lyx_mirror/lib/lyx2lyx/lyx_1_6.py
Uwe Stöhr a5c7679887 lyx_1_6.py: fix for previous commit
git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@21360 a592a061-630c-0410-9148-cb99ea01b6c8
2007-11-01 22:40:31 +00:00

816 lines
26 KiB
Python

# This file is part of lyx2lyx
# -*- coding: utf-8 -*-
# Copyright (C) 2007 José Matos <jamatos@lyx.org>
#
# 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 1.6"""
import re
import unicodedata
import sys, os
from parser_tools import find_token, find_end_of, find_tokens, get_value
####################################################################
# Private helper functions
def find_end_of_inset(lines, i):
" Find end of inset, where lines[i] is included."
return find_end_of(lines, i, "\\begin_inset", "\\end_inset")
####################################################################
def fix_wrong_tables(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 + 1)
if j == -1:
document.warning("Malformed LyX document: Could not find end of tabular.")
continue
m = i + 1
nrows = int(document.body[i+1].split('"')[3])
ncols = int(document.body[i+1].split('"')[5])
for l in range(nrows):
prev_multicolumn = 0
for k in range(ncols):
m = find_token(document.body, '<cell', m)
if document.body[m].find('multicolumn') != -1:
multicol_cont = int(document.body[m].split('"')[1])
if multicol_cont == 2 and (k == 0 or prev_multicolumn == 0):
document.body[m] = document.body[m][:5] + document.body[m][21:]
prev_multicolumn = 0
else:
prev_multicolumn = multicol_cont
else:
prev_multicolumn = 0
i = j + 1
def close_begin_deeper(document):
i = 0
depth = 0
while True:
i = find_tokens(document.body, ["\\begin_deeper", "\\end_deeper"], i)
if i == -1:
break
if document.body[i][:13] == "\\begin_deeper":
depth += 1
else:
depth -= 1
i += 1
document.body[-2:-2] = ['\\end_deeper' for i in range(depth)]
def long_charstyle_names(document):
i = 0
while True:
i = find_token(document.body, "\\begin_inset CharStyle", i)
if i == -1:
return
document.body[i] = document.body[i].replace("CharStyle ", "CharStyle CharStyle:")
i += 1
def revert_long_charstyle_names(document):
i = 0
while True:
i = find_token(document.body, "\\begin_inset CharStyle", i)
if i == -1:
return
document.body[i] = document.body[i].replace("CharStyle CharStyle:", "CharStyle")
i += 1
def axe_show_label(document):
i = 0
while True:
i = find_token(document.body, "\\begin_inset CharStyle", i)
if i == -1:
return
if document.body[i + 1].find("show_label") != -1:
if document.body[i + 1].find("true") != -1:
document.body[i + 1] = "status open"
del document.body[ i + 2]
else:
if document.body[i + 1].find("false") != -1:
document.body[i + 1] = "status collapsed"
del document.body[ i + 2]
else:
document.warning("Malformed LyX document: show_label neither false nor true.")
else:
document.warning("Malformed LyX document: show_label missing in CharStyle.")
i += 1
def revert_show_label(document):
i = 0
while True:
i = find_token(document.body, "\\begin_inset CharStyle", i)
if i == -1:
return
if document.body[i + 1].find("status open") != -1:
document.body.insert(i + 1, "show_label true")
else:
if document.body[i + 1].find("status collapsed") != -1:
document.body.insert(i + 1, "show_label false")
else:
document.warning("Malformed LyX document: no legal status line in CharStyle.")
i += 1
def revert_begin_modules(document):
i = 0
while True:
i = find_token(document.header, "\\begin_modules", i)
if i == -1:
return
j = find_end_of(document.header, i, "\\begin_modules", "\\end_modules")
if j == -1:
# this should not happen
break
document.header[i : j + 1] = []
def convert_flex(document):
"Convert CharStyle to Flex"
i = 0
while True:
i = find_token(document.body, "\\begin_inset CharStyle", i)
if i == -1:
return
document.body[i] = document.body[i].replace('\\begin_inset CharStyle', '\\begin_inset Flex')
def revert_flex(document):
"Convert Flex to CharStyle"
i = 0
while True:
i = find_token(document.body, "\\begin_inset Flex", i)
if i == -1:
return
document.body[i] = document.body[i].replace('\\begin_inset Flex', '\\begin_inset CharStyle')
# Discard PDF options for hyperref
def revert_pdf_options(document):
"Revert PDF options for hyperref."
