kpeter@836: #! /usr/bin/env python
kpeter@743: """
kpeter@743:   BibTeX to Doxygen converter
kpeter@743:   Usage: python bib2dox.py bibfile.bib > bibfile.dox
kpeter@743: 
kpeter@836:   This file is a part of LEMON, a generic C++ optimization library.
kpeter@836: 
kpeter@836:   **********************************************************************
kpeter@836: 
kpeter@743:   This code is the modification of the BibTeX to XML converter
kpeter@836:   by Vidar Bronken Gundersen et al.
kpeter@836:   See the original copyright notices below. 
kpeter@743: 
kpeter@743:   **********************************************************************
kpeter@743: 
kpeter@743:   Decoder for bibliographic data, BibTeX
kpeter@743:   Usage: python bibtex2xml.py bibfile.bib > bibfile.xml
kpeter@743: 
kpeter@743:   v.8
kpeter@743:   (c)2002-06-23 Vidar Bronken Gundersen
kpeter@743:   http://bibtexml.sf.net/
kpeter@743:   Reuse approved as long as this notification is kept.
kpeter@743:   Licence: GPL.
kpeter@743: 
kpeter@743:   Contributions/thanks to:
kpeter@743:   Egon Willighagen, http://sf.net/projects/jreferences/
kpeter@743:   Richard Mahoney (for providing a test case)
kpeter@743: 
kpeter@743:   Editted by Sara Sprenkle to be more robust and handle more bibtex features.
kpeter@743:   (c) 2003-01-15
kpeter@743: 
kpeter@743:   1.  Changed bibtex: tags to bibxml: tags.
kpeter@743:   2.  Use xmlns:bibxml="http://bibtexml.sf.net/"
kpeter@743:   3.  Allow spaces between @type and first {
kpeter@743:   4.  "author" fields with multiple authors split by " and "
kpeter@743:       are put in separate xml "bibxml:author" tags.
kpeter@743:   5.  Option for Titles: words are capitalized
kpeter@743:       only if first letter in title or capitalized inside braces
kpeter@743:   6.  Removes braces from within field values
kpeter@743:   7.  Ignores comments in bibtex file (including @comment{ or % )
kpeter@743:   8.  Replaces some special latex tags, e.g., replaces ~ with ' '
kpeter@743:   9.  Handles bibtex @string abbreviations
kpeter@743:         --> includes bibtex's default abbreviations for months
kpeter@743:         --> does concatenation of abbr # " more " and " more " # abbr
kpeter@743:   10. Handles @type( ... ) or @type{ ... }
kpeter@743:   11. The keywords field is split on , or ; and put into separate xml
kpeter@743:       "bibxml:keywords" tags
kpeter@743:   12. Ignores @preamble
kpeter@743: 
kpeter@743:   Known Limitations
kpeter@743:   1.  Does not transform Latex encoding like math mode and special
kpeter@743:       latex symbols.
kpeter@743:   2.  Does not parse author fields into first and last names.
kpeter@743:       E.g., It does not do anything special to an author whose name is
kpeter@743:       in the form LAST_NAME, FIRST_NAME
kpeter@743:       In "author" tag, will show up as
kpeter@743:       <bibxml:author>LAST_NAME, FIRST_NAME</bibxml:author>
kpeter@743:   3.  Does not handle "crossref" fields other than to print
kpeter@743:       <bibxml:crossref>...</bibxml:crossref>
kpeter@743:   4.  Does not inform user of the input's format errors.  You just won't
kpeter@743:       be able to transform the file later with XSL
kpeter@743: 
kpeter@743:   You will have to manually edit the XML output if you need to handle
kpeter@743:   these (and unknown) limitations.
kpeter@743: 
kpeter@743: """
kpeter@743: 
kpeter@743: import string, re
kpeter@743: 
kpeter@743: # set of valid name characters
kpeter@743: valid_name_chars = '[\w\-:]'
kpeter@743: 
kpeter@743: #
kpeter@743: # define global regular expression variables
kpeter@743: #
kpeter@743: author_rex = re.compile('\s+and\s+')
kpeter@743: rembraces_rex = re.compile('[{}]')
kpeter@754: capitalize_rex = re.compile('({[^}]*})')
kpeter@743: 
kpeter@743: # used by bibtexkeywords(data)
kpeter@743: keywords_rex = re.compile('[,;]')
kpeter@743: 
kpeter@743: # used by concat_line(line)
kpeter@743: concatsplit_rex = re.compile('\s*#\s*')
kpeter@743: 
kpeter@743: # split on {, }, or " in verify_out_of_braces
kpeter@743: delimiter_rex = re.compile('([{}"])',re.I)
kpeter@743: 
kpeter@743: field_rex = re.compile('\s*(\w*)\s*=\s*(.*)')
kpeter@743: data_rex = re.compile('\s*(\w*)\s*=\s*([^,]*),?')
