Commit 92b6380bb699cdb843cb3994e24ac49ec1789d70
1 parent
f6d5f22e
Exists in
master
and in
39 other branches
removing solrutils file
Showing
1 changed file
with
0 additions
and
254 deletions
Show diff stats
src/colab/deprecated/solrutils.py
... | ... | @@ -1,254 +0,0 @@ |
1 | -#!/usr/bin/env python | |
2 | -# encoding: utf-8 | |
3 | - | |
4 | -import math | |
5 | -import json | |
6 | -import urllib | |
7 | -import socket | |
8 | -import logging | |
9 | -import httplib | |
10 | - | |
11 | -from dateutil.parser import parse as parse_timestamp | |
12 | - | |
13 | -from django.conf import settings | |
14 | - | |
15 | -from super_archives.models import EmailAddress | |
16 | - | |
17 | - | |
18 | -def build_query(user_query, filters=None): | |
19 | - """Build the query that will be sent to Solr""" | |
20 | - | |
21 | - if not user_query: | |
22 | - user_query = '*' | |
23 | - | |
24 | - query = settings.SOLR_BASE_QUERY.strip() + ' AND ' + user_query | |
25 | - if filters: | |
26 | - query = "(%s)" % query | |
27 | - | |
28 | - for (key, value) in filters.items(): | |
29 | - if value: | |
30 | - query += " AND %s:%s" % (key, value) | |
31 | - | |
32 | - logging.info(query) | |
33 | - return query.encode('utf-8') | |
34 | - | |
35 | - | |
36 | -def parse_document_timestamps(doc, date_attrs=('modified', 'created')): | |
37 | - """Converts the `modified' and `created' dates from | |
38 | - ISO 8601 format to a date time object for the given | |
39 | - document. | |
40 | - | |
41 | - """ | |
42 | - | |
43 | - for date in date_attrs: | |
44 | - date_str = doc.get(date) | |
45 | - try: | |
46 | - date_obj = parse_timestamp(date_str) | |
47 | - except ValueError: | |
48 | - logging.error('Error trying to parse "%s"', date_str) | |
49 | - date_obj = None | |
50 | - doc.update({date: date_obj}) | |
51 | - | |
52 | - return doc | |
53 | - | |
54 | - | |
55 | -def get_document_url(doc): | |
56 | - """Set the url attribute for a document using the path_string. | |
57 | - In case the resource comes from an external domain it will | |
58 | - be prepended to this URL. | |
59 | - | |
60 | - """ | |
61 | - doc_type = doc.get('Type') | |
62 | - | |
63 | - url = '' | |
64 | - if settings.SOLR_COLAB_URI: | |
65 | - url += settings.SOLR_COLAB_URI | |
66 | - | |
67 | - url += doc.get('path_string', '') | |
68 | - doc.update({'url': url}) | |
69 | - | |
70 | - return doc | |
71 | - | |
72 | - | |
73 | -def get_document_from_addr(doc): | |
74 | - """Get a EmailAddress instance for the given document if | |
75 | - its available. | |
76 | - | |
77 | - """ | |
78 | - | |
79 | - username = doc.get('last_author') | |
80 | - if not username: | |
81 | - username = doc.get('Creator') | |
82 | - from_addresses = EmailAddress.objects.filter(user__username=username) | |
83 | - if username and from_addresses: | |
84 | - doc.update({'from_address': from_addresses[0]}) | |
85 | - | |
86 | - | |
87 | -def add_attrs_to_doc(doc): | |
88 | - """Wraps the call of functions that adds or modifies keys | |
89 | - of the giving doc (which should be a dict). | |
90 | - | |
91 | - """ | |
92 | - get_document_url(doc) | |
93 | - parse_document_timestamps(doc) | |
94 | - get_document_from_addr(doc) | |
95 | - | |
96 | - | |
97 | -class SolrPaginator(list): | |
98 | - | |
99 | - def __init__(self, response_dict, current_page): | |
100 | - super(SolrPaginator, self).__init__() | |
101 | - | |
102 | - responseHeader = response_dict.get('responseHeader', {}) | |
103 | - response = response_dict.get('response', {}) | |
104 | - request_params = responseHeader.get('params', {}) | |
105 | - | |
106 | - docs = response.get('docs', []) | |
107 | - self.extend(docs) | |
108 | - | |
109 | - self.QTime = int(responseHeader.get('QTime', 1)) / 1000.0 | |
110 | - | |
111 | - self.per_page = int(request_params.get('rows', 10)) | |
112 | - self.numFound = int(response.get('numFound', 0)) | |
113 | - self.page_num = current_page | |
114 | - | |
115 | - self.num_of_pages = int(math.ceil(self.numFound / float(self.per_page))) | |
116 | - | |
117 | - self.has_previous = self.page_num > 1 | |
118 | - if self.has_previous: | |
119 | - self.previous_page_number = self.page_num - 1 | |
120 | - else: | |
121 | - self.previous_page_number = None | |
122 | - | |
123 | - self.has_next = self.page_num < self.num_of_pages | |
