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 currentFormat = 8
  33
  34
  35 def usage(prog_name):
  36     return ("Usage: %s inputfile outputfile\n" % prog_name +
  37             "or     %s <inputfile >outputfile" % prog_name)
  38
  39
  40 def error(message):
  41     sys.stderr.write(message + '\n')
  42     sys.exit(1)
  43
  44
  45 def trim_eol(line):
  46     " Remove end of line char(s)."
  47     if line[-2:-1] == '\r':
  48         return line[:-2]
  49     elif line[-1:] == '\r' or line[-1:] == '\n':
  50         return line[:-1]
  51     else:
  52         # file with no EOL in last line
  53         return line
  54
  55
  56 def read(input):
  57     " Read input file and strip lineendings."
  58     lines = list()
  59     while 1:
  60         line = input.readline()
  61         if not line:
  62             break
  63         lines.append(trim_eol(line))
  64     return lines
  65
  66
  67 def write(output, lines):
  68     " Write output file with native lineendings."
  69     for line in lines:
  70         output.write(line + os.linesep)
  71
  72
  73 # Concatenates old and new in an intelligent way:
  74 # If old is wrapped in ", they are stripped. The result is wrapped in ".
  75 def concatenate_label(old, new):
  76     # Don't use strip as long as we support python 1.5.2
  77     if old[0] == '"':
  78         return old[0:-1] + new + '"'
  79     else:
  80         return '"' + old + new + '"'
  81
  82
  83 def convert(lines):
  84     " Convert to new format."
  85     re_Comment = re.compile(r'^(\s*)#')
  86     re_Empty = re.compile(r'^(\s*)$')
  87     re_Format = re.compile(r'^(\s*)(Format)(\s+)(\S+)', re.IGNORECASE)
  88     re_Preamble = re.compile(r'^(\s*)Preamble', re.IGNORECASE)
  89     re_EndPreamble = re.compile(r'^(\s*)EndPreamble', re.IGNORECASE)
  90     re_MaxCounter = re.compile(r'^(\s*)(MaxCounter)(\s+)(\S+)', re.IGNORECASE)
  91     re_LabelType = re.compile(r'^(\s*)(LabelType)(\s+)(\S+)', re.IGNORECASE)
  92     re_LabelString = re.compile(r'^(\s*)(LabelString)(\s+)(("[^"]+")|(\S+))', re.IGNORECASE)
  93     re_LabelStringAppendix = re.compile(r'^(\s*)(LabelStringAppendix)(\s+)(("[^"]+")|(\S+))', re.IGNORECASE)
  94     re_LatexType = re.compile(r'^(\s*)(LatexType)(\s+)(\S+)', re.IGNORECASE)
  95     re_Style = re.compile(r'^(\s*)(Style)(\s+)(\S+)', re.IGNORECASE)
  96     re_CopyStyle = re.compile(r'^(\s*)(CopyStyle)(\s+)(\S+)', re.IGNORECASE)
  97     re_NoStyle = re.compile(r'^(\s*)(NoStyle)(\s+)(\S+)', re.IGNORECASE)
  98     re_End = re.compile(r'^(\s*)(End)(\s*)$', re.IGNORECASE)
  99     re_Provides = re.compile(r'^(\s*)Provides(\S+)(\s+)(\S+)', re.IGNORECASE)
 100     re_CharStyle = re.compile(r'^(\s*)CharStyle(\s+)(\S+)$', re.IGNORECASE)
 101
 102     # counters for sectioning styles (hardcoded in 1.3)
 103     counters = {"part"          : "\\Roman{part}",
 104                 "chapter"       : "\\arabic{chapter}",
 105                 "section"       : "\\arabic{section}",
 106                 "subsection"    : "\\arabic{section}.\\arabic{subsection}",
 107                 "subsubsection" : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}",
 108                 "paragraph"     : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}",
 109                 "subparagraph"  : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}.\\arabic{subparagraph}"}
 110
 111     # counters for sectioning styles in appendix (hardcoded in 1.3)
 112     appendixcounters = {"chapter"       : "\\Alph{chapter}",
 113                         "section"       : "\\Alph{section}",
 114                         "subsection"    : "\\arabic{section}.\\arabic{subsection}",
 115                         "subsubsection" : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}",
 116                         "paragraph"     : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}",
 117                         "subparagraph"  : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}.\\arabic{subparagraph}"}
 118
