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 DEFAULT_CONFIG='''
  49 {{
  50     "meta": {{
  51         "title": "Page title",
  52         "blurb": "Page description"
  53     }},
  54
  55     "views": [
  56         {{
  57             "title": "View title",
  58             "comment": "View description",
  59             "query": [ "{query}" ]
  60         }}
  61     ]
  62 }}'''
  63
  64
  65 if not hasattr(collections, 'OrderedDict'):  # Python 2.6 or earlier
  66     class _OrderedDict (dict):
  67         "Just enough of a stub to get through Page._get_threads"
  68         def __init__(self, *args, **kwargs):
  69             super(_OrderedDict, self).__init__(*args, **kwargs)
  70             self._keys = []  # record key order
  71
  72         def __setitem__(self, key, value):
  73             super(_OrderedDict, self).__setitem__(key, value)
  74             self._keys.append(key)
  75
  76         def values(self):
  77             for key in self._keys:
  78                 yield self[key]
  79
  80
  81     collections.OrderedDict = _OrderedDict
  82
  83
  84 def read_config(path, encoding=None):
  85     "Read config from json file"
  86     if not encoding:
  87         encoding = _ENCODING
  88     fp = open(path)
  89     return json.load(fp)
  90
  91
  92 class Thread (list):
  93     def __init__(self):
  94         self.running_data = {}
  95
  96
  97 class Page (object):
  98     def __init__(self, header=None, footer=None):
  99         self.header = header
 100         self.footer = footer
 101
 102     def write(self, database, views, stream=None):
 103         if not stream:
 104             try:  # Python 3
 105                 byte_stream = sys.stdout.buffer
 106             except AttributeError:  # Python 2
 107                 byte_stream = sys.stdout
 108             stream = codecs.getwriter(encoding=_ENCODING)(stream=byte_stream)
 109         self._write_header(views=views, stream=stream)
 110         for view in views:
 111             self._write_view(database=database, view=view, stream=stream)
 112         self._write_footer(views=views, stream=stream)
 113
 114     def _write_header(self, views, stream):
 115         if self.header:
 116             stream.write(self.header)
 117
 118     def _write_footer(self, views, stream):
 119         if self.footer:
 120             stream.write(self.footer)
 121
 122     def _write_view(self, database, view, stream):
 123         if 'query-string' not in view:
 124             query = view['query']
 125             view['query-string'] = ' and '.join(query)
 126         q = notmuch.Query(database, view['query-string'])
 127         q.set_sort(notmuch.Query.SORT.OLDEST_FIRST)
 128         threads = self._get_threads(messages=q.search_messages())
 129         self._write_view_header(view=view, stream=stream)
 130         self._write_threads(threads=threads, stream=stream)
 131
 132     def _get_threads(self, messages):
 133         threads = collections.OrderedDict()
 134         for message in messages:
 135             thread_id = message.get_thread_id()
 136             if thread_id in threads:
 137                 thread = threads[thread_id]
 138             else:
 139                 thread = Thread()
 140                 threads[thread_id] = thread
 141             thread.running_data, display_data = self._message_display_data(
 142                 running_data=thread.running_data, message=message)
 143             thread.append(display_data)
 144         return list(threads.values())
 145
 146     def _write_view_header(self, view, stream):
 147         pass
 148
 149     def _write_threads(self, threads, stream):
 150         for thread in threads:
 151             for message_display_data in thread:
 152                 stream.write(
 153                     ('{date:10.10s} {from:20.20s} {subject:40.40s}\n'
 154                      '{message-id-term:>72}\n'
 155                      ).format(**message_display_data))
 156             if thread != threads[-1]:
 157                 stream.write('\n')
 158
 159     def _message_display_data(self, running_data, message):
 160         headers = ('thread-id', 'message-id', 'date', 'from', 'subject')
 161         data = {}
 162         for header in headers:
 163             if header == 'thread-id':
 164                 value = message.get_thread_id()
 165             elif header == 'message-id':
 166                 value = message.get_message_id()
 167                 data['message-id-term'] = 'id:"{0}"'.format(value)
 168             elif header == 'date':
 169                 value = str(datetime.datetime.utcfromtimestamp(
 170                     message.get_date()).date())
 171             else:
 172                 value = message.get_header(header)
 173             if header == 'from':
 174                 (value, addr) = email.utils.parseaddr(value)
 175                 if not value:
 176                     value = addr.split('@')[0]
 177             data[header] = value
 178         next_running_data = data.copy()
 179         for header, value in data.items():
 180             if header in ['message-id', 'subject']:
 181                 continue
 182             if value == running_data.get(header, None):
 183                 data[header] = ''
 184         return (next_running_data, data)
 185
 186
 187 class HtmlPage (Page):
 188     _slug_regexp = re.compile('\W+')
 189
 190     def _write_header(self, views, stream):
 191         super(HtmlPage, self)._write_header(views=views, stream=stream)
 192         stream.write('<ul>\n')
 193         for view in views:
 194             if 'id' not in view:
 195                 view['id'] = self._slug(view['title'])
 196             stream.write(
 197                 '<li><a href="#{id}">{title}</a></li>\n'.format(**view))
 198         stream.write('</ul>\n')
 199
 200     def _write_view_header(self, view, stream):
 201         stream.write('<h3 id="{id}">{title}</h3>\n'.format(**view))
