]> git.lyx.org Git - features.git/blob - lib/scripts/layout2layout.py
Introduce a "formatted counter" for use with formatted reference during
[features.git] / lib / scripts / layout2layout.py
1 #! /usr/bin/env python
2 # -*- coding: utf-8 -*-
3
4 # file layout2layout.py
5 # This file is part of LyX, the document processor.
6 # Licence details can be found in the file COPYING.
7
8 # author Georg Baum
9
10 # Full author contact details are available in file CREDITS
11
12 # This script will update a .layout file to current format
13
14
15 import os, re, string, sys
16
17 # Incremented to format 4, 6 April 2007, lasgouttes
18 # Introduction of generic "Provides" declaration
19
20 # Incremented to format 5, 22 August 2007 by vermeer
21 # InsetLayout material
22
23 # Incremented to format 6, 7 January 2008 by spitz
24 # Requires tag added to layout files
25
26 # Incremented to format 7, 24 March 2008 by rgh
27 # AddToPreamble tag added to layout files
28
29 # Incremented to format 8, 25 July 2008 by rgh
30 # UseModule tag added to layout files
31 # CopyStyle added to InsetLayout
32
33 # Incremented to format 9, 5 October 2008 by rgh
34 # ForcePlain and CustomPars tags added to InsetLayout
35
36 # Incremented to format 10, 6 October 2008 by rgh
37 # Change format of counters
38
39 # Incremented to format 11, 14 October 2008 by rgh
40 # Add ProvidesModule, ExcludesModule tags
41
42 # Incremented to format 12, 10 January 2009 by gb
43 # Add I18NPreamble tag
44
45 # Incremented to format 13, 5 February 2009 by rgh
46 # Add InToc tag for InsetLayout
47
48 # Incremented to format 14, 14 February 2009 by gb
49 # Rename I18NPreamble to BabelPreamble and add LangPreamble
50
51 # Incremented to format 15, 28 May 2009 by lasgouttes
52 # Add new tag OutputFormat; modules can be conditioned on feature 
53 # "from->to".
54
55 # Incremented to format 16, 5 June 2009 by rgh
56 # Add new tags for Text Class:
57 #   HTMLPreamble, HTMLAddToPreamble
58 # For Layout:
59 #   HTMLTag, HTMLAttr, HTMLLabel, HTMLLabelAttr, HTMLItem, HTMLItemAttr
60 #   HTMLStyle, and HTMLPreamble
61 # For InsetLayout:
62 #   HTMLTag, HTMLAttr, HTMLStyle, and HTMLPreamble
63 # For Floats:
64 #   HTMLType, HTMLClass, HTMLStyle
65
66 # Incremented to format 17, 12 August 2009 by rgh
67 # Add IfStyle and IfCounter tags for layout.
68
69 # Incremented to format 18, 27 October 2009 by rgh
70 # Added some new tags for HTML output.
71
72 # Incremented to format 19, 17 November 2009 by rgh
73 # Added InPreamble tag.
74
75 # Incremented to format 20, 17 December 2009 by rgh
76 # Added ContentAsLabel tag.
77
78 # Incremented to format 21, 12 January 2010 by rgh
79 # Added HTMLTocLayout and HTMLTitle tags.
80         
81 # Incremented to format 22, 20 January 2010 by rgh
82 # Added HTMLFormat tag to Counters.
83
84 # Do not forget to document format change in Customization
85 # Manual (section "Declaring a new text class").
86
87 # You might also want to consider running the
88 # development/tools/updatelayouts.sh script to update all
89 # layout files to the new format.
90
91 currentFormat = 22
92
93
94 def usage(prog_name):
95     return ("Usage: %s inputfile outputfile\n" % prog_name +
96             "or     %s <inputfile >outputfile" % prog_name)
97
98
99 def error(message):
100     sys.stderr.write(message + '\n')
101     sys.exit(1)
102
103
104 def trim_bom(line):
105     " Remove byte order mark."
106     if line[0:3] == "\357\273\277":
107         return line[3:]
108     else:
109         return line
110
111
112 def read(source):
113     " Read input file and strip lineendings."
114     lines = source.read().splitlines()
115     lines[0] = trim_bom(lines[0])
116     return lines
117
118
119 def write(output, lines):
120     " Write output file with native lineendings."
121     output.write(os.linesep.join(lines) + os.linesep)
122
123
124 # Concatenates old and new in an intelligent way:
125 # If old is wrapped in ", they are stripped. The result is wrapped in ".