124 | - if self.has_next: | |
125 | - self.next_page_number = self.page_num + 1 | |
126 | - else: | |
127 | - self.next_page_number = None | |
128 | - | |
129 | - @property | |
130 | - def last_page(self): | |
131 | - return self.num_of_pages | |
132 | - | |
133 | - | |
134 | -def select(query, results_per_page=None, page_number=None, sort=None, fields=None, link_attrs=True): | |
135 | - """Perform a select in a Solr instance using the configuration | |
136 | - set in settings.py. | |
137 | - | |
138 | - """ | |
139 | - | |
140 | - if not settings.SOLR_HOSTNAME: | |
141 | - return {} | |
142 | - | |
143 | - data = { | |
144 | - 'q': query, | |
145 | - 'wt': 'json', | |
146 | - } | |
147 | - | |
148 | - # Number of results per page | |
149 | - if results_per_page: | |
150 | - data.update({'rows': results_per_page}) | |
151 | - | |
152 | - # Page number | |
153 | - if page_number: | |
154 | - data.update({ | |
155 | - 'start': (page_number - 1) * results_per_page, | |
156 | - }) | |
157 | - | |
158 | - # Sort order | |
159 | - if sort: | |
160 | - data.update({ | |
161 | - 'sort': sort, | |
162 | - }) | |
163 | - | |
164 | - # Only select those fields | |
165 | - if fields: | |
166 | - data.update({ | |
167 | - 'fl': ','.join(fields), | |
168 | - }) | |
169 | - # First version of this was implemented using urllib2 and was | |
170 | - # a milion times easier but unfortunatelly urllib2.urlopen | |
171 | - # does not support http headers. Without setting http headers | |
172 | - # for charset the solr server tries to decode utf-8 params | |
173 | - # as ASCII causing it to crash. HTTPConnection deals with | |
174 | - # encodings automagically. | |
175 | - solr_conn = httplib.HTTPConnection(settings.SOLR_HOSTNAME, | |
176 | - settings.SOLR_PORT) | |
177 | - query_params = urllib.urlencode(data) | |
178 | - solr_select_uri = settings.SOLR_SELECT_PATH + '?' + query_params | |
179 | - | |
180 | - try: | |
181 | - solr_conn.request('GET', solr_select_uri) | |
182 | - solr_response = solr_conn.getresponse() | |
183 | - except socket.error as err: | |
184 | - solr_response = None | |
185 | - logging.exception(err) | |
186 | - | |
187 | - if solr_response and solr_response.status == 200: | |
188 | - #TODO: Log error connecting to solr | |
189 | - solr_json_resp = solr_response.read() | |
190 | - solr_dict_resp = json.loads(solr_json_resp) | |
191 | - else: | |
192 | - solr_dict_resp = {} | |
193 | - | |
194 | - docs = solr_dict_resp.get('response', {}).get("docs", []) | |
195 | - | |
196 | - if link_attrs: | |
197 | - # Loop over all documents adding or linking its information | |
198 | - # with the data from this app or database | |
199 | - map(add_attrs_to_doc, docs) | |
200 | - | |
201 | - return solr_dict_resp | |
202 | - | |
203 | - | |
204 | -def get_latest_collaborations(number=10, username=None): | |
205 | - """Get the n documents recently modified that this username | |
206 | - has helped in somehow. | |
207 | - | |
208 | - """ | |
209 | - | |
210 | - if username: | |
211 | - filters = {'collaborator': username} | |
212 | - else: | |
213 | - filters = None | |
214 | - | |
215 | - query = build_query('*', filters) | |
216 | - solr_response = select( | |
217 | - query=query, | |
218 | - results_per_page=number, | |
219 | - sort='modified desc' | |
220 | - ) | |
221 | - | |
222 | - return solr_response.get('response', {}).get('docs', []) | |
223 | - | |
224 | - | |
225 | -def count_types(sample=100, filters=None): | |
226 | - """Count the type of the last modifications returning the | |
227 | - results in dict. | |
228 | - | |
229 | - Example: { | |
230 | - 'wiki' 30, | |
231 | - 'thread': 40, | |
232 | - 'ticket', 10, | |
233 | - 'changeset' 20, | |
234 | - } | |
235 | - | |
236 | - """ | |
237 | - | |
238 | - query = build_query('*', filters) | |
239 | - solr_response = select( | |
240 | - query=query, | |
241 | - results_per_page=sample, | |
242 | - sort='modified desc', | |
243 | - link_attrs=False, | |
244 | - ) | |
245 | - | |
246 | - docs = solr_response.get('response', {}).get('docs', []) | |
247 | - | |
248 | - type_count = {} | |
249 | - for doc in docs: | |
250 | - doc_type = doc.get('Type') | |
251 | - count = type_count.get(doc_type, 0) + 1 | |
252 | - type_count.update({doc_type: count}) | |
253 | - | |
254 | - return type_count |