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 currentFormat = 9
  37
  38
  39 def usage(prog_name):
  40     return ("Usage: %s inputfile outputfile\n" % prog_name +
  41             "or     %s <inputfile >outputfile" % prog_name)
  42
  43
  44 def error(message):
  45     sys.stderr.write(message + '\n')
  46     sys.exit(1)
  47
  48
  49 def trim_eol(line):
  50     " Remove end of line char(s)."
  51     if line[-2:-1] == '\r':
  52         return line[:-2]
  53     elif line[-1:] == '\r' or line[-1:] == '\n':
  54         return line[:-1]
  55     else:
  56         # file with no EOL in last line
  57         return line
  58
  59
  60 def read(input):
  61     " Read input file and strip lineendings."
  62     lines = list()
  63     while 1:
  64         line = input.readline()
  65         if not line:
  66             break
  67         lines.append(trim_eol(line))
  68     return lines
  69
  70
  71 def write(output, lines):
  72     " Write output file with native lineendings."
  73     for line in lines:
  74         output.write(line + os.linesep)
  75
  76
  77 # Concatenates old and new in an intelligent way:
  78 # If old is wrapped in ", they are stripped. The result is wrapped in ".
  79 def concatenate_label(old, new):
  80     # Don't use strip as long as we support python 1.5.2
  81     if old[0] == '"':
  82         return old[0:-1] + new + '"'
  83     else:
  84         return '"' + old + new + '"'
  85
  86
  87 def convert(lines):
  88     " Convert to new format."
  89     re_Comment = re.compile(r'^(\s*)#')
  90     re_Empty = re.compile(r'^(\s*)$')
  91     re_Format = re.compile(r'^(\s*)(Format)(\s+)(\S+)', re.IGNORECASE)
  92     re_Preamble = re.compile(r'^(\s*)Preamble', re.IGNORECASE)
  93     re_EndPreamble = re.compile(r'^(\s*)EndPreamble', re.IGNORECASE)
  94     re_MaxCounter = re.compile(r'^(\s*)(MaxCounter)(\s+)(\S+)', re.IGNORECASE)
  95     re_LabelType = re.compile(r'^(\s*)(LabelType)(\s+)(\S+)', re.IGNORECASE)
  96     re_LabelString = re.compile(r'^(\s*)(LabelString)(\s+)(("[^"]+")|(\S+))', re.IGNORECASE)
  97     re_LabelStringAppendix = re.compile(r'^(\s*)(LabelStringAppendix)(\s+)(("[^"]+")|(\S+))', re.IGNORECASE)
  98     re_LatexType = re.compile(r'^(\s*)(LatexType)(\s+)(\S+)', re.IGNORECASE)
  99     re_Style = re.compile(r'^(\s*)(Style)(\s+)(\S+)', re.IGNORECASE)
 100     re_CopyStyle = re.compile(r'^(\s*)(CopyStyle)(\s+)(\S+)', re.IGNORECASE)
 101     re_NoStyle = re.compile(r'^(\s*)(NoStyle)(\s+)(\S+)', re.IGNORECASE)
 102     re_End = re.compile(r'^(\s*)(End)(\s*)$', re.IGNORECASE)
 103     re_Provides = re.compile(r'^(\s*)Provides(\S+)(\s+)(\S+)', re.IGNORECASE)
 104     re_CharStyle = re.compile(r'^(\s*)CharStyle(\s+)(\S+)$', re.IGNORECASE)
 105
 106     # counters for sectioning styles (hardcoded in 1.3)
 107     counters = {"part"          : "\\Roman{part}",
 108                 "chapter"       : "\\arabic{chapter}",
 109                 "section"       : "\\arabic{section}",
 110                 "subsection"    : "\\arabic{section}.\\arabic{subsection}",
 111                 "subsubsection" : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}",
 112                 "paragraph"     : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}",
 113                 "subparagraph"  : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}.\\arabic{subparagraph}"}
 114
 115     # counters for sectioning styles in appendix (hardcoded in 1.3)
 116     appendixcounters = {"chapter"       : "\\Alph{chapter}",
 117                         "section"       : "\\Alph{section}",
 118                         "subsection"    : "\\arabic{section}.\\arabic{subsection}",
 119                         "subsubsection" : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}",
 120                         "paragraph"     : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}",
 121                         "subparagraph"  : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}.\\arabic{subparagraph}"}
