scripts/bib2dox.py
author Alpar Juttner <alpar@cs.elte.hu>
Fri, 03 Dec 2010 13:26:38 +0100
changeset 1031 06491fd08efd
parent 801 2de0fc630899
child 1220 eb2f9d453070
permissions -rwxr-xr-x
Add contrib dir (#401)
kpeter@905
     1
#! /usr/bin/env python
kpeter@790
     2
"""
kpeter@790
     3
  BibTeX to Doxygen converter
kpeter@790
     4
  Usage: python bib2dox.py bibfile.bib > bibfile.dox
kpeter@790
     5
kpeter@905
     6
  This file is a part of LEMON, a generic C++ optimization library.
kpeter@905
     7
kpeter@905
     8
  **********************************************************************
kpeter@905
     9
kpeter@790
    10
  This code is the modification of the BibTeX to XML converter
kpeter@905
    11
  by Vidar Bronken Gundersen et al.
kpeter@905
    12
  See the original copyright notices below. 
kpeter@790
    13
kpeter@790
    14
  **********************************************************************
kpeter@790
    15
kpeter@790
    16
  Decoder for bibliographic data, BibTeX
kpeter@790
    17
  Usage: python bibtex2xml.py bibfile.bib > bibfile.xml
kpeter@790
    18
kpeter@790
    19
  v.8
kpeter@790
    20
  (c)2002-06-23 Vidar Bronken Gundersen
kpeter@790
    21
  http://bibtexml.sf.net/
kpeter@790
    22
  Reuse approved as long as this notification is kept.
kpeter@790
    23
  Licence: GPL.
kpeter@790
    24
kpeter@790
    25
  Contributions/thanks to:
kpeter@790
    26
  Egon Willighagen, http://sf.net/projects/jreferences/
kpeter@790
    27
  Richard Mahoney (for providing a test case)
kpeter@790
    28
kpeter@790
    29
  Editted by Sara Sprenkle to be more robust and handle more bibtex features.
kpeter@790
    30
  (c) 2003-01-15
kpeter@790
    31
kpeter@790
    32
  1.  Changed bibtex: tags to bibxml: tags.
kpeter@790
    33
  2.  Use xmlns:bibxml="http://bibtexml.sf.net/"
kpeter@790
    34
  3.  Allow spaces between @type and first {
kpeter@790
    35
  4.  "author" fields with multiple authors split by " and "
kpeter@790
    36
      are put in separate xml "bibxml:author" tags.
kpeter@790
    37
  5.  Option for Titles: words are capitalized
kpeter@790
    38
      only if first letter in title or capitalized inside braces
kpeter@790
    39
  6.  Removes braces from within field values
kpeter@790
    40
  7.  Ignores comments in bibtex file (including @comment{ or % )
kpeter@790
    41
  8.  Replaces some special latex tags, e.g., replaces ~ with '&#160;'
kpeter@790
    42
  9.  Handles bibtex @string abbreviations
kpeter@790
    43
        --> includes bibtex's default abbreviations for months
kpeter@790
    44
        --> does concatenation of abbr # " more " and " more " # abbr
kpeter@790
    45
  10. Handles @type( ... ) or @type{ ... }
kpeter@790
    46
  11. The keywords field is split on , or ; and put into separate xml
kpeter@790
    47
      "bibxml:keywords" tags
kpeter@790
    48
  12. Ignores @preamble
kpeter@790
    49
kpeter@790
    50
  Known Limitations
kpeter@790
    51
  1.  Does not transform Latex encoding like math mode and special
kpeter@790
    52
      latex symbols.
kpeter@790
    53
  2.  Does not parse author fields into first and last names.
kpeter@790
    54
      E.g., It does not do anything special to an author whose name is
kpeter@790
    55
      in the form LAST_NAME, FIRST_NAME
kpeter@790
    56
      In "author" tag, will show up as
kpeter@790
    57
      <bibxml:author>LAST_NAME, FIRST_NAME</bibxml:author>
kpeter@790
    58
  3.  Does not handle "crossref" fields other than to print
kpeter@790
    59
      <bibxml:crossref>...</bibxml:crossref>
kpeter@790
    60
  4.  Does not inform user of the input's format errors.  You just won't
kpeter@790
    61
      be able to transform the file later with XSL
kpeter@790
    62
kpeter@790
    63
  You will have to manually edit the XML output if you need to handle
kpeter@790
    64
  these (and unknown) limitations.
kpeter@790
    65
kpeter@790
    66
"""
kpeter@790
    67
kpeter@790
    68
import string, re
kpeter@790
    69
kpeter@790
    70
# set of valid name characters
kpeter@790
    71
valid_name_chars = '[\w\-:]'
kpeter@790
    72
kpeter@790
    73
#
kpeter@790
    74
# define global regular expression variables
kpeter@790
    75
#
kpeter@790
    76
author_rex = re.compile('\s+and\s+')
kpeter@790
    77
rembraces_rex = re.compile('[{}]')
kpeter@801
    78
capitalize_rex = re.compile('({[^}]*})')
kpeter@790
    79
kpeter@790
    80
# used by bibtexkeywords(data)
kpeter@790
    81
keywords_rex = re.compile('[,;]')
kpeter@790
    82
kpeter@790
    83
# used by concat_line(line)
kpeter@790
    84
concatsplit_rex = re.compile('\s*#\s*')
kpeter@790
    85
kpeter@790
    86
# split on {, }, or " in verify_out_of_braces
kpeter@790
    87
delimiter_rex = re.compile('([{}"])',re.I)
kpeter@790
    88
kpeter@790
    89
field_rex = re.compile('\s*(\w*)\s*=\s*(.*)')
kpeter@790
    90
data_rex = re.compile('\s*(\w*)\s*=\s*([^,]*),?')