i = 0
i = find_token(document.header, "\\use_hyperref", i)
if i != -1:
del document.header[i]
i = find_token(document.header, "\\pdf_store_options", i)
if i != -1:
del document.header[i]
i = find_token(document.header, "\\pdf_title", 0)
if i != -1:
del document.header[i]
i = find_token(document.header, "\\pdf_author", 0)
if i != -1:
del document.header[i]
i = find_token(document.header, "\\pdf_subject", 0)
if i != -1:
del document.header[i]
i = find_token(document.header, "\\pdf_keywords", 0)
if i != -1:
del document.header[i]
i = find_token(document.header, "\\pdf_bookmarks", 0)
if i != -1:
del document.header[i]
i = find_token(document.header, "\\pdf_bookmarksnumbered", i)
if i != -1:
del document.header[i]
i = find_token(document.header, "\\pdf_bookmarksopen", i)
if i != -1:
del document.header[i]
i = find_token(document.header, "\\pdf_bookmarksopenlevel", i)
if i != -1:
del document.header[i]
i = find_token(document.header, "\\pdf_breaklinks", i)
if i != -1:
del document.header[i]
i = find_token(document.header, "\\pdf_pdfborder", i)
if i != -1:
del document.header[i]
i = find_token(document.header, "\\pdf_colorlinks", i)
if i != -1:
del document.header[i]
i = find_token(document.header, "\\pdf_backref", i)
if i != -1:
del document.header[i]
i = find_token(document.header, "\\pdf_pagebackref", i)
if i != -1:
del document.header[i]
i = find_token(document.header, "\\pdf_pagemode", 0)
if i != -1:
del document.header[i]
i = find_token(document.header, "\\pdf_quoted_options", 0)
if i != -1:
del document.header[i]
def remove_inzip_options(document):
"Remove inzipName and embed options from the Graphics inset"
i = 0
while 1:
i = find_token(document.body, "\\begin_inset Graphics", i)
if i == -1:
return
j = find_end_of_inset(document.body, i + 1)
if j == -1:
# should not happen
document.warning("Malformed LyX document: Could not find end of graphics inset.")
# If there's a inzip param, just remove that
k = find_token(document.body, "\tinzipName", i + 1, j)
if k != -1:
del document.body[k]
# embed option must follow the inzipName option
del document.body[k+1]
i = i + 1
def convert_inset_command(document):
"""
Convert:
\begin_inset LatexCommand cmd
to
\begin_inset CommandInset InsetType
LatexCommand cmd
"""
i = 0
while 1:
i = find_token(document.body, "\\begin_inset LatexCommand", i)
if i == -1:
return
line = document.body[i]
r = re.compile(r'\\begin_inset LatexCommand (.*)$')
m = r.match(line)
cmdName = m.group(1)
insetName = ""
#this is adapted from factory.cpp
if cmdName[0:4].lower() == "cite":
insetName = "citation"
elif cmdName == "url" or cmdName == "htmlurl":
insetName = "url"
elif cmdName[-3:] == "ref":
insetName = "ref"
elif cmdName == "tableofcontents":
insetName = "toc"
elif cmdName == "printnomenclature":
insetName = "nomencl_print"
elif cmdName == "printindex":
insetName = "index_print"
else:
insetName = cmdName
insertion = ["\\begin_inset CommandInset " + insetName, "LatexCommand " + cmdName]
document.body[i : i+1] = insertion
def revert_inset_command(document):
"""
Convert:
\begin_inset CommandInset InsetType
LatexCommand cmd
to
\begin_inset LatexCommand cmd
Some insets may end up being converted to insets earlier versions of LyX
will not be able to recognize. Not sure what to do about that.
"""
i = 0
while 1:
i = find_token(document.body, "\\begin_inset CommandInset", i)
if i == -1:
return
nextline = document.body[i+1]
r = re.compile(r'LatexCommand\s+(.*)$')
m = r.match(nextline)
if not m:
document.warning("Malformed LyX document: Missing LatexCommand in " + document.body[i] + ".")
continue
cmdName = m.group(1)
insertion = ["\\begin_inset LatexCommand " + cmdName]
document.body[i : i+2] = insertion
def convert_wrapfig_options(document):
"Convert optional options for wrap floats (wrapfig)."
