1 # This file is part of lyx2lyx
2 # -*- coding: utf-8 -*-
3 # Copyright (C) 2002-2004 Dekel Tsur <dekel@lyx.org>
4 # Copyright (C) 2002-2006 José Matos <jamatos@lyx.org>
6 # This program is free software; you can redistribute it and/or
7 # modify it under the terms of the GNU General Public License
8 # as published by the Free Software Foundation; either version 2
9 # of the License, or (at your option) any later version.
11 # This program is distributed in the hope that it will be useful,
12 # but WITHOUT ANY WARRANTY; without even the implied warranty of
13 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
14 # GNU General Public License for more details.
16 # You should have received a copy of the GNU General Public License
17 # along with this program; if not, write to the Free Software
18 # Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
20 " The LyX module has all the rules related with different lyx file formats."
22 from parser_tools import get_value, check_token, find_token, \
23 find_tokens, find_end_of
32 import lyx2lyx_version
33 version__ = lyx2lyx_version.version
34 except: # we are running from build directory so assume the last version
35 version__ = '1.6.0svn'
39 ####################################################################
40 # Private helper functions
42 def find_end_of_inset(lines, i):
43 " Find beginning of inset, where lines[i] is included."
44 return find_end_of(lines, i, "\\begin_inset", "\\end_inset")
46 def minor_versions(major, last_minor_version):
47 """ Generate minor versions, using major as prefix and minor
48 versions from 0 until last_minor_version, plus the generic version.
52 minor_versions("1.2", 4) ->
53 [ "1.2", "1.2.0", "1.2.1", "1.2.2", "1.2.3"]
55 return [major] + [major + ".%d" % i for i in range(last_minor_version + 1)]
58 # End of helper functions
59 ####################################################################
62 # Regular expressions used
63 format_re = re.compile(r"(\d)[\.,]?(\d\d)")
64 fileformat = re.compile(r"\\lyxformat\s*(\S*)")
65 original_version = re.compile(r".*?LyX ([\d.]*)")
68 # file format information:
69 # file, supported formats, stable release versions
70 format_relation = [("0_06", [200], minor_versions("0.6" , 4)),
71 ("0_08", [210], minor_versions("0.8" , 6) + ["0.7"]),
72 ("0_10", [210], minor_versions("0.10", 7) + ["0.9"]),
73 ("0_12", [215], minor_versions("0.12", 1) + ["0.11"]),
74 ("1_0", [215], minor_versions("1.0" , 4)),
75 ("1_1", [215], minor_versions("1.1" , 4)),
76 ("1_1_5", [216], ["1.1.5","1.1.5.1","1.1.5.2","1.1"]),
77 ("1_1_6_0", [217], ["1.1.6","1.1.6.1","1.1.6.2","1.1"]),
78 ("1_1_6_3", [218], ["1.1.6.3","1.1.6.4","1.1"]),
79 ("1_2", [220], minor_versions("1.2" , 4)),
80 ("1_3", [221], minor_versions("1.3" , 7)),
81 ("1_4", range(222,246), minor_versions("1.4" , 5)),
82 ("1_5", range(246,277), minor_versions("1.5" , 2)),
83 ("1_6", range(277,314), minor_versions("1.6" , 0))] # Richard Heck: conversion of module representations
87 " Returns a list with supported file formats."
89 for version in format_relation:
90 for format in version[1]:
91 if format not in formats:
92 formats.append(format)
97 " Returns the more recent file format available."
98 return format_relation[-1][1][-1]
101 def get_backend(textclass):
102 " For _textclass_ returns its backend."
103 if textclass == "linuxdoc" or textclass == "manpage":
105 if textclass[:7] == "docbook":
111 " Remove end of line char(s)."
112 if line[-2:-1] == '\r':
118 def get_encoding(language, inputencoding, format, cjk_encoding):
119 " Returns enconding of the lyx file"
122 # CJK-LyX encodes files using the current locale encoding.
123 # This means that files created by CJK-LyX can only be converted using
124 # the correct locale settings unless the encoding is given as commandline
126 if cjk_encoding == 'auto':
127 return locale.getpreferredencoding()
130 from lyx2lyx_lang import lang
131 if inputencoding == "auto" or inputencoding == "default":
132 return lang[language][3]
133 if inputencoding == "":
135 # python does not know the alias latin9
136 if inputencoding == "latin9":
144 """This class carries all the information of the LyX file."""