kpeter@790
    91
kpeter@790
    92
url_rex = re.compile('\\\url\{([^}]*)\}')
kpeter@790
    93
kpeter@792
    94
#
kpeter@792
    95
# styles for html formatting
kpeter@792
    96
#
kpeter@792
    97
divstyle = 'margin-top: -4ex; margin-left: 8em;'
kpeter@790
    98
kpeter@790
    99
#
kpeter@790
   100
# return the string parameter without braces
kpeter@790
   101
#
kpeter@790
   102
def transformurls(str):
kpeter@790
   103
    return url_rex.sub(r'<a href="\1">\1</a>', str)
kpeter@790
   104
kpeter@790
   105
#
kpeter@790
   106
# return the string parameter without braces
kpeter@790
   107
#
kpeter@790
   108
def removebraces(str):
kpeter@790
   109
    return rembraces_rex.sub('', str)
kpeter@790
   110
kpeter@790
   111
#
kpeter@790
   112
# latex-specific replacements
kpeter@790
   113
# (do this after braces were removed)
kpeter@790
   114
#
kpeter@790
   115
def latexreplacements(line):
kpeter@790
   116
    line = string.replace(line, '~', '&nbsp;')
kpeter@790
   117
    line = string.replace(line, '\\\'a', '&aacute;')
kpeter@790
   118
    line = string.replace(line, '\\"a', '&auml;')
kpeter@790
   119
    line = string.replace(line, '\\\'e', '&eacute;')
kpeter@790
   120
    line = string.replace(line, '\\"e', '&euml;')
kpeter@790
   121
    line = string.replace(line, '\\\'i', '&iacute;')
kpeter@790
   122
    line = string.replace(line, '\\"i', '&iuml;')
kpeter@790
   123
    line = string.replace(line, '\\\'o', '&oacute;')
kpeter@790
   124
    line = string.replace(line, '\\"o', '&ouml;')
kpeter@790
   125
    line = string.replace(line, '\\\'u', '&uacute;')
kpeter@790
   126
    line = string.replace(line, '\\"u', '&uuml;')
kpeter@790
   127
    line = string.replace(line, '\\H o', '&otilde;')
kpeter@790
   128
    line = string.replace(line, '\\H u', '&uuml;')   # &utilde; does not exist
kpeter@790
   129
    line = string.replace(line, '\\\'A', '&Aacute;')
kpeter@790
   130
    line = string.replace(line, '\\"A', '&Auml;')
kpeter@790
   131
    line = string.replace(line, '\\\'E', '&Eacute;')
kpeter@790
   132
    line = string.replace(line, '\\"E', '&Euml;')
kpeter@790
   133
    line = string.replace(line, '\\\'I', '&Iacute;')
kpeter@790
   134
    line = string.replace(line, '\\"I', '&Iuml;')
kpeter@790
   135
    line = string.replace(line, '\\\'O', '&Oacute;')
kpeter@790
   136
    line = string.replace(line, '\\"O', '&Ouml;')
kpeter@790
   137
    line = string.replace(line, '\\\'U', '&Uacute;')
kpeter@790
   138
    line = string.replace(line, '\\"U', '&Uuml;')
kpeter@790
   139
    line = string.replace(line, '\\H O', '&Otilde;')
kpeter@790
   140
    line = string.replace(line, '\\H U', '&Uuml;')   # &Utilde; does not exist
kpeter@790
   141
kpeter@790
   142
    return line
kpeter@790
   143
kpeter@790
   144
#
kpeter@790
   145
# copy characters form a string decoding html expressions (&xyz;)
kpeter@790
   146
#
kpeter@790
   147
def copychars(str, ifrom, count):
kpeter@790
   148
    result = ''
kpeter@790
   149
    i = ifrom
kpeter@790
   150
    c = 0
kpeter@790
   151
    html_spec = False
kpeter@790
   152
    while (i < len(str)) and (c < count):
kpeter@790
   153
        if str[i] == '&':
kpeter@790
   154
            html_spec = True;
kpeter@790
   155
            if i+1 < len(str):
kpeter@790
   156
                result += str[i+1]
kpeter@790
   157
            c += 1
kpeter@790
   158
            i += 2
kpeter@790
   159
        else:
kpeter@790
   160
            if not html_spec:
kpeter@790
   161
                if ((str[i] >= 'A') and (str[i] <= 'Z')) or \
kpeter@790
   162
                   ((str[i] >= 'a') and (str[i] <= 'z')):
kpeter@790
   163
                    result += str[i]
kpeter@790
   164
                    c += 1
kpeter@790
   165
            elif str[i] == ';':
kpeter@790
   166
                html_spec = False;
kpeter@790
   167
            i += 1
kpeter@790
   168
    
