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 # Incremented to format 16, 5 June 2009 by rgh
  56 # Add new tags for Text Class:
  57 #   HTML Preamble
  58 # For Layout:
  59 #   HTMLTag, HTMLAttr, HTMLLabel, HTMLLabelAttr, HTMLItem, HTMLItemAttr
  60 #   HTMLStyle, and HTMLPreamble
  61 # For InsetLayout:
  62 #   HTMLTag, HTMLAttr, HTMLStyle, and HTMLPreamble
  63 # For Floats:
  64 #   HTMLType, HTMLClass, HTMLStyle
  65 # These are still to be documented, once everything stabilizes.
  66
  67 # Do not forget to document format change in Customization
  68 # Manual (section "Declaring a new text class").
  69
  70 currentFormat = 16
  71
  72
  73 def usage(prog_name):
  74     return ("Usage: %s inputfile outputfile\n" % prog_name +
  75             "or     %s <inputfile >outputfile" % prog_name)
  76
  77
  78 def error(message):
  79     sys.stderr.write(message + '\n')
  80     sys.exit(1)
  81
  82
  83 def trim_eol(line):
  84     " Remove end of line char(s)."
  85     if line[-2:-1] == '\r':
  86         return line[:-2]
  87     elif line[-1:] == '\r' or line[-1:] == '\n':
  88         return line[:-1]
  89     else:
  90         # file with no EOL in last line
  91         return line
  92
  93
  94 def read(input):
  95     " Read input file and strip lineendings."
  96     lines = list()
  97     while 1:
  98         line = input.readline()
  99         if not line:
 100             break
 101         lines.append(trim_eol(line))
 102     return lines
 103
 104
 105 def write(output, lines):
 106     " Write output file with native lineendings."
 107     for line in lines:
 108         output.write(line + os.linesep)
 109
 110
 111 # Concatenates old and new in an intelligent way:
 112 # If old is wrapped in ", they are stripped. The result is wrapped in ".
 113 def concatenate_label(old, new):
 114     # Don't use strip as long as we support python 1.5.2
 115     if old[0] == '"':
 116         return old[0:-1] + new + '"'
 117     else:
 118         return '"' + old + new + '"'
 119
 120 # appends a string to a list unless it's already there
 121 def addstring(s, l):
 122     if l.count(s) > 0:
 123         return
 124     l.append(s)
 125
 126
 127 def convert(lines):
 128     " Convert to new format."
 129     re_Comment = re.compile(r'^(\s*)#')
 130     re_Counter = re.compile(r'\s*Counter\s*', re.IGNORECASE)
 131     re_Name = re.compile(r'\s*Name\s+(\S+)\s*', re.IGNORECASE)
 132     re_UseMod = re.compile(r'^\s*UseModule\s+(.*)', re.IGNORECASE)
 133     re_Empty = re.compile(r'^(\s*)$')
 134     re_Format = re.compile(r'^(\s*)(Format)(\s+)(\S+)', re.IGNORECASE)
 135     re_Preamble = re.compile(r'^(\s*)Preamble', re.IGNORECASE)
 136     re_EndPreamble = re.compile(r'^(\s*)EndPreamble', re.IGNORECASE)
 137     re_LangPreamble = re.compile(r'^(\s*)LangPreamble', re.IGNORECASE)
 138     re_EndLangPreamble = re.compile(r'^(\s*)EndLangPreamble', re.IGNORECASE)
 139     re_BabelPreamble = re.compile(r'^(\s*)BabelPreamble', re.IGNORECASE)
 140     re_EndBabelPreamble = re.compile(r'^(\s*)EndBabelPreamble', re.IGNORECASE)
 141     re_MaxCounter = re.compile(r'^(\s*)(MaxCounter)(\s+)(\S+)', re.IGNORECASE)
 142     re_LabelType = re.compile(r'^(\s*)(LabelType)(\s+)(\S+)', re.IGNORECASE)
 143     re_LabelString = re.compile(r'^(\s*)(LabelString)(\s+)(("[^"]+")|(\S+))', re.IGNORECASE)
 144     re_LabelStringAppendix = re.compile(r'^(\s*)(LabelStringAppendix)(\s+)(("[^"]+")|(\S+))', re.IGNORECASE)
 145     re_LatexType = re.compile(r'^(\s*)(LatexType)(\s+)(\S+)', re.IGNORECASE)
 146     re_Style = re.compile(r'^(\s*)(Style)(\s+)(\S+)', re.IGNORECASE)
 147     re_CopyStyle = re.compile(r'^(\s*)(CopyStyle)(\s+)(\S+)', re.IGNORECASE)
 148     re_NoStyle = re.compile(r'^(\s*)(NoStyle)(\s+)(\S+)', re.IGNORECASE)
 149     re_End = re.compile(r'^(\s*)(End)(\s*)$', re.IGNORECASE)
 150     re_Provides = re.compile(r'^(\s*)Provides(\S+)(\s+)(\S+)', re.IGNORECASE)
 151     re_CharStyle = re.compile(r'^(\s*)CharStyle(\s+)(\S+)$', re.IGNORECASE)
 152     re_AMSMaths = re.compile(r'^\s*Input amsmaths.inc\s*')
 153     re_AMSMathsPlain = re.compile(r'^\s*Input amsmaths-plain.inc\s*')
 154     re_AMSMathsSeq = re.compile(r'^\s*Input amsmaths-seq.inc\s*')
 155     re_TocLevel = re.compile(r'^(\s*)(TocLevel)(\s+)(\S+)', re.IGNORECASE)
 156     re_I18nPreamble = re.compile(r'^(\s*)I18nPreamble', re.IGNORECASE)