kpeter@743: 
kpeter@743: url_rex = re.compile('\\\url\{([^}]*)\}')
kpeter@743: 
kpeter@745: #
kpeter@745: # styles for html formatting
kpeter@745: #
kpeter@745: divstyle = 'margin-top: -4ex; margin-left: 8em;'
kpeter@743: 
kpeter@743: #
kpeter@743: # return the string parameter without braces
kpeter@743: #
kpeter@743: def transformurls(str):
kpeter@743:     return url_rex.sub(r'<a href="\1">\1</a>', str)
kpeter@743: 
kpeter@743: #
kpeter@743: # return the string parameter without braces
kpeter@743: #
kpeter@743: def removebraces(str):
kpeter@743:     return rembraces_rex.sub('', str)
kpeter@743: 
kpeter@743: #
kpeter@743: # latex-specific replacements
kpeter@743: # (do this after braces were removed)
kpeter@743: #
kpeter@743: def latexreplacements(line):
kpeter@743:     line = string.replace(line, '~', '&nbsp;')
kpeter@743:     line = string.replace(line, '\\\'a', '&aacute;')
kpeter@743:     line = string.replace(line, '\\"a', '&auml;')
kpeter@743:     line = string.replace(line, '\\\'e', '&eacute;')
kpeter@743:     line = string.replace(line, '\\"e', '&euml;')
kpeter@743:     line = string.replace(line, '\\\'i', '&iacute;')
kpeter@743:     line = string.replace(line, '\\"i', '&iuml;')
kpeter@743:     line = string.replace(line, '\\\'o', '&oacute;')
kpeter@743:     line = string.replace(line, '\\"o', '&ouml;')
kpeter@743:     line = string.replace(line, '\\\'u', '&uacute;')
kpeter@743:     line = string.replace(line, '\\"u', '&uuml;')
kpeter@743:     line = string.replace(line, '\\H o', '&otilde;')
kpeter@743:     line = string.replace(line, '\\H u', '&uuml;')   # &utilde; does not exist
kpeter@743:     line = string.replace(line, '\\\'A', '&Aacute;')
kpeter@743:     line = string.replace(line, '\\"A', '&Auml;')
kpeter@743:     line = string.replace(line, '\\\'E', '&Eacute;')
kpeter@743:     line = string.replace(line, '\\"E', '&Euml;')
kpeter@743:     line = string.replace(line, '\\\'I', '&Iacute;')
kpeter@743:     line = string.replace(line, '\\"I', '&Iuml;')
kpeter@743:     line = string.replace(line, '\\\'O', '&Oacute;')
kpeter@743:     line = string.replace(line, '\\"O', '&Ouml;')
kpeter@743:     line = string.replace(line, '\\\'U', '&Uacute;')
kpeter@743:     line = string.replace(line, '\\"U', '&Uuml;')
kpeter@743:     line = string.replace(line, '\\H O', '&Otilde;')
kpeter@743:     line = string.replace(line, '\\H U', '&Uuml;')   # &Utilde; does not exist
kpeter@743: 
kpeter@743:     return line
kpeter@743: 
kpeter@743: #
kpeter@743: # copy characters form a string decoding html expressions (&xyz;)
kpeter@743: #
kpeter@743: def copychars(str, ifrom, count):
kpeter@743:     result = ''
kpeter@743:     i = ifrom
kpeter@743:     c = 0
kpeter@743:     html_spec = False
kpeter@743:     while (i < len(str)) and (c < count):
kpeter@743:         if str[i] == '&':
kpeter@743:             html_spec = True;
kpeter@743:             if i+1 < len(str):
kpeter@743:                 result += str[i+1]
kpeter@743:             c += 1
kpeter@743:             i += 2
kpeter@743:         else:
kpeter@743:             if not html_spec:
kpeter@743:                 if ((str[i] >= 'A') and (str[i] <= 'Z')) or \
kpeter@743:                    ((str[i] >= 'a') and (str[i] <= 'z')):
kpeter@743:                     result += str[i]
kpeter@743:                     c += 1
kpeter@743:             elif str[i] == ';':
kpeter@743:                 html_spec = False;
kpeter@743:             i += 1
kpeter@743:     
kpeter@743:     return result
kpeter@743: 
kpeter@743: 
kpeter@743: # 
kpeter@743: # Handle a list of authors (separated by 'and').