146 def __init__(self, end_format = 0, input = "", output = "", error = "",
147 debug = default_debug__, try_hard = 0, cjk_encoding = '',
148 language = "english", encoding = "auto"):
151 end_format: final format that the file should be converted. (integer)
152 input: the name of the input source, if empty resort to standard input.
153 output: the name of the output file, if empty use the standard output.
154 error: the name of the error file, if empty use the standard error.
155 debug: debug level, O means no debug, as its value increases be more verbose.
157 self.choose_io(input, output)
160 self.err = open(error, "w")
162 self.err = sys.stderr
165 self.try_hard = try_hard
166 self.cjk_encoding = cjk_encoding
169 self.end_format = self.lyxformat(end_format)
171 self.end_format = get_end_format()
173 self.backend = "latex"
174 self.textclass = "article"
175 # This is a hack: We use '' since we don't know the default
176 # layout of the text class. LyX will parse it as default layout.
177 # FIXME: Read the layout file and use the real default layout
178 self.default_layout = ''
183 self.encoding = encoding
184 self.language = language
187 def warning(self, message, debug_level= default_debug__):
188 """ Emits warning to self.error, if the debug_level is less
189 than the self.debug."""
190 if debug_level <= self.debug:
191 self.err.write("Warning: " + message + "\n")
194 def error(self, message):
195 " Emits a warning and exits if not in try_hard mode."
196 self.warning(message)
197 if not self.try_hard:
198 self.warning("Quiting.")
205 """Reads a file into the self.header and
206 self.body parts, from self.input."""
209 line = self.input.readline()
211 self.error("Invalid LyX file.")
213 line = trim_eol(line)
214 if check_token(line, '\\begin_preamble'):
216 line = self.input.readline()
218 self.error("Invalid LyX file.")
220 line = trim_eol(line)
221 if check_token(line, '\\end_preamble'):
224 if line.split()[:0] in ("\\layout",
225 "\\begin_layout", "\\begin_body"):
227 self.warning("Malformed LyX file:"
228 "Missing '\\end_preamble'."
229 "\nAdding it now and hoping"
232 self.preamble.append(line)
234 if check_token(line, '\\end_preamble'):
241 if line.split()[0] in ("\\layout", "\\begin_layout",
242 "\\begin_body", "\\begin_deeper"):
243 self.body.append(line)
246 self.header.append(line)
248 self.textclass = get_value(self.header, "\\textclass", 0)
249 self.backend = get_backend(self.textclass)
250 self.format = self.read_format()
251 self.language = get_value(self.header, "\\language", 0,
253 self.inputencoding = get_value(self.header, "\\inputencoding",
255 self.encoding = get_encoding(self.language,
256 self.inputencoding, self.format,
258 self.initial_version = self.read_version()
260 # Second pass over header and preamble, now we know the file encoding
261 for i in range(len(self.header)):
262 self.header[i] = self.header[i].decode(self.encoding)
263 for i in range(len(self.preamble)):
264 self.preamble[i] = self.preamble[i].decode(self.encoding)
268 line = self.input.readline().decode(self.encoding)
271 self.body.append(trim_eol(line))
275 " Writes the LyX file to self.output."
279 if self.encoding == "auto":
280 self.encoding = get_encoding(self.language, self.encoding,
281 self.format, self.cjk_encoding)
283 i = find_token(self.header, '\\textclass', 0) + 1
284 preamble = ['\\begin_preamble'] + self.preamble + ['\\end_preamble']
286 self.error("Malformed LyX file: Missing '\\textclass'.")
288 header = self.header[:i] + preamble + self.header[i:]
292 for line in header + [''] + self.body:
293 self.output.write(line.encode(self.encoding)+"\n")
296 def choose_io(self, input, output):
297 """Choose input and output streams, dealing transparently with
301 self.output = open(output, "wb")
303 self.output = sys.stdout
305 if input and input != '-':
306 self.dir = os.path.dirname(os.path.abspath(input))
308 gzip.open(input).readline()
309 self.input = gzip.open(input)
310 self.output = gzip.GzipFile(mode="wb", fileobj=self.output)
312 self.input = open(input)
315 self.input = sys.stdin
318 def lyxformat(self, format):
319 " Returns the file format representation, an integer."
320 result = format_re.match(format)
322 format = int(result.group(1) + result.group(2))
326 self.error(str(format) + ": " + "Invalid LyX file.")
328 if format in formats_list():
331 self.error(str(format) + ": " + "Format not supported.")
335 def read_version(self):
336 """ Searchs for clues of the LyX version used to write the
337 file, returns the most likely value, or None otherwise."""
