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 format 6
  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 currentFormat = 11
  43
  44
  45 def usage(prog_name):
  46     return ("Usage: %s inputfile outputfile\n" % prog_name +
  47             "or     %s <inputfile >outputfile" % prog_name)
  48
  49
  50 def error(message):
  51     sys.stderr.write(message + '\n')
  52     sys.exit(1)
  53
  54
  55 def trim_eol(line):
  56     " Remove end of line char(s)."
  57     if line[-2:-1] == '\r':
  58         return line[:-2]
  59     elif line[-1:] == '\r' or line[-1:] == '\n':
  60         return line[:-1]
  61     else:
  62         # file with no EOL in last line
  63         return line
  64
  65
  66 def read(input):
  67     " Read input file and strip lineendings."
  68     lines = list()
  69     while 1:
  70         line = input.readline()
  71         if not line:
  72             break
  73         lines.append(trim_eol(line))
  74     return lines
  75
  76
  77 def write(output, lines):
  78     " Write output file with native lineendings."
  79     for line in lines:
  80         output.write(line + os.linesep)
  81
  82
  83 # Concatenates old and new in an intelligent way:
  84 # If old is wrapped in ", they are stripped. The result is wrapped in ".
  85 def concatenate_label(old, new):
  86     # Don't use strip as long as we support python 1.5.2
  87     if old[0] == '"':
  88         return old[0:-1] + new + '"'
  89     else:
  90         return '"' + old + new + '"'
  91
  92 # appends a string to a list unless it's already there
  93 def addstring(s, l):
  94     if l.count(s) > 0:
  95         return
  96     l.append(s)
  97
  98
  99 def convert(lines):
 100     " Convert to new format."
 101     re_Comment = re.compile(r'^(\s*)#')
 102     re_Counter = re.compile(r'\s*Counter\s*', re.IGNORECASE)
 103     re_Name = re.compile(r'\s*Name\s+(\S+)\s*', re.IGNORECASE)
 104     re_UseMod = re.compile(r'^\s*UseModule\s+(.*)', re.IGNORECASE)
 105     re_Empty = re.compile(r'^(\s*)$')
 106     re_Format = re.compile(r'^(\s*)(Format)(\s+)(\S+)', re.IGNORECASE)
 107     re_Preamble = re.compile(r'^(\s*)Preamble', re.IGNORECASE)
 108     re_EndPreamble = re.compile(r'^(\s*)EndPreamble', re.IGNORECASE)
 109     re_MaxCounter = re.compile(r'^(\s*)(MaxCounter)(\s+)(\S+)', re.IGNORECASE)
 110     re_LabelType = re.compile(r'^(\s*)(LabelType)(\s+)(\S+)', re.IGNORECASE)
 111     re_LabelString = re.compile(r'^(\s*)(LabelString)(\s+)(("[^"]+")|(\S+))', re.IGNORECASE)
 112     re_LabelStringAppendix = re.compile(r'^(\s*)(LabelStringAppendix)(\s+)(("[^"]+")|(\S+))', re.IGNORECASE)
 113     re_LatexType = re.compile(r'^(\s*)(LatexType)(\s+)(\S+)', re.IGNORECASE)
 114     re_Style = re.compile(r'^(\s*)(Style)(\s+)(\S+)', re.IGNORECASE)
 115     re_CopyStyle = re.compile(r'^(\s*)(CopyStyle)(\s+)(\S+)', re.IGNORECASE)
 116     re_NoStyle = re.compile(r'^(\s*)(NoStyle)(\s+)(\S+)', re.IGNORECASE)
 117     re_End = re.compile(r'^(\s*)(End)(\s*)$', re.IGNORECASE)
 118     re_Provides = re.compile(r'^(\s*)Provides(\S+)(\s+)(\S+)', re.IGNORECASE)
 119     re_CharStyle = re.compile(r'^(\s*)CharStyle(\s+)(\S+)$', re.IGNORECASE)
 120     re_AMSMaths = re.compile(r'^\s*Input amsmaths.inc\s*')
 121     re_AMSMathsPlain = re.compile(r'^\s*Input amsmaths-plain.inc\s*')
 122     re_AMSMathsSeq = re.compile(r'^\s*Input amsmaths-seq.inc\s*')
 123
 124     # counters for sectioning styles (hardcoded in 1.3)
 125     counters = {"part"          : "\\Roman{part}",
 126                 "chapter"       : "\\arabic{chapter}",
 127                 "section"       : "\\arabic{section}",
 128                 "subsection"    : "\\arabic{section}.\\arabic{subsection}",
 129                 "subsubsection" : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}",
 130                 "paragraph"     : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}",
 131                 "subparagraph"  : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}.\\arabic{subparagraph}"}
 132
 133     # counters for sectioning styles in appendix (hardcoded in 1.3)
 134     appendixcounters = {"chapter"       : "\\Alph{chapter}",
 135                         "section"       : "\\Alph{section}",
 136                         "subsection"    : "\\arabic{section}.\\arabic{subsection}",
 137                         "subsubsection" : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}",
 138                         "paragraph"     : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}",
 139                         "subparagraph"  : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}.\\arabic{subparagraph}"}
 140
 141     # Value of TocLevel for sectioning styles
 142     toclevels = {"part"          : 0,
 143                  "chapter"       : 0,
 144                  "section"       : 1,
 145                  "subsection"    : 2,
 146                  "subsubsection" : 3,
 147                  "paragraph"     : 4,
 148                  "subparagraph"  : 5}
 149
 150     i = 0
 151     only_comment = 1
 152     counter = ""
 153     label = ""
 154     labelstring = ""
 155     labelstringappendix = ""
 156     space1 = ""
 157     labelstring_line = -1
 158     labelstringappendix_line = -1
 159     labeltype_line = -1
 160     latextype = ""
 161     latextype_line = -1
 162     style = ""
 163     maxcounter = 0
