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 currentFormat = 10
  40
  41
  42 def usage(prog_name):
  43     return ("Usage: %s inputfile outputfile\n" % prog_name +
  44             "or     %s <inputfile >outputfile" % prog_name)
  45
  46
  47 def error(message):
  48     sys.stderr.write(message + '\n')
  49     sys.exit(1)
  50
  51
  52 def trim_eol(line):
  53     " Remove end of line char(s)."
  54     if line[-2:-1] == '\r':
  55         return line[:-2]
  56     elif line[-1:] == '\r' or line[-1:] == '\n':
  57         return line[:-1]
  58     else:
  59         # file with no EOL in last line
  60         return line
  61
  62
  63 def read(input):
  64     " Read input file and strip lineendings."
  65     lines = list()
  66     while 1:
  67         line = input.readline()
  68         if not line:
  69             break
  70         lines.append(trim_eol(line))
  71     return lines
  72
  73
  74 def write(output, lines):
  75     " Write output file with native lineendings."
  76     for line in lines:
  77         output.write(line + os.linesep)
  78
  79
  80 # Concatenates old and new in an intelligent way:
  81 # If old is wrapped in ", they are stripped. The result is wrapped in ".
  82 def concatenate_label(old, new):
  83     # Don't use strip as long as we support python 1.5.2
  84     if old[0] == '"':
  85         return old[0:-1] + new + '"'
  86     else:
  87         return '"' + old + new + '"'
  88
  89 # appends a string to a list unless it's already there
  90 def addstring(s, l):
  91     if l.count(s) > 0:
  92         return
  93     l.append(s)
  94
  95
  96 def convert(lines):
  97     " Convert to new format."
  98     re_Comment = re.compile(r'^(\s*)#')
  99     re_Counter = re.compile(r'\s*Counter\s*', re.IGNORECASE)
 100     re_Name = re.compile(r'\s*Name\s+(\S+)\s*', re.IGNORECASE)
 101     re_Empty = re.compile(r'^(\s*)$')
 102     re_Format = re.compile(r'^(\s*)(Format)(\s+)(\S+)', re.IGNORECASE)
 103     re_Preamble = re.compile(r'^(\s*)Preamble', re.IGNORECASE)
 104     re_EndPreamble = re.compile(r'^(\s*)EndPreamble', re.IGNORECASE)
 105     re_MaxCounter = re.compile(r'^(\s*)(MaxCounter)(\s+)(\S+)', re.IGNORECASE)
 106     re_LabelType = re.compile(r'^(\s*)(LabelType)(\s+)(\S+)', re.IGNORECASE)
 107     re_LabelString = re.compile(r'^(\s*)(LabelString)(\s+)(("[^"]+")|(\S+))', re.IGNORECASE)
 108     re_LabelStringAppendix = re.compile(r'^(\s*)(LabelStringAppendix)(\s+)(("[^"]+")|(\S+))', re.IGNORECASE)
 109     re_LatexType = re.compile(r'^(\s*)(LatexType)(\s+)(\S+)', re.IGNORECASE)
 110     re_Style = re.compile(r'^(\s*)(Style)(\s+)(\S+)', re.IGNORECASE)
 111     re_CopyStyle = re.compile(r'^(\s*)(CopyStyle)(\s+)(\S+)', re.IGNORECASE)
 112     re_NoStyle = re.compile(r'^(\s*)(NoStyle)(\s+)(\S+)', re.IGNORECASE)
 113     re_End = re.compile(r'^(\s*)(End)(\s*)$', re.IGNORECASE)
 114     re_Provides = re.compile(r'^(\s*)Provides(\S+)(\s+)(\S+)', re.IGNORECASE)
 115     re_CharStyle = re.compile(r'^(\s*)CharStyle(\s+)(\S+)$', re.IGNORECASE)
 116     re_AMSMaths = re.compile(r'^\s*Input amsmaths.inc\s*')
 117     re_AMSMathsPlain = re.compile(r'^\s*Input amsmaths-plain.inc\s*')
 118     re_AMSMathsSeq = re.compile(r'^\s*Input amsmaths-seq.inc\s*')
 119
 120     # counters for sectioning styles (hardcoded in 1.3)
 121     counters = {"part"          : "\\Roman{part}",
 122                 "chapter"       : "\\arabic{chapter}",
 123                 "section"       : "\\arabic{section}",
 124                 "subsection"    : "\\arabic{section}.\\arabic{subsection}",
 125                 "subsubsection" : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}",
 126                 "paragraph"     : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}",
 127                 "subparagraph"  : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}.\\arabic{subparagraph}"}
 128
 129     # counters for sectioning styles in appendix (hardcoded in 1.3)
 130     appendixcounters = {"chapter"       : "\\Alph{chapter}",
 131                         "section"       : "\\Alph{section}",
 132                         "subsection"    : "\\arabic{section}.\\arabic{subsection}",
 133                         "subsubsection" : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}",
 134                         "paragraph"     : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}",
 135                         "subparagraph"  : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}.\\arabic{subparagraph}"}
 136
 137     # Value of TocLevel for sectioning styles
 138     toclevels = {"part"          : 0,
 139                  "chapter"       : 0,
 140                  "section"       : 1,
 141                  "subsection"    : 2,
 142                  "subsubsection" : 3,
 143                  "paragraph"     : 4,
 144                  "subparagraph"  : 5}
 145
 146     i = 0
 147     only_comment = 1
 148     counter = ""
 149     label = ""
 150     labelstring = ""
 151     labelstringappendix = ""
 152     space1 = ""
 153     labelstring_line = -1
 154     labelstringappendix_line = -1
 155     labeltype_line = -1
 156     latextype = ""