 157     re_EndI18nPreamble = re.compile(r'^(\s*)EndI18nPreamble', re.IGNORECASE)
 158
 159     # counters for sectioning styles (hardcoded in 1.3)
 160     counters = {"part"          : "\\Roman{part}",
 161                 "chapter"       : "\\arabic{chapter}",
 162                 "section"       : "\\arabic{section}",
 163                 "subsection"    : "\\arabic{section}.\\arabic{subsection}",
 164                 "subsubsection" : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}",
 165                 "paragraph"     : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}",
 166                 "subparagraph"  : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}.\\arabic{subparagraph}"}
 167
 168     # counters for sectioning styles in appendix (hardcoded in 1.3)
 169     appendixcounters = {"chapter"       : "\\Alph{chapter}",
 170                         "section"       : "\\Alph{section}",
 171                         "subsection"    : "\\arabic{section}.\\arabic{subsection}",
 172                         "subsubsection" : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}",
 173                         "paragraph"     : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}",
 174                         "subparagraph"  : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}.\\arabic{subparagraph}"}
 175
 176     # Value of TocLevel for sectioning styles
 177     toclevels = {"part"          : 0,
 178                  "chapter"       : 0,
 179                  "section"       : 1,
 180                  "subsection"    : 2,
 181                  "subsubsection" : 3,
 182                  "paragraph"     : 4,
 183                  "subparagraph"  : 5}
 184
 185     i = 0
 186     only_comment = 1
 187     counter = ""
 188     toclevel = ""
 189     label = ""
 190     labelstring = ""
 191     labelstringappendix = ""
 192     space1 = ""
 193     labelstring_line = -1
 194     labelstringappendix_line = -1
 195     labeltype_line = -1
 196     latextype = ""
 197     latextype_line = -1
 198     style = ""
 199     maxcounter = 0
 200     format = 1
 201     formatline = 0
 202     usemodules = []
 203
 204     while i < len(lines):
 205         # Skip comments and empty lines
 206         if re_Comment.match(lines[i]) or re_Empty.match(lines[i]):
 207             i += 1
 208             continue
 209
 210         # insert file format if not already there
 211         if (only_comment):
 212             match = re_Format.match(lines[i])
 213             if match:
 214                 formatline = i
 215                 format = int(match.group(4))
 216                 if format > 1 and format < currentFormat:
 217                     lines[i] = "Format %d" % (format + 1)
 218                     only_comment = 0
 219                 elif format == currentFormat:
 220                     # nothing to do
 221                     return format
 222                 else:
 223                     error('Cannot convert file format %s' % format)
 224             else:
 225                 lines.insert(i, "Format 2")
 226                 only_comment = 0
 227                 continue
 228
 229         # Don't get confused by LaTeX code
 230         if re_Preamble.match(lines[i]):
 231             i += 1
 232             while i < len(lines) and not re_EndPreamble.match(lines[i]):
 233                 i += 1
 234             continue
 235         if re_LangPreamble.match(lines[i]):
 236             i += 1
 237             while i < len(lines) and not re_EndLangPreamble.match(lines[i]):
 238                 i += 1
 239             continue
 240         if re_BabelPreamble.match(lines[i]):
 241             i += 1
 242             while i < len(lines) and not re_EndBabelPreamble.match(lines[i]):
 243                 i += 1
 244             continue
 245
 246         # This just involved new features, not any changes to old ones
 247         if format == 14 or format == 15:
 248           i += 1
 249           continue
 250
 251         # Rename I18NPreamble to BabelPreamble
 252         if format == 13:
 253             match = re_I18nPreamble.match(lines[i])
 254             if match:
 255                 lines[i] = match.group(1) + "BabelPreamble"
 256                 i += 1
 257                 match = re_EndI18nPreamble.match(lines[i])
 258                 while i < len(lines) and not match:
 259                     i += 1
