses of put_cmd_in_ert. -def old_put_cmd_in_ert(string): - for rep in unicode_reps: - string = string.replace(rep[1], rep[0].replace('\\\\', '\\')) - string = string.replace('\\', "\\backslash\n") - string = "\\begin_inset ERT\nstatus collapsed\n\\begin_layout Plain Layout\n" \ - + string + "\n\\end_layout\n\\end_inset" - return string - - ############################################################################### ### ### Conversion and reversion routines @@ -98,10 +88,7 @@ def revert_tabularvalign(document): if p != -1: q = document.body[fline].find("tabularvalignment") if q != -1: - # FIXME - # This seems wrong: It removes everything after - # tabularvalignment, too. - document.body[fline] = document.body[fline][:q - 1] + '>' + document.body[fline] = re.sub(r' tabularvalignment=\"[a-z]+\"', "", document.body[fline]) i += 1 continue @@ -117,16 +104,13 @@ def revert_tabularvalign(document): # delete tabularvalignment q = document.body[fline].find("tabularvalignment") if q != -1: - # FIXME - # This seems wrong: It removes everything after - # tabularvalignment, too. - document.body[fline] = document.body[fline][:q - 1] + '>' + document.body[fline] = re.sub(r' tabularvalignment=\"[a-z]+\"', "", document.body[fline]) # don't add a box when centered if tabularvalignment == 'c': i = end continue - subst = ['\\end_layout', '\\end_inset'] + subst = ['\\end_inset', '\\end_layout'] document.body[end:end] = subst # just inserts those lines subst = ['\\begin_inset Box Frameless', 'position "' + tabularvalignment +'"', @@ -165,13 +149,13 @@ def revert_phantom_types(document, ptype, cmd): document.warning("Can't find layout for inset at line " + str(i)) i = end continue - bend = find_token(document.body, "\\end_layout", blay, end) + bend = find_end_of_layout(document.body, blay) if bend == -1: document.warning("Malformed LyX document: Could not find end of Phantom inset's layout.") i = end continue substi = ["\\begin_inset ERT", "status collapsed", "", - "\\begin_layout Plain Layout", "", "", "\\backslash", + "\\begin_layout Plain Layout", "", "", "\\backslash", cmd + "{", "\\end_layout", "", "\\end_inset"] substj = ["\\size default", "", "\\begin_inset ERT", "status collapsed", "", "\\begin_layout Plain Layout", "", "}", "\\end_layout", "", "\\end_inset"] @@ -183,7 +167,7 @@ def revert_phantom_types(document, ptype, cmd): def revert_phantom(document): revert_phantom_types(document, "Phantom", "phantom") - + def revert_hphantom(document): revert_phantom_types(document, "HPhantom", "hphantom") @@ -193,102 +177,117 @@ def revert_vphantom(document): def revert_xetex(document): " Reverts documents that use XeTeX " + i = find_token(document.header, '\\use_xetex', 0) if i == -1: document.warning("Malformed LyX document: Missing \\use_xetex.") return - if get_value(document.header, "\\use_xetex", i) == 'false': + if not str2bool(get_value(document.header, "\\use_xetex", i)): del document.header[i] return del document.header[i] + # 1.) set doc encoding to utf8-plain i = find_token(document.header, "\\inputencoding", 0) if i == -1: document.warning("Malformed LyX document: Missing \\inputencoding.") - document.header[i] = "\\inputencoding utf8-plain" + else: + document.header[i] = "\\inputencoding utf8-plain" + # 2.) check font settings - l = find_token(document.header, "\\font_roman", 0) - if l == -1: - document.warning("Malformed LyX document: Missing \\font_roman.") - line = document.header[l] - l = re.compile(r'\\font_roman (.*)$') - m = l.match(line) - roman = m.group(1) - l = find_token(document.header, "\\font_sans", 0) - if l == -1: - document.warning("Malformed LyX document: Missing \\font_sans.") - line = document.header[l] - l = re.compile(r'\\font_sans (.*)$') - m = l.match(line) - sans = m.group(1) - l = find_token(document.header, "\\font_typewriter", 0) - if l == -1: - document.warning("Malformed LyX document: Missing \\font_typewriter.") - line = document.header[l] - l = re.compile(r'\\font_typewriter (.*)$') - m = l.match(line) - typewriter = m.group(1) - osf = get_value(document.header, '\\font_osf', 0) == "true" - sf_scale = float(get_value(document.header, '\\font_sf_scale', 0)) - tt_scale = float(get_value(document.header, '\\font_tt_scale', 0)) - # 3.) set preamble stuff - pretext = '%% This document must be processed with xelatex!\n' - pretext += '\\usepackage{fontspec}\n' - if roman != "default": - pretext += '\\setmainfont[Mapping=tex-text]{' + roman + '}\n' - if sans != "default": - pretext += '\\setsansfont[' - if sf_scale != 100: - pretext += 'Scale=' + str(sf_scale / 100) + ',' - pretext += 'Mapping=tex-text]{' + sans + '}\n' - if typewriter != "default": - pretext += '\\setmonofont' - if tt_scale != 100: - pretext += '[Scale=' + str(tt_scale / 100) + ']' - pretext += '{' + typewriter + '}\n' - if osf: - pretext += '\\defaultfontfeatures{Numbers=OldStyle}\n' - pretext += '\usepackage{xunicode}\n' - pretext += '\usepackage{xltxtra}\n' - insert_to_preamble(0, document, pretext) - # 4.) reset font settings + # defaults + roman = sans = typew = "default" + osf = False + sf_scale = tt_scale = 100.0 + i = find_token(document.header, "\\font_roman", 0) if i == -1: document.warning("Malformed LyX document: Missing \\font_roman.") - document.header[i] = "\\font_roman default" + else: + roman = get_value(document.header, "\\font_roman", i) + document.header[i] = "\\font_roman default" + i = find_token(document.header, "\\font_sans", 0) if i == -1: document.warning("Malformed LyX document: Missing \\font_sans.") - document.header[i] = "\\font_sans default" + else: + sans = get_value(document.header, "\\font_sans", i) + document.header[i] = "\\font_sans default" + i = find_token(document.header, "\\font_typewriter", 0) if i == -1: document.warning("Malformed LyX document: Missing \\font_typewriter.") - document.header[i] = "\\font_typewriter default" + else: + typew = get_value(document.header, "\\font_typewriter", i) + document.header[i] = "\\font_typewriter default" + i = find_token(document.header, "\\font_osf", 0) if i == -1: document.warning("Malformed LyX document: Missing \\font_osf.") - document.header[i] = "\\font_osf false" + else: + osf = str2bool(get_value(document.header, "\\font_osf", i)) + document.header[i] = "\\font_osf false" + i = find_token(document.header, "\\font_sc", 0) if i == -1: document.warning("Malformed LyX document: Missing \\font_sc.") - document.header[i] = "\\font_sc false" + else: + # we do not need this value. + document.header[i] = "\\font_sc false" + i = find_token(document.header, "\\font_sf_scale", 0) if i == -1: document.warning("Malformed LyX document: Missing \\font_sf_scale.") - document.header[i] = "\\font_sf_scale 100" + else: + val = get_value(document.header, '\\font_sf_scale', i) + try: + # float() can throw + sf_scale = float(val) + except: + document.warning("Invalid font_sf_scale value: " + val) + document.header[i] = "\\font_sf_scale 100" + i = find_token(document.header, "\\font_tt_scale", 0) if i == -1: document.warning("Malformed LyX document: Missing \\font_tt_scale.") - document.header[i] = "\\font_tt_scale 100" + else: + val = get_value(document.header, '\\font_tt_scale', i) + try: + # float() can throw + tt_scale = float(val) + except: + document.warning("Invalid font_tt_scale value: " + val) + document.header[i] = "\\font_tt_scale 100" + + # 3.) set preamble stuff + pretext = ['%% This document must be processed with xelatex!'] + pretext.append('\\usepackage{fontspec}') + if roman != "default": + pretext.append('\\setmainfont[Mapping=tex-text]{' + roman + '}') + if sans != "default": + sf = '\\setsansfont[' + if sf_scale != 100.0: + sf += 'Scale=' + str(sf_scale / 100.0) + ',' + sf += 'Mapping=tex-text]{' + sans + '}' + pretext.append(sf) + if typew != "default": + tw = '\\setmonofont' + if tt_scale != 100.0: + tw += '[Scale=' + str(tt_scale / 100.0) + ']' + tw += '{' + typew + '}' + pretext.append(tw) + if osf: + pretext.append('\\defaultfontfeatures{Numbers=OldStyle}') + pretext.append('\\usepackage{xunicode}') + pretext.append('\\usepackage{xltxtra}') + insert_to_preamble(document, pretext) def revert_outputformat(document): " Remove default output format param " - i = find_token(document.header, '\\default_output_format', 0) - if i == -1: + + if not