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