git.cworth.org Git - obsolete/notmuch-to-html/blob - notmuch-to-html

   1 #!/usr/bin/python
   2 #
   3 # Copyright (c) 2011-2012 David Bremner <david@tethera.net>
   4 #
   5 # dependencies
   6 #       - python 2.6 for json
   7 #       - argparse; either python 2.7, or install separately
   8 #
   9 # This program is free software: you can redistribute it and/or modify
  10 # it under the terms of the GNU General Public License as published by
  11 # the Free Software Foundation, either version 3 of the License, or
  12 # (at your option) any later version.
  13 #
  14 # This program is distributed in the hope that it will be useful,
  15 # but WITHOUT ANY WARRANTY; without even the implied warranty of
  16 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  17 # GNU General Public License for more details.
  18 #
  19 # You should have received a copy of the GNU General Public License
  20 # along with this program.  If not, see http://www.gnu.org/licenses/ .
  21
  22 from __future__ import print_function
  23 from __future__ import unicode_literals
  24
  25 import codecs
  26 import collections
  27 import datetime
  28 import email.utils
  29 try:  # Python 3
  30     from urllib.parse import quote
  31 except ImportError:  # Python 2
  32     from urllib import quote
  33 import json
  34 import argparse
  35 import os
  36 import re
  37 import sys
  38 import subprocess
  39 import xml.sax.saxutils
  40
  41
  42 _ENCODING = 'UTF-8'
  43 _PAGES = {}
  44
  45
  46 if not hasattr(collections, 'OrderedDict'):  # Python 2.6 or earlier
  47     class _OrderedDict (dict):
  48         "Just enough of a stub to get through Page._get_threads"
  49         def __init__(self, *args, **kwargs):
  50             super(_OrderedDict, self).__init__(*args, **kwargs)
  51             self._keys = []  # record key order
  52
  53         def __setitem__(self, key, value):
  54             super(_OrderedDict, self).__setitem__(key, value)
  55             self._keys.append(key)
  56
  57         def values(self):
  58             for key in self._keys:
  59                 yield self[key]
  60
  61
  62     collections.OrderedDict = _OrderedDict
  63
  64
  65 def read_config(path=None, encoding=None):
  66     "Read config from json file"
  67     if not encoding:
  68         encoding = _ENCODING
  69     if path:
  70         fp = open(path)
  71     else:
  72         nmbhome = os.getenv('NMBGIT', os.path.expanduser('~/.nmbug'))
  73
  74         # read only the first line from the pipe
  75         sha1_bytes = subprocess.Popen(
  76             ['git', '--git-dir', nmbhome, 'show-ref', '-s', 'config'],
  77             stdout=subprocess.PIPE).stdout.readline()
  78         sha1 = sha1_bytes.decode(encoding).rstrip()
  79
  80         fp_byte_stream = subprocess.Popen(
  81             ['git', '--git-dir', nmbhome, 'cat-file', 'blob',
  82              sha1+':status-config.json'],
  83             stdout=subprocess.PIPE).stdout
  84         fp = codecs.getreader(encoding=encoding)(stream=fp_byte_stream)
  85
  86     return json.load(fp)
  87
  88
  89 class Thread (list):
  90     def __init__(self):
  91         self.running_data = {}
  92
  93
  94 class Page (object):
  95     def __init__(self, header=None, footer=None):
  96         self.header = header
  97         self.footer = footer
  98
  99     def write(self, database, views, stream=None):
 100         if not stream:
 101             try:  # Python 3
 102                 byte_stream = sys.stdout.buffer
 103             except AttributeError:  # Python 2
 104                 byte_stream = sys.stdout
 105             stream = codecs.getwriter(encoding=_ENCODING)(stream=byte_stream)
 106         self._write_header(views=views, stream=stream)
 107         for view in views:
 108             self._write_view(database=database, view=view, stream=stream)
 109         self._write_footer(views=views, stream=stream)
 110
 111     def _write_header(self, views, stream):
 112         if self.header:
 113             stream.write(self.header)
 114
 115     def _write_footer(self, views, stream):
 116         if self.footer:
 117             stream.write(self.footer)
 118
 119     def _write_view(self, database, view, stream):
 120         if 'query-string' not in view:
 121             query = view['query']
