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 currentFormat = 13
  49
  50
  51 def usage(prog_name):
  52     return ("Usage: %s inputfile outputfile\n" % prog_name +
  53             "or     %s <inputfile >outputfile" % prog_name)
  54
  55
  56 def error(message):
  57     sys.stderr.write(message + '\n')
  58     sys.exit(1)
  59
  60
  61 def trim_eol(line):
  62     " Remove end of line char(s)."
  63     if line[-2:-1] == '\r':
  64         return line[:-2]
  65     elif line[-1:] == '\r' or line[-1:] == '\n':
  66         return line[:-1]
  67     else:
  68         # file with no EOL in last line
  69         return line
  70
  71
  72 def read(input):
  73     " Read input file and strip lineendings."
  74     lines = list()
  75     while 1:
  76         line = input.readline()
  77         if not line:
  78             break
  79         lines.append(trim_eol(line))
  80     return lines
  81
  82
  83 def write(output, lines):
  84     " Write output file with native lineendings."
  85     for line in lines:
  86         output.write(line + os.linesep)
  87
  88
  89 # Concatenates old and new in an intelligent way:
  90 # If old is wrapped in ", they are stripped. The result is wrapped in ".
  91 def concatenate_label(old, new):
  92     # Don't use strip as long as we support python 1.5.2
  93     if old[0] == '"':
  94         return old[0:-1] + new + '"'
  95     else:
  96         return '"' + old + new + '"'
  97
  98 # appends a string to a list unless it's already there
  99 def addstring(s, l):
 100     if l.count(s) > 0:
 101         return
 102     l.append(s)
 103
 104
 105 def convert(lines):
 106     " Convert to new format."
 107     re_Comment = re.compile(r'^(\s*)#')
 108     re_Counter = re.compile(r'\s*Counter\s*', re.IGNORECASE)
 109     re_Name = re.compile(r'\s*Name\s+(\S+)\s*', re.IGNORECASE)
 110     re_UseMod = re.compile(r'^\s*UseModule\s+(.*)', re.IGNORECASE)
 111     re_Empty = re.compile(r'^(\s*)$')
 112     re_Format = re.compile(r'^(\s*)(Format)(\s+)(\S+)', re.IGNORECASE)
 113     re_Preamble = re.compile(r'^(\s*)Preamble', re.IGNORECASE)
 114     re_EndPreamble = re.compile(r'^(\s*)EndPreamble', re.IGNORECASE)
 115     re_MaxCounter = re.compile(r'^(\s*)(MaxCounter)(\s+)(\S+)', re.IGNORECASE)
 116     re_LabelType = re.compile(r'^(\s*)(LabelType)(\s+)(\S+)', re.IGNORECASE)
 117     re_LabelString = re.compile(r'^(\s*)(LabelString)(\s+)(("[^"]+")|(\S+))', re.IGNORECASE)
 118     re_LabelStringAppendix = re.compile(r'^(\s*)(LabelStringAppendix)(\s+)(("[^"]+")|(\S+))', re.IGNORECASE)
 119     re_LatexType = re.compile(r'^(\s*)(LatexType)(\s+)(\S+)', re.IGNORECASE)
 120     re_Style = re.compile(r'^(\s*)(Style)(\s+)(\S+)', re.IGNORECASE)
 121     re_CopyStyle = re.compile(r'^(\s*)(CopyStyle)(\s+)(\S+)', re.IGNORECASE)
 122     re_NoStyle = re.compile(r'^(\s*)(NoStyle)(\s+)(\S+)', re.IGNORECASE)
 123     re_End = re.compile(r'^(\s*)(End)(\s*)$', re.IGNORECASE)
 124     re_Provides = re.compile(r'^(\s*)Provides(\S+)(\s+)(\S+)', re.IGNORECASE)
 125     re_CharStyle = re.compile(r'^(\s*)CharStyle(\s+)(\S+)$', re.IGNORECASE)
 126     re_AMSMaths = re.compile(r'^\s*Input amsmaths.inc\s*')
 127     re_AMSMathsPlain = re.compile(r'^\s*Input amsmaths-plain.inc\s*')
 128     re_AMSMathsSeq = re.compile(r'^\s*Input amsmaths-seq.inc\s*')
 129     re_TocLevel = re.compile(r'^(\s*)(TocLevel)(\s+)(\S+)', re.IGNORECASE)
 130
 131     # counters for sectioning styles (hardcoded in 1.3)
 132     counters = {"part"          : "\\Roman{part}",
 133                 "chapter"       : "\\arabic{chapter}",
 134                 "section"       : "\\arabic{section}",
 135                 "subsection"    : "\\arabic{section}.\\arabic{subsection}",
 136                 "subsubsection" : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}",
 137                 "paragraph"     : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}",
 138                 "subparagraph"  : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}.\\arabic{subparagraph}"}
 139
 140     # counters for sectioning styles in appendix (hardcoded in 1.3)
 141     appendixcounters = {"chapter"       : "\\Alph{chapter}",
 142                         "section"       : "\\Alph{section}",
 143                         "subsection"    : "\\arabic{section}.\\arabic{subsection}",
 144                         "subsubsection" : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}",
 145                         "paragraph"     : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}",
 146                         "subparagraph"  : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}.\\arabic{subparagraph}"}
 147
 148     # Value of TocLevel for sectioning styles
 149     toclevels = {"part"          : 0,
 150                  "chapter"       : 0,
 151                  "section"       : 1,
 152                  "subsection"    : 2,
 153                  "subsubsection" : 3,
 154                  "paragraph"     : 4,
 155                  "subparagraph"  : 5}
 156
 157     i = 0
 158     only_comment = 1
 159     counter = ""
 160     toclevel = ""
 161     label = ""
 162     labelstring = ""
 163     labelstringappendix = ""
 164     space1 = ""
 165     labelstring_line = -1
 166     labelstringappendix_line = -1
 167     labeltype_line = -1
 168     latextype = ""
 169     latextype_line = -1
 170     style = ""
 171     maxcounter = 0
 172     format = 1
 173     formatline = 0
 174     usemodules = []
