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