# adds the tokens "lines", "placement", and "overhang"
i = 0
while True:
i = find_token(document.body, "\\begin_inset Wrap figure", i)
if i == -1:
return
document.body.insert(i + 1, "lines 0")
j = find_token(document.body, "placement", i)
# placement can be already set or not; if not, set it
if j == i+2:
document.body.insert(i + 3, "overhang 0col%")
else:
document.body.insert(i + 2, "placement o")
document.body.insert(i + 3, "overhang 0col%")
i = i + 1
def revert_wrapfig_options(document):
"Revert optional options for wrap floats (wrapfig)."
i = 0
while True:
i = find_token(document.body, "lines", i)
if i == -1:
return
j = find_token(document.body, "overhang", i+1)
if j != i + 2 and j != -1:
document.warning("Malformed LyX document: Couldn't find overhang parameter of wrap float.")
if j == -1:
return
del document.body[i]
del document.body[j-1]
i = i + 1
def convert_latexcommand_index(document):
"Convert from LatexCommand form to collapsable form."
i = 0
while True:
i = find_token(document.body, "\\begin_inset CommandInset index", i)
if i == -1:
return
if document.body[i + 1] != "LatexCommand index": # Might also be index_print
return
fullcommand = document.body[i + 2]
document.body[i] = "\\begin_inset Index"
document.body[i + 1] = "status collapsed"
document.body[i + 2] = "\\begin_layout standard"
document.body.insert(i + 3, fullcommand[6:].strip('"'))
document.body.insert(i + 4, "\\end_layout")
i = i + 5
def revert_latexcommand_index(document):
"Revert from collapsable form toLatexCommand form."
i = 0
while True:
i = find_token(document.body, "\\begin_inset Index", i)
if i == -1:
return
j = find_end_of_inset(document.body, i)
del document.body[j - 1]
del document.body[j - 2] # \end_layout
document.body[i] = "\\begin_inset CommandInset index"
document.body[i + 1] = "LatexCommand index"
document.body[i + 3] = "name " + '"' + document.body[i + 3] + '"'
document.body.insert(i + 4, "")
del document.body[i + 2] # \begin_layout standard
i = i + 5
def revert_wraptable(document):
"Revert wrap table to wrap figure."
i = 0
while True:
i = find_token(document.body, "\\begin_inset Wrap table", i)
if i == -1:
return
document.body[i] = document.body[i].replace('\\begin_inset Wrap table', '\\begin_inset Wrap figure')
i = i + 1
def revert_vietnamese(document):
"Set language Vietnamese to English"
# Set document language from Vietnamese to English
i = 0
if document.language == "vietnamese":
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 vietnamese", j)
if j == -1:
return
document.body[j] = document.body[j].replace("\\lang vietnamese", "\\lang english")
j = j + 1
def revert_japanese(document):
"Set language japanese-plain to japanese"
# Set document language from japanese-plain to japanese
i = 0
if document.language == "japanese-plain":
document.language = "japanese"
i = find_token(document.header, "\\language", 0)
if i != -1:
document.header[i] = "\\language japanese"
j = 0
while True:
j = find_token(document.body, "\\lang japanese-plain", j)
if j == -1:
return
document.body[j] = document.body[j].replace("\\lang japanese-plain", "\\lang japanese")
j = j + 1
def revert_japanese_encoding(document):
"Set input encoding form EUC-JP-plain to EUC-JP etc."
# Set input encoding form EUC-JP-plain to EUC-JP etc.
i = 0
i = find_token(document.header, "\\inputencoding EUC-JP-plain", 0)
if i != -1:
document.header[i] = "\\inputencoding EUC-JP"
j = 0
j = find_token(document.header, "\\inputencoding JIS-plain", 0)
if j != -1:
document.header[j] = "\\inputencoding JIS"
k = 0
k = find_token(document.header, "\\inputencoding SJIS-plain", 0)
if k != -1: # convert to UTF8 since there is currently no SJIS encoding
document.header[k] = "\\inputencoding UTF8"
def revert_inset_info(document):
'Replace info inset with its content'
i = 0
while 1:
i = find_token(document.body, '\\begin_inset Info', i)
if i == -1:
return
j = find_end_of_inset(document.body, i + 1)
if j == -1:
# should not happen
document.warning("Malformed LyX document: Could not find end of Info inset.")