 122             view['query-string'] = ' and '.join(query)
 123         q = notmuch.Query(database, view['query-string'])
 124         q.set_sort(notmuch.Query.SORT.OLDEST_FIRST)
 125         threads = self._get_threads(messages=q.search_messages())
 126         self._write_view_header(view=view, stream=stream)
 127         self._write_threads(threads=threads, stream=stream)
 128
 129     def _get_threads(self, messages):
 130         threads = collections.OrderedDict()
 131         for message in messages:
 132             thread_id = message.get_thread_id()
 133             if thread_id in threads:
 134                 thread = threads[thread_id]
 135             else:
 136                 thread = Thread()
 137                 threads[thread_id] = thread
 138             thread.running_data, display_data = self._message_display_data(
 139                 running_data=thread.running_data, message=message)
 140             thread.append(display_data)
 141         return list(threads.values())
 142
 143     def _write_view_header(self, view, stream):
 144         pass
 145
 146     def _write_threads(self, threads, stream):
 147         for thread in threads:
 148             for message_display_data in thread:
 149                 stream.write(
 150                     ('{date:10.10s} {from:20.20s} {subject:40.40s}\n'
 151                      '{message-id-term:>72}\n'
 152                      ).format(**message_display_data))
 153             if thread != threads[-1]:
 154                 stream.write('\n')
 155
 156     def _message_display_data(self, running_data, message):
 157         headers = ('thread-id', 'message-id', 'date', 'from', 'subject')
 158         data = {}
 159         for header in headers:
 160             if header == 'thread-id':
 161                 value = message.get_thread_id()
 162             elif header == 'message-id':
 163                 value = message.get_message_id()
 164                 data['message-id-term'] = 'id:"{0}"'.format(value)
 165             elif header == 'date':
 166                 value = str(datetime.datetime.utcfromtimestamp(
 167                     message.get_date()).date())
 168             else:
 169                 value = message.get_header(header)
 170             if header == 'from':
 171                 (value, addr) = email.utils.parseaddr(value)
 172                 if not value:
 173                     value = addr.split('@')[0]
 174             data[header] = value
 175         next_running_data = data.copy()
 176         for header, value in data.items():
 177             if header in ['message-id', 'subject']:
 178                 continue
 179             if value == running_data.get(header, None):
 180                 data[header] = ''
 181         return (next_running_data, data)
 182
 183
 184 class HtmlPage (Page):
 185     _slug_regexp = re.compile('\W+')
 186
 187     def _write_header(self, views, stream):
 188         super(HtmlPage, self)._write_header(views=views, stream=stream)
 189         stream.write('<ul>\n')
 190         for view in views:
 191             if 'id' not in view:
 192                 view['id'] = self._slug(view['title'])
 193             stream.write(
 194                 '<li><a href="#{id}">{title}</a></li>\n'.format(**view))
 195         stream.write('</ul>\n')
 196
 197     def _write_view_header(self, view, stream):
 198         stream.write('<h3 id="{id}">{title}</h3>\n'.format(**view))
 199         stream.write('<p>\n')
 200         if 'comment' in view:
 201             stream.write(view['comment'])
 202             stream.write('\n')
 203         for line in [
 204                 'The view is generated from the following query:',
 205                 '</p>',
 206                 '<p>',
 207                 '  <code>',
 208                 view['query-string'],
 209                 '  </code>',
 210                 '</p>',
 211                 ]:
 212             stream.write(line)
 213             stream.write('\n')
 214
 215     def _write_threads(self, threads, stream):
 216         if not threads:
 217             return
 218         stream.write('<table>\n')
 219         for thread in threads:
 220             stream.write('  <tbody>\n')
 221             for message_display_data in thread:
 222                 stream.write((
 223                     '    <tr class="message-first">\n'