 119     # Value of TocLevel for sectioning styles
 120     toclevels = {"part"          : 0,
 121                  "chapter"       : 0,
 122                  "section"       : 1,
 123                  "subsection"    : 2,
 124                  "subsubsection" : 3,
 125                  "paragraph"     : 4,
 126                  "subparagraph"  : 5}
 127
 128     i = 0
 129     only_comment = 1
 130     counter = ""
 131     label = ""
 132     labelstring = ""
 133     labelstringappendix = ""
 134     space1 = ""
 135     labelstring_line = -1
 136     labelstringappendix_line = -1
 137     labeltype_line = -1
 138     latextype = ""
 139     latextype_line = -1
 140     style = ""
 141     maxcounter = 0
 142     format = 1
 143     while i < len(lines):
 144
 145         # Skip comments and empty lines
 146         if re_Comment.match(lines[i]) or re_Empty.match(lines[i]):
 147             i += 1
 148             continue
 149
 150         # insert file format if not already there
 151         if (only_comment):
 152                 match = re_Format.match(lines[i])
 153                 if match:
 154                         format = int(match.group(4))
 155                         if format > 1 and format < currentFormat:
 156                             lines[i] = "Format %d" % (format + 1)
 157                             only_comment = 0
 158                         elif format == currentFormat:
 159                                 # nothing to do
 160                                 return format
 161                         else:
 162                             error('Cannot convert file format %s' % format)
 163                 else:
 164                         lines.insert(i, "Format 2")
 165                         only_comment = 0
 166                         continue
 167
 168         # Don't get confused by LaTeX code
 169         if re_Preamble.match(lines[i]):
 170             i += 1
 171             while i < len(lines) and not re_EndPreamble.match(lines[i]):
 172                 i += 1
 173             continue
 174
 175         if format == 7:
 176           i += 1
 177           continue
 178
 179         if format == 6:
 180           i += 1
 181           continue
 182
 183         if format == 5:
 184           i += 1
 185           continue
 186
 187         if format == 4:
 188             # Handle conversion to long CharStyle names
 189             match = re_CharStyle.match(lines[i])
 190             if match:
 191                 lines[i] = "InsetLayout CharStyle:%s" % (match.group(3))
 192                 i += 1
 193                 lines.insert(i, "\tLyXType charstyle")
 194                 i += 1
 195                 lines.insert(i, "")
 196                 lines[i] = "\tLabelString %s" % (match.group(3))
 197             i += 1
 198             continue
 199
 200         if format == 3:
 201             # convert 'providesamsmath x',  'providesmakeidx x',  'providesnatbib x',  'providesurl x' to
 202             #         'provides amsmath x', 'provides makeidx x', 'provides natbib x', 'provides url x'
 203             # x is either 0 or 1
 204             match = re_Provides.match(lines[i])
 205             if match:
 206                 lines[i] = "%sProvides %s%s%s" % (match.group(1), match.group(2).lower(),
 207                                                   match.group(3), match.group(4))
 208             i += 1
 209             continue
 210
 211         if format == 2:
 212             caption = []
 213
 214             # delete caption styles
 215             match = re_Style.match(lines[i])
 216             if match:
 217                 style = string.lower(match.group(4))
 218                 if style == "caption":
 219                     del lines[i]
 220                     while i < len(lines) and not re_End.match(lines[i]):
 221                         caption.append(lines[i])
 222                         del lines[i]
 223                     if i == len(lines):
 224                         error('Incomplete caption style.')
