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 3
  13
  14
  15 import os, re, string, sys
  16
  17
  18 def usage(prog_name):
  19     return ("Usage: %s inputfile outputfile\n" % prog_name +
  20             "or     %s <inputfile >outputfile" % prog_name)
  21
  22
  23 def error(message):
  24     sys.stderr.write(message + '\n')
  25     sys.exit(1)
  26
  27
  28 def trim_eol(line):
  29     " Remove end of line char(s)."
  30     if line[-2:-1] == '\r':
  31         return line[:-2]
  32     elif line[-1:] == '\r' or line[-1:] == '\n':
  33         return line[:-1]
  34     else:
  35         # file with no EOL in last line
  36         return line
  37
  38
  39 def read(input):
  40     " Read input file and strip lineendings."
  41     lines = list()
  42     while 1:
  43         line = input.readline()
  44         if not line:
  45             break
  46         lines.append(trim_eol(line))
  47     return lines
  48
  49
  50 def write(output, lines):
  51     " Write output file with native lineendings."
  52     for line in lines:
  53         output.write(line + os.linesep)
  54
  55
  56 # Concatenates old and new in an intelligent way:
  57 # If old is wrapped in ", they are stripped. The result is wrapped in ".
  58 def concatenate_label(old, new):
  59     # Don't use strip as long as we support python 1.5.2
  60     if old[0] == '"':
  61         return old[0:-1] + new + '"'
  62     else:
  63         return '"' + old + new + '"'
  64
  65
  66 def convert(lines):
  67     " Convert to new format."
  68     re_Comment = re.compile(r'^(\s*)#')
  69     re_Empty = re.compile(r'^(\s*)$')
  70     re_Format = re.compile(r'^(\s*)(Format)(\s+)(\S+)', re.IGNORECASE)
  71     re_Preamble = re.compile(r'^(\s*)Preamble', re.IGNORECASE)
  72     re_EndPreamble = re.compile(r'^(\s*)EndPreamble', re.IGNORECASE)
  73     re_MaxCounter = re.compile(r'^(\s*)(MaxCounter)(\s+)(\S+)', re.IGNORECASE)
  74     re_LabelType = re.compile(r'^(\s*)(LabelType)(\s+)(\S+)', re.IGNORECASE)
  75     re_LabelString = re.compile(r'^(\s*)(LabelString)(\s+)(("[^"]+")|(\S+))', re.IGNORECASE)
  76     re_LabelStringAppendix = re.compile(r'^(\s*)(LabelStringAppendix)(\s+)(("[^"]+")|(\S+))', re.IGNORECASE)
  77     re_LatexType = re.compile(r'^(\s*)(LatexType)(\s+)(\S+)', re.IGNORECASE)
  78     re_Style = re.compile(r'^(\s*)(Style)(\s+)(\S+)', re.IGNORECASE)
  79     re_CopyStyle = re.compile(r'^(\s*)(CopyStyle)(\s+)(\S+)', re.IGNORECASE)
  80     re_NoStyle = re.compile(r'^(\s*)(NoStyle)(\s+)(\S+)', re.IGNORECASE)
  81     re_End = re.compile(r'^(\s*)(End)(\s*)$', re.IGNORECASE)
  82     re_Provides = re.compile(r'^(\s*)Provides(\S+)(\s+)(\S+)', re.IGNORECASE)
  83
  84     # counters for sectioning styles (hardcoded in 1.3)
  85     counters = {"part"          : "\\Roman{part}",
  86                 "chapter"       : "\\arabic{chapter}",
  87                 "section"       : "\\arabic{section}",
  88                 "subsection"    : "\\arabic{section}.\\arabic{subsection}",
  89                 "subsubsection" : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}",
  90                 "paragraph"     : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}",
  91                 "subparagraph"  : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}.\\arabic{subparagraph}"}
  92
  93     # counters for sectioning styles in appendix (hardcoded in 1.3)
  94     appendixcounters = {"chapter"       : "\\Alph{chapter}",
  95                         "section"       : "\\Alph{section}",
  96                         "subsection"    : "\\arabic{section}.\\arabic{subsection}",
  97                         "subsubsection" : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}",
  98                         "paragraph"     : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}",
  99                         "subparagraph"  : "\\arabic{section}.\\arabic{subsection}.\\arabic{subsubsection}.\\arabic{paragraph}.\\arabic{subparagraph}"}
 100
 101     # Value of TocLevel for sectioning styles
 102     toclevels = {"part"          : 0,
 103                  "chapter"       : 0,
 104                  "section"       : 1,
 105                  "subsection"    : 2,
 106                  "subsubsection" : 3,
 107                  "paragraph"     : 4,
 108                  "subparagraph"  : 5}
 109
 110     i = 0
 111     only_comment = 1
 112     counter = ""
 113     label = ""
 114     labelstring = ""
 115     labelstringappendix = ""
 116     space1 = ""
 117     labelstring_line = -1
 118     labelstringappendix_line = -1
 119     labeltype_line = -1
 120     latextype = ""
 121     latextype_line = -1
 122     style = ""
 123     maxcounter = 0
 124     format = 1
 125     while i < len(lines):
 126
 127         # Skip comments and empty lines
 128         if re_Comment.match(lines[i]) or re_Empty.match(lines[i]):
 129             i = i + 1
 130             continue
 131
 132         # insert file format if not already there
 133         if (only_comment):
 134                 match = re_Format.match(lines[i])
 135                 if match:
 136                         format = int(match.group(4))
 137                         if format > 1 and format < 4:
 138                             lines[i] = "Format %d" % (format + 1)
 139                             only_comment = 0
 140                         elif format == 4:
 141                                 # nothing to do
 142                                 return format
 143                         else:
 144                             error('Cannot convert file format %s' % format)
 145                 else:
 146                         lines.insert(i, "Format 2")