kpeter@743: # It gives back an array of the follwing values:
kpeter@743: #  - num: the number of authors,
kpeter@743: #  - list: the list of the author names,
kpeter@743: #  - text: the bibtex text (separated by commas and/or 'and')
kpeter@743: #  - abbrev: abbreviation that can be used for indicate the
kpeter@743: #    bibliography entries
kpeter@743: #
kpeter@743: def bibtexauthor(data):
kpeter@743:     result = {}
kpeter@743:     bibtex = ''
kpeter@743:     result['list'] = author_rex.split(data)
kpeter@743:     result['num'] = len(result['list'])
kpeter@743:     for i, author in enumerate(result['list']):
kpeter@743:         # general transformations
kpeter@743:         author = latexreplacements(removebraces(author.strip()))
kpeter@743:         # transform "Xyz, A. B." to "A. B. Xyz"
kpeter@743:         pos = author.find(',')
kpeter@743:         if pos != -1:
kpeter@743:             author = author[pos+1:].strip() + ' ' + author[:pos].strip()
kpeter@743:         result['list'][i] = author
kpeter@743:         bibtex += author + '#'
kpeter@743:     bibtex = bibtex[:-1]
kpeter@743:     if result['num'] > 1:
kpeter@743:         ix = bibtex.rfind('#')
kpeter@743:         if result['num'] == 2:
kpeter@743:             bibtex = bibtex[:ix] + ' and ' + bibtex[ix+1:]
kpeter@743:         else:
kpeter@743:             bibtex = bibtex[:ix] + ', and ' + bibtex[ix+1:]
kpeter@743:     bibtex = bibtex.replace('#', ', ')
kpeter@743:     result['text'] = bibtex
kpeter@743:     
kpeter@743:     result['abbrev'] = ''
kpeter@743:     for author in result['list']:
kpeter@743:         pos = author.rfind(' ') + 1
kpeter@743:         count = 1
kpeter@743:         if result['num'] == 1:
kpeter@743:             count = 3
kpeter@743:         result['abbrev'] += copychars(author, pos, count)
kpeter@743: 
kpeter@743:     return result
kpeter@743: 
kpeter@743: 
kpeter@743: #
kpeter@743: # data = title string
kpeter@743: # @return the capitalized title (first letter is capitalized), rest are capitalized
kpeter@743: # only if capitalized inside braces
kpeter@743: #
kpeter@743: def capitalizetitle(data):
kpeter@743:     title_list = capitalize_rex.split(data)
kpeter@743:     title = ''
kpeter@743:     count = 0
kpeter@743:     for phrase in title_list:
kpeter@743:          check = string.lstrip(phrase)
kpeter@743: 
kpeter@743:          # keep phrase's capitalization the same
kpeter@743:          if check.find('{') == 0:
kpeter@743:               title += removebraces(phrase)
kpeter@743:          else:
kpeter@743:          # first word --> capitalize first letter (after spaces)
kpeter@743:               if count == 0:
kpeter@743:                   title += check.capitalize()
kpeter@743:               else:
kpeter@743:                   title += phrase.lower()
kpeter@743:          count = count + 1
kpeter@743: 
kpeter@743:     return title
kpeter@743: 
kpeter@743: 
kpeter@743: #
kpeter@743: # @return the bibtex for the title
kpeter@743: # @param data --> title string
kpeter@743: # braces are removed from title
kpeter@743: #
kpeter@743: def bibtextitle(data, entrytype):
kpeter@743:     if entrytype in ('book', 'inbook'):
kpeter@743:         title = removebraces(data.strip())
kpeter@743:     else:
kpeter@743:         title = removebraces(capitalizetitle(data.strip()))
kpeter@743:     bibtex = title
kpeter@743:     return bibtex
kpeter@743: 
kpeter@743: 
kpeter@743: #
kpeter@743: # function to compare entry lists
kpeter@743: #
kpeter@743: def entry_cmp(x, y):
kpeter@743:     return cmp(x[0], y[0])
kpeter@743: 
kpeter@743: 
kpeter@743: #
kpeter@743: # print the XML for the transformed "filecont_source"
kpeter@743: #
kpeter@743: def bibtexdecoder(filecont_source):
kpeter@743:     filecont = []
kpeter@743:     file = []
kpeter@743:     
kpeter@743:     # want @<alphanumeric chars><spaces>{<spaces><any chars>,
kpeter@743:     pubtype_rex = re.compile('@(\w*)\s*{\s*(.*),')
kpeter@743:     endtype_rex = re.compile('}\s*$')
kpeter@743:     endtag_rex = re.compile('^\s*}\s*$')
kpeter@743: 
kpeter@743:     bracefield_rex = re.compile('\s*(\w*)\s*=\s*(.*)')
kpeter@743:     bracedata_rex = re.compile('\s*(\w*)\s*=\s*{(.*)},?')
kpeter@743: 
kpeter@743:     quotefield_rex = re.compile('\s*(\w*)\s*=\s*(.*)')
kpeter@743:     quotedata_rex = re.compile('\s*(\w*)\s*=\s*"(.*)",?')