kpeter@790
   169
    return result
kpeter@790
   170
kpeter@790
   171
kpeter@790
   172
# 
kpeter@790
   173
# Handle a list of authors (separated by 'and').
kpeter@790
   174
# It gives back an array of the follwing values:
kpeter@790
   175
#  - num: the number of authors,
kpeter@790
   176
#  - list: the list of the author names,
kpeter@790
   177
#  - text: the bibtex text (separated by commas and/or 'and')
kpeter@790
   178
#  - abbrev: abbreviation that can be used for indicate the
kpeter@790
   179
#    bibliography entries
kpeter@790
   180
#
kpeter@790
   181
def bibtexauthor(data):
kpeter@790
   182
    result = {}
kpeter@790
   183
    bibtex = ''
kpeter@790
   184
    result['list'] = author_rex.split(data)
kpeter@790
   185
    result['num'] = len(result['list'])
kpeter@790
   186
    for i, author in enumerate(result['list']):
kpeter@790
   187
        # general transformations
kpeter@790
   188
        author = latexreplacements(removebraces(author.strip()))
kpeter@790
   189
        # transform "Xyz, A. B." to "A. B. Xyz"
kpeter@790
   190
        pos = author.find(',')
kpeter@790
   191
        if pos != -1:
kpeter@790
   192
            author = author[pos+1:].strip() + ' ' + author[:pos].strip()
kpeter@790
   193
        result['list'][i] = author
kpeter@790
   194
        bibtex += author + '#'
kpeter@790
   195
    bibtex = bibtex[:-1]
kpeter@790
   196
    if result['num'] > 1:
kpeter@790
   197
        ix = bibtex.rfind('#')
kpeter@790
   198
        if result['num'] == 2:
kpeter@790
   199
            bibtex = bibtex[:ix] + ' and ' + bibtex[ix+1:]
kpeter@790
   200
        else:
kpeter@790
   201
            bibtex = bibtex[:ix] + ', and ' + bibtex[ix+1:]
kpeter@790
   202
    bibtex = bibtex.replace('#', ', ')
kpeter@790
   203
    result['text'] = bibtex
kpeter@790
   204
    
kpeter@790
   205
    result['abbrev'] = ''
kpeter@790
   206
    for author in result['list']:
kpeter@790
   207
        pos = author.rfind(' ') + 1
kpeter@790
   208
        count = 1
kpeter@790
   209
        if result['num'] == 1:
kpeter@790
   210
            count = 3
kpeter@790
   211
        result['abbrev'] += copychars(author, pos, count)
kpeter@790
   212
kpeter@790
   213
    return result
kpeter@790
   214
kpeter@790
   215
kpeter@790
   216
#
kpeter@790
   217
# data = title string
kpeter@790
   218
# @return the capitalized title (first letter is capitalized), rest are capitalized
kpeter@790
   219
# only if capitalized inside braces
kpeter@790
   220
#
kpeter@790
   221
def capitalizetitle(data):
kpeter@790
   222
    title_list = capitalize_rex.split(data)
kpeter@790
   223
    title = ''
kpeter@790
   224
    count = 0
kpeter@790
   225
    for phrase in title_list:
kpeter@790
   226
         check = string.lstrip(phrase)
kpeter@790
   227
kpeter@790
   228
         # keep phrase's capitalization the same
kpeter@790
   229
         if check.find('{') == 0:
kpeter@790
   230
              title += removebraces(phrase)
kpeter@790
   231
         else:
kpeter@790
   232
         # first word --> capitalize first letter (after spaces)
kpeter@790
   233
              if count == 0:
kpeter@790
   234
                  title += check.capitalize()
kpeter@790
   235
              else:
kpeter@790
   236
                  title += phrase.lower()
kpeter@790
   237
         count = count + 1
kpeter@790
   238
kpeter@790
   239
    return title
kpeter@790
   240
kpeter@790
   241
kpeter@790
   242
#
kpeter@790
   243
# @return the bibtex for the title
kpeter@790
   244
# @param data --> title string
kpeter@790
   245
# braces are removed from title
kpeter@790
   246
#
kpeter@790
   247
def bibtextitle(data, entrytype):
kpeter@790
   248
    if entrytype in ('book', 'inbook'):
kpeter@790
   249
        title = removebraces(data.strip())
kpeter@790
   250
    else:
kpeter@790
   251
        title = removebraces(capitalizetitle(data.strip()))
kpeter@790
   252
    bibtex = title
kpeter@790
   253
    return bibtex
kpeter@790
   254
kpeter@790
   255
kpeter@790
   256
#
kpeter@790
   257
# function to compare entry lists
kpeter@790
   258
#
kpeter@790
   259
def entry_cmp(x, y):
kpeter@790
   260
    return cmp(x[0], y[0])
kpeter@790
   261
kpeter@790
   262
kpeter@790
   263
#
kpeter@790
   264
# print the XML for the transformed "filecont_source"
kpeter@790
   265
#
kpeter@790
   266
def bibtexdecoder(filecont_source):
kpeter@790
   267
    filecont = []
kpeter@790
   268
    file = []
kpeter@790
   269
    
kpeter@790
   270
    # want @<alphanumeric chars><spaces>{<spaces><any chars>,
kpeter@790
   271
    pubtype_rex = re.compile('@(\w*)\s*{\s*(.*),')
kpeter@790
   272
    endtype_rex = re.compile('}\s*$')
kpeter@790
   273
    endtag_rex = re.compile('^\s*}\s*$')
kpeter@790
   274
kpeter@790
   275
    bracefield_rex = re.compile('\s*(\w*)\s*=\s*(.*)')
kpeter@790
   276
    bracedata_rex = re.compile('\s*(\w*)\s*=\s*{(.*)},?')
kpeter@790
   277
kpeter@790
   278
    quotefield_rex = re.compile('\s*(\w*)\s*=\s*(.*)')
kpeter@790
   279
    quotedata_rex = re.compile('\s*(\w*)\s*=\s*"(.*)",?')
