1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
|
#
# Copyright (c) 2014 ThoughtWorks, Inc.
#
# Pixelated is free software: you can redistribute it and/or modify
# it under the terms of the GNU Affero General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# Pixelated is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU Affero General Public License for more details.
#
# You should have received a copy of the GNU Affero General Public License
# along with Pixelated. If not, see <http://www.gnu.org/licenses/>.
from pixelated.support.encrypted_file_storage import EncryptedFileStorage
import os
import re
from pixelated.adapter.model.status import Status
from pixelated.adapter.search.contacts import contacts_suggestions
from whoosh.index import FileIndex
from whoosh.fields import Schema, ID, KEYWORD, TEXT, NUMERIC
from whoosh.qparser import QueryParser
from whoosh.qparser import MultifieldParser
from whoosh.writing import AsyncWriter
from whoosh import sorting
from pixelated.support.functional import unique
from pixelated.support.date import milliseconds
import traceback
class SearchEngine(object):
DEFAULT_INDEX_HOME = os.path.join(os.environ['HOME'], '.leap')
DEFAULT_TAGS = ['inbox', 'sent', 'drafts', 'trash']
def __init__(self, key, agent_home=DEFAULT_INDEX_HOME):
self.key = key
self.index_folder = os.path.join(agent_home, 'search_index')
if not os.path.exists(self.index_folder):
os.makedirs(self.index_folder)
self._index = self._create_index()
def _add_to_tags(self, tags, group, skip_default_tags, count_type, query=None):
query_matcher = re.compile(".*%s.*" % query.lower()) if query else re.compile(".*")
for tag, count in group.iteritems():
if skip_default_tags and tag in self.DEFAULT_TAGS or not query_matcher.match(tag):
continue
if not tags.get(tag):
tags[tag] = {'ident': tag, 'name': tag, 'default': False, 'counts': {'total': 0, 'read': 0},
'mails': []}
tags[tag]['counts'][count_type] += count
def _search_tag_groups(self, is_filtering_tags):
seen = None
query_parser = QueryParser('tag', self._index.schema)
options = {'limit': None, 'groupedby': sorting.FieldFacet('tag', allow_overlap=True), 'maptype': sorting.Count}
with self._index.searcher() as searcher:
total = searcher.search(query_parser.parse('*'), **options).groups()
if not is_filtering_tags:
seen = searcher.search(query_parser.parse("* AND flags:%s" % Status.SEEN), **options).groups()
return seen, total
def _init_tags_defaults(self):
tags = {}
for default_tag in self.DEFAULT_TAGS:
tags[default_tag] = {
'ident': default_tag,
'name': default_tag,
'default': True,
'counts': {
'total': 0,
'read': 0
},
'mails': []
}
return tags
def _build_tags(self, seen, total, skip_default_tags, query):
tags = {}
if not skip_default_tags:
tags = self._init_tags_defaults()
self._add_to_tags(tags, total, skip_default_tags, count_type='total', query=query)
if seen:
self._add_to_tags(tags, seen, skip_default_tags, count_type='read')
return tags.values()
def tags(self, query, skip_default_tags):
is_filtering_tags = True if query else False
seen, total = self._search_tag_groups(is_filtering_tags=is_filtering_tags)
return self._build_tags(seen, total, skip_default_tags, query)
def _mail_schema(self):
return Schema(
ident=ID(stored=True, unique=True),
sender=ID(stored=False),
to=KEYWORD(stored=False, commas=True),
cc=KEYWORD(stored=False, commas=True),
bcc=KEYWORD(stored=False, commas=True),
bounced=KEYWORD(stored=False, commas=True),
subject=TEXT(stored=False),
date=NUMERIC(stored=False, sortable=True, bits=64, signed=False),
body=TEXT(stored=False),
tag=KEYWORD(stored=True, commas=True),
flags=KEYWORD(stored=True, commas=True),
raw=TEXT(stored=False))
def _create_index(self):
storage = EncryptedFileStorage(self.index_folder, self.key)
return FileIndex.create(storage, self._mail_schema(), indexname='mails')
def index_mail(self, mail):
with AsyncWriter(self._index) as writer:
self._index_mail(writer, mail)
def _index_mail(self, writer, mail):
mdict = mail.as_dict()
header = mdict['header']
tags = set(mdict.get('tags', {}))
tags.add(mail.mailbox_name.lower())
bounced = mail.bounced if mail.bounced else ['']
index_data = {
'sender': self._unicode_header_field(header.get('from', '')),
'subject': self._unicode_header_field(header.get('subject', '')),
'date': milliseconds(header.get('date', '')),
'to': u','.join([h.decode('utf-8') for h in header.get('to', [''])]),
'cc': u','.join([h.decode('utf-8') for h in header.get('cc', [''])]),
'bcc': u','.join([h.decode('utf-8') for h in header.get('bcc', [''])]),
'tag': u','.join(unique(tags)),
'bounced': u','.join(bounced),
'body': unicode(mdict['textPlainBody'] if 'textPlainBody' in mdict else mdict['body']),
'ident': unicode(mdict['ident']),
'flags': unicode(','.join(unique(mail.flags))),
'raw': unicode(mail.raw.decode('utf-8'))
}
writer.update_document(**index_data)
def _unicode_header_field(self, field_value):
if not field_value:
return None
return unicode(field_value.decode('utf-8'))
def index_mails(self, mails, callback=None):
try:
with AsyncWriter(self._index) as writer:
for mail in mails:
self._index_mail(writer, mail)
if callback:
callback()
except Exception, e:
traceback.print_exc(e)
raise
def _search_with_options(self, options, query):
with self._index.searcher() as searcher:
query = QueryParser('raw', self._index.schema).parse(query)
results = searcher.search(query, **options)
return results
def search(self, query, window=25, page=1, all_mails=False):
query = self.prepare_query(query)
return self._search_all_mails(query) if all_mails else self._paginated_search_mails(query, window, page)
def _search_all_mails(self, query):
with self._index.searcher() as searcher:
sorting_facet = sorting.FieldFacet('date', reverse=True)
results = searcher.search(query, sortedby=sorting_facet, reverse=True, limit=None)
return unique([mail['ident'] for mail in results])
def _paginated_search_mails(self, query, window, page):
page = int(page) if page is not None and int(page) > 1 else 1
window = int(window) if window is not None else 25
with self._index.searcher() as searcher:
tags_facet = sorting.FieldFacet('tag', allow_overlap=True, maptype=sorting.Count)
sorting_facet = sorting.FieldFacet('date', reverse=True)
results = searcher.search_page(query, page, pagelen=window, groupedby=tags_facet, sortedby=sorting_facet)
return unique([mail['ident'] for mail in results]), sum(results.results.groups().values())
def prepare_query(self, query):
query = (
query
.replace('-in:', 'AND NOT tag:')
.replace('in:all', '*')
)
return MultifieldParser(['raw', 'body'], self._index.schema).parse(query)
def remove_from_index(self, mail_id):
with AsyncWriter(self._index) as writer:
writer.delete_by_term('ident', mail_id)
def contacts(self, query):
with self._index.searcher() as searcher:
return contacts_suggestions(query, searcher)
|