kpeter@743: 
kpeter@743:     for line in filecont_source:
kpeter@743:         line = line[:-1]
kpeter@743: 
kpeter@743:         # encode character entities
kpeter@743:         line = string.replace(line, '&', '&amp;')
kpeter@743:         line = string.replace(line, '<', '&lt;')
kpeter@743:         line = string.replace(line, '>', '&gt;')
kpeter@743: 
kpeter@743:         # start entry: publication type (store for later use)
kpeter@743:         if pubtype_rex.match(line):
kpeter@743:         # want @<alphanumeric chars><spaces>{<spaces><any chars>,
kpeter@743:             entrycont = {}
kpeter@743:             entry = []
kpeter@743:             entrytype = pubtype_rex.sub('\g<1>',line)
kpeter@743:             entrytype = string.lower(entrytype)
kpeter@745:             entryid   = pubtype_rex.sub('\g<2>', line)
kpeter@743: 
kpeter@743:         # end entry if just a }
kpeter@743:         elif endtype_rex.match(line):
kpeter@743:             # generate doxygen code for the entry
kpeter@743: 
kpeter@743:             # enty type related formattings
kpeter@743:             if entrytype in ('book', 'inbook'):
kpeter@743:                 entrycont['title'] = '<em>' + entrycont['title'] + '</em>'
kpeter@743:                 if not entrycont.has_key('author'):
kpeter@743:                     entrycont['author'] = entrycont['editor']
kpeter@743:                     entrycont['author']['text'] += ', editors'
kpeter@743:             elif entrytype == 'article':
kpeter@743:                 entrycont['journal'] = '<em>' + entrycont['journal'] + '</em>'
kpeter@743:             elif entrytype in ('inproceedings', 'incollection', 'conference'):
kpeter@743:                 entrycont['booktitle'] = '<em>' + entrycont['booktitle'] + '</em>'
kpeter@743:             elif entrytype == 'techreport':
kpeter@743:                 if not entrycont.has_key('type'):
kpeter@743:                     entrycont['type'] = 'Technical report'
kpeter@743:             elif entrytype == 'mastersthesis':
kpeter@743:                 entrycont['type'] = 'Master\'s thesis'
kpeter@743:             elif entrytype == 'phdthesis':
kpeter@743:                 entrycont['type'] = 'PhD thesis'
kpeter@743: 
kpeter@743:             for eline in entrycont:
kpeter@743:                 if eline != '':
kpeter@743:                     eline = latexreplacements(eline)
kpeter@743: 
kpeter@743:             if entrycont.has_key('pages') and (entrycont['pages'] != ''):
kpeter@743:                 entrycont['pages'] = string.replace(entrycont['pages'], '--', '-')
kpeter@743: 
kpeter@743:             if entrycont.has_key('author') and (entrycont['author'] != ''):
kpeter@743:                 entry.append(entrycont['author']['text'] + '.')
kpeter@743:             if entrycont.has_key('title') and (entrycont['title'] != ''):
kpeter@743:                 entry.append(entrycont['title'] + '.')
kpeter@743:             if entrycont.has_key('journal') and (entrycont['journal'] != ''):
kpeter@743:                 entry.append(entrycont['journal'] + ',')
kpeter@743:             if entrycont.has_key('booktitle') and (entrycont['booktitle'] != ''):
kpeter@743:                 entry.append('In ' + entrycont['booktitle'] + ',')
kpeter@743:             if entrycont.has_key('type') and (entrycont['type'] != ''):
kpeter@743:                 eline = entrycont['type']
kpeter@743:                 if entrycont.has_key('number') and (entrycont['number'] != ''):
kpeter@743:                     eline += ' ' + entrycont['number']
kpeter@743:                 eline += ','
kpeter@743:                 entry.append(eline)
kpeter@743:             if entrycont.has_key('institution') and (entrycont['institution'] != ''):
kpeter@743:                 entry.append(entrycont['institution'] + ',')
kpeter@743:             if entrycont.has_key('publisher') and (entrycont['publisher'] != ''):
kpeter@743:                 entry.append(entrycont['publisher'] + ',')
kpeter@743:             if entrycont.has_key('school') and (entrycont['school'] != ''):
kpeter@743:                 entry.append(entrycont['school'] + ',')
kpeter@743:             if entrycont.has_key('address') and (entrycont['address'] != ''):
kpeter@743:                 entry.append(entrycont['address'] + ',')
kpeter@743:             if entrycont.has_key('edition') and (entrycont['edition'] != ''):
kpeter@743:                 entry.append(entrycont['edition'] + ' edition,')
kpeter@743:             if entrycont.has_key('howpublished') and (entrycont['howpublished'] != ''):
kpeter@743:                 entry.append(entrycont['howpublished'] + ',')
kpeter@743:             if entrycont.has_key('volume') and (entrycont['volume'] != ''):
kpeter@743:                 eline = entrycont['volume'];
kpeter@743:                 if entrycont.has_key('number') and (entrycont['number'] != ''):
kpeter@743:                     eline += '(' + entrycont['number'] + ')'
kpeter@743:                 if entrycont.has_key('pages') and (entrycont['pages'] != ''):
kpeter@743:                     eline += ':' + entrycont['pages']
kpeter@743:                 eline += ','
kpeter@743:                 entry.append(eline)
kpeter@743:             else:
kpeter@743:                 if entrycont.has_key('pages') and (entrycont['pages'] != ''):
kpeter@743:                     entry.append('pages ' + entrycont['pages'] + ',')
kpeter@743:             if entrycont.has_key('year') and (entrycont['year'] != ''):
kpeter@743:                 if entrycont.has_key('month') and (entrycont['month'] != ''):
kpeter@743:                     entry.append(entrycont['month'] + ' ' + entrycont['year'] + '.')
kpeter@743:                 else:
kpeter@743:                     entry.append(entrycont['year'] + '.')
kpeter@743:             if entrycont.has_key('note') and (entrycont['note'] != ''):
kpeter@743:                 entry.append(entrycont['note'] + '.')
kpeter@754:             if entrycont.has_key('url') and (entrycont['url'] != ''):
kpeter@754:                 entry.append(entrycont['url'] + '.')