 122
 123     # Value of TocLevel for sectioning styles
 124     toclevels = {"part"          : 0,
 125                  "chapter"       : 0,
 126                  "section"       : 1,
 127                  "subsection"    : 2,
 128                  "subsubsection" : 3,
 129                  "paragraph"     : 4,
 130                  "subparagraph"  : 5}
 131
 132     i = 0
 133     only_comment = 1
 134     counter = ""
 135     label = ""
 136     labelstring = ""
 137     labelstringappendix = ""
 138     space1 = ""
 139     labelstring_line = -1
 140     labelstringappendix_line = -1
 141     labeltype_line = -1
 142     latextype = ""
 143     latextype_line = -1
 144     style = ""
 145     maxcounter = 0
 146     format = 1
 147     while i < len(lines):
 148
 149         # Skip comments and empty lines
 150         if re_Comment.match(lines[i]) or re_Empty.match(lines[i]):
 151             i += 1
 152             continue
 153
 154         # insert file format if not already there
 155         if (only_comment):
 156                 match = re_Format.match(lines[i])
 157                 if match:
 158                         format = int(match.group(4))
 159                         if format > 1 and format < currentFormat:
 160                             lines[i] = "Format %d" % (format + 1)
 161                             only_comment = 0
 162                         elif format == currentFormat:
 163                                 # nothing to do
 164                                 return format
 165                         else:
 166                             error('Cannot convert file format %s' % format)
 167                 else:
 168                         lines.insert(i, "Format 2")
 169                         only_comment = 0
 170                         continue
 171
 172         # Don't get confused by LaTeX code
 173         if re_Preamble.match(lines[i]):
 174             i += 1
 175             while i < len(lines) and not re_EndPreamble.match(lines[i]):
 176                 i += 1
 177             continue
 178
 179         # These just involved new features, not any changes to old ones
 180         if format >= 5 and format <= 8:
 181           i += 1
 182           continue
 183
 184         if format == 4:
 185             # Handle conversion to long CharStyle names
 186             match = re_CharStyle.match(lines[i])
 187             if match:
 188                 lines[i] = "InsetLayout CharStyle:%s" % (match.group(3))
 189                 i += 1
 190                 lines.insert(i, "\tLyXType charstyle")
 191                 i += 1
 192                 lines.insert(i, "")
 193                 lines[i] = "\tLabelString %s" % (match.group(3))
 194             i += 1
 195             continue
 196
 197         if format == 3:
 198             # convert 'providesamsmath x',  'providesmakeidx x',  'providesnatbib x',  'providesurl x' to
 199             #         'provides amsmath x', 'provides makeidx x', 'provides natbib x', 'provides url x'
 200             # x is either 0 or 1
 201             match = re_Provides.match(lines[i])
 202             if match:
 203                 lines[i] = "%sProvides %s%s%s" % (match.group(1), match.group(2).lower(),
 204                                                   match.group(3), match.group(4))
 205             i += 1
 206             continue
 207
 208         if format == 2:
 209             caption = []
 210
 211             # delete caption styles
 212             match = re_Style.match(lines[i])
 213             if match:
 214                 style = string.lower(match.group(4))
 215                 if style == "caption":
 216                     del lines[i]
 217                     while i < len(lines) and not re_End.match(lines[i]):
 218                         caption.append(lines[i])
 219                         del lines[i]
 220                     if i == len(lines):
 221                         error('Incomplete caption style.')
