Tag release lyx-1_6_0
[lyx.git] / lib / lyx2lyx / LyX.py
blobe5175355c7f7fe52b3accf933e8c5291bc1b5a0f
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
24 import os.path
25 import gzip
26 import locale
27 import sys
28 import re
29 import time
31 try:
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'
37 default_debug__ = 2
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.
50 Example:
52 minor_versions("1.2", 4) ->
53 [ "1.2", "1.2.0", "1.2.1", "1.2.2", "1.2.3"]
54 """
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" , 6)),
83 ("1_6", range(277,346), minor_versions("1.6" , 0))]
85 ####################################################################
86 # This is useful just for development versions #
87 # if the list of supported formats is empty get it from last step #
88 if not format_relation[-1][1]:
89 step, mode = format_relation[-1][0], "convert"
90 convert = getattr(__import__("lyx_" + step), mode)
91 format_relation[-1] = (step,
92 [conv[0] for conv in convert],
93 format_relation[-1][2])
94 # #
95 ####################################################################
97 def formats_list():
98 " Returns a list with supported file formats."
99 formats = []
100 for version in format_relation:
101 for format in version[1]:
102 if format not in formats:
103 formats.append(format)
104 return formats
107 def get_end_format():
108 " Returns the more recent file format available."
109 return format_relation[-1][1][-1]
112 def get_backend(textclass):
113 " For _textclass_ returns its backend."
114 if textclass == "linuxdoc" or textclass == "manpage":
115 return "linuxdoc"
116 if textclass.startswith("docbook") or textclass.startswith("agu-"):
117 return "docbook"
118 return "latex"
121 def trim_eol(line):
122 " Remove end of line char(s)."
123 if line[-2:-1] == '\r':
124 return line[:-2]
125 else:
126 return line[:-1]
129 def get_encoding(language, inputencoding, format, cjk_encoding):
130 " Returns enconding of the lyx file"
131 if format > 248:
132 return "utf8"
133 # CJK-LyX encodes files using the current locale encoding.
134 # This means that files created by CJK-LyX can only be converted using
135 # the correct locale settings unless the encoding is given as commandline
136 # argument.
137 if cjk_encoding == 'auto':
138 return locale.getpreferredencoding()
139 elif cjk_encoding:
140 return cjk_encoding
141 from lyx2lyx_lang import lang
142 if inputencoding == "auto" or inputencoding == "default":
143 return lang[language][3]
144 if inputencoding == "":
145 return "latin1"
146 if inputencoding == "utf8x":
147 return "utf8"
148 # python does not know the alias latin9
149 if inputencoding == "latin9":
150 return "iso-8859-15"
151 return inputencoding
154 # Class
156 class LyX_base:
157 """This class carries all the information of the LyX file."""
159 def __init__(self, end_format = 0, input = "", output = "", error = "",
160 debug = default_debug__, try_hard = 0, cjk_encoding = '',
161 language = "english", encoding = "auto"):
163 """Arguments:
164 end_format: final format that the file should be converted. (integer)
165 input: the name of the input source, if empty resort to standard input.
166 output: the name of the output file, if empty use the standard output.
167 error: the name of the error file, if empty use the standard error.
168 debug: debug level, O means no debug, as its value increases be more verbose.
170 self.choose_io(input, output)
172 if error:
173 self.err = open(error, "w")
174 else:
175 self.err = sys.stderr
177 self.debug = debug
178 self.try_hard = try_hard
179 self.cjk_encoding = cjk_encoding
181 if end_format:
182 self.end_format = self.lyxformat(end_format)
183 else:
184 self.end_format = get_end_format()
186 self.backend = "latex"
187 self.textclass = "article"
188 # This is a hack: We use '' since we don't know the default
189 # layout of the text class. LyX will parse it as default layout.
190 # FIXME: Read the layout file and use the real default layout
191 self.default_layout = ''
192 self.header = []
193 self.preamble = []
194 self.body = []
195 self.status = 0
196 self.encoding = encoding
197 self.language = language
200 def warning(self, message, debug_level= default_debug__):
201 """ Emits warning to self.error, if the debug_level is less
202 than the self.debug."""