 157     latextype_line = -1
 158     style = ""
 159     maxcounter = 0
 160     format = 1
 161     formatline = 0
 162     usemodules = []
 163
 164     while i < len(lines):
 165         # Skip comments and empty lines
 166         if re_Comment.match(lines[i]) or re_Empty.match(lines[i]):
 167             i += 1
 168             continue
 169
 170         # insert file format if not already there
 171         if (only_comment):
 172             match = re_Format.match(lines[i])
 173             if match:
 174                 formatline = i
 175                 format = int(match.group(4))
 176                 if format > 1 and format < currentFormat:
 177                     lines[i] = "Format %d" % (format + 1)
 178                     only_comment = 0
 179                 elif format == currentFormat:
 180                     # nothing to do
 181                     return format
 182                 else:
 183                     error('Cannot convert file format %s' % format)
 184             else:
 185                 lines.insert(i, "Format 2")
 186                 only_comment = 0
 187                 continue
 188
 189         # Don't get confused by LaTeX code
 190         if re_Preamble.match(lines[i]):
 191             i += 1
 192             while i < len(lines) and not re_EndPreamble.match(lines[i]):
 193                 i += 1
 194             continue
 195
 196         if format == 9:
 197             match = re_Counter.match(lines[i])
 198             if match:
 199                 counterline = i
 200                 i += 1
 201                 while i < len(lines):
 202                     namem = re_Name.match(lines[i])
 203                     if namem:
 204                         name = namem.group(1)
 205                         lines.pop(i)
 206                         lines[counterline] = "Counter %s" % name
 207                         # we don't need to increment i
 208                         continue
 209                     endem = re_End.match(lines[i])
 210                     if endem:
 211                         i += 1
 212                         break
 213                     i += 1
 214             i += 1
 215             continue
 216
 217         if format == 8:
 218             # We want to scan for ams-type includes and, if we find them,
 219             # add corresponding UseModule tags to the layout.
 220             match = re_AMSMaths.match(lines[i])
 221             if match:
 222                 addstring("theorems-ams", usemodules)
 223                 addstring("theorems-ams-extended", usemodules)
 224                 addstring("theorems-sec", usemodules)
 225                 lines.pop(i)
 226                 continue
 227             match = re_AMSMathsPlain.match(lines[i])
 228             if match:
 229                 addstring("theorems-starred", usemodules)
 230                 lines.pop(i)
 231                 continue
 232             match = re_AMSMathsSeq.match(lines[i])
 233             if match:
 234                 addstring("theorems-ams", usemodules)
 235                 addstring("theorems-ams-extended", usemodules)
 236                 lines.pop(i)
 237                 continue
 238             i += 1
 239             continue
 240
 241         # These just involved new features, not any changes to old ones
 242         if format >= 5 and format <= 7:
 243           i += 1
 244           continue
 245
 246         if format == 4:
 247             # Handle conversion to long CharStyle names
 248             match = re_CharStyle.match(lines[i])
 249             if match:
 250                 lines[i] = "InsetLayout CharStyle:%s" % (match.group(3))
 251                 i += 1
 252                 lines.insert(i, "\tLyXType charstyle")
 253                 i += 1
 254                 lines.insert(i, "")
 255                 lines[i] = "\tLabelString %s" % (match.group(3))
 256             i += 1
 257             continue
 258
 259         if format == 3:
 260             # convert 'providesamsmath x',  'providesmakeidx x',  'providesnatbib x',  'providesurl x' to
 261             #         'provides amsmath x', 'provides makeidx x', 'provides natbib x', 'provides url x'
 262             # x is either 0 or 1
 263             match = re_Provides.match(lines[i])
 264             if match:
 265                 lines[i] = "%sProvides %s%s%s" % (match.group(1), match.group(2).lower(),
 266                                                   match.group(3), match.group(4))
 267             i += 1
 268             continue
 269
 270         if format == 2:
 271             caption = []
 272
 273             # delete caption styles
 274             match = re_Style.match(lines[i])
 275             if match:
 276                 style = string.lower(match.group(4))
 277                 if style == "caption":
 278                     del lines[i]
 279                     while i < len(lines) and not re_End.match(lines[i]):
 280                         caption.append(lines[i])
 281                         del lines[i]
 282                     if i == len(lines):
 283                         error('Incomplete caption style.')