 164     format = 1
 165     formatline = 0
 166     usemodules = []
 167
 168     while i < len(lines):
 169         # Skip comments and empty lines
 170         if re_Comment.match(lines[i]) or re_Empty.match(lines[i]):
 171             i += 1
 172             continue
 173
 174         # insert file format if not already there
 175         if (only_comment):
 176             match = re_Format.match(lines[i])
 177             if match:
 178                 formatline = i
 179                 format = int(match.group(4))
 180                 if format > 1 and format < currentFormat:
 181                     lines[i] = "Format %d" % (format + 1)
 182                     only_comment = 0
 183                 elif format == currentFormat:
 184                     # nothing to do
 185                     return format
 186                 else:
 187                     error('Cannot convert file format %s' % format)
 188             else:
 189                 lines.insert(i, "Format 2")
 190                 only_comment = 0
 191                 continue
 192
 193         # Don't get confused by LaTeX code
 194         if re_Preamble.match(lines[i]):
 195             i += 1
 196             while i < len(lines) and not re_EndPreamble.match(lines[i]):
 197                 i += 1
 198             continue
 199
 200         if format == 10:
 201             match = re_UseMod.match(lines[i])
 202             if match:
 203                 module = match.group(1)
 204                 lines[i] = "DefaultModule " + module
 205             i += 1
 206             continue
 207
 208         if format == 9:
 209             match = re_Counter.match(lines[i])
 210             if match:
 211                 counterline = i
 212                 i += 1
 213                 while i < len(lines):
 214                     namem = re_Name.match(lines[i])
 215                     if namem:
 216                         name = namem.group(1)
 217                         lines.pop(i)
 218                         lines[counterline] = "Counter %s" % name
 219                         # we don't need to increment i
 220                         continue
 221                     endem = re_End.match(lines[i])
 222                     if endem:
 223                         i += 1
 224                         break
 225                     i += 1
 226             i += 1
 227             continue
 228
 229         if format == 8:
 230             # We want to scan for ams-type includes and, if we find them,
 231             # add corresponding UseModule tags to the layout.
 232             match = re_AMSMaths.match(lines[i])
 233             if match:
 234                 addstring("theorems-ams", usemodules)
 235                 addstring("theorems-ams-extended", usemodules)
 236                 addstring("theorems-sec", usemodules)
 237                 lines.pop(i)
 238                 continue
 239             match = re_AMSMathsPlain.match(lines[i])
 240             if match:
 241                 addstring("theorems-starred", usemodules)
 242                 lines.pop(i)
 243                 continue
 244             match = re_AMSMathsSeq.match(lines[i])
 245             if match:
 246                 addstring("theorems-ams", usemodules)
 247                 addstring("theorems-ams-extended", usemodules)
 248                 lines.pop(i)
 249                 continue
 250             i += 1
 251             continue
 252
 253         # These just involved new features, not any changes to old ones
 254         if format >= 5 and format <= 7:
 255           i += 1
 256           continue
 257
 258         if format == 4:
 259             # Handle conversion to long CharStyle names
 260             match = re_CharStyle.match(lines[i])
 261             if match:
 262                 lines[i] = "InsetLayout CharStyle:%s" % (match.group(3))
 263                 i += 1
 264                 lines.insert(i, "\tLyXType charstyle")
 265                 i += 1
 266                 lines.insert(i, "")
 267                 lines[i] = "\tLabelString %s" % (match.group(3))
 268             i += 1
 269             continue
 270
 271         if format == 3:
 272             # convert 'providesamsmath x',  'providesmakeidx x',  'providesnatbib x',  'providesurl x' to
 273             #         'provides amsmath x', 'provides makeidx x', 'provides natbib x', 'provides url x'
 274             # x is either 0 or 1
 275             match = re_Provides.match(lines[i])
 276             if match:
 277                 lines[i] = "%sProvides %s%s%s" % (match.group(1), match.group(2).lower(),
 278                                                   match.group(3), match.group(4))
 279             i += 1
 280             continue
 281
 282         if format == 2:
 283             caption = []
 284
 285             # delete caption styles
 286             match = re_Style.match(lines[i])
 287             if match:
 288                 style = string.lower(match.group(4))
 289                 if style == "caption":
 290                     del lines[i]
 291                     while i < len(lines) and not re_End.match(lines[i]):
 292                         caption.append(lines[i])
 293                         del lines[i]
 294                     if i == len(lines):
 295                         error('Incomplete caption style.')