kpeter@790
   280
kpeter@790
   281
    for line in filecont_source:
kpeter@790
   282
        line = line[:-1]
kpeter@790
   283
kpeter@790
   284
        # encode character entities
kpeter@790
   285
        line = string.replace(line, '&', '&amp;')
kpeter@790
   286
        line = string.replace(line, '<', '&lt;')
kpeter@790
   287
        line = string.replace(line, '>', '&gt;')
kpeter@790
   288
kpeter@790
   289
        # start entry: publication type (store for later use)
kpeter@790
   290
        if pubtype_rex.match(line):
kpeter@790
   291
        # want @<alphanumeric chars><spaces>{<spaces><any chars>,
kpeter@790
   292
            entrycont = {}
kpeter@790
   293
            entry = []
kpeter@790
   294
            entrytype = pubtype_rex.sub('\g<1>',line)
kpeter@790
   295
            entrytype = string.lower(entrytype)
kpeter@792
   296
            entryid   = pubtype_rex.sub('\g<2>', line)
kpeter@790
   297
kpeter@790
   298
        # end entry if just a }
kpeter@790
   299
        elif endtype_rex.match(line):
kpeter@790
   300
            # generate doxygen code for the entry
kpeter@790
   301
kpeter@790
   302
            # enty type related formattings
kpeter@790
   303
            if entrytype in ('book', 'inbook'):
kpeter@790
   304
                entrycont['title'] = '<em>' + entrycont['title'] + '</em>'
kpeter@790
   305
                if not entrycont.has_key('author'):
kpeter@790
   306
                    entrycont['author'] = entrycont['editor']
kpeter@790
   307
                    entrycont['author']['text'] += ', editors'
kpeter@790
   308
            elif entrytype == 'article':
kpeter@790
   309
                entrycont['journal'] = '<em>' + entrycont['journal'] + '</em>'
kpeter@790
   310
            elif entrytype in ('inproceedings', 'incollection', 'conference'):
kpeter@790
   311
                entrycont['booktitle'] = '<em>' + entrycont['booktitle'] + '</em>'
kpeter@790
   312
            elif entrytype == 'techreport':
kpeter@790
   313
                if not entrycont.has_key('type'):
kpeter@790
   314
                    entrycont['type'] = 'Technical report'
kpeter@790
   315
            elif entrytype == 'mastersthesis':
kpeter@790
   316
                entrycont['type'] = 'Master\'s thesis'
kpeter@790
   317
            elif entrytype == 'phdthesis':
kpeter@790
   318
                entrycont['type'] = 'PhD thesis'
kpeter@790
   319
kpeter@790
   320
            for eline in entrycont:
kpeter@790
   321
                if eline != '':
kpeter@790
   322
                    eline = latexreplacements(eline)
kpeter@790
   323
kpeter@790
   324
            if entrycont.has_key('pages') and (entrycont['pages'] != ''):
kpeter@790
   325
                entrycont['pages'] = string.replace(entrycont['pages'], '--', '-')
kpeter@790
   326
kpeter@790
   327
            if entrycont.has_key('author') and (entrycont['author'] != ''):
kpeter@790
   328
                entry.append(entrycont['author']['text'] + '.')
kpeter@790
   329
            if entrycont.has_key('title') and (entrycont['title'] != ''):
kpeter@790
   330
                entry.append(entrycont['title'] + '.')
kpeter@790
   331
            if entrycont.has_key('journal') and (entrycont['journal'] != ''):
kpeter@790
   332
                entry.append(entrycont['journal'] + ',')
kpeter@790
   333
            if entrycont.has_key('booktitle') and (entrycont['booktitle'] != ''):
kpeter@790
   334
                entry.append('In ' + entrycont['booktitle'] + ',')
kpeter@790
   335
            if entrycont.has_key('type') and (entrycont['type'] != ''):
kpeter@790
   336
                eline = entrycont['type']
kpeter@790
   337
                if entrycont.has_key('number') and (entrycont['number'] != ''):
kpeter@790
   338
                    eline += ' ' + entrycont['number']
kpeter@790
   339
                eline += ','
kpeter@790
   340
                entry.append(eline)
kpeter@790
   341
            if entrycont.has_key('institution') and (entrycont['institution'] != ''):
kpeter@790
   342
                entry.append(entrycont['institution'] + ',')
kpeter@790
   343
            if entrycont.has_key('publisher') and (entrycont['publisher'] != ''):
kpeter@790
   344
                entry.append(entrycont['publisher'] + ',')
kpeter@790
   345
            if entrycont.has_key('school') and (entrycont['school'] != ''):
kpeter@790
   346
                entry.append(entrycont['school'] + ',')
kpeter@790
   347
            if entrycont.has_key('address') and (entrycont['address'] != ''):
kpeter@790
   348
                entry.append(entrycont['address'] + ',')
kpeter@790
   349
            if entrycont.has_key('edition') and (entrycont['edition'] != ''):
kpeter@790
   350
                entry.append(entrycont['edition'] + ' edition,')
kpeter@790
   351
            if entrycont.has_key('howpublished') and (entrycont['howpublished'] != ''):
kpeter@790
   352
                entry.append(entrycont['howpublished'] + ',')
kpeter@790
   353
            if entrycont.has_key('volume') and (entrycont['volume'] != ''):
kpeter@790
   354
                eline = entrycont['volume'];
kpeter@790
   355
                if entrycont.has_key('number') and (entrycont['number'] != ''):
kpeter@790
   356
                    eline += '(' + entrycont['number'] + ')'
kpeter@790
   357
                if entrycont.has_key('pages') and (entrycont['pages'] != ''):
kpeter@790
   358
                    eline += ':' + entrycont['pages']
kpeter@790
   359
                eline += ','
kpeter@790
   360
                entry.append(eline)
kpeter@790
   361
            else:
kpeter@790
   362
                if entrycont.has_key('pages') and (entrycont['pages'] != ''):
kpeter@790
   363
                    entry.append('pages ' + entrycont['pages'] + ',')
kpeter@790
   364
            if entrycont.has_key('year') and (entrycont['year'] != ''):
kpeter@790
   365
                if entrycont.has_key('month') and (entrycont['month'] != ''):
kpeter@790
   366
                    entry.append(entrycont['month'] + ' ' + entrycont['year'] + '.')
kpeter@790
   367
                else:
kpeter@790
   368
                    entry.append(entrycont['year'] + '.')
kpeter@790
   369
            if entrycont.has_key('note') and (entrycont['note'] != ''):
kpeter@790
   370
                entry.append(entrycont['note'] + '.')
kpeter@801
   371
            if entrycont.has_key('url') and (entrycont['url'] != ''):
kpeter@801
   372
                entry.append(entrycont['url'] + '.')