 147                         only_comment = 0
 148                         continue
 149
 150         # Don't get confused by LaTeX code
 151         if re_Preamble.match(lines[i]):
 152             i = i + 1
 153             while i < len(lines) and not re_EndPreamble.match(lines[i]):
 154                 i = i + 1
 155             continue
 156
 157         if format == 3:
 158             # convert 'providesamsmath x',  'providesmakeidx x',  'providesnatbib x',  'providesurl x' to
 159             #         'provides amsmath x', 'provides makeidx x', 'provides natbib x', 'provides url x'
 160             # x is either 0 or 1
 161             match = re_Provides.match(lines[i])
 162             if match:
 163                 lines[i] = "%sProvides %s%s%s" % (match.group(1), match.group(2).lower(),
 164                                                   match.group(3), match.group(4))
 165             i = i + 1
 166             continue
 167
 168         if format == 2:
 169             caption = []
 170
 171             # delete caption styles
 172             match = re_Style.match(lines[i])
 173             if match:
 174                 style = string.lower(match.group(4))
 175                 if style == "caption":
 176                     del lines[i]
 177                     while i < len(lines) and not re_End.match(lines[i]):
 178                         caption.append(lines[i])
 179                         del lines[i]
 180                     if i == len(lines):
 181                         error('Incomplete caption style.')
 182                     else:
 183                         del lines[i]
 184                         continue
 185
 186             # delete undefinition of caption styles
 187             match = re_NoStyle.match(lines[i])
 188             if match:
 189                 style = string.lower(match.group(4))
 190                 if style == "caption":
 191                     del lines[i]
 192                     continue
 193
 194             # replace the CopyStyle statement with the definition of the real
 195             # style. This may result in duplicate statements, but that is OK
 196             # since the second one will overwrite the first one.
 197             match = re_CopyStyle.match(lines[i])
 198             if match:
 199                 style = string.lower(match.group(4))
 200                 if style == "caption":
 201                     if len(caption) > 0:
 202                         lines[i:i+1] = caption
 203                     else:
 204                         # FIXME: This style comes from an include file, we
 205                         # should replace the real style and not this default.
 206                         lines[i:i+1] = ['       Margin                First_Dynamic',
 207                                         '       LatexType             Command',
 208                                         '       LatexName             caption',
 209                                         '       NeedProtect           1',
 210                                         '       LabelSep              xx',
 211                                         '       ParSkip               0.4',
 212                                         '       TopSep                0.5',
 213                                         '       Align                 Center',
 214                                         '       AlignPossible         Center',
 215                                         '       LabelType             Sensitive',
 216                                         '       LabelString           "Senseless!"',
 217                                         '       OptionalArgs          1',
 218                                         '       LabelFont',
 219                                         '         Series              Bold',
 220                                         '       EndFont']
 221
 222             i = i + 1
 223             continue
 224
 225         # Delete MaxCounter and remember the value of it
 226         match = re_MaxCounter.match(lines[i])
 227         if match:
 228             level = match.group(4)
 229             if string.lower(level) == "counter_chapter":
 230                 maxcounter = 0
 231             elif string.lower(level) == "counter_section":
 232                 maxcounter = 1
 233             elif string.lower(level) == "counter_subsection":
 234                 maxcounter = 2
 235             elif string.lower(level) == "counter_subsubsection":
 236                 maxcounter = 3
 237             elif string.lower(level) == "counter_paragraph":
 238                 maxcounter = 4
 239             elif string.lower(level) == "counter_subparagraph":
 240                 maxcounter = 5
 241             elif string.lower(level) == "counter_enumi":
 242                 maxcounter = 6
 243             elif string.lower(level) == "counter_enumii":
 244                 maxcounter = 7
 245             elif string.lower(level) == "counter_enumiii":
 246                 maxcounter = 8
 247             del lines[i]
 248             continue
 249
 250         # Replace line
 251         #
 252         # LabelType Counter_EnumI
 253         #
 254         # with two lines
 255         #
 256         # LabelType Counter
 257         # LabelCounter EnumI
 258         #
 259         match = re_LabelType.match(lines[i])
 260         if match:
 261             label = match.group(4)
 262             # Remember indenting space for later reuse in added lines
 263             space1 = match.group(1)
 264             # Remember the line for adding the LabelCounter later.
 265             # We can't do it here because it could shift latextype_line etc.