type = 'unknown'
arg = ''
for k in range(i, j+1):
if document.body[k].startswith("arg"):
arg = document.body[k][3:].strip().strip('"')
if document.body[k].startswith("type"):
type = document.body[k][4:].strip().strip('"')
# I think there is a newline after \\end_inset, which should be removed.
if document.body[j + 1].strip() == "":
document.body[i : (j + 2)] = [type + ':' + arg]
else:
document.body[i : (j + 1)] = [type + ':' + arg]
def convert_pdf_options(document):
# Set the pdfusetitle tag, delete the pdf_store_options,
# set quotes for bookmarksopenlevel"
has_hr = get_value(document.header, "\\use_hyperref", 0, default = "0")
if has_hr == "1":
k = find_token(document.header, "\\use_hyperref", 0)
document.header.insert(k + 1, "\\pdf_pdfusetitle true")
k = find_token(document.header, "\\pdf_store_options", 0)
if k != -1:
del document.header[k]
i = find_token(document.header, "\\pdf_bookmarksopenlevel", k)
if i == -1: return
document.header[i] = document.header[i].replace('"', '')
def revert_pdf_options_2(document):
# reset the pdfusetitle tag, set quotes for bookmarksopenlevel"
k = find_token(document.header, "\\use_hyperref", 0)
i = find_token(document.header, "\\pdf_pdfusetitle", k)
if i != -1:
del document.header[i]
i = find_token(document.header, "\\pdf_bookmarksopenlevel", k)
if i == -1: return
values = document.header[i].split()
values[1] = ' "' + values[1] + '"'
document.header[i] = ''.join(values)
def convert_htmlurl(document):
'Convert "htmlurl" to "href" insets for docbook'
if document.backend != "docbook":
return
i = 0
while True:
i = find_token(document.body, "\\begin_inset CommandInset url", i)
if i == -1:
return
document.body[i] = "\\begin_inset CommandInset href"
document.body[i + 1] = "LatexCommand href"
i = i + 1
def convert_url(document):
'Convert url insets to url charstyles'
if document.backend == "docbook":
return
didone = False
i = 0
while True:
i = find_token(document.body, "\\begin_inset CommandInset url", i)
if i == -1:
break
n = find_token(document.body, "name", i)
if n == i + 2:
# place the URL name in typewriter before the new URL insert
# grab the name 'bla' from the e.g. the line 'name "bla"',
# therefore start with the 6th character
name = document.body[n][6:-1]
newname = [name + " "]
document.body[i:i] = newname
i = i + 1
j = find_token(document.body, "target", i)
if j == -1:
document.warning("Malformed LyX document: Can't find target for url inset")
i = j
continue
target = document.body[j][8:-1]
k = find_token(document.body, "\\end_inset", j)
if k == -1:
document.warning("Malformed LyX document: Can't find end of url inset")
i = k
continue
newstuff = ["\\begin_inset Flex URL",
"status collapsed", "",
"\\begin_layout Standard",
"",
target,
"\\end_layout",
""]
document.body[i:k] = newstuff
didone = True
i = k
#If we did one, we need to add URL to the modules
if didone:
i = find_token(document.header, "\\begin_modules", 0)
if i == -1:
#No modules yet included
i = find_token(document.header, "\\textclass", 0)
if i == -1:
document.warning("Malformed LyX document: No \\textclass!!")
return
modinfo = ["\\begin_modules", "URL", "\\end_modules"]
document.header[i + 1: i + 1] = modinfo
return
j = find_token(document.header, "\\end_modules", i)
if j == -1:
document.warning("Malformed LyX document: No \\end_modules.")
return
k = find_token(document.header, "URL", i)
if k != -1 and k < j:
return
document.header.insert(i + 1, "URL")
def revert_href(document):
'Reverts hyperlink insets (href) to url insets (url)'
i = 0
while True:
i = find_token(document.body, "\\begin_inset CommandInset href", i)
if i == -1:
return
document.body[i : i + 2] = \
["\\begin_inset CommandInset url", "LatexCommand url"]
i = i + 2
def convert_include(document):
'Converts include insets to new format.'
i = 0
r = re.compile(r'\\begin_inset Include\s+\\([^{]+){([^}]*)}(?:\[(.*)\])?')
while True:
i = find_token(document.body, "\\begin_inset Include", i)
if i == -1:
return
line = document.body[i]
previewline = document.body[i + 1]
m = r.match(line)
if m == None:
document.warning("Unable to match line " + str(i) + " of body!")
i += 1
continue
cmd = m.group(1)
fn = m.group(2)
opt = m.group(3)
insertion = ["\\begin_inset CommandInset include",
"LatexCommand " + cmd, previewline,
"filename \"" + fn + "\""]
newlines = 2
if opt:
insertion.append("lstparams " + '"' + opt + '"')
newlines += 1
document.body[i : i + 2] = insertion
i += newlines
def revert_include(document):
'Reverts include insets to old format.'
i = 0
r1 = re.compile('LatexCommand (.+)')
r2 = re.compile('filename (.+)')
r3 = re.compile('options (.*)')
while True:
i = find_token(document.body, "\\begin_inset CommandInset include", i)
if i == -1:
return
previewline = document.body[i + 1]
m = r1.match(document.body[i + 2])
if m == None:
document.warning("Malformed LyX document: No LatexCommand line for `" +
document.body[i] + "' on line " + str(i) + ".")
