""" Convert files to the file format generated by lyx 2.0"""
-import re
+import re, string
+import unicodedata
+import sys, os
from parser_tools import find_token, find_end_of, find_tokens, get_value, get_value_string
document.preamble.insert(index, text)
+def read_unicodesymbols():
+ " Read the unicodesymbols list of unicode characters and corresponding commands."
+ pathname = os.path.abspath(os.path.dirname(sys.argv[0]))
+ fp = open(os.path.join(pathname.strip('lyx2lyx'), 'unicodesymbols'))
+ spec_chars = []
+ # Two backslashes, followed by some non-word character, and then a character
+ # in brackets. The idea is to check for constructs like: \"{u}, which is how
+ # they are written in the unicodesymbols file; but they can also be written
+ # as: \"u or even \" u.
+ r = re.compile(r'\\\\(\W)\{(\w)\}')
+ for line in fp.readlines():
+ if line[0] != '#' and line.strip() != "":
+ line=line.replace(' "',' ') # remove all quotation marks with spaces before
+ line=line.replace('" ',' ') # remove all quotation marks with spaces after
+ line=line.replace(r'\"','"') # replace \" by " (for characters with diaeresis)
+ try:
+ [ucs4,command,dead] = line.split(None,2)
+ if command[0:1] != "\\":
+ continue
+ spec_chars.append([command, unichr(eval(ucs4))])
+ except:
+ continue
+ m = r.match(command)
+ if m != None:
+ command = "\\\\"
+ # If the character is a double-quote, then we need to escape it, too,
+ # since it is done that way in the LyX file.
+ if m.group(1) == "\"":
+ command += "\\"
+ commandbl = command
+ command += m.group(1) + m.group(2)
+ commandbl += m.group(1) + ' ' + m.group(2)
+ spec_chars.append([command, unichr(eval(ucs4))])
+ spec_chars.append([commandbl, unichr(eval(ucs4))])
+ fp.close()
+ return spec_chars
+
+
+unicode_reps = read_unicodesymbols()
+
+
+def 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 Standard\n" \
+ + string + "\n\\end_layout\n\\end_inset"
+ return string
+
+
+def lyx2latex(document, lines):
+ 'Convert some LyX stuff into corresponding LaTeX stuff, as best we can.'
+ # clean up multiline stuff
+ content = ""
+ ert_end = 0
+
+ for curline in range(len(lines)):
+ line = lines[curline]
+ if line.startswith("\\begin_inset ERT"):
+ # We don't want to replace things inside ERT, so figure out
+ # where the end of the inset is.
+ ert_end = find_end_of_inset(lines, curline + 1)
+ continue
+ elif line.startswith("\\begin_inset Formula"):
+ line = line[20:]
+ elif line.startswith("\\begin_inset Quotes"):
+ # For now, we do a very basic reversion. Someone who understands
+ # quotes is welcome to fix it up.
+ qtype = line[20:].strip()
+ # lang = qtype[0]
+ side = qtype[1]
+ dbls = qtype[2]
+ if side == "l":
+ if dbls == "d":
+ line = "``"
+ else:
+ line = "`"
+ else:
+ if dbls == "d":
+ line = "''"
+ else:
+ line = "'"
+ elif line.isspace() or \
+ line.startswith("\\begin_layout") or \
+ line.startswith("\\end_layout") or \
+ line.startswith("\\begin_inset") or \
+ line.startswith("\\end_inset") or \
+ line.startswith("\\lang") or \
+ line.strip() == "status collapsed" or \
+ line.strip() == "status open":
+ #skip all that stuff
+ continue
+
+ # this needs to be added to the preamble because of cases like
+ # \textmu, \textbackslash, etc.
