bib2html.py

   1 #!/usr/bin/env python3
   2 """Creates a webpage with all entries of a .bib file"""
   3
   4 __version__ = "1.1"
   5
   6 __author__ = "Stefan Huber"
   7 __email__ = "shuber@sthu.org"
   8 __copyright__ = "Copyright 2013, Stefan Huber"
   9
  10 __license__ = "MIT"
  11
  12 # Permission is hereby granted, free of charge, to any person
  13 # obtaining a copy of this software and associated documentation
  14 # files (the "Software"), to deal in the Software without
  15 # restriction, including without limitation the rights to use,
  16 # copy, modify, merge, publish, distribute, sublicense, and/or sell
  17 # copies of the Software, and to permit persons to whom the
  18 # Software is furnished to do so, subject to the following
  19 # conditions:
  20 #
  21 # The above copyright notice and this permission notice shall be
  22 # included in all copies or substantial portions of the Software.
  23 #
  24 # THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
  25 # EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
  26 # OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
  27 # NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
  28 # HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
  29 # WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
  30 # FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
  31 # OTHER DEALINGS IN THE SOFTWARE.
  32
  33
  34 import os, sys, getopt, re
  35
  36
  37 def format_latex(text):
  38     # Get rid of matching dollar signs
  39     text = re.sub(r'\$([^\$]*)\$', r'\1', text)
  40     return text.replace('\mathcal', '').replace('{', '').replace('}', '').replace('\\', '')
  41
  42 def format_field_span(type, value):
  43     return "<span class=bibentry_" + type + ">" + format_latex(value) + "</span>"
  44
  45 def format_field(bibentry, field, pre='', post=''):
  46     if field in bibentry.fields:
  47         if bibentry.fields[field] != "":
  48             return format_field_span(field, pre + bibentry.fields[field] + post)
  49     return ""
  50
  51 def format_author(a):
  52     return  ' '.join(' '.join(p) for p in (a.first_names, a.middle_names, a.prelast_names, a.last_names, a.lineage_names) if p)
  53
  54 def format_authors(entry):
  55     return ", ".join([format_author(a) for a in entry.persons['author']])
  56
  57
  58 def format_details_article(entry):
  59
  60     where = format_field(entry, 'journal')
  61
  62     line = []
  63     line.append(format_field(entry, 'pages', pre='pp. '))
  64     line.append(format_field(entry, 'volume', pre='vol. ') + \
  65             format_field(entry, 'number', pre='(', post=')'))
  66     line.append(format_field(entry, 'month', post=' ') + \
  67             format_field(entry, 'year'))
  68     line.append(format_field(entry, 'note'))
  69
  70     line = filter(lambda l: l != "", line)
  71     return [where, ", ".join(line)]
  72
  73 def format_details_inproceedings(entry):
  74     where = format_field(entry, 'booktitle')
  75
  76     line = []
  77     line.append(format_field(entry, 'pages', pre='pp. '))
  78     line.append(format_field(entry, 'address'))
  79     line.append(format_field(entry, 'month', post=' ') + \
  80             format_field(entry, 'year'))
  81     line.append(format_field(entry, 'isbn', pre='ISBN '))
  82     line.append(format_field(entry, 'note'))
  83
  84     line = filter(lambda l: l != "", line)
  85     return [where, ", ".join(line)]
  86
  87 def format_details_thesis(entry):
  88     line = []
  89     line.append(format_field(entry, 'school'))
  90     line.append(format_field(entry, 'month', post=' ') + \
  91             format_field(entry, 'year'))
  92     line.append(format_field(entry, 'note'))
  93
  94     line = filter(lambda l: l != "", line)
  95     return [", ".join(line)]
  96
  97 def format_details_book(entry):
  98     line = []
  99     line.append(format_field(entry, 'publisher'))
 100     line.append(format_field(entry, 'isbn', pre='ISBN '))
 101     line.append(format_field(entry, 'month', post=' ') + \
 102             format_field(entry, 'year'))