339 for line in self.header:
343 line = line.replace("fix",".")
344 result = original_version.match(line)
346 # Special know cases: reLyX and KLyX
347 if line.find("reLyX") != -1 or line.find("KLyX") != -1:
350 res = result.group(1)
353 #self.warning("Version %s" % result.group(1))
355 self.warning(str(self.header[:2]))
359 def set_version(self):
360 " Set the header with the version used."
361 self.header[0] = " ".join(["#LyX %s created this file." % version__,
362 "For more info see http://www.lyx.org/"])
363 if self.header[1][0] == '#':
367 def read_format(self):
368 " Read from the header the fileformat of the present LyX file."
369 for line in self.header:
370 result = fileformat.match(line)
372 return self.lyxformat(result.group(1))
374 self.error("Invalid LyX File.")
378 def set_format(self):
379 " Set the file format of the file, in the header."
380 if self.format <= 217:
381 format = str(float(self.format)/100)
383 format = str(self.format)
384 i = find_token(self.header, "\\lyxformat", 0)
385 self.header[i] = "\\lyxformat %s" % format
388 def set_textclass(self):
389 i = find_token(self.header, "\\textclass", 0)
390 self.header[i] = "\\textclass %s" % self.textclass
393 #Note that the module will be added at the END of the extant ones
394 def add_module(self, module):
395 i = find_token(self.header, "\\begin_modules", 0)
397 #No modules yet included
398 i = find_token(self.header, "\\textclass", 0)
400 self.warning("Malformed LyX document: No \\textclass!!")
402 modinfo = ["\\begin_modules", module, "\\end_modules"]
403 self.header[i + 1: i + 1] = modinfo
405 j = find_token(self.header, "\\end_modules", i)
407 self.warning("Malformed LyX document: No \\end_modules.")
409 k = find_token(self.header, module, i)
410 if k != -1 and k < j:
412 self.header.insert(j, module)
415 def get_module_list(self):
416 i = find_token(self.header, "\\begin_modules", 0)
419 j = find_token(self.header, "\\end_modules", i)
420 return self.header[i + 1 : j]
423 def set_module_list(self, mlist):
424 modbegin = find_token(self.header, "\\begin_modules", 0)
426 #No modules yet included
427 modbegin = find_token(self.header, "\\textclass", 0)
429 self.warning("Malformed LyX document: No \\textclass!!")
431 modend = find_token(self.header, "\\end_modules", modbegin)
433 self.warning("Malformed LyX document: No \\end_modules.")
435 newmodlist = ['\\begin_modules'] + mlist + ['\\end_modules']
436 self.header[modbegin:modend + 1] = newmodlist
439 def set_parameter(self, param, value):
440 " Set the value of the header parameter."
441 i = find_token(self.header, '\\' + param, 0)
443 self.warning('Parameter not found in the header: %s' % param, 3)
445 self.header[i] = '\\%s %s' % (param, str(value))
448 def is_default_layout(self, layout):
449 " Check whether a layout is the default layout of this class."
450 # FIXME: Check against the real text class default layout
451 if layout == 'Standard' or layout == self.default_layout:
457 "Convert from current (self.format) to self.end_format."
458 mode, convertion_chain = self.chain()
459 self.warning("convertion chain: " + str(convertion_chain), 3)
461 for step in convertion_chain:
462 steps = getattr(__import__("lyx_" + step), mode)
464 self.warning("Convertion step: %s - %s" % (step, mode),
467 self.error("The convertion to an older "
468 "format (%s) is not implemented." % self.format)
470 multi_conv = len(steps) != 1
471 for version, table in steps:
473 (self.format >= version and mode == "convert") or\
474 (self.format <= version and mode == "revert"):
482 self.warning("An error ocurred in %s, %s" %
483 (version, str(conv)),
485 if not self.try_hard:
489 self.warning("%lf: Elapsed time on %s" %
490 (time.time() - init_t,
491 str(conv)), default_debug__ +
493 self.format = version
494 if self.end_format == self.format:
499 """ This is where all the decisions related with the
500 convertion are taken. It returns a list of modules needed to
501 convert the LyX file from self.format to self.end_format"""
503 self.start = self.format
507 for rel in format_relation:
508 if self.initial_version in rel[2]:
510 initial_step = rel[0]
514 if not correct_version:
516 self.warning("Version does not match file format, "
517 "discarding it. (Version %s, format %d)" %
518 (self.initial_version, self.format))
519 for rel in format_relation:
521 initial_step = rel[0]
524 # This should not happen, really.