kpeter@743: 
kpeter@743:             # generate keys for sorting and for the output
kpeter@743:             sortkey = ''
kpeter@743:             bibkey = ''
kpeter@743:             if entrycont.has_key('author'):
kpeter@743:                 for author in entrycont['author']['list']:
kpeter@743:                     sortkey += copychars(author, author.rfind(' ')+1, len(author))
kpeter@743:                 bibkey = entrycont['author']['abbrev']
kpeter@743:             else:
kpeter@743:                 bibkey = 'x'
kpeter@743:             if entrycont.has_key('year'):
kpeter@743:                 sortkey += entrycont['year']
kpeter@743:                 bibkey += entrycont['year'][-2:]
kpeter@743:             if entrycont.has_key('title'):
kpeter@743:                 sortkey += entrycont['title']
kpeter@743:             if entrycont.has_key('key'):
kpeter@743:                 sortkey = entrycont['key'] + sortkey
kpeter@743:                 bibkey = entrycont['key']
kpeter@743:             entry.insert(0, sortkey)
kpeter@743:             entry.insert(1, bibkey)
kpeter@745:             entry.insert(2, entryid)
kpeter@743:            
kpeter@743:             # add the entry to the file contents
kpeter@743:             filecont.append(entry)
kpeter@743: 
kpeter@743:         else:
kpeter@743:             # field, publication info
kpeter@743:             field = ''
kpeter@743:             data = ''
kpeter@743:             
kpeter@743:             # field = {data} entries
kpeter@743:             if bracedata_rex.match(line):
kpeter@743:                 field = bracefield_rex.sub('\g<1>', line)
kpeter@743:                 field = string.lower(field)
kpeter@743:                 data =  bracedata_rex.sub('\g<2>', line)
kpeter@743: 
kpeter@743:             # field = "data" entries
kpeter@743:             elif quotedata_rex.match(line):
kpeter@743:                 field = quotefield_rex.sub('\g<1>', line)
kpeter@743:                 field = string.lower(field)
kpeter@743:                 data =  quotedata_rex.sub('\g<2>', line)
kpeter@743: 
kpeter@743:             # field = data entries
kpeter@743:             elif data_rex.match(line):
kpeter@743:                 field = field_rex.sub('\g<1>', line)
kpeter@743:                 field = string.lower(field)
kpeter@743:                 data =  data_rex.sub('\g<2>', line)
kpeter@754: 
kpeter@754:             if field == 'url':
kpeter@754:                 data = '\\url{' + data.strip() + '}'
kpeter@743:             
kpeter@743:             if field in ('author', 'editor'):
kpeter@743:                 entrycont[field] = bibtexauthor(data)
kpeter@743:                 line = ''
kpeter@743:             elif field == 'title':
kpeter@743:                 line = bibtextitle(data, entrytype)
kpeter@743:             elif field != '':
kpeter@743:                 line = removebraces(transformurls(data.strip()))
kpeter@743: 
kpeter@743:             if line != '':
kpeter@743:                 line = latexreplacements(line)
kpeter@743:                 entrycont[field] = line
kpeter@743: 
kpeter@743: 
kpeter@743:     # sort entries
kpeter@743:     filecont.sort(entry_cmp)
kpeter@743:     
kpeter@743:     # count the bibtex keys
kpeter@743:     keytable = {}
kpeter@743:     counttable = {}
kpeter@743:     for entry in filecont:
kpeter@743:         bibkey = entry[1]
kpeter@743:         if not keytable.has_key(bibkey):
kpeter@743:             keytable[bibkey] = 1
kpeter@743:         else:
kpeter@743:             keytable[bibkey] += 1
kpeter@743: 
kpeter@743:     for bibkey in keytable.keys():
kpeter@743:         counttable[bibkey] = 0
kpeter@743:     
kpeter@743:     # generate output
kpeter@743:     for entry in filecont:
kpeter@743:         # generate output key form the bibtex key
kpeter@743:         bibkey = entry[1]
kpeter@745:         entryid = entry[2]
kpeter@743:         if keytable[bibkey] == 1:
kpeter@743:             outkey = bibkey
kpeter@743:         else:
kpeter@743:             outkey = bibkey + chr(97 + counttable[bibkey])
kpeter@743:         counttable[bibkey] += 1
kpeter@743:         
kpeter@743:         # append the entry code to the output
kpeter@745:         file.append('\\section ' + entryid + ' [' + outkey + ']')
kpeter@745:         file.append('<div style="' + divstyle + '">')
kpeter@745:         for line in entry[3:]:
kpeter@743:             file.append(line)
kpeter@745:         file.append('</div>')
kpeter@743:         file.append('')
kpeter@743: 
kpeter@743:     return file
kpeter@743: 
kpeter@743: 
kpeter@743: #