203 if debug_level <= self.debug:
204 self.err.write("Warning: " + message + "\n")
207 def error(self, message):
208 " Emits a warning and exits if not in try_hard mode."
209 self.warning(message)
210 if not self.try_hard:
211 self.warning("Quiting.")
212 sys.exit(1)
214 self.status = 2
217 def read(self):
218 """Reads a file into the self.header and
219 self.body parts, from self.input."""
221 while True:
222 line = self.input.readline()
223 if not line:
224 self.error("Invalid LyX file.")
226 line = trim_eol(line)
227 if check_token(line, '\\begin_preamble'):
228 while 1:
229 line = self.input.readline()
230 if not line:
231 self.error("Invalid LyX file.")
233 line = trim_eol(line)
234 if check_token(line, '\\end_preamble'):
235 break
237 if line.split()[:0] in ("\\layout",
238 "\\begin_layout", "\\begin_body"):
240 self.warning("Malformed LyX file:"
241 "Missing '\\end_preamble'."
242 "\nAdding it now and hoping"
243 "for the best.")
245 self.preamble.append(line)
247 if check_token(line, '\\end_preamble'):
248 continue
250 line = line.strip()
251 if not line:
252 continue
254 if line.split()[0] in ("\\layout", "\\begin_layout",
255 "\\begin_body", "\\begin_deeper"):
256 self.body.append(line)
257 break
259 self.header.append(line)
261 i = find_token(self.header, '\\textclass', 0)
262 if i == -1:
263 self.warning("Malformed LyX file: Missing '\\textclass'.")
264 i = find_token(self.header, '\\lyxformat', 0) + 1
265 self.header[i:i] = ['\\textclass article']
267 self.textclass = get_value(self.header, "\\textclass", 0)
268 self.backend = get_backend(self.textclass)
269 self.format = self.read_format()
270 self.language = get_value(self.header, "\\language", 0,
271 default = "english")
272 self.inputencoding = get_value(self.header, "\\inputencoding",
273 0, default = "auto")
274 self.encoding = get_encoding(self.language,
275 self.inputencoding, self.format,
276 self.cjk_encoding)
277 self.initial_version = self.read_version()
279 # Second pass over header and preamble, now we know the file encoding
280 for i in range(len(self.header)):
281 self.header[i] = self.header[i].decode(self.encoding)
282 for i in range(len(self.preamble)):
283 self.preamble[i] = self.preamble[i].decode(self.encoding)
285 # Read document body
286 while 1:
287 line = self.input.readline().decode(self.encoding)
288 if not line:
289 break
290 self.body.append(trim_eol(line))
293 def write(self):
294 " Writes the LyX file to self.output."
295 self.set_version()
296 self.set_format()
297 self.set_textclass()
298 if self.encoding == "auto":
299 self.encoding = get_encoding(self.language, self.encoding,
300 self.format, self.cjk_encoding)
301 if self.preamble:
302 i = find_token(self.header, '\\textclass', 0) + 1
303 preamble = ['\\begin_preamble'] + self.preamble + ['\\end_preamble']
304 header = self.header[:i] + preamble + self.header[i:]
305 else:
306 header = self.header
308 for line in header + [''] + self.body:
309 self.output.write(line.encode(self.encoding)+"\n")
312 def choose_io(self, input, output):
313 """Choose input and output streams, dealing transparently with
314 compressed files."""
316 if output:
317 self.output = open(output, "wb")
318 else:
319 self.output = sys.stdout
321 if input and input != '-':
322 self.dir = os.path.dirname(os.path.abspath(input))
323 try:
324 gzip.open(input).readline()
325 self.input = gzip.open(input)
326 self.output = gzip.GzipFile(mode="wb", fileobj=self.output)
327 except:
328 self.input = open(input)
329 else:
330 self.dir = ''
331 self.input = sys.stdin
334 def lyxformat(self, format):
335 " Returns the file format representation, an integer."