 222                     else:
 223                         del lines[i]
 224                         continue
 225
 226             # delete undefinition of caption styles
 227             match = re_NoStyle.match(lines[i])
 228             if match:
 229                 style = string.lower(match.group(4))
 230                 if style == "caption":
 231                     del lines[i]
 232                     continue
 233
 234             # replace the CopyStyle statement with the definition of the real
 235             # style. This may result in duplicate statements, but that is OK
 236             # since the second one will overwrite the first one.
 237             match = re_CopyStyle.match(lines[i])
 238             if match:
 239                 style = string.lower(match.group(4))
 240                 if style == "caption":
 241                     if len(caption) > 0:
 242                         lines[i:i+1] = caption
 243                     else:
 244                         # FIXME: This style comes from an include file, we
 245                         # should replace the real style and not this default.
 246                         lines[i:i+1] = ['       Margin                First_Dynamic',
 247                                         '       LatexType             Command',
 248                                         '       LatexName             caption',
 249                                         '       NeedProtect           1',
 250                                         '       LabelSep              xx',
 251                                         '       ParSkip               0.4',
 252                                         '       TopSep                0.5',
 253                                         '       Align                 Center',
 254                                         '       AlignPossible         Center',
 255                                         '       LabelType             Sensitive',
 256                                         '       LabelString           "Senseless!"',
 257                                         '       OptionalArgs          1',
 258                                         '       LabelFont',
 259                                         '         Series              Bold',
 260                                         '       EndFont']
 261
 262             i += 1
 263             continue
 264
 265         # Delete MaxCounter and remember the value of it
 266         match = re_MaxCounter.match(lines[i])
 267         if match:
 268             level = match.group(4)
 269             if string.lower(level) == "counter_chapter":
 270                 maxcounter = 0
 271             elif string.lower(level) == "counter_section":
 272                 maxcounter = 1
 273             elif string.lower(level) == "counter_subsection":
 274                 maxcounter = 2
 275             elif string.lower(level) == "counter_subsubsection":
 276                 maxcounter = 3
 277             elif string.lower(level) == "counter_paragraph":
 278                 maxcounter = 4
 279             elif string.lower(level) == "counter_subparagraph":
 280                 maxcounter = 5
 281             elif string.lower(level) == "counter_enumi":
 282                 maxcounter = 6
 283             elif string.lower(level) == "counter_enumii":
 284                 maxcounter = 7
 285             elif string.lower(level) == "counter_enumiii":
 286                 maxcounter = 8
 287             del lines[i]
 288             continue
 289
 290         # Replace line
 291         #
 292         # LabelType Counter_EnumI
 293         #
 294         # with two lines
 295         #
 296         # LabelType Counter
 297         # LabelCounter EnumI
 298         #
 299         match = re_LabelType.match(lines[i])
 300         if match:
 301             label = match.group(4)
 302             # Remember indenting space for later reuse in added lines
 303             space1 = match.group(1)
 304             # Remember the line for adding the LabelCounter later.
 305             # We can't do it here because it could shift latextype_line etc.