kpeter@790
   373
kpeter@790
   374
            # generate keys for sorting and for the output
kpeter@790
   375
            sortkey = ''
kpeter@790
   376
            bibkey = ''
kpeter@790
   377
            if entrycont.has_key('author'):
kpeter@790
   378
                for author in entrycont['author']['list']:
kpeter@790
   379
                    sortkey += copychars(author, author.rfind(' ')+1, len(author))
kpeter@790
   380
                bibkey = entrycont['author']['abbrev']
kpeter@790
   381
            else:
kpeter@790
   382
                bibkey = 'x'
kpeter@790
   383
            if entrycont.has_key('year'):
kpeter@790
   384
                sortkey += entrycont['year']
kpeter@790
   385
                bibkey += entrycont['year'][-2:]
kpeter@790
   386
            if entrycont.has_key('title'):
kpeter@790
   387
                sortkey += entrycont['title']
kpeter@790
   388
            if entrycont.has_key('key'):
kpeter@790
   389
                sortkey = entrycont['key'] + sortkey
kpeter@790
   390
                bibkey = entrycont['key']
kpeter@790
   391
            entry.insert(0, sortkey)
kpeter@790
   392
            entry.insert(1, bibkey)
kpeter@792
   393
            entry.insert(2, entryid)
kpeter@790
   394
           
kpeter@790
   395
            # add the entry to the file contents
kpeter@790
   396
            filecont.append(entry)
kpeter@790
   397
kpeter@790
   398
        else:
kpeter@790
   399
            # field, publication info
kpeter@790
   400
            field = ''
kpeter@790
   401
            data = ''
kpeter@790
   402
            
kpeter@790
   403
            # field = {data} entries
kpeter@790
   404
            if bracedata_rex.match(line):
kpeter@790
   405
                field = bracefield_rex.sub('\g<1>', line)
kpeter@790
   406
                field = string.lower(field)
kpeter@790
   407
                data =  bracedata_rex.sub('\g<2>', line)
kpeter@790
   408
kpeter@790
   409
            # field = "data" entries
kpeter@790
   410
            elif quotedata_rex.match(line):
kpeter@790
   411
                field = quotefield_rex.sub('\g<1>', line)
kpeter@790
   412
                field = string.lower(field)
kpeter@790
   413
                data =  quotedata_rex.sub('\g<2>', line)
kpeter@790
   414
kpeter@790
   415
            # field = data entries
kpeter@790
   416
            elif data_rex.match(line):
kpeter@790
   417
                field = field_rex.sub('\g<1>', line)
kpeter@790
   418
                field = string.lower(field)
kpeter@790
   419
                data =  data_rex.sub('\g<2>', line)
kpeter@801
   420
kpeter@801
   421
            if field == 'url':
kpeter@801
   422
                data = '\\url{' + data.strip() + '}'
kpeter@790
   423
            
kpeter@790
   424
            if field in ('author', 'editor'):
kpeter@790
   425
                entrycont[field] = bibtexauthor(data)
kpeter@790
   426
                line = ''
kpeter@790
   427
            elif field == 'title':
kpeter@790
   428
                line = bibtextitle(data, entrytype)
kpeter@790
   429
            elif field != '':
kpeter@790
   430
                line = removebraces(transformurls(data.strip()))
kpeter@790
   431
kpeter@790
   432
            if line != '':
kpeter@790
   433
                line = latexreplacements(line)
kpeter@790
   434
                entrycont[field] = line
kpeter@790
   435
kpeter@790
   436
kpeter@790
   437
    # sort entries
kpeter@790
   438
    filecont.sort(entry_cmp)
kpeter@790
   439
    
kpeter@790
   440
    # count the bibtex keys
kpeter@790
   441
    keytable = {}
kpeter@790
   442
    counttable = {}
kpeter@790
   443
    for entry in filecont:
kpeter@790
   444
        bibkey = entry[1]
kpeter@790
   445
        if not keytable.has_key(bibkey):
kpeter@790
   446
            keytable[bibkey] = 1
kpeter@790
   447
        else:
kpeter@790
   448
            keytable[bibkey] += 1
kpeter@790
   449
kpeter@790
   450
    for bibkey in keytable.keys():
kpeter@790
   451
        counttable[bibkey] = 0
kpeter@790
   452
    
kpeter@790
   453
    # generate output
kpeter@790
   454
    for entry in filecont:
kpeter@790
   455
        # generate output key form the bibtex key
kpeter@790
   456
        bibkey = entry[1]
kpeter@792
   457
        entryid = entry[2]
kpeter@790
   458
        if keytable[bibkey] == 1:
kpeter@790
   459
            outkey = bibkey
kpeter@790
   460
        else:
kpeter@790
   461
            outkey = bibkey + chr(97 + counttable[bibkey])
kpeter@790
   462
        counttable[bibkey] += 1
kpeter@790
   463
        
kpeter@790
   464
        # append the entry code to the output
kpeter@792
   465
        file.append('\\section ' + entryid + ' [' + outkey + ']')
kpeter@792
   466
        file.append('<div style="' + divstyle + '">')
kpeter@792
   467
        for line in entry[3:]:
kpeter@790
   468
            file.append(line)
kpeter@792
   469
        file.append('</div>')
kpeter@790
   470
        file.append('')
