2 # -*- coding: utf-8 -*-
4 # stats.py - part of the FDroid server tools
5 # Copyright (C) 2010-13, Ciaran Gultnieks, ciaran@ciarang.com
7 # This program is free software: you can redistribute it and/or modify
8 # it under the terms of the GNU Affero General Public License as published by
9 # the Free Software Foundation, either version 3 of the License, or
10 # (at your option) any later version.
12 # This program is distributed in the hope that it will be useful,
13 # but WITHOUT ANY WARRANTY; without even the implied warranty of
14 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
15 # GNU Affero General Public License for more details.
17 # You should have received a copy of the GNU Affero General Public License
18 # along with this program. If not, see <http://www.gnu.org/licenses/>.
27 from optparse import OptionParser
34 from collections import Counter
37 def carbon_send(key, value):
39 s.connect((config['carbon_host'], config['carbon_port']))
40 msg = '%s %d %d\n' % (key, value, int(time.time()))
50 global options, config
52 # Parse command line...
53 parser = OptionParser()
54 parser.add_option("-v", "--verbose", action="store_true", default=False,
55 help="Spew out even more information than normal")
56 parser.add_option("-q", "--quiet", action="store_true", default=False,
57 help="Restrict output to warnings and errors")
58 parser.add_option("-d", "--download", action="store_true", default=False,
59 help="Download logs we don't have")
60 parser.add_option("--recalc", action="store_true", default=False,
61 help="Recalculate aggregate stats - use when changes "
62 "have been made that would invalidate old cached data.")
63 parser.add_option("--nologs", action="store_true", default=False,
64 help="Don't do anything logs-related")
65 (options, args) = parser.parse_args()
67 config = common.read_config(options)
69 if not config['update_stats']:
70 logging.info("Stats are disabled - check your configuration")
73 # Get all metadata-defined apps...
74 metadata.read_srclibs()
75 metaapps = metadata.read_metadata()
78 logsdir = os.path.join(statsdir, 'logs')
79 datadir = os.path.join(statsdir, 'data')
80 if not os.path.exists(statsdir):
82 if not os.path.exists(logsdir):
84 if not os.path.exists(datadir):
88 # Get any access logs we don't have...
92 logging.info('Retrieving logs')
93 ssh = paramiko.SSHClient()
94 ssh.load_system_host_keys()
95 ssh.connect('f-droid.org', username='fdroid', timeout=10,
96 key_filename=config['webserver_keyfile'])
98 ftp.get_channel().settimeout(60)
99 logging.info("...connected")
102 files = ftp.listdir()
104 if f.startswith('access-') and f.endswith('.log.gz'):
106 destpath = os.path.join(logsdir, f)
107 destsize = ftp.stat(f).st_size
108 if (not os.path.exists(destpath) or
109 os.path.getsize(destpath) != destsize):
110 logging.debug("...retrieving " + f)
113 traceback.print_exc()
122 knownapks = common.KnownApks()
125 if not options.nologs:
127 logging.info('Processing logs...')
128 appscount = Counter()
129 appsvercount = Counter()
130 logexpr = '(?P<ip>[.:0-9a-fA-F]+) - - \[(?P<time>.*?)\] ' + \
131 '"GET (?P<uri>.*?) HTTP/1.\d" (?P<statuscode>\d+) ' + \
132 '\d+ "(?P<referral>.*?)" "(?P<useragent>.*?)"'
133 logsearch = re.compile(logexpr).search
134 for logfile in glob.glob(os.path.join(logsdir, 'access-*.log.gz')):
135 logging.debug('...' + logfile)
137 # Get the date for this log - e.g. 2012-02-28
138 thisdate = os.path.basename(logfile)[7:-7]
140 agg_path = os.path.join(datadir, thisdate + '.json')
141 if not options.recalc and os.path.exists(agg_path):
142 # Use previously calculated aggregate data
143 with open(agg_path, 'r') as f:
147 # Calculate from logs...