+ add_to_preamble(document, ['% added by lyx2lyx for converted index entries',
+ '\\@ifundefined{textmu}',
+ ' {\\usepackage{textcomp}}{}'])
+ # a lossless reversion is not possible
+ # try at least to handle some common insets and settings
+ if ert_end >= curline:
+ line = line.replace(r'\backslash', r'\\')
+ else:
+ line = line.replace('&', '\\&{}')
+ line = line.replace('#', '\\#{}')
+ line = line.replace('^', '\\^{}')
+ line = line.replace('%', '\\%{}')
+ line = line.replace('_', '\\_{}')
+ line = line.replace('$', '\\${}')
+
+ # Do the LyX text --> LaTeX conversion
+ for rep in unicode_reps:
+ line = line.replace(rep[1], rep[0] + "{}")
+ line = line.replace(r'\backslash', r'\textbackslash{}')
+ line = line.replace(r'\series bold', r'\bfseries{}').replace(r'\series default', r'\mdseries{}')
+ line = line.replace(r'\shape italic', r'\itshape{}').replace(r'\shape smallcaps', r'\scshape{}')
+ line = line.replace(r'\shape slanted', r'\slshape{}').replace(r'\shape default', r'\upshape{}')
+ line = line.replace(r'\emph on', r'\em{}').replace(r'\emph default', r'\em{}')
+ line = line.replace(r'\noun on', r'\scshape{}').replace(r'\noun default', r'\upshape{}')
+ line = line.replace(r'\bar under', r'\underbar{').replace(r'\bar default', r'}')
+ line = line.replace(r'\family sans', r'\sffamily{}').replace(r'\family default', r'\normalfont{}')
+ line = line.replace(r'\family typewriter', r'\ttfamily{}').replace(r'\family roman', r'\rmfamily{}')
+ line = line.replace(r'\InsetSpace ', r'').replace(r'\SpecialChar ', r'')
+ content += line
+ return content
+
+
####################################################################
" 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':
+ del document.header[i]
return
del document.header[i]
# 1.) set doc encoding to utf8-plain
# 3.) set preamble stuff
pretext = '%% This document must be processed with xelatex!\n'
pretext += '\\usepackage{fontspec}\n'
- pretext += '\\setmainfont[Mapping=tex-text]{' + roman + '}\n'
- pretext += '\\setsansfont['
- if sf_scale != 100:
- pretext += 'Scale=' + str(sf_scale / 100) + ','
- pretext += 'Mapping=tex-text]{' + sans + '}\n'
- pretext += '\\setmonofont'
- if tt_scale != 100:
- pretext += '[Scale=' + str(tt_scale / 100) + ']'
- pretext += '{' + typewriter + '}\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
i = find_token(document.header, "\\font_roman", 0)
document.header[i] = "\\font_tt_scale 100"
+def revert_outputformat(document):
+ " Remove default output format param "
+ i = find_token(document.header, '\\default_output_format', 0)
+ if i == -1:
+ document.warning("Malformed LyX document: Missing \\default_output_format.")
+ return
+ del document.header[i]
+
+
+def revert_backgroundcolor(document):
+ " Reverts background color to preamble code "
+ i = 0
+ colorcode = ""
+ while True:
+ i = find_token(document.header, "\\backgroundcolor", i)
+ if i == -1:
+ return
+ colorcode = get_value(document.header, '\\backgroundcolor', 0)
+ del document.header[i]
+ # don't clutter the preamble if backgroundcolor is not set
+ if colorcode == "#ffffff":
+ continue
+ # the color code is in the form #rrggbb where every character denotes a hex number
+ # convert the string to an int
+ red = string.atoi(colorcode[1:3],16)
+ # we want the output "0.5" for the value "127" therefore add here
+ if red != 0:
+ red = red + 1
+ redout = float(red) / 256
+ green = string.atoi(colorcode[3:5],16)
+ if green != 0:
+ green = green + 1
+ greenout = float(green) / 256
+ blue = string.atoi(colorcode[5:7],16)
+ if blue != 0:
+ blue = blue + 1
+ blueout = float(blue) / 256
+ # write the preamble
+ insert_to_preamble(0, document,
+ '% Commands inserted by lyx2lyx to set the background color\n'
+ + '\\@ifundefined{definecolor}{\\usepackage{color}}{}\n'
+ + '\\definecolor{page_backgroundcolor}{rgb}{'
+ + str(redout) + ', ' + str(greenout)
+ + ', ' + str(blueout) + '}\n'
+ + '\\pagecolor{page_backgroundcolor}\n')
+
+
+def revert_splitindex(document):
+ " Reverts splitindex-aware documents "
+ i = find_token(document.header, '\\use_indices', 0)
+ if i == -1:
+ document.warning("Malformed LyX document: Missing \\use_indices.")
