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/>.
26 from optparse import OptionParser
30 import common, metadata
32 from collections import Counter
34 def carbon_send(key, value):
36 s.connect((config['carbon_host'], config['carbon_port']))
37 msg = '%s %d %d\n' % (key, value, int(time.time()))
46 global options, config
48 # Parse command line...
49 parser = OptionParser()
50 parser.add_option("-v", "--verbose", action="store_true", default=False,
51 help="Spew out even more information than normal")
52 parser.add_option("-d", "--download", action="store_true", default=False,
53 help="Download logs we don't have")
54 parser.add_option("--nologs", action="store_true", default=False,
55 help="Don't do anything logs-related")
56 (options, args) = parser.parse_args()
58 config = common.read_config(options)
60 if not config['update_stats']:
61 logging.info("Stats are disabled - check your configuration")
64 # Get all metadata-defined apps...
65 metaapps = metadata.read_metadata(options.verbose)
68 logsdir = os.path.join(statsdir, 'logs')
69 datadir = os.path.join(statsdir, 'data')
70 if not os.path.exists(statsdir):
72 if not os.path.exists(logsdir):
74 if not os.path.exists(datadir):
78 # Get any access logs we don't have...
82 logging.info('Retrieving logs')
83 ssh = paramiko.SSHClient()
84 ssh.load_system_host_keys()
85 ssh.connect('f-droid.org', username='fdroid', timeout=10,
86 key_filename=config['webserver_keyfile'])
88 ftp.get_channel().settimeout(60)
89 logging.info("...connected")
94 if f.startswith('access-') and f.endswith('.log.gz'):
96 destpath = os.path.join(logsdir, f)
97 destsize = ftp.stat(f).st_size
98 if (not os.path.exists(destpath) or
99 os.path.getsize(destpath) != destsize):
100 logging.info("...retrieving " + f)
103 traceback.print_exc()
112 knownapks = common.KnownApks()
115 if not options.nologs:
117 logging.info('Processing logs...')
118 appscount = Counter()
119 appsvercount = Counter()
120 logexpr = '(?P<ip>[.:0-9a-fA-F]+) - - \[(?P<time>.*?)\] "GET (?P<uri>.*?) HTTP/1.\d" (?P<statuscode>\d+) \d+ "(?P<referral>.*?)" "(?P<useragent>.*?)"'
121 logsearch = re.compile(logexpr).search
122 for logfile in glob.glob(os.path.join(logsdir,'access-*.log.gz')):
123 logging.info('...' + logfile)
125 print '...' + logfile
126 p = subprocess.Popen(["zcat", logfile], stdout = subprocess.PIPE)
127 matches = (logsearch(line) for line in p.stdout)
128 for match in matches:
129 if match and match.group('statuscode') == '200':
130 uri = match.group('uri')
131 if uri.endswith('.apk'):
132 _, apkname = os.path.split(uri)
133 app = knownapks.getapp(apkname)
136 appscount[appid] += 1
137 # Strip the '.apk' from apkname
138 appver = apkname[:-4]
139 appsvercount[appver] += 1
141 if not apkname in unknownapks:
142 unknownapks.append(apkname)
144 # Calculate and write stats for total downloads...
147 for appid in appscount:
148 count = appscount[appid]
149 lst.append(app + " " + str(count))
150 if config['stats_to_carbon']:
151 carbon_send('fdroid.download.' + app.replace('.', '_'), count)
152 alldownloads += count
153 lst.append("ALL " + str(alldownloads))
154 f = open('stats/total_downloads_app.txt', 'w')
155 f.write('# Total downloads by application, since October 2011\n')
156 for line in sorted(lst):
160 f = open('stats/total_downloads_app_version.txt', 'w')
161 f.write('# Total downloads by application and version, since October 2011\n')
163 for appver in appsvercount:
164 count = appsvercount[appver]
165 lst.append(appver + " " + str(count))
166 for line in sorted(lst):
170 # Calculate and write stats for repo types...
171 logging.info("Processing repo types...")
174 if len(app['Repo Type']) == 0:
177 if app['Repo Type'] == 'srclib':
178 rtype = common.getsrclibvcs(app['Repo'])
180 rtype = app['Repo Type']
181 if rtype in repotypes:
182 repotypes[rtype] += 1;
185 f = open('stats/repotypes.txt', 'w')
186 for rtype, count in repotypes.iteritems():
187 f.write(rtype + ' ' + str(count) + '\n')
190 # Calculate and write stats for update check modes...
191 logging.info("Processing update check modes...")
194 checkmode = app['Update Check Mode']
195 if checkmode.startswith('RepoManifest/'):
196 checkmode = checkmode[:12]
197 if checkmode.startswith('Tags '):
198 checkmode = checkmode[:4]
199 if checkmode in ucms:
200 ucms[checkmode] += 1;
203 f = open('stats/update_check_modes.txt', 'w')
204 for checkmode, count in ucms.iteritems():
205 f.write(checkmode + ' ' + str(count) + '\n')
208 logging.info("Processing categories...")
211 if app['Categories'] is None:
213 categories = [c.strip() for c in app['Categories'].split(',')]
214 for category in categories:
219 f = open('stats/categories.txt', 'w')
220 for category, count in ctgs.iteritems():
221 f.write(category + ' ' + str(count) + '\n')
224 logging.info("Processing antifeatures...")
227 if app['AntiFeatures'] is None:
229 antifeatures = [a.strip() for a in app['AntiFeatures'].split(',')]
230 for antifeature in antifeatures:
231 if antifeature in afs:
232 afs[antifeature] += 1;
235 f = open('stats/antifeatures.txt', 'w')
236 for antifeature, count in afs.iteritems():
237 f.write(antifeature + ' ' + str(count) + '\n')
240 # Calculate and write stats for licenses...
241 logging.info("Processing licenses...")
244 license = app['License']
245 if license in licenses:
246 licenses[license] += 1;
248 licenses[license] = 1
249 f = open('stats/licenses.txt', 'w')
250 for license, count in licenses.iteritems():
251 f.write(license + ' ' + str(count) + '\n')
254 # Write list of latest apps added to the repo...
255 logging.info("Processing latest apps...")
256 latest = knownapks.getlatest(10)
257 f = open('stats/latestapps.txt', 'w')
263 logging.info('\nUnknown apks:')
264 for apk in unknownapks:
267 logging.info("Finished.")
269 if __name__ == "__main__":