del_token(document.header, '\\default_output_format', 0): document.warning("Malformed LyX document: Missing \\default_output_format.") - return - del document.header[i] def revert_backgroundcolor(document): @@ -304,27 +303,47 @@ def revert_backgroundcolor(document): red = hex2ratio(colorcode[1:3]) green = hex2ratio(colorcode[3:5]) blue = hex2ratio(colorcode[5:7]) - insert_to_preamble(0, document, - '% Commands inserted by lyx2lyx to set the background color\n' - + '\\@ifundefined{definecolor}{\\usepackage{color}}{}\n' - + '\\definecolor{page_backgroundcolor}{rgb}{' - + red + ',' + green + ',' + blue + '}\n' - + '\\pagecolor{page_backgroundcolor}\n') + insert_to_preamble(document, \ + ['% To set the background color', + '\\@ifundefined{definecolor}{\\usepackage{color}}{}', + '\\definecolor{page_backgroundcolor}{rgb}{' + red + ',' + green + ',' + blue + '}', + '\\pagecolor{page_backgroundcolor}']) -def revert_splitindex(document): - " Reverts splitindex-aware documents " +def add_use_indices(document): + " Add \\use_indices if it is missing " i = find_token(document.header, '\\use_indices', 0) + if i != -1: + return i + i = find_token(document.header, '\\use_bibtopic', 0) + if i == -1: + i = find_token(document.header, '\\cite_engine', 0) + if i == -1: + i = find_token(document.header, '\\use_mathdots', 0) + if i == -1: + i = find_token(document.header, '\\use_mhchem', 0) + if i == -1: + i = find_token(document.header, '\\use_esint', 0) + if i == -1: + i = find_token(document.header, '\\use_amsmath', 0) if i == -1: document.warning("Malformed LyX document: Missing \\use_indices.") + return -1 + document.header.insert(i + 1, '\\use_indices 0') + return i + 1 + + +def revert_splitindex(document): + " Reverts splitindex-aware documents " + i = add_use_indices(document) + if i == -1: return - indices = get_value(document.header, "\\use_indices", i) - preamble = "" - useindices = (indices == "true") - if useindices: - preamble += "\\usepackage{splitidx}\n" + useindices = str2bool(get_value(document.header, "\\use_indices", i)) del document.header[i] - + preamble = [] + if useindices: + preamble.append("\\usepackage{splitidx})") + # deal with index declarations in the preamble i = 0 while True: @@ -335,18 +354,18 @@ def revert_splitindex(document): if k == -1: document.warning("Malformed LyX document: Missing \\end_index.") return - if useindices: + if useindices: line = document.header[i] l = re.compile(r'\\index (.*)$') m = l.match(line) iname = m.group(1) ishortcut = get_value(document.header, '\\shortcut', i, k) if ishortcut != "": - preamble += "\\newindex[" + iname + "]{" + ishortcut + "}\n" + preamble.append("\\newindex[" + iname + "]{" + ishortcut + "}") del document.header[i:k + 1] - if preamble != "": - insert_to_preamble(0, document, preamble) - + if preamble: + insert_to_preamble(document, preamble) + # deal with index insets # these need to have the argument removed i = 0 @@ -372,7 +391,7 @@ def revert_splitindex(document): subst = put_cmd_in_ert("\\sindex[" + itype + "]{" + content + "}") document.body[i:k + 1] = subst i = i + 1 - + # deal with index_print insets i = 0 while True: @@ -380,7 +399,7 @@ def revert_splitindex(document): if i == -1: return k = find_end_of_inset(document.body, i) - ptype = get_value(document.body, 'type', i, k).strip('"') + ptype = get_quoted_value(document.body, 'type', i, k) if ptype == "idx": j = find_token(document.body, "type", i, k) del document.body[j] @@ -394,6 +413,7 @@ def revert_splitindex(document): def convert_splitindex(document): " Converts index and printindex insets to splitindex-aware format " + add_use_indices(document) i = 0 while True: i = find_token(document.body, "\\begin_inset Index", i) @@ -410,7 +430,7 @@ def convert_splitindex(document): if document.body[i + 1].find('LatexCommand printindex') == -1: document.warning("Malformed LyX document: Incomplete printindex inset.") return - subst = ["LatexCommand printindex", + subst = ["LatexCommand printindex", "type \"idx\""] document.body[i + 1:i + 2] = subst i = i + 1 @@ -418,12 +438,10 @@ def convert_splitindex(document): def revert_subindex(document): " Reverts \\printsubindex CommandInset types " - i = find_token(document.header, '\\use_indices', 0) + i = add_use_indices(document) if i == -1: - document.warning("Malformed LyX document: Missing \\use_indices.") return - indices = get_value(document.header, "\\use_indices", i) - useindices = (indices == "true") + useindices = str2bool(get_value(document.header, "\\use_indices", i)) i = 0 while True: i = find_token(document.body, "\\begin_inset CommandInset index_print", i) @@ -434,7 +452,7 @@ def revert_subindex(document): if ctype != "printsubindex": i = k + 1 continue - ptype = get_value(document.body, 'type', i, k).strip('"') + ptype = get_quoted_value(document.body, 'type', i, k) if not useindices: del document.body[i:k + 1] else: @@ -445,12 +463,10 @@ def revert_subindex(document): def revert_printindexall(document): " Reverts \\print[sub]index* CommandInset types " - i = find_token(document.header, '\\use_indices', 0) + i = add_use_indices(document) if i == -1: - document.warning("Malformed LyX document: Missing \\use_indices.") return - indices = get_value(document.header, "\\use_indices", i) - useindices = (indices == "true") + useindices = str2bool(get_value(document.header, "\\use_indices", i)) i = 0 while True: i = find_token(document.body, "\\begin_inset CommandInset index_print", i) @@ -471,14 +487,14 @@ def revert_printindexall(document): def revert_strikeout(document): " Reverts \\strikeout font attribute " - changed = revert_font_attrs(document, "\\uuline", "\\uuline") - changed = revert_font_attrs(document, "\\uwave", "\\uwave") or changed - changed = revert_font_attrs(document, "\\strikeout", "\\sout") or changed + changed = revert_font_attrs(document.body, "\\uuline", "\\uuline") + changed = revert_font_attrs(document.body, "\\uwave", "\\uwave") or changed + changed = revert_font_attrs(document.body, "\\strikeout", "\\sout") or changed if changed == True: - insert_to_preamble(0, document, - '% Commands inserted by lyx2lyx for proper underlining\n' - + '\\PassOptionsToPackage{normalem}{ulem}\n' - + '\\usepackage{ulem}\n') + insert_to_preamble(document, \ + ['% for proper underlining', + '\\PassOptionsToPackage{normalem}{ulem}', + '\\usepackage{ulem}']) def revert_ulinelatex(document): @@ -486,28 +502,24 @@ def revert_ulinelatex(document): i = find_token(document.body, '\\bar under', 0) if i == -1: return - insert_to_preamble(0, document, - '% Commands inserted by lyx2lyx for proper underlining\n' - + '\\PassOptionsToPackage{normalem}{ulem}\n' - + '\\usepackage{ulem}\n' - + '\\let\\cite@rig\\cite\n' - + '\\newcommand{\\b@xcite}[2][\\%]{\\def\\def@pt{\\%}\\def\\pas@pt{#1}\n' - + ' \\mbox{\\ifx\\def@pt\\pas@pt\\cite@rig{#2}\\else\\cite@rig[#1]{#2}\\fi}}\n' - + '\\renewcommand{\\underbar}[1]{{\\let\\cite\\b@xcite\\uline{#1}}}\n') + insert_to_preamble(document,\ + ['% for proper underlining', + '\\PassOptionsToPackage{normalem}{ulem}', + '\\usepackage{ulem}', + '\\let\\cite@rig\\cite', + '\\newcommand{\\b@xcite}[2][\\%]{\\def\\def@pt{\\%}\\def\\pas@pt{#1}', + ' \\mbox{\\ifx\\def@pt\\pas@pt\\cite@rig{#2}\\else\\cite@rig[#1]{#2}\\fi}}', + '\\renewcommand{\\underbar}[1]{{\\let\\cite\\b@xcite\\uline{#1}}}']) def revert_custom_processors(document): " Remove bibtex_command and index_command params " - i = find_token(document.header, '\\bibtex_command', 0) - if i == -1: + + if not del_token(document.header, '\\bibtex_command', 0): document.warning("Malformed LyX document: Missing \\bibtex_command.") - else: - del document.header[i] - i = find_token(document.header, '\\index_command', 0) - if i == -1: + + if not del_token(document.header, '\\index_command', 0): document.warning("Malformed LyX document: Missing \\index_command.") - else: - del document.header[i] def convert_nomencl_width(document): @@ -529,13 +541,9 @@ def revert_nomencl_width(document): if i == -1: break j = find_end_of_inset(document.body, i) - l = find_token(document.body, "set_width", i, j) - if l == -1: - document.warning("Can't find set_width option for nomencl_print!") - i = j - continue - del document.body[l] - i = j - 1 + if not del_token(document.body, "set_width", i, j): + document.warning("Can't