 224                     '      <td>{from}</td>\n'
 225                     '      <td>{date}</td>\n'
 226                     '      <td>{subject}</td>\n'
 227                     '    </tr>\n'
 228                     ).format(**message_display_data))
 229             stream.write('  </tbody>\n')
 230             if thread != threads[-1]:
 231                 stream.write(
 232                     '  <tbody><tr><td colspan="2"><br /></td></tr></tbody>\n')
 233         stream.write('</table>\n')
 234
 235     def _message_display_data(self, *args, **kwargs):
 236         running_data, display_data = super(
 237             HtmlPage, self)._message_display_data(
 238                 *args, **kwargs)
 239         if 'subject' in display_data and 'message-id' in display_data:
 240             d = {
 241                 'message-id': quote(display_data['message-id']),
 242                 'subject': xml.sax.saxutils.escape(display_data['subject']),
 243                 }
 244             display_data['subject'] = (
 245                 '<a href="http://mid.gmane.org/{message-id}">{subject}</a>'
 246                 ).format(**d)
 247         for key in ['message-id', 'from']:
 248             if key in display_data:
 249                 display_data[key] = xml.sax.saxutils.escape(display_data[key])
 250         return (running_data, display_data)
 251
 252     def _slug(self, string):
 253         return self._slug_regexp.sub('-', string)
 254
 255 parser = argparse.ArgumentParser()
 256 parser.add_argument('--text', help='output plain text format',
 257                     action='store_true')
 258 parser.add_argument('--config', help='load config from given file',
 259                     metavar='PATH')
 260 parser.add_argument('--list-views', help='list views',
 261                     action='store_true')
 262 parser.add_argument('--get-query', help='get query for view',
 263                     metavar='VIEW')
 264
 265 args = parser.parse_args()
 266
 267 config = read_config(path=args.config)
 268
 269 _PAGES['text'] = Page()
 270 _PAGES['html'] = HtmlPage(
 271     header='''<!DOCTYPE html>
 272 <html lang="en">
 273 <head>
 274   <meta http-equiv="Content-Type" content="text/html; charset={encoding}" />
 275   <title>{title}</title>
 276   <style media="screen" type="text/css">
 277     table {{
 278       border-spacing: 0;
 279     }}
 280     tr.message-first td {{
 281       padding-top: {inter_message_padding};
 282     }}
 283     tr.message-last td {{
 284       padding-bottom: {inter_message_padding};
 285     }}
 286     td {{
 287       padding-left: {border_radius};
 288       padding-right: {border_radius};
 289     }}
 290     tr:first-child td:first-child {{
 291       border-top-left-radius: {border_radius};
 292     }}
 293     tr:first-child td:last-child {{
 294       border-top-right-radius: {border_radius};
 295     }}
 296     tr:last-child td:first-child {{
 297       border-bottom-left-radius: {border_radius};
 298     }}
 299     tr:last-child td:last-child {{
 300       border-bottom-right-radius: {border_radius};
 301     }}
 302     tbody:nth-child(4n+1) tr td {{
 303       background-color: #ffd96e;
 304     }}
 305     tbody:nth-child(4n+3) tr td {{
 306       background-color: #bce;
 307     }}
 308   </style>
 309 </head>
 310 <body>
 311 <h2>{title}</h2>
 312 <p>
 313 Generated: {date}<br />
 314 {blurb}
 315 </p>
 316 <h3>Views</h3>
 317 '''.format(date=datetime.datetime.utcnow().date(),
 318            title=config['meta']['title'],
 319            blurb=config['meta']['blurb'],
 320            encoding=_ENCODING,
 321            inter_message_padding='0.25em',
 322            border_radius='0.5em'),
 323     footer='</body>\n</html>\n',
 324     )
 325
 326 if args.list_views:
 327     for view in config['views']:
 328         print(view['title'])
 329     sys.exit(0)
 330 elif args.get_query != None:
 331     for view in config['views']:
 332         if args.get_query == view['title']:
 333             print(' and '.join(view['query']))
 334     sys.exit(0)
 335 else:
 336     # only import notmuch if needed
 337     import notmuch
 338
 339 if args.text:
 340     page = _PAGES['text']
 341 else:
 342     page = _PAGES['html']
 343
 344 db = notmuch.Database(mode=notmuch.Database.MODE.READ_ONLY)
 345 page.write(database=db, views=config['views'])