126 def concatenate_label(old, new):
127     # Don't use strip as long as we support python 1.5.2
128     if old[0] == '"':
129         return old[0:-1] + new + '"'
130     else:
131         return '"' + old + new + '"'
132
133 # appends a string to a list unless it's already there
134 def addstring(s, l):
135     if l.count(s) > 0:
136         return
137     l.append(s)
138
139
140 def convert(lines):
141     " Convert to new format."
142     re_Comment = re.compile(r'^(\s*)#')
143     re_Counter = re.compile(r'\s*Counter\s*', re.IGNORECASE)
144     re_Name = re.compile(r'\s*Name\s+(\S+)\s*', re.IGNORECASE)
145     re_UseMod = re.compile(r'^\s*UseModule\s+(.*)', re.IGNORECASE)
146     re_Empty = re.compile(r'^(\s*)$')
147     re_Format = re.compile(r'^(\s*)(Format)(\s+)(\S+)', re.IGNORECASE)
148     re_Preamble = re.compile(r'^(\s*)Preamble', re.IGNORECASE)
149     re_EndPreamble = re.compile(r'^(\s*)EndPreamble', re.IGNORECASE)
150     re_LangPreamble = re.compile(r'^(\s*)LangPreamble', re.IGNORECASE)
151     re_EndLangPreamble = re.compile(r'^(\s*)EndLangPreamble', re.IGNORECASE)
152     re_BabelPreamble = re.compile(r'^(\s*)BabelPreamble', re.IGNORECASE)
153     re_EndBabelPreamble = re.compile(r'^(\s*)EndBabelPreamble', re.IGNORECASE)
154     re_MaxCounter = re.compile(r'^(\s*)(MaxCounter)(\s+)(\S+)', re.IGNORECASE)
155     re_LabelType = re.compile(r'^(\s*)(LabelType)(\s+)(\S+)', re.IGNORECASE)
156     re_LabelString = re.compile(r'^(\s*)(LabelString)(\s+)(("[^"]+")|(\S+))', re.IGNORECASE)
157     re_LabelStringAppendix = re.compile(r'^(\s*)(LabelStringAppendix)(\s+)(("[^"]+")|(\S+))', re.IGNORECASE)
158     re_LatexType = re.compile(r'^(\s*)(LatexType)(\s+)(\S+)', re.IGNORECASE)
159     re_Style = re.compile(r'^(\s*)(Style)(\s+)(\S+)', re.IGNORECASE)
160     re_CopyStyle = re.compile(r'^(\s*)(CopyStyle)(\s+)(\S+)', re.IGNORECASE)
161     re_NoStyle = re.compile(r'^(\s*)(NoStyle)(\s+)(\S+)', re.IGNORECASE)
162     re_End = re.compile(r'^(\s*)(End)(\s*)$', re.IGNORECASE)
163     re_Provides = re.compile(r'^(\s*)Provides(\S+)(\s+)(\S+)', re.IGNORECASE)
164     re_CharStyle = re.compile(r'^(\s*)CharStyle(\s+)(\S+)$', re.IGNORECASE)
165     re_AMSMaths = re.compile(r'^\s*Input ams(?:math|def)s.inc\s*')
166     re_AMSMathsPlain = re.compile(r'^\s*Input amsmaths-plain.inc\s*')
167     re_AMSMathsSeq = re.compile(r'^\s*Input amsmaths-seq.inc\s*')
168     re_TocLevel = re.compile(r'^(\s*)(TocLevel)(\s+)(\S+)', re.IGNORECASE)
169     re_I18nPreamble = re.compile(r'^(\s*)I18nPreamble', re.IGNORECASE)
170     re_EndI18nPreamble = re.compile(r'^(\s*)EndI18nPreamble', re.IGNORECASE)
171
172     # counters for sectioning styles (hardcoded in 1.3)
173     counters = {"part"          : "\\Roman{part}",
174                 "chapter"       : "\\arabic{chapter}",
175                 "section"       : "\\arabic{section}",
176                 "subsection"    : "\\arabic{section}.\\arabic{subsection}",
177                 "subsubsection" : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}",
178                 "paragraph"     : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}",
179                 "subparagraph"  : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}.\\arabic{subparagraph}"}
180
181     # counters for sectioning styles in appendix (hardcoded in 1.3)
182     appendixcounters = {"chapter"       : "\\Alph{chapter}",
183                         "section"       : "\\Alph{section}",
184                         "subsection"    : "\\arabic{section}.\\arabic{subsection}",
185                         "subsubsection" : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}",
186                         "paragraph"     : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}",
187                         "subparagraph"  : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}.\\arabic{subparagraph}"}
188
189     # Value of TocLevel for sectioning styles
190     toclevels = {"part"          : 0,
191                  "chapter"       : 0,
192                  "section"       : 1,
193                  "subsection"    : 2,
194                  "subsubsection" : 3,
195                  "paragraph"     : 4,
196                  "subparagraph"  : 5}
197