336 result = format_re.match(format)
337 if result:
338 format = int(result.group(1) + result.group(2))
339 elif format == '2':
340 format = 200
341 else:
342 self.error(str(format) + ": " + "Invalid LyX file.")
344 if format in formats_list():
345 return format
347 self.error(str(format) + ": " + "Format not supported.")
348 return None
351 def read_version(self):
352 """ Searchs for clues of the LyX version used to write the
353 file, returns the most likely value, or None otherwise."""
355 for line in self.header:
356 if line[0] != "#":
357 return None
359 line = line.replace("fix",".")
360 result = original_version.match(line)
361 if result:
362 # Special know cases: reLyX and KLyX
363 if line.find("reLyX") != -1 or line.find("KLyX") != -1:
364 return "0.12"
366 res = result.group(1)
367 if not res:
368 self.warning(line)
369 #self.warning("Version %s" % result.group(1))
370 return res
371 self.warning(str(self.header[:2]))
372 return None
375 def set_version(self):
376 " Set the header with the version used."
377 self.header[0] = " ".join(["#LyX %s created this file." % version__,
378 "For more info see http://www.lyx.org/"])
379 if self.header[1][0] == '#':
380 del self.header[1]
383 def read_format(self):
384 " Read from the header the fileformat of the present LyX file."
385 for line in self.header:
386 result = fileformat.match(line)
387 if result:
388 return self.lyxformat(result.group(1))
389 else:
390 self.error("Invalid LyX File.")
391 return None
394 def set_format(self):
395 " Set the file format of the file, in the header."
396 if self.format <= 217:
397 format = str(float(self.format)/100)
398 else:
399 format = str(self.format)
400 i = find_token(self.header, "\\lyxformat", 0)
401 self.header[i] = "\\lyxformat %s" % format
404 def set_textclass(self):
405 i = find_token(self.header, "\\textclass", 0)
406 self.header[i] = "\\textclass %s" % self.textclass
409 #Note that the module will be added at the END of the extant ones
410 def add_module(self, module):
411 i = find_token(self.header, "\\begin_modules", 0)
412 if i == -1:
413 #No modules yet included
414 i = find_token(self.header, "\\textclass", 0)
415 if i == -1:
416 self.warning("Malformed LyX document: No \\textclass!!")
417 return
418 modinfo = ["\\begin_modules", module, "\\end_modules"]
419 self.header[i + 1: i + 1] = modinfo
420 return
421 j = find_token(self.header, "\\end_modules", i)
422 if j == -1:
423 self.warning("(add_module)Malformed LyX document: No \\end_modules.")
424 return
425 k = find_token(self.header, module, i)
426 if k != -1 and k < j:
427 return
428 self.header.insert(j, module)
431 def get_module_list(self):
432 i = find_token(self.header, "\\begin_modules", 0)
433 if (i == -1):
434 return []
435 j = find_token(self.header, "\\end_modules", i)
436 return self.header[i + 1 : j]
439 def set_module_list(self, mlist):
440 modbegin = find_token(self.header, "\\begin_modules", 0)
441 newmodlist = ['\\begin_modules'] + mlist + ['\\end_modules']
442 if (modbegin == -1):
443 #No modules yet included
444 tclass = find_token(self.header, "\\textclass", 0)
445 if tclass == -1:
446 self.warning("Malformed LyX document: No \\textclass!!")
447 return
448 modbegin = tclass + 1
449 self.header[modbegin:modbegin] = newmodlist
450 return
451 modend = find_token(self.header, "\\end_modules", modbegin)
452 if modend == -1:
453 self.warning("(set_module_list)Malformed LyX document: No \\end_modules.")
454 return
455 newmodlist = ['\\begin_modules'] + mlist + ['\\end_modules']
456 self.header[modbegin:modend + 1] = newmodlist
459 def set_parameter(self, param, value):
460 " Set the value of the header parameter."