 175
 176     while i < len(lines):
 177         # Skip comments and empty lines
 178         if re_Comment.match(lines[i]) or re_Empty.match(lines[i]):
 179             i += 1
 180             continue
 181
 182         # insert file format if not already there
 183         if (only_comment):
 184             match = re_Format.match(lines[i])
 185             if match:
 186                 formatline = i
 187                 format = int(match.group(4))
 188                 if format > 1 and format < currentFormat:
 189                     lines[i] = "Format %d" % (format + 1)
 190                     only_comment = 0
 191                 elif format == currentFormat:
 192                     # nothing to do
 193                     return format
 194                 else:
 195                     error('Cannot convert file format %s' % format)
 196             else:
 197                 lines.insert(i, "Format 2")
 198                 only_comment = 0
 199                 continue
 200
 201         # Don't get confused by LaTeX code
 202         if re_Preamble.match(lines[i]):
 203             i += 1
 204             while i < len(lines) and not re_EndPreamble.match(lines[i]):
 205                 i += 1
 206             continue
 207
 208         # These just involved new features, not any changes to old ones
 209         if format == 11 or format == 12:
 210           i += 1
 211           continue
 212
 213         if format == 10:
 214             match = re_UseMod.match(lines[i])
 215             if match:
 216                 module = match.group(1)
 217                 lines[i] = "DefaultModule " + module
 218             i += 1
 219             continue
 220
 221         if format == 9:
 222             match = re_Counter.match(lines[i])
 223             if match:
 224                 counterline = i
 225                 i += 1
 226                 while i < len(lines):
 227                     namem = re_Name.match(lines[i])
 228                     if namem:
 229                         name = namem.group(1)
 230                         lines.pop(i)
 231                         lines[counterline] = "Counter %s" % name
 232                         # we don't need to increment i
 233                         continue
 234                     endem = re_End.match(lines[i])
 235                     if endem:
 236                         i += 1
 237                         break
 238                     i += 1
 239             i += 1
 240             continue
 241
 242         if format == 8:
 243             # We want to scan for ams-type includes and, if we find them,
 244             # add corresponding UseModule tags to the layout.
 245             match = re_AMSMaths.match(lines[i])
 246             if match:
 247                 addstring("theorems-ams", usemodules)
 248                 addstring("theorems-ams-extended", usemodules)
 249                 addstring("theorems-sec", usemodules)
 250                 lines.pop(i)
 251                 continue
 252             match = re_AMSMathsPlain.match(lines[i])
 253             if match:
 254                 addstring("theorems-starred", usemodules)
 255                 lines.pop(i)
 256                 continue
 257             match = re_AMSMathsSeq.match(lines[i])
 258             if match:
 259                 addstring("theorems-ams", usemodules)
 260                 addstring("theorems-ams-extended", usemodules)
 261                 lines.pop(i)
 262                 continue
 263             i += 1
 264             continue
 265
 266         # These just involved new features, not any changes to old ones
 267         if format >= 5 and format <= 7:
 268           i += 1
 269           continue
 270
 271         if format == 4:
 272             # Handle conversion to long CharStyle names
 273             match = re_CharStyle.match(lines[i])
 274             if match:
 275                 lines[i] = "InsetLayout CharStyle:%s" % (match.group(3))
 276                 i += 1
 277                 lines.insert(i, "\tLyXType charstyle")
 278                 i += 1
 279                 lines.insert(i, "")
 280                 lines[i] = "\tLabelString %s" % (match.group(3))
 281             i += 1
 282             continue
 283
 284         if format == 3:
 285             # convert 'providesamsmath x',  'providesmakeidx x',  'providesnatbib x',  'providesurl x' to
 286             #         'provides amsmath x', 'provides makeidx x', 'provides natbib x', 'provides url x'
 287             # x is either 0 or 1
 288             match = re_Provides.match(lines[i])
 289             if match:
 290                 lines[i] = "%sProvides %s%s%s" % (match.group(1), match.group(2).lower(),
 291                                                   match.group(3), match.group(4))
 292             i += 1
 293             continue
 294
 295         if format == 2:
 296             caption = []
 297
 298             # delete caption styles
 299             match = re_Style.match(lines[i])
 300             if match:
 301                 style = string.lower(match.group(4))
 302                 if style == "caption":
 303                     del lines[i]
 304                     while i < len(lines) and not re_End.match(lines[i]):
 305                         caption.append(lines[i])
 306                         del lines[i]
 307                     if i == len(lines):
 308                         error('Incomplete caption style.')
