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

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