525 self.error("Format not supported.")
527 # Find the final step
528 for rel in format_relation:
529 if self.end_format in rel[1]:
533 self.error("Format not supported.")
535 # Convertion mode, back or forth
537 if (initial_step, self.start) < (final_step, self.end_format):
540 for step in format_relation:
541 if initial_step <= step[0] <= final_step:
542 if first_step and len(step[1]) == 1:
545 steps.append(step[0])
548 relation_format = format_relation[:]
549 relation_format.reverse()
552 for step in relation_format:
553 if final_step <= step[0] <= initial_step:
554 steps.append(step[0])
557 if last_step[1][-1] == self.end_format:
560 self.warning("Convertion mode: %s\tsteps%s" %(mode, steps), 10)
564 def get_toc(self, depth = 4):
565 " Returns the TOC of this LyX document."
566 paragraphs_filter = {'Title' : 0,'Chapter' : 1, 'Section' : 2,
567 'Subsection' : 3, 'Subsubsection': 4}
568 allowed_insets = ['Quotes']
569 allowed_parameters = ('\\paragraph_spacing', '\\noindent',
570 '\\align', '\\labelwidthstring',
571 "\\start_of_appendix", "\\leftindent")
573 for section in paragraphs_filter.keys():
574 sections.append('\\begin_layout %s' % section)
579 i = find_tokens(self.body, sections, i)
583 j = find_end_of(self.body, i + 1, '\\begin_layout', '\\end_layout')
585 self.warning('Incomplete file.', 0)
588 section = self.body[i].split()[1]
589 if section[-1] == '*':
590 section = section[:-1]
595 # skip paragraph parameters
596 while not self.body[k].strip() or self.body[k].split()[0] \
597 in allowed_parameters:
601 if check_token(self.body[k], '\\begin_inset'):
602 inset = self.body[k].split()[1]
603 end = find_end_of_inset(self.body, k)
604 if end == -1 or end > j:
605 self.warning('Malformed file.', 0)
607 if inset in allowed_insets:
608 par.extend(self.body[k: end+1])
611 par.append(self.body[k])
614 # trim empty lines in the end.
615 while par and par[-1].strip() == '':
618 toc_par.append(Paragraph(section, par))
625 class File(LyX_base):
626 " This class reads existing LyX files."
628 def __init__(self, end_format = 0, input = "", output = "", error = "",
629 debug = default_debug__, try_hard = 0, cjk_encoding = ''):
630 LyX_base.__init__(self, end_format, input, output, error,
631 debug, try_hard, cjk_encoding)
635 class NewFile(LyX_base):
636 " This class is to create new LyX files."
637 def set_header(self, **params):
640 "#LyX xxxx created this file."
641 "For more info see http://www.lyx.org/",
645 "\\textclass article",
646 "\\language english",
647 "\\inputencoding auto",
648 "\\font_roman default",
649 "\\font_sans default",
650 "\\font_typewriter default",
651 "\\font_default_family default",
654 "\\font_sf_scale 100",
655 "\\font_tt_scale 100",
656 "\\graphics default",
657 "\\paperfontsize default",
658 "\\papersize default",
659 "\\use_geometry false",
661 "\\cite_engine basic",
662 "\\use_bibtopic false",
663 "\\paperorientation portrait",
666 "\\paragraph_separation indent",
668 "\\quotes_language english",
671 "\\paperpagestyle default",
672 "\\tracking_changes false",
675 self.format = get_end_format()
677 self.set_parameter(param, params[param])
680 def set_body(self, paragraphs):
681 self.body.extend(['\\begin_body',''])
683 for par in paragraphs:
684 self.body.extend(par.asLines())
686 self.body.extend(['','\\end_body', '\\end_document'])
690 # unfinished implementation, it is missing the Text and Insets
692 " This class represents the LyX paragraphs."
693 def __init__(self, name, body=[], settings = [], child = []):
695 name: paragraph name.
696 body: list of lines of body text.
697 child: list of paragraphs that descend from this paragraph.
701 self.settings = settings
705 """ Converts the paragraph to a list of strings, representing
706 it in the LyX file."""
708 result = ['','\\begin_layout %s' % self.name]
709 result.extend(self.settings)
711 result.extend(self.body)
712 result.append('\\end_layout')
717 result.append('\\begin_deeper')
718 for node in self.child:
719 result.extend(node.asLines())
720 result.append('\\end_deeper')