 306             labeltype_line = i
 307             if string.lower(label[:8]) == "counter_":
 308                 counter = string.lower(label[8:])
 309                 lines[i] = re_LabelType.sub(r'\1\2\3Counter', lines[i])
 310
 311         # Remember the LabelString line
 312         match = re_LabelString.match(lines[i])
 313         if match:
 314             labelstring = match.group(4)
 315             labelstring_line = i
 316
 317         # Remember the LabelStringAppendix line
 318         match = re_LabelStringAppendix.match(lines[i])
 319         if match:
 320             labelstringappendix = match.group(4)
 321             labelstringappendix_line = i
 322
 323         # Remember the LatexType line
 324         match = re_LatexType.match(lines[i])
 325         if match:
 326             latextype = string.lower(match.group(4))
 327             latextype_line = i
 328
 329         # Reset variables at the beginning of a style definition
 330         match = re_Style.match(lines[i])
 331         if match:
 332             style = string.lower(match.group(4))
 333             counter = ""
 334             label = ""
 335             space1 = ""
 336             labelstring = ""
 337             labelstringappendix = ""
 338             labelstring_line = -1
 339             labelstringappendix_line = -1
 340             labeltype_line = -1
 341             latextype = ""
 342             latextype_line = -1
 343
 344         if re_End.match(lines[i]):
 345
 346             # Add a line "LatexType Bib_Environment" if LabelType is Bibliography
 347             # (or change the existing LatexType)
 348             if string.lower(label) == "bibliography":
 349                 if (latextype_line < 0):
 350                     lines.insert(i, "%sLatexType Bib_Environment" % space1)
 351                     i += 1
 352                 else:
 353                     lines[latextype_line] = re_LatexType.sub(r'\1\2\3Bib_Environment', lines[latextype_line])
 354
 355             # Change "LabelType Static" to "LabelType Itemize" for itemize environments
 356             if latextype == "item_environment" and string.lower(label) == "static":
 357                 lines[labeltype_line] = re_LabelType.sub(r'\1\2\3Itemize', lines[labeltype_line])
 358
 359             # Change "LabelType Counter_EnumI" to "LabelType Enumerate" for enumerate environments
 360             if latextype == "item_environment" and string.lower(label) == "counter_enumi":
 361                 lines[labeltype_line] = re_LabelType.sub(r'\1\2\3Enumerate', lines[labeltype_line])
 362                 # Don't add the LabelCounter line later
 363                 counter = ""
 364
 365             # Replace
 366             #
 367             # LabelString "Chapter"
 368             #
 369             # with
 370             #
 371             # LabelString "Chapter \arabic{chapter}"
 372             #
 373             # if this style has a counter. Ditto for LabelStringAppendix.
 374             # This emulates the hardcoded article style numbering of 1.3
 375             #
 376             if counter != "":
 377                 if counters.has_key(style):
 378                     if labelstring_line < 0:
 379                         lines.insert(i, '%sLabelString "%s"' % (space1, counters[style]))
 380                         i += 1
 381                     else:
 382                         new_labelstring = concatenate_label(labelstring, counters[style])
 383                         lines[labelstring_line] = re_LabelString.sub(
 384                                 r'\1\2\3%s' % new_labelstring.replace("\\", "\\\\"),
 385                                 lines[labelstring_line])
 386                 if appendixcounters.has_key(style):
 387                     if labelstringappendix_line < 0:
 388                         lines.insert(i, '%sLabelStringAppendix "%s"' % (space1, appendixcounters[style]))
 389                         i += 1
 390                     else:
 391                         new_labelstring = concatenate_label(labelstring, appendixcounters[style])
 392                         lines[labelstringappendix_line] = re_LabelStringAppendix.sub(
 393                                 r'\1\2\3%s' % new_labelstring.replace("\\", "\\\\"),
 394                                 lines[labelstringappendix_line])
 395
 396                 # Now we can safely add the LabelCounter line
 397                 lines.insert(labeltype_line + 1, "%sLabelCounter %s" % (space1, counter))
 398                 i += 1
 399
 400             # Add the TocLevel setting for sectioning styles
 401             if toclevels.has_key(style) and maxcounter <= toclevels[style]:
 402                 lines.insert(i, '%sTocLevel %d' % (space1, toclevels[style]))
 403                 i += 1
 404
 405         i += 1
 406
 407     return format + 1
 408
 409
 410 def main(argv):
 411
 412     # Open files
 413     if len(argv) == 1:
 414         input = sys.stdin
 415         output = sys.stdout
 416     elif len(argv) == 3:
 417         input = open(argv[1], 'rb')
 418         output = open(argv[2], 'wb')
 419     else:
 420         error(usage(argv[0]))
 421
 422     # Do the real work
 423     lines = read(input)
 424     format = 1
 425     while (format < currentFormat):
 426         format = convert(lines)
 427     write(output, lines)
 428
 429     # Close files
 430     if len(argv) == 3:
 431         input.close()
 432         output.close()
 433
 434     return 0
 435
 436
 437 if __name__ == "__main__":
 438     main(sys.argv)