151 'appsver': Counter(),
155 p = subprocess.Popen(["zcat", logfile], stdout=subprocess.PIPE)
156 matches = (logsearch(line) for line in p.stdout)
157 for match in matches:
158 if match and match.group('statuscode') == '200':
159 uri = match.group('uri')
160 if uri.endswith('.apk'):
161 _, apkname = os.path.split(uri)
162 app = knownapks.getapp(apkname)
165 today['apps'][appid] += 1
166 # Strip the '.apk' from apkname
167 appver = apkname[:-4]
168 today['appsver'][appver] += 1
170 if apkname not in today['unknown']:
171 today['unknown'].append(apkname)
173 # Save calculated aggregate data for today to cache
174 with open(agg_path, 'w') as f:
177 # Add today's stats (whether cached or recalculated) to the total
178 for appid in today['apps']:
179 appscount[appid] += today['apps'][appid]
180 for appid in today['appsver']:
181 appsvercount[appid] += today['appsver'][appid]
182 for uk in today['unknown']:
183 if uk not in unknownapks:
184 unknownapks.append(uk)
186 # Calculate and write stats for total downloads...
189 for appid in appscount:
190 count = appscount[appid]
191 lst.append(appid + " " + str(count))
192 if config['stats_to_carbon']:
193 carbon_send('fdroid.download.' + appid.replace('.', '_'),
195 alldownloads += count
196 lst.append("ALL " + str(alldownloads))
197 f = open('stats/total_downloads_app.txt', 'w')
198 f.write('# Total downloads by application, since October 2011\n')
199 for line in sorted(lst):
203 f = open('stats/total_downloads_app_version.txt', 'w')
204 f.write('# Total downloads by application and version, '
205 'since October 2011\n')
207 for appver in appsvercount:
208 count = appsvercount[appver]
209 lst.append(appver + " " + str(count))
210 for line in sorted(lst):
214 # Calculate and write stats for repo types...
215 logging.info("Processing repo types...")
216 repotypes = Counter()
218 if len(app['Repo Type']) == 0:
221 if app['Repo Type'] == 'srclib':
222 rtype = common.getsrclibvcs(app['Repo'])
224 rtype = app['Repo Type']
225 repotypes[rtype] += 1
226 f = open('stats/repotypes.txt', 'w')
227 for rtype in repotypes:
228 count = repotypes[rtype]
229 f.write(rtype + ' ' + str(count) + '\n')
232 # Calculate and write stats for update check modes...
233 logging.info("Processing update check modes...")
236 checkmode = app['Update Check Mode']
237 if checkmode.startswith('RepoManifest/'):
238 checkmode = checkmode[:12]
239 if checkmode.startswith('Tags '):
240 checkmode = checkmode[:4]
242 f = open('stats/update_check_modes.txt', 'w')
243 for checkmode in ucms:
244 count = ucms[checkmode]
245 f.write(checkmode + ' ' + str(count) + '\n')
248 logging.info("Processing categories...")
251 for category in app['Categories']:
253 f = open('stats/categories.txt', 'w')
254 for category in ctgs:
255 count = ctgs[category]
256 f.write(category + ' ' + str(count) + '\n')
259 logging.info("Processing antifeatures...")
262 if app['AntiFeatures'] is None:
264 antifeatures = [a.strip() for a in app['AntiFeatures'].split(',')]
265 for antifeature in antifeatures:
266 afs[antifeature] += 1
267 f = open('stats/antifeatures.txt', 'w')
268 for antifeature in afs:
269 count = afs[antifeature]
270 f.write(antifeature + ' ' + str(count) + '\n')
273 # Calculate and write stats for licenses...
274 logging.info("Processing licenses...")
277 license = app['License']
278 licenses[license] += 1
279 f = open('stats/licenses.txt', 'w')
280 for license in licenses:
281 count = licenses[license]
282 f.write(license + ' ' + str(count) + '\n')
285 # Write list of latest apps added to the repo...
286 logging.info("Processing latest apps...")
287 latest = knownapks.getlatest(10)
288 f = open('stats/latestapps.txt', 'w')
294 logging.info('\nUnknown apks:')
295 for apk in unknownapks:
298 logging.info("Finished.")
300 if __name__ == "__main__":