find set_width option for nomencl_print!") + i = j def revert_nomencl_cwidth(document): @@ -548,13 +556,11 @@ def revert_nomencl_cwidth(document): j = find_end_of_inset(document.body, i) l = find_token(document.body, "width", i, j) if l == -1: - document.warning("Can't find width option for nomencl_print!") i = j continue - width = get_value(document.body, "width", i, j).strip('"') + width = get_quoted_value(document.body, "width", i, j) del document.body[l] - add_to_preamble(document, ["% this command was inserted by lyx2lyx"]) - add_to_preamble(document, ["\\setlength{\\nomlabelwidth}{" + width + "}"]) + insert_to_preamble(document, ["\\setlength{\\nomlabelwidth}{" + width + "}"]) i = j - 1 @@ -589,10 +595,10 @@ def revert_longtable_align(document): if j == -1: i += 1 continue - # FIXME Is this correct? It wipes out everything after the + # FIXME Is this correct? It wipes out everything after the # one we found. document.body[fline] = document.body[fline][:j - 1] + '>' - # since there could be a tabular inside this one, we + # since there could be a tabular inside this one, we # cannot jump to end. i += 1 @@ -617,8 +623,7 @@ def revert_paragraph_indentation(document): if length != "default": # handle percent lengths length = latex_length(length)[1] - add_to_preamble(document, ["% this command was inserted by lyx2lyx"]) - add_to_preamble(document, ["\\setlength{\\parindent}{" + length + "}"]) + insert_to_preamble(document, ["\\setlength{\\parindent}{" + length + "}"]) del document.header[i] @@ -635,8 +640,7 @@ def revert_percent_skip_lengths(document): # handle percent lengths percent, length = latex_length(length) if percent: - add_to_preamble(document, ["% this command was inserted by lyx2lyx"]) - add_to_preamble(document, ["\\setlength{\\parskip}{" + length + "}"]) + insert_to_preamble(document, ["\\setlength{\\parskip}{" + length + "}"]) # set defskip to medskip as default document.header[i] = "\\defskip medskip" @@ -676,7 +680,7 @@ def revert_percent_hspace_lengths(document): " Revert relative HSpace lengths to ERT " i = 0 while True: - i = find_token(document.body, "\\begin_inset space \\hspace", i) + i = find_token_exact(document.body, "\\begin_inset space \\hspace", i) if i == -1: break j = find_end_of_inset(document.body, i) @@ -707,7 +711,7 @@ def revert_hspace_glue_lengths(document): " Revert HSpace glue lengths to ERT " i = 0 while True: - i = find_token(document.body, "\\begin_inset space \\hspace", i) + i = find_token_exact(document.body, "\\begin_inset space \\hspace", i) if i == -1: break j = find_end_of_inset(document.body, i) @@ -738,7 +742,7 @@ def convert_author_id(document): i = 0 anum = 1 re_author = re.compile(r'(\\author) (\".*\")\s*(.*)$') - + while True: i = find_token(document.header, "\\author", i) if i == -1: @@ -748,10 +752,9 @@ def convert_author_id(document): name = m.group(2) email = m.group(3) document.header[i] = "\\author %i %s %s" % (anum, name, email) - # FIXME Should this really be incremented if we didn't match? anum += 1 i += 1 - + i = 0 while True: i = find_token(document.body, "\\change_", i) @@ -770,7 +773,7 @@ def revert_author_id(document): " Remove the author_id from the \\author definition " i = 0 anum = 0 - rx = re.compile(r'(\\author)\s+(\d+)\s+(\".*\")\s*(.*)$') + rx = re.compile(r'(\\author)\s+(-?\d+)\s+(\".*\")\s*(.*)$') idmap = dict() while True: @@ -809,13 +812,34 @@ def revert_suppress_date(document): return # remove the preamble line and write to the preamble # when suppress_date was true - date = get_value(document.header, "\\suppress_date", i) - if date == "true": - add_to_preamble(document, ["% this command was inserted by lyx2lyx"]) + date = str2bool(get_value(document.header, "\\suppress_date", i)) + if date: add_to_preamble(document, ["\\date{}"]) del document.header[i] +def convert_mhchem(document): + "Set mhchem to off for versions older than 1.6.x" + if document.initial_format < 277: + # LyX 1.5.x and older did never load mhchem. + # Therefore we must switch it off: Documents that use mhchem have + # a manual \usepackage anyway, and documents not using mhchem but + # custom macros with the same names as mhchem commands might get + # corrupted if mhchem is automatically loaded. + mhchem = 0 # off + else: + # LyX 1.6.x did always load mhchem automatically. + mhchem = 1 # auto + i = find_token(document.header, "\\use_esint", 0) + if i == -1: + # pre-1.5.x document + i = find_token(document.header, "\\use_amsmath", 0) + if i == -1: + document.warning("Malformed LyX document: Could not find amsmath os esint setting.") + return + document.header.insert(i + 1, "\\use_mhchem %d" % mhchem) + + def revert_mhchem(document): "Revert mhchem loading to preamble code" @@ -832,6 +856,10 @@ def revert_mhchem(document): mhchem = "on" del document.header[i] + if mhchem == "off": + # don't load case + return + if mhchem == "auto": i = 0 while True: @@ -839,25 +867,21 @@ def revert_mhchem(document): if i == -1: break line = document.body[i] - if line.find("\\ce{") != -1 or line.find("\\cf{") != 1: + if line.find("\\ce{") != -1 or line.find("\\cf{") != -1: mhchem = "on" break i += 1 if mhchem == "on": - pre = ["% lyx2lyx mhchem commands", - "\\PassOptionsToPackage{version=3}{mhchem}", + pre = ["\\PassOptionsToPackage{version=3}{mhchem}", "\\usepackage{mhchem}"] - add_to_preamble(document, pre) + insert_to_preamble(document, pre) def revert_fontenc(document): " Remove fontencoding param " - i = find_token(document.header, '\\fontencoding', 0) - if i == -1: + if not del_token(document.header, '\\fontencoding', 0): document.warning("Malformed LyX document: Missing \\fontencoding.") - return - del document.header[i] def merge_gbrief(document): @@ -891,7 +915,7 @@ def merge_gbrief(document): "Verteiler": "cc", "Gruss": "Closing"} i = 0 - while 1: + while True: i = find_token(document.body, "\\begin_layout", i) if i == -1: break @@ -901,7 +925,7 @@ def merge_gbrief(document): document.body[i] = "\\begin_layout " + obsoletedby[layout] i += 1 - + document.textclass = "g-brief" document.set_textclass() @@ -915,12 +939,8 @@ def revert_gbrief(document): def revert_html_options(document): " Remove html options " - i = find_token(document.header, '\\html_use_mathml', 0) - if i != -1: - del document.header[i] - i = find_token(document.header, '\\html_be_strict', 0) - if i != -1: - del document.header[i] + del_token(document.header, '\\html_use_mathml', 0) + del_token(document.header, '\\html_be_strict', 0) def revert_includeonly(document): @@ -938,72 +958,135 @@ def revert_includeonly(document): def revert_includeall(document): " Remove maintain_unincluded_children param " - i = find_token(document.header, '\\maintain_unincluded_children', 0) - if i != -1: - del document.header[i] + del_token(document.header, '\\maintain_unincluded_children', 0) def revert_multirow(document): " Revert multirow cells in tables to TeX-code" - i = 0 - multirow = False + + # first, let's find out if we need to do anything + # cell type 3 is multirow begin cell + i = find_token(document.body, '", i) - if cend == -1: - document.warning("Malformed LyX document: Could not find end of tabular cell.") - i += 1 - continue - blay = find_token(document.body, "\\begin_layout", i, cend) - if blay == -1: - document.warning("Can't find layout for cell!") - i = j - continue - bend = find_end_of_layout(document.body, blay) - if blay == -1: - document.warning("Can't find end of layout for cell!") - i = cend - continue + # find begin/end of table + begin_table = find_token(document.body, '') + if end_table == -1: + document.warning("Malformed LyX document: Could not find end of table.") + begin_table += 1 + continue + # does this table have multirow? + i = find_token(document.body, '', begin_row, end_table) + if begin_row == -1: + document.warning("Can't find row " + str(row + 1)) break - # remove the multirow tag, set the valignment to top - # and remove the top line - # FIXME Are we sure these always have space around them? - document.body[k] = document.body[k].replace(' multirow="4" ', ' ') - document.body[k] = document.body[k].replace('valignment="middle"', 'valignment="top"') - document.body[k] = document.body[k].replace(' topline="true" ', ' ') - k += 1 - # this will always be ok - i = cend - - if multirow == True: - add_to_preamble(document, - ["% lyx2lyx multirow additions ", "\\usepackage{multirow}"]) + end_row = find_end_of(document.body, begin_row, '', '') + if end_row == -1: + document.warning("Can't find end of row " + str(row + 1)) + break + begin_cell = begin_row + multirows.append([]) + for column in range(numcols): + begin_cell = find_token(document.body, '" in a cell, but + # that seems fairly unlikely. + end_cell = find_end_of(document.body, begin_cell, '') + if end_cell == -1: + document.warning("Can't find end of column " + str(column + 1) + \ + "in row " + str(row + 1)) + break + multirows[row].append([begin_cell, end_cell, 0]) + if document.body[begin_cell].find('multirow="3"') != -1: + multirows[row][column][2] = 3 # begin multirow + mrstarts.append([row, column]) + elif document.body[begin_cell].find('multirow="4"') != -1: + multirows[row][column][2] = 4 # in multirow + begin_cell = end_cell + begin_row = end_row + # end of table info collection + + # work from the back to avoid messing up numbering + mrstarts.reverse() + for m in mrstarts: + row = m[0] + col = m[1] + # get column width + col_width = get_option_value(document.body[begin_table + 2 + col], "width") + # "0pt" means that no width is specified + if not col_width or col_width == "0pt": + col_width = "*" + # determine the number of cells that are part of the multirow + nummrs = 1 + for r in range(row + 1, numrows): + if multirows[r][col][2] != 4: + break + nummrs += 1 + # take the opportunity to revert this line + lineno = multirows[r][col][0] + document.body[lineno] = document.body[lineno].\ + replace(' multirow="4" ', ' ').\ + replace('valignment="middle"', 'valignment="top"').\ + replace(' topline="true" ', ' ') + # remove bottom line of previous multirow-part cell + lineno = multirows[r-1][col][0] + document.body[lineno] = document.body[lineno].replace(' bottomline="true" ', ' ') + # revert beginning cell + bcell = multirows[row][col][0] + ecell = multirows[row][col][1] + document.body[bcell] = document.body[bcell].\ + replace(' multirow="3" ', ' ').\ + replace('valignment="middle"', 'valignment="top"') + blay = find_token(document.body, "\\begin_layout", bcell, ecell) + if blay == -1: + document.warning("Can't find layout for cell!") + continue + bend = find_end_of_layout(document.body, blay) + if bend == -1: + document.warning("Can't find end of layout for cell!") + continue + # do the later one first, so as not to mess up the numbering + # we are wrapping the whole cell in this ert + # so before the end of the layout... + document.body[bend:bend] = put_cmd_in_ert("}") + # ...and after the beginning + document.body[blay + 1:blay + 1] = \ + put_cmd_in_ert("\\multirow{" + str(nummrs) + "}{" + col_width + "}{") + + begin_table = end_table def convert_math_output(document): @@ -1015,8 +1098,8 @@ def convert_math_output(document): m = rgx.match(document.header[i]) newval = "0" # MathML if m: - val = m.group(1) - if val != "true": + val = str2bool(m.group(1)) + if not val: newval = "2" # Images else: document.warning("Can't match " + document.header[i]) @@ -1038,7 +1121,7 @@ def revert_math_output(document): else: document.warning("Unable to match " + document.header[i]) document.header[i] = "\\html_use_mathml " + newval - + def revert_inset_preview(document): @@ -1053,7 +1136,7 @@ def revert_inset_preview(document): document.warning("Malformed LyX document: Could not find end of Preview inset.") i += 1 continue - + # This has several issues. # We need to do something about the layouts inside InsetPreview. # If we just leave the first one, then we have something like: @@ -1062,16 +1145,16 @@ def revert_inset_preview(document): # \begin_layout Standard # and we get a "no \end_layout" error. So something has to be done. # Ideally, we would check if it is the same as the layout we are in. - # If so, we just remove it; if not, we end the active one. But it is + # If so, we just remove it; if not, we end the active one. But it is # not easy to know what layout we are in, due to depth changes, etc, # and it is not clear to me how much work it is worth doing. In most # cases, the layout will probably be the same. - # + # # For the same reason, we have to remove the \end_layout tag at the # end of the last layout in the inset. Again, that will sometimes be # wrong, but it will usually be right. To know what to do, we would # again have to know what layout the inset is in. - + blay = find_token(document.body, "\\begin_layout", i, iend) if blay == -1: document.warning("Can't find layout for preview inset!") @@ -1083,13 +1166,13 @@ def revert_inset_preview(document): # This is where we would check what layout we are in. # The check for Standard is definitely wrong. - # + # # lay = document.body[blay].split(None, 1)[1] # if lay != oldlayout: # # record a boolean to tell us what to do later.... # # better to do it later, since (a) it won't mess up # # the numbering and (b) we only modify at the end. - + # we want to delete the last \\end_layout in this inset, too. # note that this may not be the \\end_layout that goes with blay!! bend = find_end_of_layout(document.body, blay) @@ -1109,7 +1192,7 @@ def revert_inset_preview(document): del document.body[bend] del document.body[i:blay + 1] # we do not need to reset i - + def revert_equalspacing_xymatrix(document): " Revert a Formula with xymatrix@! to an ERT inset " @@ -1126,12 +1209,12 @@ def revert_equalspacing_xymatrix(document): document.warning("Malformed LyX document: Could not find end of Formula inset.") i += 1 continue - + for curline in range(i,j): found = document.body[curline].find("\\xymatrix@!") if found != -1: break - + if found != -1: has_equal_spacing = True content = [document.body[i][21:]] @@ -1146,9 +1229,9 @@ def revert_equalspacing_xymatrix(document): has_preamble = True; break; i = j + 1 - + if has_equal_spacing and not has_preamble: - add_to_preamble(document, ['% lyx2lyx xymatrix addition', '\\usepackage[all]{xy}']) + add_to_preamble(document, ['\\usepackage[all]{xy}']) def revert_notefontcolor(document): @@ -1158,22 +1241,22 @@ def revert_notefontcolor(document): if i == -1: return + colorcode = get_value(document.header, '\\notefontcolor', i) + del document.header[i] + # are there any grey notes? if find_token(document.body, "\\begin_inset Note Greyedout", 0) == -1: - # no need to do anything, and \renewcommand will throw an error - # since lyxgreyedout will not exist. + # no need to do anything else, and \renewcommand will throw + # an error since lyxgreyedout will not exist. return - colorcode = get_value(document.header, '\\notefontcolor', i) - del document.header[i] # the color code is in the form #rrggbb where every character denotes a hex number red = hex2ratio(colorcode[1:3]) green = hex2ratio(colorcode[3:5]) blue = hex2ratio(colorcode[5:7]) # write the preamble - insert_to_preamble(0, document, - ['% Commands inserted by lyx2lyx to set the font color', - '% for greyed-out notes', + insert_to_preamble(document, + [ '% for greyed-out notes', '\\@ifundefined{definecolor}{\\usepackage{color}}{}' '\\definecolor{note_fontcolor}{rgb}{%s,%s,%s}' % (red, green, blue), '\\renewenvironment{lyxgreyedout}', @@ -1181,21 +1264,21 @@ def revert_notefontcolor(document): def revert_turkmen(document): - "Set language Turkmen to English" + "Set language Turkmen to English" - if document.language == "turkmen": - document.language = "english" - i = find_token(document.header, "\\language", 0) - if i != -1: - document.header[i] = "\\language english" + if document.language == "turkmen": + document.language = "english" + i = find_token(document.header, "\\language", 0) + if i != -1: + document.header[i] = "\\language english" - j = 0 - while True: - j = find_token(document.body, "\\lang turkmen", j) - if j == -1: - return - document.body[j] = document.body[j].replace("\\lang turkmen", "\\lang english") - j += 1 + j = 0 + while True: + j = find_token(document.body, "\\lang turkmen", j) + if j == -1: + return + document.body[j] = document.body[j].replace("\\lang turkmen", "\\lang english") + j += 1 def revert_fontcolor(document): @@ -1213,8 +1296,8 @@ def revert_fontcolor(document): green = hex2ratio(colorcode[3:5]) blue = hex2ratio(colorcode[5:7]) # write the preamble - insert_to_preamble(0, document, - ['% Commands inserted by lyx2lyx to set the font color', + insert_to_preamble(document, + ['% Set the font color', '\\@ifundefined{definecolor}{\\usepackage{color}}{}', '\\definecolor{document_fontcolor}{rgb}{%s,%s,%s}' % (red, green, blue), '\\color{document_fontcolor}']) @@ -1232,8 +1315,8 @@ def revert_shadedboxcolor(document): green = hex2ratio(colorcode[3:5]) blue = hex2ratio(colorcode[5:7]) # write the preamble - insert_to_preamble(0, document, - ['% Commands inserted by lyx2lyx to set the color of boxes with shaded background', + insert_to_preamble(document, + ['% Set the color of boxes with shaded background', '\\@ifundefined{definecolor}{\\usepackage{color}}{}', "\\definecolor{shadecolor}{rgb}{%s,%s,%s}" % (red, green, blue)]) @@ -1248,7 +1331,7 @@ def revert_lyx_version(document): pass i = 0 - while 1: + while True: i = find_token(document.body, '\\begin_inset Info', i) if i == -1: return @@ -1263,13 +1346,8 @@ def revert_lyx_version(document): # type "lyxinfo" # arg "version" # \end_inset - # but we shall try to be forgiving. - arg = typ = "" - for k in range(i, j): - if document.body[k].startswith("arg"): - arg = document.body[k][3:].strip().strip('"') - if document.body[k].startswith("type"): - typ = document.body[k][4:].strip().strip('"') + typ = get_quoted_value(document.body, "type", i, j) + arg = get_quoted_value(document.body, "arg", i, j) if arg != "version" or typ != "lyxinfo": i = j + 1 continue @@ -1288,15 +1366,9 @@ def revert_lyx_version(document): def revert_math_scale(document): " Remove math scaling and LaTeX options " - i = find_token(document.header, '\\html_math_img_scale', 0) - if i != -1: - del document.header[i] - i = find_token(document.header, '\\html_latex_start', 0) - if i != -1: - del document.header[i] - i = find_token(document.header, '\\html_latex_end', 0) - if i != -1: - del document.header[i] + del_token(document.header, '\\html_math_img_scale', 0) + del_token(document.header, '\\html_latex_start', 0) + del_token(document.header, '\\html_latex_end', 0) def revert_pagesizes(document): @@ -1333,7 +1405,7 @@ def convert_html_quotes(document): m = l.match(line) if m: document.header[i] = "\\html_latex_start " + m.group(1) - + i = find_token(document.header, '\\html_latex_end', 0) if i != -1: line = document.header[i] @@ -1341,11 +1413,11 @@ def convert_html_quotes(document): m = l.match(line) if m: document.header[i] = "\\html_latex_end " + m.group(1) - + def revert_html_quotes(document): " Remove quotes around html_latex_start and html_latex_end " - + i = find_token(document.header, '\\html_latex_start', 0) if i != -1: line = document.header[i] @@ -1356,7 +1428,7 @@ def revert_html_quotes(document): del document.header[i] else: document.header[i] = "\\html_latex_start \"" + m.group(1) + "\"" - + i = find_token(document.header, '\\html_latex_end', 0) if i != -1: line = document.header[i] @@ -1371,29 +1443,44 @@ def revert_html_quotes(document): def revert_output_sync(document): " Remove forward search options " - i = find_token(document.header, '\\output_sync_macro', 0) - if i != -1: - del document.header[i] - i = find_token(document.header, '\\output_sync', 0) - if i != -1: - del document.header[i] + del_token(document.header, '\\output_sync_macro', 0) + del_token(document.header, '\\output_sync', 0) -# FIXME This doesn't do anything!! def revert_align_decimal(document): - l = 0 + i = 0 while True: - l = document.body[l].find('alignment=decimal') - if l == -1: - break - remove_option(document, l, 'decimal_point') - document.body[l].replace('decimal', 'center') + i = find_token(document.body, "\\begin_inset Tabular", i) + if i == -1: + return + j = find_end_of_inset(document.body, i) + if j == -1: + document.warning("Unable to find end of Tabular inset at line " + str(i)) + i += 1 + continue + cell = find_token(document.body, " cmdloc: - continue - endins = find_end_of_inset(document.body, stins) - if endins == -1: - document.warning("Can't find end of inset at line " + stins + "!!") - continue - if endins < cmdloc: - continue - refline = find_token(document.body, "reference", stins) - if refline == -1 or refline > endins: + # A normal line could begin with "LatexCommand nameref"! + val = is_in_inset(document.body, cmdloc, \ + "\\begin_inset CommandInset ref") + if not val: + continue + stins, endins = val + + # ok, so it is in an InsetRef + refline = find_token(document.body, "reference", stins, endins) + if refline == -1: document.warning("Can't find reference for inset at line " + stinst + "!!") continue m = rx.match(document.body[refline]) @@ -1609,18 +1694,17 @@ def revert_nameref(document): continue foundone = True ref = m.group(1) - newcontent = ['\\begin_inset ERT', 'status collapsed', '', \ - '\\begin_layout Plain Layout', '', '\\backslash', \ - cmd + '{' + ref + '}', '\\end_layout', '', '\\end_inset'] + newcontent = put_cmd_in_ert('\\' + cmd + '{' + ref + '}') document.body[stins:endins + 1] = newcontent + if foundone: - add_to_preamble(document, "\usepackage{nameref}") + add_to_preamble(document, ["\\usepackage{nameref}"]) def remove_Nameref(document): " Convert Nameref commands to nameref commands " i = 0 - while 1: + while True: # It seems better to look for this, as most of the reference # insets won't be ones we care about. i = find_token(document.body, "LatexCommand Nameref" , i) @@ -1628,22 +1712,11 @@ def remove_Nameref(document): break cmdloc = i i += 1 - + # Make sure it is actually in an inset! - # We could just check document.lines[i-1], but that relies - # upon something that might easily change. - # We'll look back a few lines. - stins = cmdloc - 10 - if stins < 0: - stins = 0 - stins = find_token(document.body, "\\begin_inset CommandInset ref", stins) - if stins == -1 or stins > cmdloc: - continue - endins = find_end_of_inset(document.body, stins) - if endins == -1: - document.warning("Can't find end of inset at line " + stins + "!!") - continue - if endins < cmdloc: + val = is_in_inset(document.body, cmdloc, \ + "\\begin_inset CommandInset ref") + if not val: continue document.body[cmdloc] = "LatexCommand nameref" @@ -1651,21 +1724,15 @@ def remove_Nameref(document): def revert_mathrsfs(document): " Load mathrsfs if \mathrsfs us use in the document " i = 0 - end = len(document.body) - 1 - while True: - j = document.body[i].find("\\mathscr{") - if j != -1: - add_to_preamble(document, ["% this command was inserted by lyx2lyx"]) + for line in document.body: + if line.find("\\mathscr{") != -1: add_to_preamble(document, ["\\usepackage{mathrsfs}"]) - break - if i == end: - break - i += 1 + return def convert_flexnames(document): "Convert \\begin_inset Flex Custom:Style to \\begin_inset Flex Style and similarly for CharStyle and Element." - + i = 0 rx = re.compile(r'^\\begin_inset Flex (?:Custom|CharStyle|Element):(.+)$') while True: @@ -1678,66 +1745,66 @@ def convert_flexnames(document): i += 1 -flex_insets = [ - ["Alert", "CharStyle:Alert"], - ["Code", "CharStyle:Code"], - ["Concepts", "CharStyle:Concepts"], - ["E-Mail", "CharStyle:E-Mail"], - ["Emph", "CharStyle:Emph"], - ["Expression", "CharStyle:Expression"], - ["Initial", "CharStyle:Initial"], - ["Institute", "CharStyle:Institute"], - ["Meaning", "CharStyle:Meaning"], - ["Noun", "CharStyle:Noun"], - ["Strong", "CharStyle:Strong"], - ["Structure", "CharStyle:Structure"], - ["ArticleMode", "Custom:ArticleMode"], - ["Endnote", "Custom:Endnote"], - ["Glosse", "Custom:Glosse"], - ["PresentationMode", "Custom:PresentationMode"], - ["Tri-Glosse", "Custom:Tri-Glosse"] -] - -flex_elements = [ - ["Abbrev", "Element:Abbrev"], - ["CCC-Code", "Element:CCC-Code"], - ["Citation-number", "Element:Citation-number"], - ["City", "Element:City"], - ["Code", "Element:Code"], - ["CODEN", "Element:CODEN"], - ["Country", "Element:Country"], - ["Day", "Element:Day"], - ["Directory", "Element:Directory"], - ["Dscr", "Element:Dscr"], - ["Email", "Element:Email"], - ["Emph", "Element:Emph"], - ["Filename", "Element:Filename"], - ["Firstname", "Element:Firstname"], - ["Fname", "Element:Fname"], - ["GuiButton", "Element:GuiButton"], - ["GuiMenu", "Element:GuiMenu"], - ["GuiMenuItem", "Element:GuiMenuItem"], - ["ISSN", "Element:ISSN"], - ["Issue-day", "Element:Issue-day"], - ["Issue-months", "Element:Issue-months"], - ["Issue-number", "Element:Issue-number"], - ["KeyCap", "Element:KeyCap"], - ["KeyCombo", "Element:KeyCombo"], - ["Keyword", "Element:Keyword"], - ["Literal", "Element:Literal"], - ["MenuChoice", "Element:MenuChoice"], - ["Month", "Element:Month"], - ["Orgdiv", "Element:Orgdiv"], - ["Orgname", "Element:Orgname"], - ["Postcode", "Element:Postcode"], - ["SS-Code", "Element:SS-Code"], - ["SS-Title", "Element:SS-Title"], - ["State", "Element:State"], - ["Street", "Element:Street"], - ["Surname", "Element:Surname"], - ["Volume", "Element:Volume"], - ["Year", "Element:Year"] -] +flex_insets = { + "Alert" : "CharStyle:Alert", + "Code" : "CharStyle:Code", + "Concepts" : "CharStyle:Concepts", + "E-Mail" : "CharStyle:E-Mail", + "Emph" : "CharStyle:Emph", + "Expression" : "CharStyle:Expression", + "Initial" : "CharStyle:Initial", + "Institute" : "CharStyle:Institute", + "Meaning" : "CharStyle:Meaning", + "Noun" : "CharStyle:Noun", + "Strong" : "CharStyle:Strong", + "Structure" : "CharStyle:Structure", + "ArticleMode" : "Custom:ArticleMode", + "Endnote" : "Custom:Endnote", + "Glosse" : "Custom:Glosse", + "PresentationMode" : "Custom:PresentationMode", + "Tri-Glosse" : "Custom:Tri-Glosse" +} + +flex_elements = { + "Abbrev" : "Element:Abbrev", + "CCC-Code" : "Element:CCC-Code", + "Citation-number" : "Element:Citation-number", + "City" : "Element:City", + "Code" : "Element:Code", + "CODEN" : "Element:CODEN", + "Country" : "Element:Country", + "Day" : "Element:Day", + "Directory" : "Element:Directory", + "Dscr" : "Element:Dscr", + "Email" : "Element:Email", + "Emph" : "Element:Emph", + "Filename" : "Element:Filename", + "Firstname" : "Element:Firstname", + "Fname" : "Element:Fname", + "GuiButton" : "Element:GuiButton", + "GuiMenu" : "Element:GuiMenu", + "GuiMenuItem" : "Element:GuiMenuItem", + "ISSN" : "Element:ISSN", + "Issue-day" : "Element:Issue-day", + "Issue-months" : "Element:Issue-months", + "Issue-number" : "Element:Issue-number", + "KeyCap" : "Element:KeyCap", + "KeyCombo" : "Element:KeyCombo", + "Keyword" : "Element:Keyword", + "Literal" : "Element:Literal", + "MenuChoice" : "Element:MenuChoice", + "Month" : "Element:Month", + "Orgdiv" : "Element:Orgdiv", + "Orgname" : "Element:Orgname", + "Postcode" : "Element:Postcode", + "SS-Code" : "Element:SS-Code", + "SS-Title" : "Element:SS-Title", + "State" : "Element:State", + "Street" : "Element:Street", + "Surname" : "Element:Surname", + "Volume" : "Element:Volume", + "Year" : "Element:Year" +} def revert_flexnames(document): @@ -1745,7 +1812,7 @@ def revert_flexnames(document): flexlist = flex_insets else: flexlist = flex_elements - + rx = re.compile(r'^\\begin_inset Flex\s+(.+)$') i = 0 while True: @@ -1757,146 +1824,156 @@ def revert_flexnames(document): document.warning("Illegal flex inset: " + document.body[i]) i += 1 continue - style = m.group(1) - for f in flexlist: - if f[0] == style: - document.body[i] = "\\begin_inset Flex " + f[1] - break - + if style in flexlist: + document.body[i] = "\\begin_inset Flex " + flexlist[style] i += 1 def convert_mathdots(document): " Load mathdots automatically " - while True: - i = find_token(document.header, "\\use_esint" , 0) - if i != -1: - document.header.insert(i + 1, "\\use_mathdots 1") - break + i = find_token(document.header, "\\use_mhchem" , 0) + if i == -1: + i = find_token(document.header, "\\use_esint" , 0) + if i == -1: + document.warning("Malformed LyX document: Can't find \\use_mhchem.") + return; + j = find_token(document.preamble, "\\usepackage{mathdots}", 0) + if j == -1: + document.header.insert(i + 1, "\\use_mathdots 0") + else: + document.header.insert(i + 1, "\\use_mathdots 2") + del document.preamble[j] def revert_mathdots(document): " Load mathdots if used in the document " - i = 0 - ddots = re.compile(r'\\begin_inset Formula .*\\ddots', re.DOTALL) - vdots = re.compile(r'\\begin_inset Formula .*\\vdots', re.DOTALL) - iddots = re.compile(r'\\begin_inset Formula .*\\iddots', re.DOTALL) + mathdots = find_token(document.header, "\\use_mathdots" , 0) - no = find_token(document.header, "\\use_mathdots 0" , 0) - auto = find_token(document.header, "\\use_mathdots 1" , 0) - yes = find_token(document.header, "\\use_mathdots 2" , 0) - if mathdots != -1: + if mathdots == -1: + document.warning("No \\use_mathdots line. Assuming auto.") + else: + val = get_value(document.header, "\\use_mathdots", mathdots) del document.header[mathdots] + try: + usedots = int(val) + except: + document.warning("Invalid \\use_mathdots value: " + val + ". Assuming auto.") + # probably usedots has not been changed, but be safe. + usedots = 1 + + if usedots == 0: + # do not load case + return + if usedots == 2: + # force load case + add_to_preamble(document, ["\\usepackage{mathdots}"]) + return + + # so we are in the auto case. we want to load mathdots if \iddots is used. + i = 0 while True: i = find_token(document.body, '\\begin_inset Formula', i) if i == -1: return j = find_end_of_inset(document.body, i) if j == -1: - document.warning("Malformed LyX document: Can't find end of Formula inset.") - return - k = ddots.search("\n".join(document.body[i:j])) - l = vdots.search("\n".join(document.body[i:j])) - m = iddots.search("\n".join(document.body[i:j])) - if (yes == -1) and ((no != -1) or (not k and not l and not m) or (auto != -1 and not m)): + document.warning("Malformed LyX document: Can't find end of Formula inset at line " + str(i)) i += 1 continue - # use \@ifundefined to catch also the "auto" case - add_to_preamble(document, ["% this command was inserted by lyx2lyx"]) - add_to_preamble(document, ["\\@ifundefined{iddots}{\\usepackage{mathdots}}\n"]) - return + code = "\n".join(document.body[i:j]) + if code.find("\\iddots") != -1: + add_to_preamble(document, ["\\@ifundefined{iddots}{\\usepackage{mathdots}}"]) + return + i = j def convert_rule(document): - " Convert \\lyxline to CommandInset line " + " Convert \\lyxline to CommandInset line. " i = 0 + + inset = ['\\begin_inset CommandInset line', + 'LatexCommand rule', + 'offset "0.5ex"', + 'width "100line%"', + 'height "1pt"', '', + '\\end_inset', '', ''] + + # if paragraphs are indented, we may have to unindent to get the + # line to be full-width. + indent = get_value(document.header, "\\paragraph_separation", 0) + have_indent = (indent == "indent") + while True: i = find_token(document.body, "\\lyxline" , i) if i == -1: return - - j = find_token(document.body, "\\color" , i - 2) - if j == i - 2: - color = document.body[j] + '\n' - else: - color = '' - k = find_token(document.body, "\\begin_layout Standard" , i - 4) - # we need to handle the case that \lyxline is in a separate paragraph and that it is colored - # the result is then an extra empty paragraph which we get by adding an empty ERT inset - if k == i - 4 and j == i - 2 and document.body[i - 1] == '': - layout = '\\begin_inset ERT\nstatus collapsed\n\n\\begin_layout Plain Layout\n\n\n\\end_layout\n\n\\end_inset\n' \ - + '\\end_layout\n\n' \ - + '\\begin_layout Standard\n' - elif k == i - 2 and document.body[i - 1] == '': - layout = '' - else: - layout = '\\end_layout\n\n' \ - + '\\begin_layout Standard\n' - l = find_token(document.body, "\\begin_layout Standard" , i + 4) - if l == i + 4 and document.body[i + 1] == '': - layout2 = '' + + # we need to find out if this line follows other content + # in its paragraph. find its layout.... + lastlay = find_token_backwards(document.body, "\\begin_layout", i) + if lastlay == -1: + document.warning("Can't find layout for line at " + str(i)) + # do the best we can. + document.body[i:i+1] = inset + i += len(inset) + continue + + # ...and look for other content before it. + lineisfirst = True + for line in document.body[lastlay + 1:i]: + # is it empty or a paragraph option? + if not line or line[0] == '\\': + continue + lineisfirst = False + break + + if lineisfirst: + document.body[i:i+1] = inset + if indent: + # we need to unindent, lest the line be too long + document.body.insert(lastlay + 1, "\\noindent") + i += len(inset) else: - layout2 = '\\end_layout\n' \ - + '\n\\begin_layout Standard\n' - subst = layout \ - + '\\noindent\n\n' \ - + color \ - + '\\begin_inset CommandInset line\n' \ - + 'LatexCommand rule\n' \ - + 'offset "0.5ex"\n' \ - + 'width "100line%"\n' \ - + 'height "1pt"\n' \ - + '\n\\end_inset\n\n\n' \ - + layout2 - document.body[i] = subst - i += 1 + # so our line is in the middle of a paragraph + # we need to add a new line, lest this line follow the + # other content on that line and run off the side of the page + document.body[i:i+1] = inset + document.body[i:i] = ["\\begin_inset Newline newline", "\\end_inset", ""] + i += len(inset) def revert_rule(document): " Revert line insets to Tex code " i = 0 - while 1: + while True: i = find_token(document.body, "\\begin_inset CommandInset line" , i) if i == -1: return # find end of inset j = find_token(document.body, "\\end_inset" , i) - # assure we found the end_inset of the current inset - if j > i + 6 or j == -1: + if j == -1: document.warning("Malformed LyX document: Can't find end of line inset.") return # determine the optional offset - k = find_token(document.body, 'offset', i, j) - if k != -1: - offset = document.body[k][8:-1] - else: - offset = "" + offset = get_quoted_value(document.body, 'offset', i, j) + if offset: + offset = '[' + offset + ']' # determine the width - l = find_token(document.body, 'width', i, j) - if l != -1: - width = document.body[l][7:-1] - else: - width = "100col%" + width = get_quoted_value(document.body, 'width', i, j, "100col%") + width = latex_length(width)[1] # determine the height - m = find_token(document.body, 'height', i, j) - if m != -1: - height = document.body[m][8:-1] - else: - height = "1pt" + height = get_quoted_value(document.body, 'height', i, j, "1pt") + height = latex_length(height)[1] # output the \rule command - if offset: - subst = "\\rule[" + offset + "]{" + width + "}{" + height + "}" - else: - subst = "\\rule{" + width + "}{" + height + "}" + subst = "\\rule[" + offset + "]{" + width + "}{" + height + "}" document.body[i:j + 1] = put_cmd_in_ert(subst) - i += 1 + i += len(subst) - (j - i) def revert_diagram(document): " Add the feyn package if \\Diagram is used in math " i = 0 - re_diagram = re.compile(r'\\begin_inset Formula .*\\Diagram', re.DOTALL) while True: i = find_token(document.body, '\\begin_inset Formula', i) if i == -1: @@ -1904,25 +1981,42 @@ def revert_diagram(document): j = find_end_of_inset(document.body, i) if j == -1: document.warning("Malformed LyX document: Can't find end of Formula inset.") - return - m = re_diagram.search("\n".join(document.body[i:j])) - if not m: - i += 1 + return + lines = "\n".join(document.body[i:j]) + if lines.find("\\Diagram") == -1: + i = j continue - add_to_preamble(document, ["% this command was inserted by lyx2lyx"]) - add_to_preamble(document, "\\usepackage{feyn}") + add_to_preamble(document, ["\\usepackage{feyn}"]) # only need to do it once! return +chapters = ("amsbook", "book", "docbook-book", "elsart", "extbook", "extreport", + "jbook", "jreport", "jsbook", "literate-book", "literate-report", "memoir", + "mwbk", "mwrep", "recipebook", "report", "scrbook", "scrreprt", "svmono", + "svmult", "tbook", "treport", "tufte-book") def convert_bibtex_clearpage(document): " insert a clear(double)page bibliographystyle if bibtotoc option is used " + if document.textclass not in chapters: + return + i = find_token(document.header, '\\papersides', 0) + sides = 0 if i == -1: document.warning("Malformed LyX document: Can't find papersides definition.") - return - sides = int(document.header[i][12]) + document.warning("Assuming single sided.") + sides = 1 + else: + val = get_value(document.header, "\\papersides", i) + try: + sides = int(val) + except: + pass + if sides != 1 and sides != 2: + document.warning("Invalid papersides value: " + val) + document.warning("Assuming single sided.") + sides = 1 j = 0 while True: @@ -1937,47 +2031,440 @@ def convert_bibtex_clearpage(document): continue # only act if there is the option "bibtotoc" - m = find_token(document.body, 'options', j, k) - if m == -1: + val = get_value(document.body, 'options', j, k) + if not val: document.warning("Can't find options for bibliography inset at line " + str(j)) j = k continue - - optline = document.body[m] - idx = optline.find("bibtotoc") - if idx == -1: + + if val.find("bibtotoc") == -1: j = k continue - - # so we want to insert a new page right before the paragraph that - # this bibliography thing is in. we'll look for it backwards. - lay = j - 1 - while lay >= 0: - if document.body[lay].startswith("\\begin_layout"): - break - lay -= 1 - if lay < 0: + # so we want to insert a new page right before the paragraph that + # this bibliography thing is in. + lay = find_token_backwards(document.body, "\\begin_layout", j) + if lay == -1: document.warning("Can't find layout containing bibliography inset at line " + str(j)) j = k continue - subst1 = '\\begin_layout Standard\n' \ - + '\\begin_inset Newpage clearpage\n' \ - + '\\end_inset\n\n\n' \ - + '\\end_layout\n' - subst2 = '\\begin_layout Standard\n' \ - + '\\begin_inset Newpage cleardoublepage\n' \ - + '\\end_inset\n\n\n' \ - + '\\end_layout\n' if sides == 1: - document.body.insert(lay, subst1) - document.warning(subst1) + cmd = "clearpage" + else: + cmd = "cleardoublepage" + subst = ['\\begin_layout Standard', + '\\begin_inset Newpage ' + cmd, + '\\end_inset', '', '', + '\\end_layout', ''] + document.body[lay:lay] = subst + j = k + len(subst) + + +def check_passthru(document): + tc = document.textclass + ok = (tc == "literate-article" or tc == "literate-book" or tc == "literate-report") + if not ok: + mods = document.get_module_list() + for mod in mods: + if mod == "sweave" or mod == "noweb": + ok = True + break + return ok + + +def convert_passthru(document): + " http://www.mail-archive.com/lyx-devel@lists.lyx.org/msg161298.html " + if not check_passthru: + return + + rx = re.compile("\\\\begin_layout \s*(\w+)") + beg = 0 + for lay in ["Chunk", "Scrap"]: + while True: + beg = find_token(document.body, "\\begin_layout " + lay, beg) + if beg == -1: + break + end = find_end_of_layout(document.body, beg) + if end == -1: + document.warning("Can't find end of layout at line " + str(beg)) + beg += 1 + continue + + # we are now going to replace newline insets within this layout + # by new instances of this layout. so we have repeated layouts + # instead of newlines. + + # if the paragraph has any customization, however, we do not want to + # do the replacement. + if document.body[beg + 1].startswith("\\"): + beg = end + 1 + continue + + ns = beg + while True: + ns = find_token(document.body, "\\begin_inset Newline newline", ns, end) + if ns == -1: + break + ne = find_end_of_inset(document.body, ns) + if ne == -1 or ne > end: + document.warning("Can't find end of inset at line " + str(nb)) + ns += 1 + continue + if document.body[ne + 1] == "": + ne += 1 + subst = ["\\end_layout", "", "\\begin_layout " + lay] + document.body[ns:ne + 1] = subst + # now we need to adjust end, in particular, but might as well + # do ns properly, too + newlines = (ne - ns) - len(subst) + ns += newlines + 2 + end += newlines + 2 + + # ok, we now want to find out if the next layout is the + # same as this one. if so, we will insert an extra copy of it + didit = False + next = find_token(document.body, "\\begin_layout", end) + if next != -1: + m = rx.match(document.body[next]) + if m: + nextlay = m.group(1) + if nextlay == lay: + subst = ["\\begin_layout " + lay, "", "\\end_layout", ""] + document.body[next:next] = subst + didit = True + beg = end + 1 + if didit: + beg += 4 # for the extra layout + + +def revert_passthru(document): + " http://www.mail-archive.com/lyx-devel@lists.lyx.org/msg161298.html " + if not check_passthru: + return + rx = re.compile("\\\\begin_layout \s*(\w+)") + beg = 0 + for lay in ["Chunk", "Scrap"]: + while True: + beg = find_token(document.body, "\\begin_layout " + lay, beg) + if beg == -1: + break + end = find_end_of_layout(document.body, beg) + if end == -1: + document.warning("Can't find end of layout at line " + str(beg)) + beg += 1 + continue + + # we now want to find out if the next layout is the + # same as this one. but we will need to do this over and + # over again. + while True: + next = find_token(document.body, "\\begin_layout", end) + if next == -1: + break + m = rx.match(document.body[next]) + if not m: + break + nextlay = m.group(1) + if nextlay != lay: + break + # so it is the same layout again. we