kpeter@790
   471
kpeter@790
   472
    return file
kpeter@790
   473
kpeter@790
   474
kpeter@790
   475
#
kpeter@790
   476
# return 1 iff abbr is in line but not inside braces or quotes
kpeter@790
   477
# assumes that abbr appears only once on the line (out of braces and quotes)
kpeter@790
   478
#
kpeter@790
   479
def verify_out_of_braces(line, abbr):
kpeter@790
   480
kpeter@790
   481
    phrase_split = delimiter_rex.split(line)
kpeter@790
   482
kpeter@790
   483
    abbr_rex = re.compile( '\\b' + abbr + '\\b', re.I)
kpeter@790
   484
kpeter@790
   485
    open_brace = 0
kpeter@790
   486
    open_quote = 0
kpeter@790
   487
kpeter@790
   488
    for phrase in phrase_split:
kpeter@790
   489
        if phrase == "{":
kpeter@790
   490
            open_brace = open_brace + 1
kpeter@790
   491
        elif phrase == "}":
kpeter@790
   492
            open_brace = open_brace - 1
kpeter@790
   493
        elif phrase == '"':
kpeter@790
   494
            if open_quote == 1:
kpeter@790
   495
                open_quote = 0
kpeter@790
   496
            else:
kpeter@790
   497
                open_quote = 1
kpeter@790
   498
        elif abbr_rex.search(phrase):
kpeter@790
   499
            if open_brace == 0 and open_quote == 0:
kpeter@790
   500
                return 1
kpeter@790
   501
kpeter@790
   502
    return 0
kpeter@790
   503
kpeter@790
   504
kpeter@790
   505
#
kpeter@790
   506
# a line in the form phrase1 # phrase2 # ... # phrasen
kpeter@790
   507
# is returned as phrase1 phrase2 ... phrasen
kpeter@790
   508
# with the correct punctuation
kpeter@790
   509
# Bug: Doesn't always work with multiple abbreviations plugged in
kpeter@790
   510
#
kpeter@790
   511
def concat_line(line):
kpeter@790
   512
    # only look at part after equals
kpeter@790
   513
    field = field_rex.sub('\g<1>',line)
kpeter@790
   514
    rest = field_rex.sub('\g<2>',line)
kpeter@790
   515
kpeter@790
   516
    concat_line = field + ' ='
kpeter@790
   517
kpeter@790
   518
    pound_split = concatsplit_rex.split(rest)
kpeter@790
   519
kpeter@790
   520
    phrase_count = 0
kpeter@790
   521
    length = len(pound_split)
kpeter@790
   522
kpeter@790
   523
    for phrase in pound_split:
kpeter@790
   524
        phrase = phrase.strip()
kpeter@790
   525
        if phrase_count != 0:
kpeter@790
   526
            if phrase.startswith('"') or phrase.startswith('{'):
kpeter@790
   527
                phrase = phrase[1:]
kpeter@790
   528
        elif phrase.startswith('"'):
kpeter@790
   529
            phrase = phrase.replace('"','{',1)
kpeter@790
   530
kpeter@790
   531
        if phrase_count != length-1:
kpeter@790
   532
            if phrase.endswith('"') or phrase.endswith('}'):
kpeter@790
   533
                phrase = phrase[:-1]
kpeter@790
   534
        else:
kpeter@790
   535
            if phrase.endswith('"'):
kpeter@790
   536
                phrase = phrase[:-1]
kpeter@790
   537
                phrase = phrase + "}"
kpeter@790
   538
            elif phrase.endswith('",'):
kpeter@790
   539
                phrase = phrase[:-2]
kpeter@790
   540
                phrase = phrase + "},"
kpeter@790
   541
kpeter@790
   542
        # if phrase did have \#, add the \# back
kpeter@790
   543
        if phrase.endswith('\\'):
kpeter@790
   544
            phrase = phrase + "#"
kpeter@790
   545
        concat_line = concat_line + ' ' + phrase
kpeter@790
   546
kpeter@790
   547
        phrase_count = phrase_count + 1
kpeter@790
   548
kpeter@790
   549
    return concat_line
kpeter@790
   550
kpeter@790
   551
kpeter@790
   552
#
kpeter@790
   553
# substitute abbreviations into filecont
kpeter@790
   554
# @param filecont_source - string of data from file
kpeter@790
   555
#
kpeter@790
   556
def bibtex_replace_abbreviations(filecont_source):
kpeter@790
   557
    filecont = filecont_source.splitlines()
kpeter@790
   558
kpeter@790
   559
    #  These are defined in bibtex, so we'll define them too
kpeter@790
   560
    abbr_list = ['jan','feb','mar','apr','may','jun',
kpeter@790
   561
                 'jul','aug','sep','oct','nov','dec']
kpeter@790
   562
    value_list = ['January','February','March','April',
kpeter@790
   563
                  'May','June','July','August','September',
kpeter@790
   564
                  'October','November','December']
kpeter@790
   565
kpeter@790
   566
    abbr_rex = []
kpeter@790
   567
    total_abbr_count = 0
kpeter@790
   568
kpeter@790
   569
    front = '\\b'
kpeter@790
   570
    back = '(,?)\\b'
kpeter@790
   571
kpeter@790
   572
    for x in abbr_list:
kpeter@790
   573
        abbr_rex.append( re.compile( front + abbr_list[total_abbr_count] + back, re.I ) )
kpeter@790
   574
        total_abbr_count = total_abbr_count + 1
kpeter@790
   575
kpeter@790
   576
kpeter@790
   577
    abbrdef_rex = re.compile('\s*@string\s*{\s*('+ valid_name_chars +'*)\s*=(.*)',
kpeter@790
   578
                             re.I)
kpeter@790
   579
kpeter@790
   580
    comment_rex = re.compile('@comment\s*{',re.I)