 284                     else:
 285                         del lines[i]
 286                         continue
 287
 288             # delete undefinition of caption styles
 289             match = re_NoStyle.match(lines[i])
 290             if match:
 291                 style = string.lower(match.group(4))
 292                 if style == "caption":
 293                     del lines[i]
 294                     continue
 295
 296             # replace the CopyStyle statement with the definition of the real
 297             # style. This may result in duplicate statements, but that is OK
 298             # since the second one will overwrite the first one.
 299             match = re_CopyStyle.match(lines[i])
 300             if match:
 301                 style = string.lower(match.group(4))
 302                 if style == "caption":
 303                     if len(caption) > 0:
 304                         lines[i:i+1] = caption
 305                     else:
 306                         # FIXME: This style comes from an include file, we
 307                         # should replace the real style and not this default.
 308                         lines[i:i+1] = ['       Margin                First_Dynamic',
 309                                         '       LatexType             Command',
 310                                         '       LatexName             caption',
 311                                         '       NeedProtect           1',
 312                                         '       LabelSep              xx',
 313                                         '       ParSkip               0.4',
 314                                         '       TopSep                0.5',
 315                                         '       Align                 Center',
 316                                         '       AlignPossible         Center',
 317                                         '       LabelType             Sensitive',
 318                                         '       LabelString           "Senseless!"',
 319                                         '       OptionalArgs          1',
 320                                         '       LabelFont',
 321                                         '         Series              Bold',
 322                                         '       EndFont']
 323
 324             i += 1
 325             continue
 326
 327         # Delete MaxCounter and remember the value of it
 328         match = re_MaxCounter.match(lines[i])
 329         if match:
 330             level = match.group(4)
 331             if string.lower(level) == "counter_chapter":
 332                 maxcounter = 0
 333             elif string.lower(level) == "counter_section":
 334                 maxcounter = 1
 335             elif string.lower(level) == "counter_subsection":
 336                 maxcounter = 2
 337             elif string.lower(level) == "counter_subsubsection":
 338                 maxcounter = 3
 339             elif string.lower(level) == "counter_paragraph":
 340                 maxcounter = 4
 341             elif string.lower(level) == "counter_subparagraph":
 342                 maxcounter = 5
 343             elif string.lower(level) == "counter_enumi":
 344                 maxcounter = 6
 345             elif string.lower(level) == "counter_enumii":
 346                 maxcounter = 7
 347             elif string.lower(level) == "counter_enumiii":
 348                 maxcounter = 8
 349             del lines[i]
 350             continue
 351
 352         # Replace line
 353         #
 354         # LabelType Counter_EnumI
 355         #
 356         # with two lines
 357         #
 358         # LabelType Counter
 359         # LabelCounter EnumI
 360         #
 361         match = re_LabelType.match(lines[i])
 362         if match:
 363             label = match.group(4)
 364             # Remember indenting space for later reuse in added lines
 365             space1 = match.group(1)
 366             # Remember the line for adding the LabelCounter later.
 367             # We can't do it here because it could shift latextype_line etc.