 225                     else:
 226                         del lines[i]
 227                         continue
 228
 229             # delete undefinition of caption styles
 230             match = re_NoStyle.match(lines[i])
 231             if match:
 232                 style = string.lower(match.group(4))
 233                 if style == "caption":
 234                     del lines[i]
 235                     continue
 236
 237             # replace the CopyStyle statement with the definition of the real
 238             # style. This may result in duplicate statements, but that is OK
 239             # since the second one will overwrite the first one.
 240             match = re_CopyStyle.match(lines[i])
 241             if match:
 242                 style = string.lower(match.group(4))
 243                 if style == "caption":
 244                     if len(caption) > 0:
 245                         lines[i:i+1] = caption
 246                     else:
 247                         # FIXME: This style comes from an include file, we
 248                         # should replace the real style and not this default.
 249                         lines[i:i+1] = ['       Margin                First_Dynamic',
 250                                         '       LatexType             Command',
 251                                         '       LatexName             caption',
 252                                         '       NeedProtect           1',
 253                                         '       LabelSep              xx',
 254                                         '       ParSkip               0.4',
 255                                         '       TopSep                0.5',
 256                                         '       Align                 Center',
 257                                         '       AlignPossible         Center',
 258                                         '       LabelType             Sensitive',
 259                                         '       LabelString           "Senseless!"',
 260                                         '       OptionalArgs          1',
 261                                         '       LabelFont',
 262                                         '         Series              Bold',
 263                                         '       EndFont']
 264
 265             i += 1
 266             continue
 267
 268         # Delete MaxCounter and remember the value of it
 269         match = re_MaxCounter.match(lines[i])
 270         if match:
 271             level = match.group(4)
 272             if string.lower(level) == "counter_chapter":
 273                 maxcounter = 0
 274             elif string.lower(level) == "counter_section":
 275                 maxcounter = 1
 276             elif string.lower(level) == "counter_subsection":
 277                 maxcounter = 2
 278             elif string.lower(level) == "counter_subsubsection":
 279                 maxcounter = 3
 280             elif string.lower(level) == "counter_paragraph":
 281                 maxcounter = 4
 282             elif string.lower(level) == "counter_subparagraph":
 283                 maxcounter = 5
 284             elif string.lower(level) == "counter_enumi":
 285                 maxcounter = 6
 286             elif string.lower(level) == "counter_enumii":
 287                 maxcounter = 7
 288             elif string.lower(level) == "counter_enumiii":
 289                 maxcounter = 8
 290             del lines[i]
 291             continue
 292
 293         # Replace line
 294         #
 295         # LabelType Counter_EnumI
 296         #
 297         # with two lines
 298         #
 299         # LabelType Counter
 300         # LabelCounter EnumI
 301         #
 302         match = re_LabelType.match(lines[i])
 303         if match:
 304             label = match.group(4)
 305             # Remember indenting space for later reuse in added lines
 306             space1 = match.group(1)
 307             # Remember the line for adding the LabelCounter later.
 308             # We can't do it here because it could shift latextype_line etc.