461 i = find_token(self.header, '\\' + param, 0)
462 if i == -1:
463 self.warning('Parameter not found in the header: %s' % param, 3)
464 return
465 self.header[i] = '\\%s %s' % (param, str(value))
468 def is_default_layout(self, layout):
469 " Check whether a layout is the default layout of this class."
470 # FIXME: Check against the real text class default layout
471 if layout == 'Standard' or layout == self.default_layout:
472 return 1
473 return 0
476 def convert(self):
477 "Convert from current (self.format) to self.end_format."
478 mode, convertion_chain = self.chain()
479 self.warning("convertion chain: " + str(convertion_chain), 3)
481 for step in convertion_chain:
482 steps = getattr(__import__("lyx_" + step), mode)
484 self.warning("Convertion step: %s - %s" % (step, mode),
485 default_debug__ + 1)
486 if not steps:
487 self.error("The convertion to an older "
488 "format (%s) is not implemented." % self.format)
490 multi_conv = len(steps) != 1
491 for version, table in steps:
492 if multi_conv and \
493 (self.format >= version and mode == "convert") or\
494 (self.format <= version and mode == "revert"):
495 continue
497 for conv in table:
498 init_t = time.time()
499 try:
500 conv(self)
501 except:
502 self.warning("An error ocurred in %s, %s" %
503 (version, str(conv)),
504 default_debug__)
505 if not self.try_hard:
506 raise
507 self.status = 2
508 else:
509 self.warning("%lf: Elapsed time on %s" %
510 (time.time() - init_t,
511 str(conv)), default_debug__ +
513 self.format = version
514 if self.end_format == self.format:
515 return
518 def chain(self):
519 """ This is where all the decisions related with the
520 convertion are taken. It returns a list of modules needed to
521 convert the LyX file from self.format to self.end_format"""
523 self.start = self.format
524 format = self.format
525 correct_version = 0
527 for rel in format_relation:
528 if self.initial_version in rel[2]:
529 if format in rel[1]:
530 initial_step = rel[0]
531 correct_version = 1
532 break
534 if not correct_version:
535 if format <= 215:
536 self.warning("Version does not match file format, "
537 "discarding it. (Version %s, format %d)" %
538 (self.initial_version, self.format))
539 for rel in format_relation:
540 if format in rel[1]:
541 initial_step = rel[0]
542 break
543 else:
544 # This should not happen, really.
545 self.error("Format not supported.")
547 # Find the final step
548 for rel in format_relation:
549 if self.end_format in rel[1]:
550 final_step = rel[0]
551 break
552 else:
553 self.error("Format not supported.")
555 # Convertion mode, back or forth
556 steps = []
557 if (initial_step, self.start) < (final_step, self.end_format):
558 mode = "convert"
559 first_step = 1
560 for step in format_relation:
561 if initial_step <= step[0] <= final_step:
562 if first_step and len(step[1]) == 1:
563 first_step = 0
564 continue
565 steps.append(step[0])
566 else:
567 mode = "revert"
568 relation_format = format_relation[:]
569 relation_format.reverse()
570 last_step = None
572 for step in relation_format:
573 if final_step <= step[0] <= initial_step:
574 steps.append(step[0])
575 last_step = step
577 if last_step[1][-1] == self.end_format:
578 steps.pop()
580 self.warning("Convertion mode: %s\tsteps%s" %(mode, steps), 10)
581 return mode, steps
584 def get_toc(self, depth = 4):
585 " Returns the TOC of this LyX document."