kpeter@790
   581
    preamble_rex = re.compile('@preamble\s*{',re.I)
kpeter@790
   582
kpeter@790
   583
    waiting_for_end_string = 0
kpeter@790
   584
    i = 0
kpeter@790
   585
    filecont2 = ''
kpeter@790
   586
kpeter@790
   587
    for line in filecont:
kpeter@790
   588
        if line == ' ' or line == '':
kpeter@790
   589
            continue
kpeter@790
   590
kpeter@790
   591
        if waiting_for_end_string:
kpeter@790
   592
            if re.search('}',line):
kpeter@790
   593
                waiting_for_end_string = 0
kpeter@790
   594
                continue
kpeter@790
   595
kpeter@790
   596
        if abbrdef_rex.search(line):
kpeter@790
   597
            abbr = abbrdef_rex.sub('\g<1>', line)
kpeter@790
   598
kpeter@790
   599
            if abbr_list.count(abbr) == 0:
kpeter@790
   600
                val = abbrdef_rex.sub('\g<2>', line)
kpeter@790
   601
                abbr_list.append(abbr)
kpeter@790
   602
                value_list.append(string.strip(val))
kpeter@790
   603
                abbr_rex.append( re.compile( front + abbr_list[total_abbr_count] + back, re.I ) )
kpeter@790
   604
                total_abbr_count = total_abbr_count + 1
kpeter@790
   605
            waiting_for_end_string = 1
kpeter@790
   606
            continue
kpeter@790
   607
kpeter@790
   608
        if comment_rex.search(line):
kpeter@790
   609
            waiting_for_end_string = 1
kpeter@790
   610
            continue
kpeter@790
   611
kpeter@790
   612
        if preamble_rex.search(line):
kpeter@790
   613
            waiting_for_end_string = 1
kpeter@790
   614
            continue
kpeter@790
   615
kpeter@790
   616
kpeter@790
   617
        # replace subsequent abbreviations with the value
kpeter@790
   618
        abbr_count = 0
kpeter@790
   619
kpeter@790
   620
        for x in abbr_list:
kpeter@790
   621
kpeter@790
   622
            if abbr_rex[abbr_count].search(line):
kpeter@790
   623
                if verify_out_of_braces(line,abbr_list[abbr_count]) == 1:
kpeter@790
   624
                    line = abbr_rex[abbr_count].sub( value_list[abbr_count] + '\g<1>', line)
kpeter@790
   625
                # Check for # concatenations
kpeter@790
   626
                if concatsplit_rex.search(line):
kpeter@790
   627
                    line = concat_line(line)
kpeter@790
   628
            abbr_count = abbr_count + 1
kpeter@790
   629
kpeter@790
   630
kpeter@790
   631
        filecont2 = filecont2 + line + '\n'
kpeter@790
   632
        i = i+1
kpeter@790
   633
kpeter@790
   634
kpeter@790
   635
    # Do one final pass over file
kpeter@790
   636
kpeter@790
   637
    # make sure that didn't end up with {" or }" after the substitution
kpeter@790
   638
    filecont2 = filecont2.replace('{"','{{')
kpeter@790
   639
    filecont2 = filecont2.replace('"}','}}')
kpeter@790
   640
kpeter@790
   641
    afterquotevalue_rex = re.compile('"\s*,\s*')
kpeter@790
   642
    afterbrace_rex = re.compile('"\s*}')
kpeter@790
   643
    afterbracevalue_rex = re.compile('(=\s*{[^=]*)},\s*')
kpeter@790
   644
kpeter@790
   645
    # add new lines to data that changed because of abbreviation substitutions
kpeter@790
   646
    filecont2 = afterquotevalue_rex.sub('",\n', filecont2)
kpeter@790
   647
    filecont2 = afterbrace_rex.sub('"\n}', filecont2)
kpeter@790
   648
    filecont2 = afterbracevalue_rex.sub('\g<1>},\n', filecont2)
kpeter@790
   649
kpeter@790
   650
    return filecont2
kpeter@790
   651
kpeter@790
   652
#
kpeter@790
   653
# convert @type( ... ) to @type{ ... }
kpeter@790
   654
#
kpeter@790
   655
def no_outer_parens(filecont):
kpeter@790
   656
kpeter@790
   657
    # do checking for open parens
kpeter@790
   658
    # will convert to braces
kpeter@790
   659
    paren_split = re.split('([(){}])',filecont)
kpeter@790
   660
kpeter@790
   661
    open_paren_count = 0
kpeter@790
   662
    open_type = 0
kpeter@790
   663
    look_next = 0
kpeter@790
   664
kpeter@790
   665
    # rebuild filecont
kpeter@790
   666
    filecont = ''
kpeter@790
   667
kpeter@790
   668
    at_rex = re.compile('@\w*')
kpeter@790
   669
kpeter@790
   670
    for phrase in paren_split:
kpeter@790
   671
        if look_next == 1:
kpeter@790
   672
            if phrase == '(':
kpeter@790
   673
                phrase = '{'
kpeter@790
   674
                open_paren_count = open_paren_count + 1
kpeter@790
   675
            else:
kpeter@790
   676
                open_type = 0
kpeter@790
   677
            look_next = 0
kpeter@790
   678
kpeter@790
   679
        if phrase == '(':
kpeter@790
   680
            open_paren_count = open_paren_count + 1
kpeter@790
   681
kpeter@790
   682
        elif phrase == ')':
kpeter@790
   683
            open_paren_count = open_paren_count - 1
kpeter@790
   684
            if open_type == 1 and open_paren_count == 0:
kpeter@790
   685
                phrase = '}'
kpeter@790
   686
                open_type = 0
kpeter@790
   687
kpeter@790
   688
        elif at_rex.search( phrase ):