 309             labeltype_line = i
 310             if string.lower(label[:8]) == "counter_":
 311                 counter = string.lower(label[8:])
 312                 lines[i] = re_LabelType.sub(r'\1\2\3Counter', lines[i])
 313
 314         # Remember the LabelString line
 315         match = re_LabelString.match(lines[i])
 316         if match:
 317             labelstring = match.group(4)
 318             labelstring_line = i
 319
 320         # Remember the LabelStringAppendix line
 321         match = re_LabelStringAppendix.match(lines[i])
 322         if match:
 323             labelstringappendix = match.group(4)
 324             labelstringappendix_line = i
 325
 326         # Remember the LatexType line
 327         match = re_LatexType.match(lines[i])
 328         if match:
 329             latextype = string.lower(match.group(4))
 330             latextype_line = i
 331
 332         # Reset variables at the beginning of a style definition
 333         match = re_Style.match(lines[i])
 334         if match:
 335             style = string.lower(match.group(4))
 336             counter = ""
 337             label = ""
 338             space1 = ""
 339             labelstring = ""
 340             labelstringappendix = ""
 341             labelstring_line = -1
 342             labelstringappendix_line = -1
 343             labeltype_line = -1
 344             latextype = ""
 345             latextype_line = -1
 346
 347         if re_End.match(lines[i]):
 348
 349             # Add a line "LatexType Bib_Environment" if LabelType is Bibliography
 350             # (or change the existing LatexType)
 351             if string.lower(label) == "bibliography":
 352                 if (latextype_line < 0):
 353                     lines.insert(i, "%sLatexType Bib_Environment" % space1)
 354                     i += 1
 355                 else:
 356                     lines[latextype_line] = re_LatexType.sub(r'\1\2\3Bib_Environment', lines[latextype_line])
 357
 358             # Change "LabelType Static" to "LabelType Itemize" for itemize environments
 359             if latextype == "item_environment" and string.lower(label) == "static":
 360                 lines[labeltype_line] = re_LabelType.sub(r'\1\2\3Itemize', lines[labeltype_line])
 361
 362             # Change "LabelType Counter_EnumI" to "LabelType Enumerate" for enumerate environments
 363             if latextype == "item_environment" and string.lower(label) == "counter_enumi":
 364                 lines[labeltype_line] = re_LabelType.sub(r'\1\2\3Enumerate', lines[labeltype_line])
 365                 # Don't add the LabelCounter line later
 366                 counter = ""
 367
 368             # Replace
 369             #
 370             # LabelString "Chapter"
 371             #
 372             # with
 373             #
 374             # LabelString "Chapter \arabic{chapter}"
 375             #
 376             # if this style has a counter. Ditto for LabelStringAppendix.
 377             # This emulates the hardcoded article style numbering of 1.3
 378             #
 379             if counter != "":
 380                 if counters.has_key(style):
 381                     if labelstring_line < 0:
 382                         lines.insert(i, '%sLabelString "%s"' % (space1, counters[style]))
 383                         i += 1
 384                     else:
 385                         new_labelstring = concatenate_label(labelstring, counters[style])
 386                         lines[labelstring_line] = re_LabelString.sub(
 387                                 r'\1\2\3%s' % new_labelstring.replace("\\", "\\\\"),
 388                                 lines[labelstring_line])
 389                 if appendixcounters.has_key(style):
 390                     if labelstringappendix_line < 0:
 391                         lines.insert(i, '%sLabelStringAppendix "%s"' % (space1, appendixcounters[style]))
 392                         i += 1
 393                     else:
 394                         new_labelstring = concatenate_label(labelstring, appendixcounters[style])
 395                         lines[labelstringappendix_line] = re_LabelStringAppendix.sub(
 396                                 r'\1\2\3%s' % new_labelstring.replace("\\", "\\\\"),
 397                                 lines[labelstringappendix_line])
 398
 399                 # Now we can safely add the LabelCounter line
 400                 lines.insert(labeltype_line + 1, "%sLabelCounter %s" % (space1, counter))
 401                 i += 1
 402
 403             # Add the TocLevel setting for sectioning styles
 404             if toclevels.has_key(style) and maxcounter <= toclevels[style]:
 405                 lines.insert(i, '%sTocLevel %d' % (space1, toclevels[style]))
 406                 i += 1
 407
 408         i += 1
 409
 410     return format + 1
 411
 412
 413 def main(argv):
 414
 415     # Open files
 416     if len(argv) == 1:
 417         input = sys.stdin
 418         output = sys.stdout
 419     elif len(argv) == 3:
 420         input = open(argv[1], 'rb')
 421         output = open(argv[2], 'wb')
 422     else:
 423         error(usage(argv[0]))
 424
 425     # Do the real work
 426     lines = read(input)
 427     format = 1
 428     while (format < currentFormat):
 429         format = convert(lines)
 430     write(output, lines)
 431
 432     # Close files
 433     if len(argv) == 3:
 434         input.close()
 435         output.close()
 436
 437     return 0
 438
 439
 440 if __name__ == "__main__":
 441     main(sys.argv)