 266             labeltype_line = i
 267             if string.lower(label[:8]) == "counter_":
 268                 counter = string.lower(label[8:])
 269                 lines[i] = re_LabelType.sub(r'\1\2\3Counter', lines[i])
 270
 271         # Remember the LabelString line
 272         match = re_LabelString.match(lines[i])
 273         if match:
 274             labelstring = match.group(4)
 275             labelstring_line = i
 276
 277         # Remember the LabelStringAppendix line
 278         match = re_LabelStringAppendix.match(lines[i])
 279         if match:
 280             labelstringappendix = match.group(4)
 281             labelstringappendix_line = i
 282
 283         # Remember the LatexType line
 284         match = re_LatexType.match(lines[i])
 285         if match:
 286             latextype = string.lower(match.group(4))
 287             latextype_line = i
 288
 289         # Reset variables at the beginning of a style definition
 290         match = re_Style.match(lines[i])
 291         if match:
 292             style = string.lower(match.group(4))
 293             counter = ""
 294             label = ""
 295             space1 = ""
 296             labelstring = ""
 297             labelstringappendix = ""
 298             labelstring_line = -1
 299             labelstringappendix_line = -1
 300             labeltype_line = -1
 301             latextype = ""
 302             latextype_line = -1
 303
 304         if re_End.match(lines[i]):
 305
 306             # Add a line "LatexType Bib_Environment" if LabelType is Bibliography
 307             # (or change the existing LatexType)
 308             if string.lower(label) == "bibliography":
 309                 if (latextype_line < 0):
 310                     lines.insert(i, "%sLatexType Bib_Environment" % space1)
 311                     i = i + 1
 312                 else:
 313                     lines[latextype_line] = re_LatexType.sub(r'\1\2\3Bib_Environment', lines[latextype_line])
 314
 315             # Change "LabelType Static" to "LabelType Itemize" for itemize environments
 316             if latextype == "item_environment" and string.lower(label) == "static":
 317                 lines[labeltype_line] = re_LabelType.sub(r'\1\2\3Itemize', lines[labeltype_line])
 318
 319             # Change "LabelType Counter_EnumI" to "LabelType Enumerate" for enumerate environments
 320             if latextype == "item_environment" and string.lower(label) == "counter_enumi":
 321                 lines[labeltype_line] = re_LabelType.sub(r'\1\2\3Enumerate', lines[labeltype_line])
 322                 # Don't add the LabelCounter line later
 323                 counter = ""
 324
 325             # Replace
 326             #
 327             # LabelString "Chapter"
 328             #
 329             # with
 330             #
 331             # LabelString "Chapter \arabic{chapter}"
 332             #
 333             # if this style has a counter. Ditto for LabelStringAppendix.
 334             # This emulates the hardcoded article style numbering of 1.3
 335             #
 336             if counter != "":
 337                 if counters.has_key(style):
 338                     if labelstring_line < 0:
 339                         lines.insert(i, '%sLabelString "%s"' % (space1, counters[style]))
 340                         i = i + 1
 341                     else:
 342                         new_labelstring = concatenate_label(labelstring, counters[style])
 343                         lines[labelstring_line] = re_LabelString.sub(
 344                                 r'\1\2\3%s' % new_labelstring.replace("\\", "\\\\"),
 345                                 lines[labelstring_line])
 346                 if appendixcounters.has_key(style):
 347                     if labelstringappendix_line < 0:
 348                         lines.insert(i, '%sLabelStringAppendix "%s"' % (space1, appendixcounters[style]))
 349                         i = i + 1
 350                     else:
 351                         new_labelstring = concatenate_label(labelstring, appendixcounters[style])
 352                         lines[labelstringappendix_line] = re_LabelStringAppendix.sub(
 353                                 r'\1\2\3%s' % new_labelstring.replace("\\", "\\\\"),
 354                                 lines[labelstringappendix_line])
 355
 356                 # Now we can safely add the LabelCounter line
 357                 lines.insert(labeltype_line + 1, "%sLabelCounter %s" % (space1, counter))
 358                 i = i + 1
 359
 360             # Add the TocLevel setting for sectioning styles
 361             if toclevels.has_key(style) and maxcounter <= toclevels[style]:
 362                 lines.insert(i, '%sTocLevel %d' % (space1, toclevels[style]))
 363                 i = i + 1
 364
 365         i = i + 1
 366
 367     return format + 1
 368
 369
 370 def main(argv):
 371
 372     # Open files
 373     if len(argv) == 1:
 374         input = sys.stdin
 375         output = sys.stdout
 376     elif len(argv) == 3:
 377         input = open(argv[1], 'rb')
 378         output = open(argv[2], 'wb')
 379     else:
 380         error(usage(argv[0]))
 381
 382     # Do the real work
 383     lines = read(input)
 384     format = 1
 385     while (format < 4):
 386         format = convert(lines)
 387     write(output, lines)
 388
 389     # Close files
 390     if len(argv) == 3:
 391         input.close()
 392         output.close()
 393
 394     return 0
 395
 396
 397 if __name__ == "__main__":
 398     main(sys.argv)