 260                     match = re_EndI18nPreamble.match(lines[i])
 261                 lines[i] = match.group(1) + "EndBabelPreamble"
 262                 i += 1
 263                 continue
 264
 265         # These just involved new features, not any changes to old ones
 266         if format == 11 or format == 12:
 267           i += 1
 268           continue
 269
 270         if format == 10:
 271             match = re_UseMod.match(lines[i])
 272             if match:
 273                 module = match.group(1)
 274                 lines[i] = "DefaultModule " + module
 275             i += 1
 276             continue
 277
 278         if format == 9:
 279             match = re_Counter.match(lines[i])
 280             if match:
 281                 counterline = i
 282                 i += 1
 283                 while i < len(lines):
 284                     namem = re_Name.match(lines[i])
 285                     if namem:
 286                         name = namem.group(1)
 287                         lines.pop(i)
 288                         lines[counterline] = "Counter %s" % name
 289                         # we don't need to increment i
 290                         continue
 291                     endem = re_End.match(lines[i])
 292                     if endem:
 293                         i += 1
 294                         break
 295                     i += 1
 296             i += 1
 297             continue
 298
 299         if format == 8:
 300             # We want to scan for ams-type includes and, if we find them,
 301             # add corresponding UseModule tags to the layout.
 302             match = re_AMSMaths.match(lines[i])
 303             if match:
 304                 addstring("theorems-ams", usemodules)
 305                 addstring("theorems-ams-extended", usemodules)
 306                 addstring("theorems-sec", usemodules)
 307                 lines.pop(i)
 308                 continue
 309             match = re_AMSMathsPlain.match(lines[i])
 310             if match:
 311                 addstring("theorems-starred", usemodules)
 312                 lines.pop(i)
 313                 continue
 314             match = re_AMSMathsSeq.match(lines[i])
 315             if match:
 316                 addstring("theorems-ams", usemodules)
 317                 addstring("theorems-ams-extended", usemodules)
 318                 lines.pop(i)
 319                 continue
 320             i += 1
 321             continue
 322
 323         # These just involved new features, not any changes to old ones
 324         if format >= 5 and format <= 7:
 325           i += 1
 326           continue
 327
 328         if format == 4:
 329             # Handle conversion to long CharStyle names
 330             match = re_CharStyle.match(lines[i])
 331             if match:
 332                 lines[i] = "InsetLayout CharStyle:%s" % (match.group(3))
 333                 i += 1
 334                 lines.insert(i, "\tLyXType charstyle")
 335                 i += 1
 336                 lines.insert(i, "")
 337                 lines[i] = "\tLabelString %s" % (match.group(3))
 338             i += 1
 339             continue
 340
 341         if format == 3:
 342             # convert 'providesamsmath x',  'providesmakeidx x',  'providesnatbib x',  'providesurl x' to
 343             #         'provides amsmath x', 'provides makeidx x', 'provides natbib x', 'provides url x'
 344             # x is either 0 or 1
 345             match = re_Provides.match(lines[i])
 346             if match:
 347                 lines[i] = "%sProvides %s%s%s" % (match.group(1), match.group(2).lower(),
 348                                                   match.group(3), match.group(4))
 349             i += 1
 350             continue
 351
 352         if format == 2:
 353             caption = []
 354
 355             # delete caption styles
 356             match = re_Style.match(lines[i])
 357             if match:
 358                 style = string.lower(match.group(4))
 359                 if style == "caption":
 360                     del lines[i]
 361                     while i < len(lines) and not re_End.match(lines[i]):
 362                         caption.append(lines[i])
 363                         del lines[i]
 364                     if i == len(lines):
 365                         error('Incomplete caption style.')