586 paragraphs_filter = {'Title' : 0,'Chapter' : 1, 'Section' : 2,
587 'Subsection' : 3, 'Subsubsection': 4}
588 allowed_insets = ['Quotes']
589 allowed_parameters = ('\\paragraph_spacing', '\\noindent',
590 '\\align', '\\labelwidthstring',
591 "\\start_of_appendix", "\\leftindent")
592 sections = []
593 for section in paragraphs_filter.keys():
594 sections.append('\\begin_layout %s' % section)
596 toc_par = []
597 i = 0
598 while 1:
599 i = find_tokens(self.body, sections, i)
600 if i == -1:
601 break
603 j = find_end_of(self.body, i + 1, '\\begin_layout', '\\end_layout')
604 if j == -1:
605 self.warning('Incomplete file.', 0)
606 break
608 section = self.body[i].split()[1]
609 if section[-1] == '*':
610 section = section[:-1]
612 par = []
614 k = i + 1
615 # skip paragraph parameters
616 while not self.body[k].strip() or self.body[k].split()[0] \
617 in allowed_parameters:
618 k += 1
620 while k < j:
621 if check_token(self.body[k], '\\begin_inset'):
622 inset = self.body[k].split()[1]
623 end = find_end_of_inset(self.body, k)
624 if end == -1 or end > j:
625 self.warning('Malformed file.', 0)
627 if inset in allowed_insets:
628 par.extend(self.body[k: end+1])
629 k = end + 1
630 else:
631 par.append(self.body[k])
632 k += 1
634 # trim empty lines in the end.
635 while par and par[-1].strip() == '':
636 par.pop()
638 toc_par.append(Paragraph(section, par))
640 i = j + 1
642 return toc_par
645 class File(LyX_base):
646 " This class reads existing LyX files."
648 def __init__(self, end_format = 0, input = "", output = "", error = "",
649 debug = default_debug__, try_hard = 0, cjk_encoding = ''):
650 LyX_base.__init__(self, end_format, input, output, error,
651 debug, try_hard, cjk_encoding)
652 self.read()
655 class NewFile(LyX_base):
656 " This class is to create new LyX files."
657 def set_header(self, **params):
658 # set default values
659 self.header.extend([
660 "#LyX xxxx created this file."
661 "For more info see http://www.lyx.org/",
662 "\\lyxformat xxx",
663 "\\begin_document",
664 "\\begin_header",
665 "\\textclass article",
666 "\\language english",
667 "\\inputencoding auto",
668 "\\font_roman default",
669 "\\font_sans default",
670 "\\font_typewriter default",
671 "\\font_default_family default",
672 "\\font_sc false",
673 "\\font_osf false",
674 "\\font_sf_scale 100",
675 "\\font_tt_scale 100",
676 "\\graphics default",
677 "\\paperfontsize default",
678 "\\papersize default",
679 "\\use_geometry false",
680 "\\use_amsmath 1",
681 "\\cite_engine basic",
682 "\\use_bibtopic false",
683 "\\paperorientation portrait",
684 "\\secnumdepth 3",
685 "\\tocdepth 3",
686 "\\paragraph_separation indent",
687 "\\defskip medskip",
688 "\\quotes_language english",
689 "\\papercolumns 1",
690 "\\papersides 1",
691 "\\paperpagestyle default",
692 "\\tracking_changes false",
693 "\\end_header"])
695 self.format = get_end_format()
696 for param in params:
697 self.set_parameter(param, params[param])
700 def set_body(self, paragraphs):
701 self.body.extend(['\\begin_body',''])
703 for par in paragraphs:
704 self.body.extend(par.asLines())
706 self.body.extend(['','\\end_body', '\\end_document'])
709 class Paragraph:
710 # unfinished implementation, it is missing the Text and Insets
711 # representation.
712 " This class represents the LyX paragraphs."
713 def __init__(self, name, body=[], settings = [], child = []):
714 """ Parameters:
715 name: paragraph name.
716 body: list of lines of body text.
717 child: list of paragraphs that descend from this paragraph.
719 self.name = name
720 self.body = body
721 self.settings = settings
722 self.child = child
724 def asLines(self):
725 """ Converts the paragraph to a list of strings, representing
726 it in the LyX file."""
728 result = ['','\\begin_layout %s' % self.name]
729 result.extend(self.settings)
730 result.append('')
731 result.extend(self.body)
732 result.append('\\end_layout')
734 if not self.child:
735 return result
737 result.append('\\begin_deeper')
738 for node in self.child:
739 result.extend(node.asLines())
740 result.append('\\end_deeper')
742 return result