 309                     else:
 310                         del lines[i]
 311                         continue
 312
 313             # delete undefinition of caption styles
 314             match = re_NoStyle.match(lines[i])
 315             if match:
 316                 style = string.lower(match.group(4))
 317                 if style == "caption":
 318                     del lines[i]
 319                     continue
 320
 321             # replace the CopyStyle statement with the definition of the real
 322             # style. This may result in duplicate statements, but that is OK
 323             # since the second one will overwrite the first one.
 324             match = re_CopyStyle.match(lines[i])
 325             if match:
 326                 style = string.lower(match.group(4))
 327                 if style == "caption":
 328                     if len(caption) > 0:
 329                         lines[i:i+1] = caption
 330                     else:
 331                         # FIXME: This style comes from an include file, we
 332                         # should replace the real style and not this default.
 333                         lines[i:i+1] = ['       Margin                First_Dynamic',
 334                                         '       LatexType             Command',
 335                                         '       LatexName             caption',
 336                                         '       NeedProtect           1',
 337                                         '       LabelSep              xx',
 338                                         '       ParSkip               0.4',
 339                                         '       TopSep                0.5',
 340                                         '       Align                 Center',
 341                                         '       AlignPossible         Center',
 342                                         '       LabelType             Sensitive',
 343                                         '       LabelString           "Senseless!"',
 344                                         '       OptionalArgs          1',
 345                                         '       LabelFont',
 346                                         '         Series              Bold',
 347                                         '       EndFont']
 348
 349             i += 1
 350             continue
 351
 352         # Delete MaxCounter and remember the value of it
 353         match = re_MaxCounter.match(lines[i])
 354         if match:
 355             level = match.group(4)
 356             if string.lower(level) == "counter_chapter":
 357                 maxcounter = 0
 358             elif string.lower(level) == "counter_section":
 359                 maxcounter = 1
 360             elif string.lower(level) == "counter_subsection":
 361                 maxcounter = 2
 362             elif string.lower(level) == "counter_subsubsection":
 363                 maxcounter = 3
 364             elif string.lower(level) == "counter_paragraph":
 365                 maxcounter = 4
 366             elif string.lower(level) == "counter_subparagraph":
 367                 maxcounter = 5
 368             elif string.lower(level) == "counter_enumi":
 369                 maxcounter = 6
 370             elif string.lower(level) == "counter_enumii":
 371                 maxcounter = 7
 372             elif string.lower(level) == "counter_enumiii":
 373                 maxcounter = 8
 374             del lines[i]
 375             continue
 376
 377         # Replace line
 378         #
 379         # LabelType Counter_EnumI
 380         #
 381         # with two lines
 382         #
 383         # LabelType Counter
 384         # LabelCounter EnumI
 385         #
 386         match = re_LabelType.match(lines[i])
 387         if match:
 388             label = match.group(4)
 389             # Remember indenting space for later reuse in added lines
 390             space1 = match.group(1)
 391             # Remember the line for adding the LabelCounter later.
 392             # We can't do it here because it could shift latextype_line etc.