198     i = 0
199     only_comment = 1
200     counter = ""
201     toclevel = ""
202     label = ""
203     labelstring = ""
204     labelstringappendix = ""
205     space1 = ""
206     labelstring_line = -1
207     labelstringappendix_line = -1
208     labeltype_line = -1
209     latextype = ""
210     latextype_line = -1
211     style = ""
212     maxcounter = 0
213     format = 1
214     formatline = 0
215     usemodules = []
216
217     while i < len(lines):
218         # Skip comments and empty lines
219         if re_Comment.match(lines[i]) or re_Empty.match(lines[i]):
220             i += 1
221             continue
222
223         # insert file format if not already there
224         if (only_comment):
225             match = re_Format.match(lines[i])
226             if match:
227                 formatline = i
228                 format = int(match.group(4))
229                 if format > 1 and format < currentFormat:
230                     lines[i] = "Format %d" % (format + 1)
231                     only_comment = 0
232                 elif format == currentFormat:
233                     # nothing to do
234                     return format
235                 else:
236                     error('Cannot convert file format %s' % format)
237             else:
238                 lines.insert(i, "Format 2")
239                 only_comment = 0
240                 continue
241
242         # Don't get confused by LaTeX code
243         if re_Preamble.match(lines[i]):
244             i += 1
245             while i < len(lines) and not re_EndPreamble.match(lines[i]):
246                 i += 1
247             continue
248         if re_LangPreamble.match(lines[i]):
249             i += 1
250             while i < len(lines) and not re_EndLangPreamble.match(lines[i]):
251                 i += 1
252             continue
253         if re_BabelPreamble.match(lines[i]):
254             i += 1
255             while i < len(lines) and not re_EndBabelPreamble.match(lines[i]):
256                 i += 1
257             continue
258
259         # This just involved new features, not any changes to old ones
260         if format >= 14 and format <= 21:
261           i += 1
262           continue
263
264         # Rename I18NPreamble to BabelPreamble
265         if format == 13:
266             match = re_I18nPreamble.match(lines[i])
267             if match:
268                 lines[i] = match.group(1) + "BabelPreamble"
269                 i += 1
270                 match = re_EndI18nPreamble.match(lines[i])
271                 while i < len(lines) and not match:
272                     i += 1
273                     match = re_EndI18nPreamble.match(lines[i])
274                 lines[i] = match.group(1) + "EndBabelPreamble"
275                 i += 1
276                 continue
277
278         # These just involved new features, not any changes to old ones
279         if format == 11 or format == 12:
280           i += 1
281           continue
282
283         if format == 10:
284             match = re_UseMod.match(lines[i])
285             if match:
286                 module = match.group(1)
287                 lines[i] = "DefaultModule " + module
288             i += 1
289             continue
290
291         if format == 9:
292             match = re_Counter.match(lines[i])
293             if match:
294                 counterline = i
295                 i += 1
296                 while i < len(lines):
297                     namem = re_Name.match(lines[i])
298                     if namem:
299                         name = namem.group(1)
300                         lines.pop(i)
301                         lines[counterline] = "Counter %s" % name
302                         # we don't need to increment i
303                         continue
304                     endem = re_End.match(lines[i])
305                     if endem:
306                         i += 1
307                         break
308                     i += 1
309             i += 1
310             continue
311
312         if format == 8:
313             # We want to scan for ams-type includes and, if we find them,
314             # add corresponding UseModule tags to the layout.