kpeter@743: # return 1 iff abbr is in line but not inside braces or quotes
kpeter@743: # assumes that abbr appears only once on the line (out of braces and quotes)
kpeter@743: #
kpeter@743: def verify_out_of_braces(line, abbr):
kpeter@743: 
kpeter@743:     phrase_split = delimiter_rex.split(line)
kpeter@743: 
kpeter@743:     abbr_rex = re.compile( '\\b' + abbr + '\\b', re.I)
kpeter@743: 
kpeter@743:     open_brace = 0
kpeter@743:     open_quote = 0
kpeter@743: 
kpeter@743:     for phrase in phrase_split:
kpeter@743:         if phrase == "{":
kpeter@743:             open_brace = open_brace + 1
kpeter@743:         elif phrase == "}":
kpeter@743:             open_brace = open_brace - 1
kpeter@743:         elif phrase == '"':
kpeter@743:             if open_quote == 1:
kpeter@743:                 open_quote = 0
kpeter@743:             else:
kpeter@743:                 open_quote = 1
kpeter@743:         elif abbr_rex.search(phrase):
kpeter@743:             if open_brace == 0 and open_quote == 0:
kpeter@743:                 return 1
kpeter@743: 
kpeter@743:     return 0
kpeter@743: 
kpeter@743: 
kpeter@743: #
kpeter@743: # a line in the form phrase1 # phrase2 # ... # phrasen
kpeter@743: # is returned as phrase1 phrase2 ... phrasen
kpeter@743: # with the correct punctuation
kpeter@743: # Bug: Doesn't always work with multiple abbreviations plugged in
kpeter@743: #
kpeter@743: def concat_line(line):
kpeter@743:     # only look at part after equals
kpeter@743:     field = field_rex.sub('\g<1>',line)
kpeter@743:     rest = field_rex.sub('\g<2>',line)
kpeter@743: 
kpeter@743:     concat_line = field + ' ='
kpeter@743: 
kpeter@743:     pound_split = concatsplit_rex.split(rest)
kpeter@743: 
kpeter@743:     phrase_count = 0
kpeter@743:     length = len(pound_split)
kpeter@743: 
kpeter@743:     for phrase in pound_split:
kpeter@743:         phrase = phrase.strip()
kpeter@743:         if phrase_count != 0:
kpeter@743:             if phrase.startswith('"') or phrase.startswith('{'):
kpeter@743:                 phrase = phrase[1:]
kpeter@743:         elif phrase.startswith('"'):
kpeter@743:             phrase = phrase.replace('"','{',1)
kpeter@743: 
kpeter@743:         if phrase_count != length-1:
kpeter@743:             if phrase.endswith('"') or phrase.endswith('}'):
kpeter@743:                 phrase = phrase[:-1]
kpeter@743:         else:
kpeter@743:             if phrase.endswith('"'):
kpeter@743:                 phrase = phrase[:-1]
kpeter@743:                 phrase = phrase + "}"
kpeter@743:             elif phrase.endswith('",'):
kpeter@743:                 phrase = phrase[:-2]
kpeter@743:                 phrase = phrase + "},"
kpeter@743: 
kpeter@743:         # if phrase did have \#, add the \# back
kpeter@743:         if phrase.endswith('\\'):
kpeter@743:             phrase = phrase + "#"
kpeter@743:         concat_line = concat_line + ' ' + phrase
kpeter@743: 
kpeter@743:         phrase_count = phrase_count + 1
kpeter@743: 
kpeter@743:     return concat_line
kpeter@743: 
kpeter@743: 
kpeter@743: #
kpeter@743: # substitute abbreviations into filecont
kpeter@743: # @param filecont_source - string of data from file
kpeter@743: #
kpeter@743: def bibtex_replace_abbreviations(filecont_source):
kpeter@743:     filecont = filecont_source.splitlines()
kpeter@743: 
kpeter@743:     #  These are defined in bibtex, so we'll define them too
kpeter@743:     abbr_list = ['jan','feb','mar','apr','may','jun',
kpeter@743:                  'jul','aug','sep','oct','nov','dec']
kpeter@743:     value_list = ['January','February','March','April',
kpeter@743:                   'May','June','July','August','September',
kpeter@743:                   'October','November','December']
kpeter@743: 
kpeter@743:     abbr_rex = []
kpeter@743:     total_abbr_count = 0
kpeter@743: 
kpeter@743:     front = '\\b'
kpeter@743:     back = '(,?)\\b'
kpeter@743: 
kpeter@743:     for x in abbr_list:
kpeter@743:         abbr_rex.append( re.compile( front + abbr_list[total_abbr_count] + back, re.I ) )
kpeter@743:         total_abbr_count = total_abbr_count + 1
kpeter@743: 
kpeter@743: 
kpeter@743:     abbrdef_rex = re.compile('\s*@string\s*{\s*('+ valid_name_chars +'*)\s*=(.*)',
kpeter@743:                              re.I)
kpeter@743: 
kpeter@743:     comment_rex = re.compile('@comment\s*{',re.I)
kpeter@743:     preamble_rex = re.compile('@preamble\s*{',re.I)
kpeter@743: 
kpeter@743:     waiting_for_end_string = 0
kpeter@743:     i = 0
kpeter@743:     filecont2 = ''
kpeter@743: 
kpeter@743:     for line in filecont:
kpeter@743:         if line == ' ' or line == '':
kpeter@743:             continue
kpeter@743: 
kpeter@743:         if waiting_for_end_string:
kpeter@743:             if re.search('}',line):
kpeter@743:                 waiting_for_end_string = 0
kpeter@743:                 continue
kpeter@743: 
kpeter@743:         if abbrdef_rex.search(line):
kpeter@743:             abbr = abbrdef_rex.sub('\g<1>', line)
kpeter@743: 
kpeter@743:             if abbr_list.count(abbr) == 0:
kpeter@743:                 val = abbrdef_rex.sub('\g<2>', line)
kpeter@743:                 abbr_list.append(abbr)
kpeter@743:                 value_list.append(string.strip(val))
kpeter@743:                 abbr_rex.append( re.compile( front + abbr_list[total_abbr_count] + back, re.I ) )
kpeter@743:                 total_abbr_count = total_abbr_count + 1
kpeter@743:             waiting_for_end_string = 1
kpeter@743:             continue
kpeter@743: 
kpeter@743:         if comment_rex.search(line):
kpeter@743:             waiting_for_end_string = 1
kpeter@743:             continue
kpeter@743: 
kpeter@743:         if preamble_rex.search(line):
kpeter@743:             waiting_for_end_string = 1
kpeter@743:             continue
kpeter@743: 
kpeter@743: 
kpeter@743:         # replace subsequent abbreviations with the value
kpeter@743:         abbr_count = 0
kpeter@743: 
kpeter@743:         for x in abbr_list:
kpeter@743: 
kpeter@743:             if abbr_rex[abbr_count].search(line):
kpeter@743:                 if verify_out_of_braces(line,abbr_list[abbr_count]) == 1:
kpeter@743:                     line = abbr_rex[abbr_count].sub( value_list[abbr_count] + '\g<1>', line)
kpeter@743:                 # Check for # concatenations
kpeter@743:                 if concatsplit_rex.search(line):
kpeter@743:                     line = concat_line(line)
kpeter@743:             abbr_count = abbr_count + 1
kpeter@743: 
kpeter@743: 
kpeter@743:         filecont2 = filecont2 + line + '\n'
kpeter@743:         i = i+1
kpeter@743: 
kpeter@743: 
kpeter@743:     # Do one final pass over file
kpeter@743: 
kpeter@743:     # make sure that didn't end up with {" or }" after the substitution
kpeter@743:     filecont2 = filecont2.replace('{"','{{')
kpeter@743:     filecont2 = filecont2.replace('"}','}}')
kpeter@743: 
kpeter@743:     afterquotevalue_rex = re.compile('"\s*,\s*')
kpeter@743:     afterbrace_rex = re.compile('"\s*}')
kpeter@743:     afterbracevalue_rex = re.compile('(=\s*{[^=]*)},\s*')
kpeter@743: 
kpeter@743:     # add new lines to data that changed because of abbreviation substitutions
kpeter@743:     filecont2 = afterquotevalue_rex.sub('",\n', filecont2)
kpeter@743:     filecont2 = afterbrace_rex.sub('"\n}', filecont2)
kpeter@743:     filecont2 = afterbracevalue_rex.sub('\g<1>},\n', filecont2)
kpeter@743: 
kpeter@743:     return filecont2
kpeter@743: 
kpeter@743: #
kpeter@743: # convert @type( ... ) to @type{ ... }
kpeter@743: #
kpeter@743: def no_outer_parens(filecont):
kpeter@743: 
kpeter@743:     # do checking for open parens
kpeter@743:     # will convert to braces
kpeter@743:     paren_split = re.split('([(){}])',filecont)
kpeter@743: 
kpeter@743:     open_paren_count = 0
kpeter@743:     open_type = 0
kpeter@743:     look_next = 0
kpeter@743: 
kpeter@743:     # rebuild filecont
kpeter@743:     filecont = ''
kpeter@743: 
kpeter@743:     at_rex = re.compile('@\w*')
kpeter@743: 
kpeter@743:     for phrase in paren_split:
kpeter@743:         if look_next == 1:
kpeter@743:             if phrase == '(':
kpeter@743:                 phrase = '{'
kpeter@743:                 open_paren_count = open_paren_count + 1
kpeter@743:             else:
kpeter@743:                 open_type = 0
kpeter@743:             look_next = 0
kpeter@743: 
kpeter@743:         if phrase == '(':
kpeter@743:             open_paren_count = open_paren_count + 1
kpeter@743: 
kpeter@743:         elif phrase == ')':
kpeter@743:             open_paren_count = open_paren_count - 1
kpeter@743:             if open_type == 1 and open_paren_count == 0:
kpeter@743:                 phrase = '}'
kpeter@743:                 open_type = 0
kpeter@743: 
kpeter@743:         elif at_rex.search( phrase ):
kpeter@743:             open_type = 1
kpeter@743:             look_next = 1
kpeter@743: 
kpeter@743:         filecont = filecont + phrase
kpeter@743: 
kpeter@743:     return filecont
kpeter@743: 
kpeter@743: 
kpeter@743: #
kpeter@743: # make all whitespace into just one space
kpeter@743: # format the bibtex file into a usable form.