 366                     else:
 367                         del lines[i]
 368                         continue
 369
 370             # delete undefinition of caption styles
 371             match = re_NoStyle.match(lines[i])
 372             if match:
 373                 style = string.lower(match.group(4))
 374                 if style == "caption":
 375                     del lines[i]
 376                     continue
 377
 378             # replace the CopyStyle statement with the definition of the real
 379             # style. This may result in duplicate statements, but that is OK
 380             # since the second one will overwrite the first one.
 381             match = re_CopyStyle.match(lines[i])
 382             if match:
 383                 style = string.lower(match.group(4))
 384                 if style == "caption":
 385                     if len(caption) > 0:
 386                         lines[i:i+1] = caption
 387                     else:
 388                         # FIXME: This style comes from an include file, we
 389                         # should replace the real style and not this default.
 390                         lines[i:i+1] = ['       Margin                First_Dynamic',
 391                                         '       LatexType             Command',
 392                                         '       LatexName             caption',
 393                                         '       NeedProtect           1',
 394                                         '       LabelSep              xx',
 395                                         '       ParSkip               0.4',
 396                                         '       TopSep                0.5',
 397                                         '       Align                 Center',
 398                                         '       AlignPossible         Center',
 399                                         '       LabelType             Sensitive',
 400                                         '       LabelString           "Senseless!"',
 401                                         '       OptionalArgs          1',
 402                                         '       LabelFont',
 403                                         '         Series              Bold',
 404                                         '       EndFont']
 405
 406             i += 1
 407             continue
 408
 409         # Delete MaxCounter and remember the value of it
 410         match = re_MaxCounter.match(lines[i])
 411         if match:
 412             level = match.group(4)
 413             if string.lower(level) == "counter_chapter":
 414                 maxcounter = 0
 415             elif string.lower(level) == "counter_section":
 416                 maxcounter = 1
 417             elif string.lower(level) == "counter_subsection":
 418                 maxcounter = 2
 419             elif string.lower(level) == "counter_subsubsection":
 420                 maxcounter = 3
 421             elif string.lower(level) == "counter_paragraph":
 422                 maxcounter = 4
 423             elif string.lower(level) == "counter_subparagraph":
 424                 maxcounter = 5
 425             elif string.lower(level) == "counter_enumi":
 426                 maxcounter = 6
 427             elif string.lower(level) == "counter_enumii":
 428                 maxcounter = 7
 429             elif string.lower(level) == "counter_enumiii":
 430                 maxcounter = 8
 431             del lines[i]
 432             continue
 433
 434         # Replace line
 435         #
 436         # LabelType Counter_EnumI
 437         #
 438         # with two lines
 439         #
 440         # LabelType Counter
 441         # LabelCounter EnumI
 442         #
 443         match = re_LabelType.match(lines[i])
 444         if match:
 445             label = match.group(4)
 446             # Remember indenting space for later reuse in added lines
 447             space1 = match.group(1)
 448             # Remember the line for adding the LabelCounter later.
 449             # We can't do it here because it could shift latextype_line etc.