315             match = re_AMSMaths.match(lines[i])
316             if match:
317                 addstring("theorems-ams", usemodules)
318                 addstring("theorems-ams-extended", usemodules)
319                 addstring("theorems-sec", usemodules)
320                 lines.pop(i)
321                 continue
322             match = re_AMSMathsPlain.match(lines[i])
323             if match:
324                 addstring("theorems-starred", usemodules)
325                 lines.pop(i)
326                 continue
327             match = re_AMSMathsSeq.match(lines[i])
328             if match:
329                 addstring("theorems-ams", usemodules)
330                 addstring("theorems-ams-extended", usemodules)
331                 lines.pop(i)
332                 continue
333             i += 1
334             continue
335
336         # These just involved new features, not any changes to old ones
337         if format >= 5 and format <= 7:
338           i += 1
339           continue
340
341         if format == 4:
342             # Handle conversion to long CharStyle names
343             match = re_CharStyle.match(lines[i])
344             if match:
345                 lines[i] = "InsetLayout CharStyle:%s" % (match.group(3))
346                 i += 1
347                 lines.insert(i, "\tLyXType charstyle")
348                 i += 1
349                 lines.insert(i, "")
350                 lines[i] = "\tLabelString %s" % (match.group(3))
351             i += 1
352             continue
353
354         if format == 3:
355             # convert 'providesamsmath x',  'providesmakeidx x',  'providesnatbib x',  'providesurl x' to
356             #         'provides amsmath x', 'provides makeidx x', 'provides natbib x', 'provides url x'
357             # x is either 0 or 1
358             match = re_Provides.match(lines[i])
359             if match:
360                 lines[i] = "%sProvides %s%s%s" % (match.group(1), match.group(2).lower(),
361                                                   match.group(3), match.group(4))
362             i += 1
363             continue
364
365         if format == 2:
366             caption = []
367
368             # delete caption styles
369             match = re_Style.match(lines[i])
370             if match:
371                 style = string.lower(match.group(4))
372                 if style == "caption":
373                     del lines[i]
374                     while i < len(lines) and not re_End.match(lines[i]):
375                         caption.append(lines[i])
376                         del lines[i]
377                     if i == len(lines):
378                         error('Incomplete caption style.')
379                     else:
380                         del lines[i]
381                         continue
382
383             # delete undefinition of caption styles
384             match = re_NoStyle.match(lines[i])
385             if match:
386                 style = string.lower(match.group(4))
387                 if style == "caption":
388                     del lines[i]
389                     continue
390
391             # replace the CopyStyle statement with the definition of the real
392             # style. This may result in duplicate statements, but that is OK
393             # since the second one will overwrite the first one.
394             match = re_CopyStyle.match(lines[i])
395             if match:
396                 style = string.lower(match.group(4))
397                 if style == "caption":
398                     if len(caption) > 0:
399                         lines[i:i+1] = caption
400                     else:
401                         # FIXME: This style comes from an include file, we
402                         # should replace the real style and not this default.
403                         lines[i:i+1] = ['       Margin                First_Dynamic',
404                                         '       LatexType             Command',
405                                         '       LatexName             caption',
406                                         '       NeedProtect           1',
407                                         '       LabelSep              xx',
408                                         '       ParSkip               0.4',
409                                         '       TopSep                0.5',
410                                         '       Align                 Center',
411                                         '       AlignPossible         Center',
412                                         '       LabelType             Sensitive',
413                                         '       LabelString           "Senseless!"',
414                                         '       OptionalArgs          1',
415                                         '       LabelFont',
416                                         '         Series              Bold',
417                                         '       EndFont']
418
419             i += 1
420             continue
421
422         # Delete MaxCounter and remember the value of it
423         match = re_MaxCounter.match(lines[i])
424         if match:
425             level = match.group(4)
426             if string.lower(level) == "counter_chapter":
427                 maxcounter = 0
428             elif string.lower(level) == "counter_section":
429                 maxcounter = 1
430             elif string.lower(level) == "counter_subsection":
431                 maxcounter = 2
432             elif string.lower(level) == "counter_subsubsection":
433                 maxcounter = 3
434             elif string.lower(level) == "counter_paragraph":
435                 maxcounter = 4
436             elif string.lower(level) == "counter_subparagraph":
437                 maxcounter = 5
438             elif string.lower(level) == "counter_enumi":
439                 maxcounter = 6
440             elif string.lower(level) == "counter_enumii":
441                 maxcounter = 7
442             elif string.lower(level) == "counter_enumiii":
443                 maxcounter = 8
444             del lines[i]
445             continue
446
447         # Replace line
448         #
449         # LabelType Counter_EnumI
450         #
451         # with two lines
452         #
453         # LabelType Counter
454         # LabelCounter EnumI
455         #
456         match = re_LabelType.match(lines[i])
457         if match:
458             label = match.group(4)
459             # Remember indenting space for later reuse in added lines
460             space1 = match.group(1)
461             # Remember the line for adding the LabelCounter later.
462             # We can't do it here because it could shift latextype_line etc.