 368             labeltype_line = i
 369             if string.lower(label[:8]) == "counter_":
 370                 counter = string.lower(label[8:])
 371                 lines[i] = re_LabelType.sub(r'\1\2\3Counter', lines[i])
 372
 373         # Remember the LabelString line
 374         match = re_LabelString.match(lines[i])
 375         if match:
 376             labelstring = match.group(4)
 377             labelstring_line = i
 378
 379         # Remember the LabelStringAppendix line
 380         match = re_LabelStringAppendix.match(lines[i])
 381         if match:
 382             labelstringappendix = match.group(4)
 383             labelstringappendix_line = i
 384
 385         # Remember the LatexType line
 386         match = re_LatexType.match(lines[i])
 387         if match:
 388             latextype = string.lower(match.group(4))
 389             latextype_line = i
 390
 391         # Reset variables at the beginning of a style definition
 392         match = re_Style.match(lines[i])
 393         if match:
 394             style = string.lower(match.group(4))
 395             counter = ""
 396             label = ""
 397             space1 = ""
 398             labelstring = ""
 399             labelstringappendix = ""
 400             labelstring_line = -1
 401             labelstringappendix_line = -1
 402             labeltype_line = -1
 403             latextype = ""
 404             latextype_line = -1
 405
 406         if re_End.match(lines[i]):
 407
 408             # Add a line "LatexType Bib_Environment" if LabelType is Bibliography
 409             # (or change the existing LatexType)
 410             if string.lower(label) == "bibliography":
 411                 if (latextype_line < 0):
 412                     lines.insert(i, "%sLatexType Bib_Environment" % space1)
 413                     i += 1
 414                 else:
 415                     lines[latextype_line] = re_LatexType.sub(r'\1\2\3Bib_Environment', lines[latextype_line])
 416
 417             # Change "LabelType Static" to "LabelType Itemize" for itemize environments
 418             if latextype == "item_environment" and string.lower(label) == "static":
 419                 lines[labeltype_line] = re_LabelType.sub(r'\1\2\3Itemize', lines[labeltype_line])
 420
 421             # Change "LabelType Counter_EnumI" to "LabelType Enumerate" for enumerate environments
 422             if latextype == "item_environment" and string.lower(label) == "counter_enumi":
 423                 lines[labeltype_line] = re_LabelType.sub(r'\1\2\3Enumerate', lines[labeltype_line])
 424                 # Don't add the LabelCounter line later
 425                 counter = ""
 426
 427             # Replace
 428             #
 429             # LabelString "Chapter"
 430             #
 431             # with
 432             #
 433             # LabelString "Chapter \arabic{chapter}"
 434             #
 435             # if this style has a counter. Ditto for LabelStringAppendix.
 436             # This emulates the hardcoded article style numbering of 1.3
 437             #
 438             if counter != "":
 439                 if counters.has_key(style):
 440                     if labelstring_line < 0:
 441                         lines.insert(i, '%sLabelString "%s"' % (space1, counters[style]))
 442                         i += 1
 443                     else:
 444                         new_labelstring = concatenate_label(labelstring, counters[style])
 445                         lines[labelstring_line] = re_LabelString.sub(
 446                                 r'\1\2\3%s' % new_labelstring.replace("\\", "\\\\"),
 447                                 lines[labelstring_line])
 448                 if appendixcounters.has_key(style):
 449                     if labelstringappendix_line < 0:
 450                         lines.insert(i, '%sLabelStringAppendix "%s"' % (space1, appendixcounters[style]))
 451                         i += 1
 452                     else:
 453                         new_labelstring = concatenate_label(labelstring, appendixcounters[style])
 454                         lines[labelstringappendix_line] = re_LabelStringAppendix.sub(
 455                                 r'\1\2\3%s' % new_labelstring.replace("\\", "\\\\"),
 456                                 lines[labelstringappendix_line])
 457
 458                 # Now we can safely add the LabelCounter line
 459                 lines.insert(labeltype_line + 1, "%sLabelCounter %s" % (space1, counter))
 460                 i += 1
 461
 462             # Add the TocLevel setting for sectioning styles
 463             if toclevels.has_key(style) and maxcounter <= toclevels[style]:
 464                 lines.insert(i, '%sTocLevel %d' % (space1, toclevels[style]))
 465                 i += 1
 466
 467         i += 1
 468
 469     if usemodules:
 470         i = formatline + 1
 471         for mod in usemodules:
 472             lines.insert(i, "UseModule " + mod)
 473             i += 1
 474
 475     return format + 1
 476
 477
 478 def main(argv):
 479
 480     # Open files
 481     if len(argv) == 1:
 482         input = sys.stdin
 483         output = sys.stdout
 484     elif len(argv) == 3:
 485         input = open(argv[1], 'rb')
 486         output = open(argv[2], 'wb')
 487     else:
 488         error(usage(argv[0]))
 489
 490     # Do the real work
 491     lines = read(input)
 492     format = 1
 493     while (format < currentFormat):
 494         format = convert(lines)
 495     write(output, lines)
 496
 497     # Close files
 498     if len(argv) == 3:
 499         input.close()
 500         output.close()
 501
 502     return 0
 503
 504
 505 if __name__ == "__main__":
 506     main(sys.argv)