kpeter@743: #
kpeter@743: def bibtexwasher(filecont_source):
kpeter@743: 
kpeter@743:     space_rex = re.compile('\s+')
kpeter@743:     comment_rex = re.compile('\s*%')
kpeter@743: 
kpeter@743:     filecont = []
kpeter@743: 
kpeter@743:     # remove trailing and excessive whitespace
kpeter@743:     # ignore comments
kpeter@743:     for line in filecont_source:
kpeter@743:         line = string.strip(line)
kpeter@743:         line = space_rex.sub(' ', line)
kpeter@743:         # ignore comments
kpeter@743:         if not comment_rex.match(line) and line != '':
kpeter@743:             filecont.append(' '+ line)
kpeter@743: 
kpeter@743:     filecont = string.join(filecont, '')
kpeter@743: 
kpeter@743:     # the file is in one long string
kpeter@743: 
kpeter@743:     filecont = no_outer_parens(filecont)
kpeter@743: 
kpeter@743:     #
kpeter@743:     # split lines according to preferred syntax scheme
kpeter@743:     #
kpeter@743:     filecont = re.sub('(=\s*{[^=]*)},', '\g<1>},\n', filecont)
kpeter@743: 
kpeter@743:     # add new lines after commas that are after values
kpeter@743:     filecont = re.sub('"\s*,', '",\n', filecont)
kpeter@743:     filecont = re.sub('=\s*([\w\d]+)\s*,', '= \g<1>,\n', filecont)
kpeter@743:     filecont = re.sub('(@\w*)\s*({(\s*)[^,\s]*)\s*,',
kpeter@743:                           '\n\n\g<1>\g<2>,\n', filecont)
kpeter@743: 
kpeter@743:     # add new lines after }
kpeter@743:     filecont = re.sub('"\s*}','"\n}\n', filecont)
kpeter@743:     filecont = re.sub('}\s*,','},\n', filecont)
kpeter@743: 
kpeter@743: 
kpeter@743:     filecont = re.sub('@(\w*)', '\n@\g<1>', filecont)
kpeter@743: 
kpeter@743:     # character encoding, reserved latex characters
kpeter@743:     filecont = re.sub('{\\\&}', '&', filecont)
kpeter@743:     filecont = re.sub('\\\&', '&', filecont)
kpeter@743: 
kpeter@743:     # do checking for open braces to get format correct
kpeter@743:     open_brace_count = 0
kpeter@743:     brace_split = re.split('([{}])',filecont)
kpeter@743: 
kpeter@743:     # rebuild filecont
kpeter@743:     filecont = ''
kpeter@743: 
kpeter@743:     for phrase in brace_split:
kpeter@743:         if phrase == '{':
kpeter@743:             open_brace_count = open_brace_count + 1
kpeter@743:         elif phrase == '}':
kpeter@743:             open_brace_count = open_brace_count - 1
kpeter@743:             if open_brace_count == 0:
kpeter@743:                 filecont = filecont + '\n'
kpeter@743: 
kpeter@743:         filecont = filecont + phrase
kpeter@743: 
kpeter@743:     filecont2 = bibtex_replace_abbreviations(filecont)
kpeter@743: 
kpeter@743:     # gather
kpeter@743:     filecont = filecont2.splitlines()
kpeter@743:     i=0
kpeter@743:     j=0         # count the number of blank lines
kpeter@743:     for line in filecont:
kpeter@743:         # ignore blank lines
kpeter@743:         if line == '' or line == ' ':
kpeter@743:             j = j+1
kpeter@743:             continue
kpeter@743:         filecont[i] = line + '\n'
kpeter@743:         i = i+1
kpeter@743: 
kpeter@743:     # get rid of the extra stuff at the end of the array
kpeter@743:     # (The extra stuff are duplicates that are in the array because
kpeter@743:     # blank lines were removed.)
kpeter@743:     length = len( filecont)
kpeter@743:     filecont[length-j:length] = []
kpeter@743: 
kpeter@743:     return filecont
kpeter@743: 
kpeter@743: 
kpeter@743: def filehandler(filepath):
kpeter@743:     try:
kpeter@743:         fd = open(filepath, 'r')
kpeter@743:         filecont_source = fd.readlines()
kpeter@743:         fd.close()
kpeter@743:     except:
kpeter@743:         print 'Could not open file:', filepath
kpeter@743:     washeddata = bibtexwasher(filecont_source)
kpeter@743:     outdata = bibtexdecoder(washeddata)
kpeter@743:     print '/**'
kpeter@743:     print '\page references References'
kpeter@743:     print
kpeter@743:     for line in outdata:
kpeter@743:         print line
kpeter@743:     print '*/'
kpeter@743: 
kpeter@743: 
kpeter@743: # main program
kpeter@743: 
kpeter@743: def main():
kpeter@743:     import sys
kpeter@743:     if sys.argv[1:]:
kpeter@743:         filepath = sys.argv[1]
kpeter@743:     else:
kpeter@743:         print "No input file"
kpeter@743:         sys.exit()
kpeter@743:     filehandler(filepath)
kpeter@743: 
kpeter@743: if __name__ == "__main__": main()
kpeter@743: 
kpeter@743: 
kpeter@743: # end python script