 103     line.append(format_field(entry, 'note'))
 104
 105     line = filter(lambda l: l != "", line)
 106     return [", ".join(line)]
 107
 108 def format_links(entry):
 109     doi = format_field(entry, 'doi', pre='<a href="http://dx.doi.org/', post='">[DOI]</a>')
 110     webpdf = format_field(entry, 'webpdf', pre='<a href="', post='">[PDF]</a>')
 111     weblink = format_field(entry, 'weblink', pre='<a href="', post='">[link]</a>')
 112     webslides = format_field(entry, 'webslides', pre='<a href="', post='">[slides]</a>')
 113     weberrata = format_field(entry, 'weberrata', pre='<a href="',
 114             post='">[errata]</a>')
 115     return " ".join([doi, webpdf, weblink, webslides, weberrata])
 116
 117 def format_entry(entry):
 118     lines = []
 119     lines.append(format_field(entry, 'title', pre="<b>", post="</b>"))
 120     lines.append(format_field_span('author', format_authors(entry)))
 121
 122     if entry.type=='article':
 123         lines.extend(format_details_article(entry))
 124     elif entry.type=='inproceedings':
 125         lines.extend(format_details_inproceedings(entry))
 126     elif entry.type=='book':
 127         lines.extend(format_details_book(entry))
 128     elif entry.type in ['mastersthesis', 'phdthesis']:
 129         lines.extend(format_details_thesis(entry))
 130     else:
 131         lines.append("Unknown type <b>'" + entry.type + "'</b>")
 132
 133     lines.append(format_field(entry, 'webnote'))
 134     lines.append(format_links(entry))
 135
 136     lines = filter(lambda l: l != "", lines)
 137     return "<br/>\n".join(lines)
 138
 139
 140 def entryDateSortKey(p):
 141     k, e = p
 142
 143     month2num = { 'jan' : '01', 'feb' : '02', 'mar' : '03', \
 144             'apr' : '04', 'may' : '05', 'jun' : '06', \
 145             'jul' : '07', 'aug' : '08', 'sep' : '09', \
 146             'oct' : '10', 'nov' : '11', 'dec' : '12'}
 147
 148     if not 'month' in e.fields:
 149         return e.fields['year']
 150
 151     month = e.fields['month'].lower()[0:3]
 152     if month in month2num:
 153         month = month2num[month]
 154     else:
 155         month = ""
 156
 157     return e.fields['year'] + "-" + month
 158
 159
 160 def usage():
 161     """Print usage text of this program"""
 162
 163     print("""Usage:
 164   {0} -i FILE
 165   {0} -h
 166
 167 OPTIONS:
 168   -h       print this text
 169   -i       .bib file
 170 """.format(sys.argv[0]))
 171
 172 if __name__ == "__main__":
 173
 174     bibfile = None
 175
 176     try:
 177         opts, args = getopt.getopt(sys.argv[1:], "hi:")
 178
 179         for opt, arg in opts:
 180             if opt == "-h":
 181                 usage()
 182                 sys.exit(os.EX_OK)
 183             elif opt == "-i":
 184                 bibfile = arg
 185             else:
 186                 print("Unknown option '", opt, "'.")
 187
 188     except getopt.GetoptError as e:
 189         print("Error parsing arguments:", e)
 190         usage()
 191
 192     if bibfile == None:
 193         print("You need to specify a bibfile")
 194         usage()
 195         sys.exit(os.EX_USAGE)
 196
 197
 198     from pybtex.database.input import bibtex
 199     parser = bibtex.Parser()
 200
 201     from pybtex.style.formatting.unsrt import Style
 202
 203     bib_data = parser.parse_file(bibfile)
 204     entries = bib_data.entries
 205
 206     years = list(set([ b.fields['year'] for b in entries.values() ]))
 207     years.sort(reverse=True)
 208
 209     for year in years:
 210
 211         print("<h2>" + year + "</h2>")
 212
 213         iteritems = list(entries.items())
 214         iteritems.sort(key=entryDateSortKey, reverse=True)
 215         for key, entry in iteritems:
 216
 217             if entry.fields['year'] != year:
 218                 continue
 219
 220             print("<div class=bibentry>")
 221             print("<a class=bibentry_key id=" + key + ">[" + key + "]</a><br/>")
 222
 223             e = format_entry(entry)
 224             print(e)
 225
 226             print("</div>\n")
 227