 393             labeltype_line = i
 394             if string.lower(label[:8]) == "counter_":
 395                 counter = string.lower(label[8:])
 396                 lines[i] = re_LabelType.sub(r'\1\2\3Counter', lines[i])
 397
 398         # Remember the LabelString line
 399         match = re_LabelString.match(lines[i])
 400         if match:
 401             labelstring = match.group(4)
 402             labelstring_line = i
 403
 404         # Remember the LabelStringAppendix line
 405         match = re_LabelStringAppendix.match(lines[i])
 406         if match:
 407             labelstringappendix = match.group(4)
 408             labelstringappendix_line = i
 409
 410         # Remember the LatexType line
 411         match = re_LatexType.match(lines[i])
 412         if match:
 413             latextype = string.lower(match.group(4))
 414             latextype_line = i
 415
 416         # Remember the TocLevel line
 417         match = re_TocLevel.match(lines[i])
 418         if match:
 419             toclevel = string.lower(match.group(4))
 420
 421         # Reset variables at the beginning of a style definition
 422         match = re_Style.match(lines[i])
 423         if match:
 424             style = string.lower(match.group(4))
 425             counter = ""
 426             toclevel = ""
 427             label = ""
 428             space1 = ""
 429             labelstring = ""
 430             labelstringappendix = ""
 431             labelstring_line = -1
 432             labelstringappendix_line = -1
 433             labeltype_line = -1
 434             latextype = ""
 435             latextype_line = -1
 436
 437         if re_End.match(lines[i]):
 438
 439             # Add a line "LatexType Bib_Environment" if LabelType is Bibliography
 440             # (or change the existing LatexType)
 441             if string.lower(label) == "bibliography":
 442                 if (latextype_line < 0):
 443                     lines.insert(i, "%sLatexType Bib_Environment" % space1)
 444                     i += 1
 445                 else:
 446                     lines[latextype_line] = re_LatexType.sub(r'\1\2\3Bib_Environment', lines[latextype_line])
 447
 448             # Change "LabelType Static" to "LabelType Itemize" for itemize environments
 449             if latextype == "item_environment" and string.lower(label) == "static":
 450                 lines[labeltype_line] = re_LabelType.sub(r'\1\2\3Itemize', lines[labeltype_line])
 451
 452             # Change "LabelType Counter_EnumI" to "LabelType Enumerate" for enumerate environments
 453             if latextype == "item_environment" and string.lower(label) == "counter_enumi":
 454                 lines[labeltype_line] = re_LabelType.sub(r'\1\2\3Enumerate', lines[labeltype_line])
 455                 # Don't add the LabelCounter line later
 456                 counter = ""
 457
 458             # Replace
 459             #
 460             # LabelString "Chapter"
 461             #
 462             # with
 463             #
 464             # LabelString "Chapter \arabic{chapter}"
 465             #
 466             # if this style has a counter. Ditto for LabelStringAppendix.
 467             # This emulates the hardcoded article style numbering of 1.3
 468             #
 469             if counter != "":
 470                 if counters.has_key(style):
 471                     if labelstring_line < 0:
 472                         lines.insert(i, '%sLabelString "%s"' % (space1, counters[style]))
 473                         i += 1
 474                     else:
 475                         new_labelstring = concatenate_label(labelstring, counters[style])
 476                         lines[labelstring_line] = re_LabelString.sub(
 477                                 r'\1\2\3%s' % new_labelstring.replace("\\", "\\\\"),
 478                                 lines[labelstring_line])
 479                 if appendixcounters.has_key(style):
 480                     if labelstringappendix_line < 0:
 481                         lines.insert(i, '%sLabelStringAppendix "%s"' % (space1, appendixcounters[style]))
 482                         i += 1
 483                     else:
 484                         new_labelstring = concatenate_label(labelstring, appendixcounters[style])
 485                         lines[labelstringappendix_line] = re_LabelStringAppendix.sub(
 486                                 r'\1\2\3%s' % new_labelstring.replace("\\", "\\\\"),
 487                                 lines[labelstringappendix_line])
 488
 489                 # Now we can safely add the LabelCounter line
 490                 lines.insert(labeltype_line + 1, "%sLabelCounter %s" % (space1, counter))
 491                 i += 1
 492
 493             # Add the TocLevel setting for sectioning styles
 494             if toclevel == "" and toclevels.has_key(style) and maxcounter <= toclevels[style]:
 495                 lines.insert(i, '%sTocLevel %d' % (space1, toclevels[style]))
 496                 i += 1
 497
 498         i += 1
 499
 500     if usemodules:
 501         i = formatline + 1
 502         for mod in usemodules:
 503             lines.insert(i, "UseModule " + mod)
 504             i += 1
 505
 506     return format + 1
 507
 508
 509 def main(argv):
 510
 511     # Open files
 512     if len(argv) == 1:
 513         input = sys.stdin
 514         output = sys.stdout
 515     elif len(argv) == 3:
 516         input = open(argv[1], 'rb')
 517         output = open(argv[2], 'wb')
 518     else:
 519         error(usage(argv[0]))
 520
 521     # Do the real work
 522     lines = read(input)
 523     format = 1
 524     while (format < currentFormat):
 525         format = convert(lines)
 526     write(output, lines)
 527
 528     # Close files
 529     if len(argv) == 3:
 530         input.close()
 531         output.close()
 532
 533     return 0
 534
 535
 536 if __name__ == "__main__":
 537     main(sys.argv)