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