463             labeltype_line = i
464             if string.lower(label[:8]) == "counter_":
465                 counter = string.lower(label[8:])
466                 lines[i] = re_LabelType.sub(r'\1\2\3Counter', lines[i])
467
468         # Remember the LabelString line
469         match = re_LabelString.match(lines[i])
470         if match:
471             labelstring = match.group(4)
472             labelstring_line = i
473
474         # Remember the LabelStringAppendix line
475         match = re_LabelStringAppendix.match(lines[i])
476         if match:
477             labelstringappendix = match.group(4)
478             labelstringappendix_line = i
479
480         # Remember the LatexType line
481         match = re_LatexType.match(lines[i])
482         if match:
483             latextype = string.lower(match.group(4))
484             latextype_line = i
485
486         # Remember the TocLevel line
487         match = re_TocLevel.match(lines[i])
488         if match:
489             toclevel = string.lower(match.group(4))
490
491         # Reset variables at the beginning of a style definition
492         match = re_Style.match(lines[i])
493         if match:
494             style = string.lower(match.group(4))
495             counter = ""
496             toclevel = ""
497             label = ""
498             space1 = ""
499             labelstring = ""
500             labelstringappendix = ""
501             labelstring_line = -1
502             labelstringappendix_line = -1
503             labeltype_line = -1
504             latextype = ""
505             latextype_line = -1
506
507         if re_End.match(lines[i]):
508
509             # Add a line "LatexType Bib_Environment" if LabelType is Bibliography
510             # (or change the existing LatexType)
511             if string.lower(label) == "bibliography":
512                 if (latextype_line < 0):
513                     lines.insert(i, "%sLatexType Bib_Environment" % space1)
514                     i += 1
515                 else:
516                     lines[latextype_line] = re_LatexType.sub(r'\1\2\3Bib_Environment', lines[latextype_line])
517
518             # Change "LabelType Static" to "LabelType Itemize" for itemize environments
519             if latextype == "item_environment" and string.lower(label) == "static":
520                 lines[labeltype_line] = re_LabelType.sub(r'\1\2\3Itemize', lines[labeltype_line])
521
522             # Change "LabelType Counter_EnumI" to "LabelType Enumerate" for enumerate environments
523             if latextype == "item_environment" and string.lower(label) == "counter_enumi":
524                 lines[labeltype_line] = re_LabelType.sub(r'\1\2\3Enumerate', lines[labeltype_line])
525                 # Don't add the LabelCounter line later
526                 counter = ""
527
528             # Replace
529             #
530             # LabelString "Chapter"
531             #
532             # with
533             #
534             # LabelString "Chapter \arabic{chapter}"
535             #
536             # if this style has a counter. Ditto for LabelStringAppendix.
537             # This emulates the hardcoded article style numbering of 1.3
538             #
539             if counter != "":
540                 if counters.has_key(style):
541                     if labelstring_line < 0:
542                         lines.insert(i, '%sLabelString "%s"' % (space1, counters[style]))
543                         i += 1
544                     else:
545                         new_labelstring = concatenate_label(labelstring, counters[style])
546                         lines[labelstring_line] = re_LabelString.sub(
547                                 r'\1\2\3%s' % new_labelstring.replace("\\", "\\\\"),
548                                 lines[labelstring_line])
549                 if appendixcounters.has_key(style):
550                     if labelstringappendix_line < 0:
551                         lines.insert(i, '%sLabelStringAppendix "%s"' % (space1, appendixcounters[style]))
552                         i += 1
553                     else:
554                         new_labelstring = concatenate_label(labelstring, appendixcounters[style])
555                         lines[labelstringappendix_line] = re_LabelStringAppendix.sub(
556                                 r'\1\2\3%s' % new_labelstring.replace("\\", "\\\\"),
557                                 lines[labelstringappendix_line])
558
559                 # Now we can safely add the LabelCounter line
560                 lines.insert(labeltype_line + 1, "%sLabelCounter %s" % (space1, counter))
561                 i += 1
562
563             # Add the TocLevel setting for sectioning styles
564             if toclevel == "" and toclevels.has_key(style) and maxcounter <= toclevels[style]:
565                 lines.insert(i, '%s\tTocLevel %d' % (space1, toclevels[style]))
566                 i += 1
567
568         i += 1
569
570     if usemodules:
571         i = formatline + 1
572         for mod in usemodules:
573             lines.insert(i, "UseModule " + mod)
574             i += 1
575
576     return format + 1
577
578
579 def main(argv):
580
581     # Open files
582     if len(argv) == 1:
583         source = sys.stdin
584         output = sys.stdout
585     elif len(argv) == 3:
586         source = open(argv[1], 'rb')
587         output = open(argv[2], 'wb')
588     else:
589         error(usage(argv[0]))
590
591     # Do the real work
592     lines = read(source)
593     format = 1
594     while (format < currentFormat):
595         format = convert(lines)
596     write(output, lines)
597
598     # Close files
599     if len(argv) == 3:
600         source.close()
601         output.close()
602
603     return 0
604
605
606 if __name__ == "__main__":
607     main(sys.argv)