 450             labeltype_line = i
 451             if string.lower(label[:8]) == "counter_":
 452                 counter = string.lower(label[8:])
 453                 lines[i] = re_LabelType.sub(r'\1\2\3Counter', lines[i])
 454
 455         # Remember the LabelString line
 456         match = re_LabelString.match(lines[i])
 457         if match:
 458             labelstring = match.group(4)
 459             labelstring_line = i
 460
 461         # Remember the LabelStringAppendix line
 462         match = re_LabelStringAppendix.match(lines[i])
 463         if match:
 464             labelstringappendix = match.group(4)
 465             labelstringappendix_line = i
 466
 467         # Remember the LatexType line
 468         match = re_LatexType.match(lines[i])
 469         if match:
 470             latextype = string.lower(match.group(4))
 471             latextype_line = i
 472
 473         # Remember the TocLevel line
 474         match = re_TocLevel.match(lines[i])
 475         if match:
 476             toclevel = string.lower(match.group(4))
 477
 478         # Reset variables at the beginning of a style definition
 479         match = re_Style.match(lines[i])
 480         if match:
 481             style = string.lower(match.group(4))
 482             counter = ""
 483             toclevel = ""
 484             label = ""
 485             space1 = ""
 486             labelstring = ""
 487             labelstringappendix = ""
 488             labelstring_line = -1
 489             labelstringappendix_line = -1
 490             labeltype_line = -1
 491             latextype = ""
 492             latextype_line = -1
 493
 494         if re_End.match(lines[i]):
 495
 496             # Add a line "LatexType Bib_Environment" if LabelType is Bibliography
 497             # (or change the existing LatexType)
 498             if string.lower(label) == "bibliography":
 499                 if (latextype_line < 0):
 500                     lines.insert(i, "%sLatexType Bib_Environment" % space1)
 501                     i += 1
 502                 else:
 503                     lines[latextype_line] = re_LatexType.sub(r'\1\2\3Bib_Environment', lines[latextype_line])
 504
 505             # Change "LabelType Static" to "LabelType Itemize" for itemize environments
 506             if latextype == "item_environment" and string.lower(label) == "static":
 507                 lines[labeltype_line] = re_LabelType.sub(r'\1\2\3Itemize', lines[labeltype_line])
 508
 509             # Change "LabelType Counter_EnumI" to "LabelType Enumerate" for enumerate environments
 510             if latextype == "item_environment" and string.lower(label) == "counter_enumi":
 511                 lines[labeltype_line] = re_LabelType.sub(r'\1\2\3Enumerate', lines[labeltype_line])
 512                 # Don't add the LabelCounter line later
 513                 counter = ""
 514
 515             # Replace
 516             #
 517             # LabelString "Chapter"
 518             #
 519             # with
 520             #
 521             # LabelString "Chapter \arabic{chapter}"
 522             #
 523             # if this style has a counter. Ditto for LabelStringAppendix.
 524             # This emulates the hardcoded article style numbering of 1.3
 525             #
 526             if counter != "":
 527                 if counters.has_key(style):
 528                     if labelstring_line < 0:
 529                         lines.insert(i, '%sLabelString "%s"' % (space1, counters[style]))
 530                         i += 1
 531                     else:
 532                         new_labelstring = concatenate_label(labelstring, counters[style])
 533                         lines[labelstring_line] = re_LabelString.sub(
 534                                 r'\1\2\3%s' % new_labelstring.replace("\\", "\\\\"),
 535                                 lines[labelstring_line])
 536                 if appendixcounters.has_key(style):
 537                     if labelstringappendix_line < 0:
 538                         lines.insert(i, '%sLabelStringAppendix "%s"' % (space1, appendixcounters[style]))
 539                         i += 1
 540                     else:
 541                         new_labelstring = concatenate_label(labelstring, appendixcounters[style])
 542                         lines[labelstringappendix_line] = re_LabelStringAppendix.sub(
 543                                 r'\1\2\3%s' % new_labelstring.replace("\\", "\\\\"),
 544                                 lines[labelstringappendix_line])
 545
 546                 # Now we can safely add the LabelCounter line
 547                 lines.insert(labeltype_line + 1, "%sLabelCounter %s" % (space1, counter))
 548                 i += 1
 549
 550             # Add the TocLevel setting for sectioning styles
 551             if toclevel == "" and toclevels.has_key(style) and maxcounter <= toclevels[style]:
 552                 lines.insert(i, '%sTocLevel %d' % (space1, toclevels[style]))
 553                 i += 1
 554
 555         i += 1
 556
 557     if usemodules:
 558         i = formatline + 1
 559         for mod in usemodules:
 560             lines.insert(i, "UseModule " + mod)
 561             i += 1
 562
 563     return format + 1
 564
 565
 566 def main(argv):
 567
 568     # Open files
 569     if len(argv) == 1:
 570         input = sys.stdin
 571         output = sys.stdout
 572     elif len(argv) == 3:
 573         input = open(argv[1], 'rb')
 574         output = open(argv[2], 'wb')
 575     else:
 576         error(usage(argv[0]))
 577
 578     # Do the real work
 579     lines = read(input)
 580     format = 1
 581     while (format < currentFormat):
 582         format = convert(lines)
 583     write(output, lines)
 584
 585     # Close files
 586     if len(argv) == 3:
 587         input.close()
 588         output.close()
 589
 590     return 0
 591
 592
 593 if __name__ == "__main__":
 594     main(sys.argv)