 296                     else:
 297                         del lines[i]
 298                         continue
 299
 300             # delete undefinition of caption styles
 301             match = re_NoStyle.match(lines[i])
 302             if match:
 303                 style = string.lower(match.group(4))
 304                 if style == "caption":
 305                     del lines[i]
 306                     continue
 307
 308             # replace the CopyStyle statement with the definition of the real
 309             # style. This may result in duplicate statements, but that is OK
 310             # since the second one will overwrite the first one.
 311             match = re_CopyStyle.match(lines[i])
 312             if match:
 313                 style = string.lower(match.group(4))
 314                 if style == "caption":
 315                     if len(caption) > 0:
 316                         lines[i:i+1] = caption
 317                     else:
 318                         # FIXME: This style comes from an include file, we
 319                         # should replace the real style and not this default.
 320                         lines[i:i+1] = ['       Margin                First_Dynamic',
 321                                         '       LatexType             Command',
 322                                         '       LatexName             caption',
 323                                         '       NeedProtect           1',
 324                                         '       LabelSep              xx',
 325                                         '       ParSkip               0.4',
 326                                         '       TopSep                0.5',
 327                                         '       Align                 Center',
 328                                         '       AlignPossible         Center',
 329                                         '       LabelType             Sensitive',
 330                                         '       LabelString           "Senseless!"',
 331                                         '       OptionalArgs          1',
 332                                         '       LabelFont',
 333                                         '         Series              Bold',
 334                                         '       EndFont']
 335
 336             i += 1
 337             continue
 338
 339         # Delete MaxCounter and remember the value of it
 340         match = re_MaxCounter.match(lines[i])
 341         if match:
 342             level = match.group(4)
 343             if string.lower(level) == "counter_chapter":
 344                 maxcounter = 0
 345             elif string.lower(level) == "counter_section":
 346                 maxcounter = 1
 347             elif string.lower(level) == "counter_subsection":
 348                 maxcounter = 2
 349             elif string.lower(level) == "counter_subsubsection":
 350                 maxcounter = 3
 351             elif string.lower(level) == "counter_paragraph":
 352                 maxcounter = 4
 353             elif string.lower(level) == "counter_subparagraph":
 354                 maxcounter = 5
 355             elif string.lower(level) == "counter_enumi":
 356                 maxcounter = 6
 357             elif string.lower(level) == "counter_enumii":
 358                 maxcounter = 7
 359             elif string.lower(level) == "counter_enumiii":
 360                 maxcounter = 8
 361             del lines[i]
 362             continue
 363
 364         # Replace line
 365         #
 366         # LabelType Counter_EnumI
 367         #
 368         # with two lines
 369         #
 370         # LabelType Counter
 371         # LabelCounter EnumI
 372         #
 373         match = re_LabelType.match(lines[i])
 374         if match:
 375             label = match.group(4)
 376             # Remember indenting space for later reuse in added lines
 377             space1 = match.group(1)
 378             # Remember the line for adding the LabelCounter later.
 379             # We can't do it here because it could shift latextype_line etc.