+ return
+ indices = get_value(document.header, "\\use_indices", i)
+ preamble = ""
+ if indices == "true":
+ preamble += "\\usepackage{splitidx}\n"
+ del document.header[i]
+ i = 0
+ while True:
+ i = find_token(document.header, "\\index", i)
+ if i == -1:
+ break
+ k = find_token(document.header, "\\end_index", i)
+ if k == -1:
+ document.warning("Malformed LyX document: Missing \\end_index.")
+ return
+ line = document.header[i]
+ l = re.compile(r'\\index (.*)$')
+ m = l.match(line)
+ iname = m.group(1)
+ ishortcut = get_value(document.header, '\\shortcut', i, k)
+ if ishortcut != "" and indices == "true":
+ preamble += "\\newindex[" + iname + "]{" + ishortcut + "}\n"
+ del document.header[i:k+1]
+ i = 0
+ if preamble != "":
+ insert_to_preamble(0, document, preamble)
+ i = 0
+ while True:
+ i = find_token(document.body, "\\begin_inset Index", i)
+ if i == -1:
+ break
+ line = document.body[i]
+ l = re.compile(r'\\begin_inset Index (.*)$')
+ m = l.match(line)
+ itype = m.group(1)
+ if itype == "idx" or indices == "false":
+ document.body[i] = "\\begin_inset Index"
+ else:
+ k = find_end_of_inset(document.body, i)
+ if k == -1:
+ return
+ content = lyx2latex(document, document.body[i:k])
+ # escape quotes
+ content = content.replace('"', r'\"')
+ subst = [put_cmd_in_ert("\\sindex[" + itype + "]{" + content + "}")]
+ document.body[i:k+1] = subst
+ i = i + 1
+ i = 0
+ while True:
+ i = find_token(document.body, "\\begin_inset CommandInset index_print", i)
+ if i == -1:
+ return
+ k = find_end_of_inset(document.body, i)
+ ptype = get_value(document.body, 'type', i, k).strip('"')
+ if ptype == "idx":
+ j = find_token(document.body, "type", i, k)
+ del document.body[j]
+ elif indices == "false":
+ del document.body[i:k+1]
+ else:
+ subst = [put_cmd_in_ert("\\printindex[" + ptype + "]{}")]
+ document.body[i:k+1] = subst
+ i = i + 1
+
+
+def convert_splitindex(document):
+ " Converts index and printindex insets to splitindex-aware format "
+ i = 0
+ while True:
+ i = find_token(document.body, "\\begin_inset Index", i)
+ if i == -1:
+ break
+ document.body[i] = document.body[i].replace("\\begin_inset Index",
+ "\\begin_inset Index idx")
+ i = i + 1
+ i = 0
+ while True:
+ i = find_token(document.body, "\\begin_inset CommandInset index_print", i)
+ if i == -1:
+ return
+ if document.body[i + 1].find('LatexCommand printindex') == -1:
+ document.warning("Malformed LyX document: Incomplete printindex inset.")
+ return
+ subst = ["LatexCommand printindex",
+ "type \"idx\""]
+ document.body[i + 1:i + 2] = subst
+ i = i + 1
+
+
+def revert_subindex(document):
+ " Reverts \\printsubindex CommandInset types "
+ i = find_token(document.header, '\\use_indices', 0)
+ if i == -1:
+ document.warning("Malformed LyX document: Missing \\use_indices.")
+ return
+ indices = get_value(document.header, "\\use_indices", i)
+ i = 0
+ while True:
+ i = find_token(document.body, "\\begin_inset CommandInset index_print", i)
+ if i == -1:
+ return
+ k = find_end_of_inset(document.body, i)
+ ctype = get_value(document.body, 'LatexCommand', i, k)
+ if ctype != "printsubindex":
+ i = i + 1
+ continue
+ ptype = get_value(document.body, 'type', i, k).strip('"')
+ if indices == "false":
+ del document.body[i:k+1]
+ else:
+ subst = [put_cmd_in_ert("\\printsubindex[" + ptype + "]{}")]
+ document.body[i:k+1] = subst
+ i = i + 1
+
+
+def revert_printindexall(document):
+ " Reverts \\print[sub]index* CommandInset types "
+ i = find_token(document.header, '\\use_indices', 0)
+ if i == -1:
+ document.warning("Malformed LyX document: Missing \\use_indices.")