 202         stream.write('<p>\n')
 203         if 'comment' in view:
 204             stream.write(view['comment'])
 205             stream.write('\n')
 206         for line in [
 207                 '<p>This view is generated from the following query:',
 208                 '</p>',
 209                 '<p>',
 210                 '  <code>',
 211                 'notmuch search ' + view['query-string'],
 212                 '  </code>',
 213                 '</p>',
 214                 ]:
 215             stream.write(line)
 216             stream.write('\n')
 217
 218     def _write_threads(self, threads, stream):
 219         if not threads:
 220             return
 221         stream.write('<table>\n')
 222         for thread in threads:
 223             stream.write('  <tbody>\n')
 224             for message_display_data in thread:
 225                 stream.write((
 226                     '    <tr class="message-first">\n'
 227                     '      <td>{from}</td>\n'
 228                     '      <td>{date}</td>\n'
 229                     '      <td>{subject}</td>\n'
 230                     '    </tr>\n'
 231                     ).format(**message_display_data))
 232             stream.write('  </tbody>\n')
 233             if thread != threads[-1]:
 234                 stream.write(
 235                     '  <tbody><tr><td colspan="2"><br /></td></tr></tbody>\n')
 236         stream.write('</table>\n')
 237
 238     def _message_display_data(self, *args, **kwargs):
 239         running_data, display_data = super(
 240             HtmlPage, self)._message_display_data(
 241                 *args, **kwargs)
 242         if 'subject' in display_data and 'message-id' in display_data:
 243             d = {
 244                 'message-id': quote(display_data['message-id']),
 245                 'subject': xml.sax.saxutils.escape(display_data['subject']),
 246                 }
 247             display_data['subject'] = (
 248                 '<a href="http://mid.gmane.org/{message-id}">{subject}</a>'
 249                 ).format(**d)
 250         for key in ['message-id', 'from']:
 251             if key in display_data:
 252                 display_data[key] = xml.sax.saxutils.escape(display_data[key])
 253         return (running_data, display_data)
 254
 255     def _slug(self, string):
 256         return self._slug_regexp.sub('-', string)
 257
 258 parser = argparse.ArgumentParser()
 259 parser.add_argument('--text', help='output plain text format',
 260                     action='store_true')
 261 group = parser.add_mutually_exclusive_group()
 262 group.add_argument('--config', help='path to configuration file',
 263                    metavar='PATH')
 264 group.add_argument('--query', help='path to configuration file',
 265                    metavar='PATH')
 266 parser.add_argument('--list-views', help='list views',
 267                     action='store_true')
 268
 269 args = parser.parse_args()
 270
 271 if (args.config):
 272     config = read_config(path=args.config)
 273 elif (args.query):
 274     config = json.loads(DEFAULT_CONFIG.format(query=args.query))
 275 else:
 276     print ('''To use notmuch-to-html, you need to provide a notmuch query. Try:
 277
 278         notmuch-to-html --query=tag:inbox
 279
 280 Or 'notmuch-to-html --help' for additional options.''')
 281     exit (0)
 282
 283 _PAGES['text'] = Page()
 284 _PAGES['html'] = HtmlPage(
 285     header='''<!DOCTYPE html>
 286 <html lang="en">
 287 <head>
 288   <meta http-equiv="Content-Type" content="text/html; charset={encoding}" />
 289   <title>{title}</title>
 290   <style media="screen" type="text/css">
 291     table {{
 292       border-spacing: 0;
 293     }}
 294     tr.message-first td {{
 295       padding-top: {inter_message_padding};
 296     }}
 297     tr.message-last td {{
 298       padding-bottom: {inter_message_padding};
 299     }}
 300     td {{
 301       padding-left: {border_radius};
 302       padding-right: {border_radius};
 303     }}
 304     tr:first-child td:first-child {{
 305       border-top-left-radius: {border_radius};
 306     }}
 307     tr:first-child td:last-child {{
 308       border-top-right-radius: {border_radius};
 309     }}
 310     tr:last-child td:first-child {{
 311       border-bottom-left-radius: {border_radius};
 312     }}
 313     tr:last-child td:last-child {{
 314       border-bottom-right-radius: {border_radius};
 315     }}
 316     tbody:nth-child(4n+1) tr td {{
 317       background-color: #ffd96e;
 318     }}
 319     tbody:nth-child(4n+3) tr td {{
 320       background-color: #bce;
 321     }}
 322   </style>
 323 </head>
 324 <body>
 325 <h2>{title}</h2>
 326 {blurb}
 327 </p>
 328 <h3>Views</h3>
 329 '''.format(title=config['meta']['title'],
 330            blurb=config['meta']['blurb'],
 331            encoding=_ENCODING,
 332            inter_message_padding='0.25em',
 333            border_radius='0.5em'),
 334     footer='''<hr>
 335 <p>Generated: {date} courtesy of <a href="http://notmuchmail.org">notmuch</a> and <a href="http://git.cworth.org/git/notmuch-to-html">notmuch-to-html</a>.
 336 </body>
 337 </html>
 338 '''.format(date=datetime.datetime.utcnow().date())
 339     )
 340
 341 if args.list_views:
 342     for view in config['views']:
 343         print(view['title'])
 344     sys.exit(0)
 345 else:
 346     # only import notmuch if needed
 347     import notmuch
 348
 349 if args.text:
 350     page = _PAGES['text']
 351 else:
 352     page = _PAGES['html']
 353
 354 db = notmuch.Database(mode=notmuch.Database.MODE.READ_ONLY)
 355 page.write(database=db, views=config['views'])
 356
 357 if (args.query):
 358     print ('''To customize the output use 'notmuch-to-html --config=CONFIG_FILE' after
 359 placing the following content into CONFIG_FILE (note that you can add
 360 additional views with their own queries):
 361
 362
 363 ''', file=sys.stderr)
 364     print (json.dumps(config, indent=4, separators=(',',':')), file=sys.stderr)