 380             labeltype_line = i
 381             if string.lower(label[:8]) == "counter_":
 382                 counter = string.lower(label[8:])
 383                 lines[i] = re_LabelType.sub(r'\1\2\3Counter', lines[i])
 384
 385         # Remember the LabelString line
 386         match = re_LabelString.match(lines[i])
 387         if match:
 388             labelstring = match.group(4)
 389             labelstring_line = i
 390
 391         # Remember the LabelStringAppendix line
 392         match = re_LabelStringAppendix.match(lines[i])
 393         if match:
 394             labelstringappendix = match.group(4)
 395             labelstringappendix_line = i
 396
 397         # Remember the LatexType line
 398         match = re_LatexType.match(lines[i])
 399         if match:
 400             latextype = string.lower(match.group(4))
 401             latextype_line = i
 402
 403         # Reset variables at the beginning of a style definition
 404         match = re_Style.match(lines[i])
 405         if match:
 406             style = string.lower(match.group(4))
 407             counter = ""
 408             label = ""
 409             space1 = ""
 410             labelstring = ""
 411             labelstringappendix = ""
 412             labelstring_line = -1
 413             labelstringappendix_line = -1
 414             labeltype_line = -1
 415             latextype = ""
 416             latextype_line = -1
 417
 418         if re_End.match(lines[i]):
 419
 420             # Add a line "LatexType Bib_Environment" if LabelType is Bibliography
 421             # (or change the existing LatexType)
 422             if string.lower(label) == "bibliography":
 423                 if (latextype_line < 0):
 424                     lines.insert(i, "%sLatexType Bib_Environment" % space1)
 425                     i += 1
 426                 else:
 427                     lines[latextype_line] = re_LatexType.sub(r'\1\2\3Bib_Environment', lines[latextype_line])
 428
 429             # Change "LabelType Static" to "LabelType Itemize" for itemize environments
 430             if latextype == "item_environment" and string.lower(label) == "static":
 431                 lines[labeltype_line] = re_LabelType.sub(r'\1\2\3Itemize', lines[labeltype_line])
 432
 433             # Change "LabelType Counter_EnumI" to "LabelType Enumerate" for enumerate environments
 434             if latextype == "item_environment" and string.lower(label) == "counter_enumi":
 435                 lines[labeltype_line] = re_LabelType.sub(r'\1\2\3Enumerate', lines[labeltype_line])
 436                 # Don't add the LabelCounter line later
 437                 counter = ""
 438
 439             # Replace
 440             #
 441             # LabelString "Chapter"
 442             #
 443             # with
 444             #
 445             # LabelString "Chapter \arabic{chapter}"
 446             #
 447             # if this style has a counter. Ditto for LabelStringAppendix.
 448             # This emulates the hardcoded article style numbering of 1.3
 449             #
 450             if counter != "":
 451                 if counters.has_key(style):
 452                     if labelstring_line < 0:
 453                         lines.insert(i, '%sLabelString "%s"' % (space1, counters[style]))
 454                         i += 1
 455                     else:
 456                         new_labelstring = concatenate_label(labelstring, counters[style])
 457                         lines[labelstring_line] = re_LabelString.sub(
 458                                 r'\1\2\3%s' % new_labelstring.replace("\\", "\\\\"),
 459                                 lines[labelstring_line])
 460                 if appendixcounters.has_key(style):
 461                     if labelstringappendix_line < 0:
 462                         lines.insert(i, '%sLabelStringAppendix "%s"' % (space1, appendixcounters[style]))
 463                         i += 1
 464                     else:
 465                         new_labelstring = concatenate_label(labelstring, appendixcounters[style])
 466                         lines[labelstringappendix_line] = re_LabelStringAppendix.sub(
 467                                 r'\1\2\3%s' % new_labelstring.replace("\\", "\\\\"),
 468                                 lines[labelstringappendix_line])
 469
 470                 # Now we can safely add the LabelCounter line
 471                 lines.insert(labeltype_line + 1, "%sLabelCounter %s" % (space1, counter))
 472                 i += 1
 473
 474             # Add the TocLevel setting for sectioning styles
 475             if toclevels.has_key(style) and maxcounter <= toclevels[style]:
 476                 lines.insert(i, '%sTocLevel %d' % (space1, toclevels[style]))
 477                 i += 1
 478
 479         i += 1
 480
 481     if usemodules:
 482         i = formatline + 1
 483         for mod in usemodules:
 484             lines.insert(i, "UseModule " + mod)
 485             i += 1
 486
 487     return format + 1
 488
 489
 490 def main(argv):
 491
 492     # Open files
 493     if len(argv) == 1:
 494         input = sys.stdin
 495         output = sys.stdout
 496     elif len(argv) == 3:
 497         input = open(argv[1], 'rb')
 498         output = open(argv[2], 'wb')
 499     else:
 500         error(usage(argv[0]))
 501
 502     # Do the real work
 503     lines = read(input)
 504     format = 1
 505     while (format < currentFormat):
 506         format = convert(lines)
 507     write(output, lines)
 508
 509     # Close files
 510     if len(argv) == 3:
 511         input.close()
 512         output.close()
 513
 514     return 0
 515
 516
 517 if __name__ == "__main__":
 518     main(sys.argv)