i += 1
continue
cmd = m.group(1)
m = r2.match(document.body[i + 3])
if m == None:
document.warning("Malformed LyX document: No filename line for `" + \
document.body[i] + "' on line " + str(i) + ".")
i += 2
continue
fn = m.group(1)
options = ""
numlines = 4
if (cmd == "lstinputlisting"):
m = r3.match(document.body[i + 4])
if m != None:
options = m.group(1)
numlines = 5
newline = "\\begin_inset Include \\" + cmd + "{" + fn + "}"
if options:
newline += ("[" + options + "]")
insertion = [newline, previewline]
document.body[i : i + numlines] = insertion
i += 2
def revert_albanian(document):
"Set language Albanian to English"
# Set document language from Albanian to English
i = 0
if document.language == "albanian":
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 albanian", j)
if j == -1:
return
document.body[j] = document.body[j].replace("\\lang albanian", "\\lang english")
j = j + 1
def revert_lowersorbian(document):
"Set language lower Sorbian to English"
# Set document language from lower Sorbian to English
i = 0
if document.language == "lowersorbian":
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 lowersorbian", j)
if j == -1:
return
document.body[j] = document.body[j].replace("\\lang lowersorbian", "\\lang english")
j = j + 1
def revert_uppersorbian(document):
"Set language uppersorbian to usorbian as this was used in LyX 1.5"
# Set document language from uppersorbian to usorbian
i = 0
if document.language == "uppersorbian":
document.language = "usorbian"
i = find_token(document.header, "\\language", 0)
if i != -1:
document.header[i] = "\\language usorbian"
j = 0
while True:
j = find_token(document.body, "\\lang uppersorbian", j)
if j == -1:
return
document.body[j] = document.body[j].replace("\\lang uppersorbian", "\\lang usorbian")
j = j + 1
def convert_usorbian(document):
"Set language uppersorbian to usorbian as this was used in LyX 1.5"
# Set document language from uppersorbian to usorbian
i = 0
if document.language == "usorbian":
document.language = "uppersorbian"
i = find_token(document.header, "\\language", 0)
if i != -1:
document.header[i] = "\\language uppersorbian"
j = 0
while True:
j = find_token(document.body, "\\lang usorbian", j)
if j == -1:
return
document.body[j] = document.body[j].replace("\\lang usorbian", "\\lang uppersorbian")
j = j + 1
def revert_macro_optional_params(document):
"Convert macro definitions with optional parameters into ERTs"
# Stub to convert macro definitions with one or more optional parameters
# into uninterpreted ERT insets
def revert_hyperlinktype(document):
'Reverts hyperlink type'
i = 0
j = 0
while True:
i = find_token(document.body, "target", i)
if i == -1:
return
j = find_token(document.body, "type", i)
if j == -1:
return
if j == i + 1:
del document.body[j]
i = i + 1
##
# Conversion hub
#
supported_versions = ["1.6.0","1.6"]
convert = [[277, [fix_wrong_tables]],
[278, [close_begin_deeper]],
[279, [long_charstyle_names]],
[280, [axe_show_label]],
[281, []],
[282, []],
[283, [convert_flex]],
[284, []],
[285, []],
[286, []],
[287, [convert_wrapfig_options]],
[288, [convert_inset_command]],
[289, [convert_latexcommand_index]],
[290, []],
[291, []],
[292, []],
[293, []],
[294, [convert_pdf_options]],
[295, [convert_htmlurl, convert_url]],
[296, [convert_include]],
[297, [convert_usorbian]],
[298, []],
[299, []]
]
revert = [[298, [revert_hyperlinktype]],
[297, [revert_macro_optional_params]],
[296, [revert_albanian, revert_lowersorbian, revert_uppersorbian]],
[295, [revert_include]],
[294, [revert_href]],
[293, [revert_pdf_options_2]],
[292, [revert_inset_info]],
[291, [revert_japanese, revert_japanese_encoding]],
[290, [revert_vietnamese]],
[289, [revert_wraptable]],
[288, [revert_latexcommand_index]],
[287, [revert_inset_command]],
[286, [revert_wrapfig_options]],
[285, [revert_pdf_options]],
[284, [remove_inzip_options]],
[283, []],
[282, [revert_flex]],
[281, []],
[280, [revert_begin_modules]],
[279, [revert_show_label]],
[278, [revert_long_charstyle_names]],
[277, []],
[276, []]
]
if __name__ == "__main__":
pass