+ return
+ indices = get_value(document.header, "\\use_indices", i)
+ i = 0
+ while True:
+ i = find_token(document.body, "\\begin_inset CommandInset index_print", i)
+ if i == -1:
+ return
+ k = find_end_of_inset(document.body, i)
+ ctype = get_value(document.body, 'LatexCommand', i, k)
+ if ctype != "printindex*" and ctype != "printsubindex*":
+ i = i + 1
+ continue
+ if indices == "false":
+ del document.body[i:k+1]
+ else:
+ subst = [put_cmd_in_ert("\\" + ctype + "{}")]
+ document.body[i:k+1] = subst
+ i = i + 1
+
+
+def revert_strikeout(document):
+ " Reverts \\strikeout character style "
+ while True:
+ i = find_token(document.body, '\\strikeout', 0)
+ if i == -1:
+ return
+ del document.body[i]
+
+
+def revert_uulinewave(document):
+ " Reverts \\uuline, and \\uwave character styles "
+ while True:
+ i = find_token(document.body, '\\uuline', 0)
+ if i == -1:
+ break
+ del document.body[i]
+ while True:
+ i = find_token(document.body, '\\uwave', 0)
+ if i == -1:
+ return
+ del document.body[i]
+
+
+def revert_ulinelatex(document):
+ " Reverts \\uline character style "
+ 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')
+
+
+def revert_custom_processors(document):
+ " Remove bibtex_command and index_command params "
+ i = find_token(document.header, '\\bibtex_command', 0)
+ if i == -1:
+ document.warning("Malformed LyX document: Missing \\bibtex_command.")
+ return
+ del document.header[i]
+ i = find_token(document.header, '\\index_command', 0)
+ if i == -1:
+ document.warning("Malformed LyX document: Missing \\index_command.")
+ return
+ del document.header[i]
+
+
+def convert_nomencl_width(document):
+ " Add set_width param to nomencl_print "
+ i = 0
+ while True:
+ i = find_token(document.body, "\\begin_inset CommandInset nomencl_print", i)
+ if i == -1:
+ break
+ document.body.insert(i + 2, "set_width \"none\"")
+ i = i + 1
+
+
+def revert_nomencl_width(document):
+ " Remove set_width param from nomencl_print "
+ i = 0
+ while True:
+ i = find_token(document.body, "\\begin_inset CommandInset nomencl_print", i)
+ 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 = i + 1
+
+
+def revert_nomencl_cwidth(document):
+ " Remove width param from nomencl_print "
+ i = 0
+ while True:
+ i = find_token(document.body, "\\begin_inset CommandInset nomencl_print", i)
+ if i == -1:
+ break
+ 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('"')
+ del document.body[l]
+ add_to_preamble(document, ["\\setlength{\\nomlabelwidth}{" + width + "}"])
+ i = i + 1
+
+
##
# Conversion hub
#
convert = [[346, []],
[347, []],
[348, []],
- [349, []]
+ [349, []],
+ [350, []],
+ [351, []],
+ [352, [convert_splitindex]],
+ [353, []],
+ [354, []],
+ [355, []],
+ [356, []],
+ [357, []],
+ [358, []],
+ [359, [convert_nomencl_width]],
+ [360, []],
+ [361, []]
]
-revert = [[348, [revert_xetex]],
+revert = [[360, []],
+ [359, [revert_nomencl_cwidth]],
+ [358, [revert_nomencl_width]],
+ [357, [revert_custom_processors]],
+ [356, [revert_ulinelatex]],
+ [355, [revert_uulinewave]],
+ [354, [revert_strikeout]],
+ [353, [revert_printindexall]],
+ [352, [revert_subindex]],
+ [351, [revert_splitindex]],
+ [350, [revert_backgroundcolor]],
+ [349, [revert_outputformat]],
+ [348, [revert_xetex]],
[347, [revert_phantom, revert_hphantom, revert_vphantom]],
[346, [revert_tabularvalign]],
[345, [revert_swiss]]