kpeter@790
   689
            open_type = 1
kpeter@790
   690
            look_next = 1
kpeter@790
   691
kpeter@790
   692
        filecont = filecont + phrase
kpeter@790
   693
kpeter@790
   694
    return filecont
kpeter@790
   695
kpeter@790
   696
kpeter@790
   697
#
kpeter@790
   698
# make all whitespace into just one space
kpeter@790
   699
# format the bibtex file into a usable form.
kpeter@790
   700
#
kpeter@790
   701
def bibtexwasher(filecont_source):
kpeter@790
   702
kpeter@790
   703
    space_rex = re.compile('\s+')
kpeter@790
   704
    comment_rex = re.compile('\s*%')
kpeter@790
   705
kpeter@790
   706
    filecont = []
kpeter@790
   707
kpeter@790
   708
    # remove trailing and excessive whitespace
kpeter@790
   709
    # ignore comments
kpeter@790
   710
    for line in filecont_source:
kpeter@790
   711
        line = string.strip(line)
kpeter@790
   712
        line = space_rex.sub(' ', line)
kpeter@790
   713
        # ignore comments
kpeter@790
   714
        if not comment_rex.match(line) and line != '':
kpeter@790
   715
            filecont.append(' '+ line)
kpeter@790
   716
kpeter@790
   717
    filecont = string.join(filecont, '')
kpeter@790
   718
kpeter@790
   719
    # the file is in one long string
kpeter@790
   720
kpeter@790
   721
    filecont = no_outer_parens(filecont)
kpeter@790
   722
kpeter@790
   723
    #
kpeter@790
   724
    # split lines according to preferred syntax scheme
kpeter@790
   725
    #
kpeter@790
   726
    filecont = re.sub('(=\s*{[^=]*)},', '\g<1>},\n', filecont)
kpeter@790
   727
kpeter@790
   728
    # add new lines after commas that are after values
kpeter@790
   729
    filecont = re.sub('"\s*,', '",\n', filecont)
kpeter@790
   730
    filecont = re.sub('=\s*([\w\d]+)\s*,', '= \g<1>,\n', filecont)
kpeter@790
   731
    filecont = re.sub('(@\w*)\s*({(\s*)[^,\s]*)\s*,',
kpeter@790
   732
                          '\n\n\g<1>\g<2>,\n', filecont)
kpeter@790
   733
kpeter@790
   734
    # add new lines after }
kpeter@790
   735
    filecont = re.sub('"\s*}','"\n}\n', filecont)
kpeter@790
   736
    filecont = re.sub('}\s*,','},\n', filecont)
kpeter@790
   737
kpeter@790
   738
kpeter@790
   739
    filecont = re.sub('@(\w*)', '\n@\g<1>', filecont)
kpeter@790
   740
kpeter@790
   741
    # character encoding, reserved latex characters
kpeter@790
   742
    filecont = re.sub('{\\\&}', '&', filecont)
kpeter@790
   743
    filecont = re.sub('\\\&', '&', filecont)
kpeter@790
   744
kpeter@790
   745
    # do checking for open braces to get format correct
kpeter@790
   746
    open_brace_count = 0
kpeter@790
   747
    brace_split = re.split('([{}])',filecont)
kpeter@790
   748
kpeter@790
   749
    # rebuild filecont
kpeter@790
   750
    filecont = ''
kpeter@790
   751
kpeter@790
   752
    for phrase in brace_split:
kpeter@790
   753
        if phrase == '{':
kpeter@790
   754
            open_brace_count = open_brace_count + 1
kpeter@790
   755
        elif phrase == '}':
kpeter@790
   756
            open_brace_count = open_brace_count - 1
kpeter@790
   757
            if open_brace_count == 0:
kpeter@790
   758
                filecont = filecont + '\n'
kpeter@790
   759
kpeter@790
   760
        filecont = filecont + phrase
kpeter@790
   761
kpeter@790
   762
    filecont2 = bibtex_replace_abbreviations(filecont)
kpeter@790
   763
kpeter@790
   764
    # gather
kpeter@790
   765
    filecont = filecont2.splitlines()
kpeter@790
   766
    i=0
kpeter@790
   767
    j=0         # count the number of blank lines
kpeter@790
   768
    for line in filecont:
kpeter@790
   769
        # ignore blank lines
kpeter@790
   770
        if line == '' or line == ' ':
kpeter@790
   771
            j = j+1
kpeter@790
   772
            continue
kpeter@790
   773
        filecont[i] = line + '\n'
kpeter@790
   774
        i = i+1
kpeter@790
   775
kpeter@790
   776
    # get rid of the extra stuff at the end of the array
kpeter@790
   777
    # (The extra stuff are duplicates that are in the array because
kpeter@790
   778
    # blank lines were removed.)
kpeter@790
   779
    length = len( filecont)
kpeter@790
   780
    filecont[length-j:length] = []
kpeter@790
   781
kpeter@790
   782
    return filecont
kpeter@790
   783
kpeter@790
   784
kpeter@790
   785
def filehandler(filepath):
kpeter@790
   786
    try:
kpeter@790
   787
        fd = open(filepath, 'r')
kpeter@790
   788
        filecont_source = fd.readlines()
kpeter@790
   789
        fd.close()
kpeter@790
   790
    except:
kpeter@790
   791
        print 'Could not open file:', filepath
kpeter@790
   792
    washeddata = bibtexwasher(filecont_source)
kpeter@790
   793
    outdata = bibtexdecoder(washeddata)
kpeter@790
   794
    print '/**'
kpeter@790
   795
    print '\page references References'
kpeter@790
   796
    print
kpeter@790
   797
    for line in outdata:
kpeter@790
   798
        print line
kpeter@790
   799
    print '*/'
kpeter@790
   800
kpeter@790
   801
kpeter@790
   802
# main program
kpeter@790
   803
kpeter@790
   804
def main():
kpeter@790
   805
    import sys
kpeter@790
   806
    if sys.argv[1:]:
kpeter@790
   807
        filepath = sys.argv[1]
kpeter@790
   808
    else:
kpeter@790
   809
        print "No input file"
kpeter@790
   810
        sys.exit()
kpeter@790
   811
    filehandler(filepath)
kpeter@790
   812
kpeter@790
   813
if __name__ == "__main__": main()
kpeter@790
   814
kpeter@790
   815
kpeter@790
   816
# end python script