now want to know if it is empty. + # but first let's check and make sure there is no content between the + # two layouts. i'm not sure if that can happen or not. + for l in range(end + 1, next): + if document.body[l] != "": + document.warning("Found content between adjacent " + lay + " layouts!") + break + nextend = find_end_of_layout(document.body, next) + if nextend == -1: + document.warning("Can't find end of layout at line " + str(next)) + break + empty = True + for l in range(next + 1, nextend): + if document.body[l] != "": + empty = False + break + if empty: + # empty layouts just get removed + # should we check if it's before yet another such layout? + del document.body[next : nextend + 1] + # and we do not want to check again. we know the next layout + # should be another Chunk and should be left as is. + break + else: + # if it's not empty, then we want to insert a newline in place + # of the layout switch + subst = ["\\begin_inset Newline newline", "\\end_inset", ""] + document.body[end : next + 1] = subst + # and now we have to find the end of the new, larger layout + newend = find_end_of_layout(document.body, beg) + if newend == -1: + document.warning("Can't find end of new layout at line " + str(beg)) + break + end = newend + beg = end + 1 + + +def revert_multirowOffset(document): + " Revert multirow cells with offset in tables to TeX-code" + # this routine is the same as the revert_multirow routine except that + # it checks additionally for the offset + + # first, let's find out if we need to do anything + i = find_token(document.body, '') + if end_table == -1: + document.warning("Malformed LyX document: Could not find end of table.") + begin_table += 1 + continue + # does this table have multirow? + i = find_token(document.body, '', begin_row, end_table) + if begin_row == -1: + document.warning("Can't find row " + str(row + 1)) + break + end_row = find_end_of(document.body, begin_row, '', '') + if end_row == -1: + document.warning("Can't find end of row " + str(row + 1)) + break + begin_cell = begin_row + multirows.append([]) + for column in range(numcols): + begin_cell = find_token(document.body, '" in a cell, but + # that seems fairly unlikely. + end_cell = find_end_of(document.body, begin_cell, '') + if end_cell == -1: + document.warning("Can't find end of column " + str(column + 1) + \ + "in row " + str(row + 1)) + break + multirows[row].append([begin_cell, end_cell, 0]) + if document.body[begin_cell].find('multirow="3" mroffset=') != -1: + multirows[row][column][2] = 3 # begin multirow + mrstarts.append([row, column]) + elif document.body[begin_cell].find('multirow="4"') != -1: + multirows[row][column][2] = 4 # in multirow + begin_cell = end_cell + begin_row = end_row + # end of table info collection + + # work from the back to avoid messing up numbering + mrstarts.reverse() + for m in mrstarts: + row = m[0] + col = m[1] + # get column width + col_width = get_option_value(document.body[begin_table + 2 + col], "width") + # "0pt" means that no width is specified + if not col_width or col_width == "0pt": + col_width = "*" + # determine the number of cells that are part of the multirow + nummrs = 1 + for r in range(row + 1, numrows): + if multirows[r][col][2] != 4: + break + nummrs += 1 + # take the opportunity to revert this line + lineno = multirows[r][col][0] + document.body[lineno] = document.body[lineno].\ + replace(' multirow="4" ', ' ').\ + replace('valignment="middle"', 'valignment="top"').\ + replace(' topline="true" ', ' ') + # remove bottom line of previous multirow-part cell + lineno = multirows[r-1][col][0] + document.body[lineno] = document.body[lineno].replace(' bottomline="true" ', ' ') + # revert beginning cell + bcell = multirows[row][col][0] + ecell = multirows[row][col][1] + offset = get_option_value(document.body[bcell], "mroffset") + document.body[bcell] = document.body[bcell].\ + replace(' multirow="3" ', ' ').\ + replace('valignment="middle"', 'valignment="top"') + # remove mroffset option + document.body[bcell] = rgx.sub('', document.body[bcell]) + + blay = find_token(document.body, "\\begin_layout", bcell, ecell) + if blay == -1: + document.warning("Can't find layout for cell!") + continue + bend = find_end_of_layout(document.body, blay) + if bend == -1: + document.warning("Can't find end of layout for cell!") + continue + # do the later one first, so as not to mess up the numbering + # we are wrapping the whole cell in this ert + # so before the end of the layout... + document.body[bend:bend] = put_cmd_in_ert("}") + # ...and after the beginning + document.body[blay + 1:blay + 1] = \ + put_cmd_in_ert("\\multirow{" + str(nummrs) + "}{" + col_width + "}[" \ + + offset + "]{") + + # on to the next table + begin_table = end_table + + +def revert_script(document): + " Convert subscript/superscript inset to TeX code " + i = 0 + foundsubscript = False + while True: + i = find_token(document.body, '\\begin_inset script', i) + if i == -1: + break + z = find_end_of_inset(document.body, i) + if z == -1: + document.warning("Malformed LyX document: Can't find end of script inset.") + i += 1 + continue + blay = find_token(document.body, "\\begin_layout", i, z) + if blay == -1: + document.warning("Malformed LyX document: Can't find layout in script inset.") + i = z + continue + + if check_token(document.body[i], "\\begin_inset script subscript"): + subst = '\\textsubscript{' + foundsubscript = True + elif check_token(document.body[i], "\\begin_inset script superscript"): + subst = '\\textsuperscript{' + else: + document.warning("Malformed LyX document: Unknown type of script inset.") + i = z + continue + bend = find_end_of_layout(document.body, blay) + if bend == -1 or bend > z: + document.warning("Malformed LyX document: Can't find end of layout in script inset.") + i = z + continue + # remove the \end_layout \end_inset pair + document.body[bend:z + 1] = put_cmd_in_ert("}") + document.body[i:blay + 1] = put_cmd_in_ert(subst) + i += 1 + # these classes provide a \textsubscript command: + # FIXME: Would be nice if we could use the information of the .layout file here + classes = ["memoir", "scrartcl", "scrbook", "scrlttr2", "scrreprt"] + if foundsubscript and find_token_exact(classes, document.textclass, 0) == -1: + add_to_preamble(document, ['\\usepackage{subscript}']) + + +def convert_use_xetex(document): + " convert \\use_xetex to \\use_non_tex_fonts " + i = find_token(document.header, "\\use_xetex", 0) + if i == -1: + document.header.insert(-1, "\\use_non_tex_fonts 0") else: - document.body.insert(lay, subst2) - document.warning(subst2) + val = get_value(document.header, "\\use_xetex", 0) + document.header[i] = "\\use_non_tex_fonts " + val + + +def revert_use_xetex(document): + " revert \\use_non_tex_fonts to \\use_xetex " + i = 0 + i = find_token(document.header, "\\use_non_tex_fonts", 0) + if i == -1: + document.warning("Malformed document. No \\use_non_tex_fonts param!") + return + + val = get_value(document.header, "\\use_non_tex_fonts", 0) + document.header[i] = "\\use_xetex " + val + + +def revert_labeling(document): + koma = ("scrartcl", "scrarticle-beamer", "scrbook", "scrlettr", + "scrlttr2", "scrreprt") + if document.textclass in koma: + return + i = 0 + while True: + i = find_token_exact(document.body, "\\begin_layout Labeling", i) + if i == -1: + return + document.body[i] = "\\begin_layout List" + + +def revert_langpack(document): + " revert \\language_package parameter " + i = 0 + i = find_token(document.header, "\\language_package", 0) + if i == -1: + document.warning("Malformed document. No \\language_package param!") + return + + del document.header[i] + + +def convert_langpack(document): + " Add \\language_package parameter " + i = find_token(document.header, "\language" , 0) + if i == -1: + document.warning("Malformed document. No \\language defined!") + return + + document.header.insert(i + 1, "\\language_package default") + + +def revert_tabularwidth(document): + i = 0 + while True: + i = find_token(document.body, "\\begin_inset Tabular", i) + if i == -1: + return + j = find_end_of_inset(document.body, i) + if j == -1: + document.warning("Unable to find end of Tabular inset at line " + str(i)) + i += 1 + continue + i += 1 + features = find_token(document.body, "