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
203
204
205
206
207
208
209
210
211
|
import os
import csv
import bz2
import gzip
import re
import urllib2
import socket
from apache_reader import ApacheLogReader
# dictionary key structure: filename, user_agent, package_name
class LocalStats(object):
"""Base class that writes the log file
"""
def _get_logs(self, logfile, file_urls):
"""Needs to return an iterator. Each entry
should be a dictionary"""
if callable(logfile):
return logfile(file_urls)
raise NotImplementedError
def _get_file_obj(self, path, mode='r', compression=None):
"""returns a file object"""
if compression == 'bz2':
return bz2.BZ2File(path, mode)
elif compression == 'gz':
return gzip.open(path, mode)
return open(path, mode)
def _build_stats(self, logfile, fileobj, files_url='/packages',
filter=None, compression=None):
"""Builds a stats file
- logfile: path to the original log file, or callable
- fileobj : a file object or a path to create a file
- files_url : a filter that define the beginnin of package urls
- filter: if given, a callable that receives the
current line. if the callable returns True,
the line is not included
"""
downloads = {}
for log in self._get_logs(logfile, files_url):
if filter is not None:
if filter(log):
continue
filename = log['filename']
user_agent = log['useragent']
package_name = log['packagename']
key = (filename, user_agent, package_name)
count = log.get('count', 1)
if key in downloads:
downloads[key] += count
else:
downloads[key] = count
self._write_stats(fileobj, downloads, compression=compression)
def _write_stats(self, fileobj, downloads, compression=None):
if isinstance(fileobj, str):
fileobj = self._get_file_obj(fileobj, 'w', compression)
file_created = True
else:
file_created = False
writer = csv.writer(fileobj)
filenames = downloads.keys()
filenames.sort()
for key in filenames:
filename, user_agent, package_name = key
count = downloads[key]
writer.writerow((package_name, filename, user_agent, count))
if file_created:
fileobj.close()
def build_daily_stats(self, year, month, day, logfile, fileobj,
files_url='/packages', compression=None):
"""creates a daily stats file using an apache log file.
- year, month, day: values for the day
- logfile : path to the log file, or callable
- fileobj : a file object or a path to create a file
- files_url : a filter that define the beginning of package urls
"""
def _filter(log):
return (day != log['day'] or month != log['month'] or
year != log['year'])
self._build_stats(logfile, fileobj, files_url, _filter, compression)
def build_monthly_stats(self, year, month, logfile, fileobj,
files_url='/packages', compression=None):
"""creates a monthly stats file using an apache log file.
- year, month: values for the month
- logfile : path to the log file
- fileobj : a file object or a path to create a file
- files_url : a filter that define the beginnin of package urls
"""
def _filter(log):
return (month != log['month'] or year != log['year'])
self._build_stats(logfile, fileobj, files_url, _filter, compression)
def read_stats(self, stats_file):
"""Returns an iterator over a stats file"""
if isinstance(stats_file, str):
ext = os.path.splitext(stats_file)[-1][1:]
stats_file = self._get_file_obj(stats_file, 'r', ext)
reader = csv.reader(stats_file)
for line in reader:
# work around user agents with commas
while len(line) > 4:
line[2] += ',' + line[3]
del line[3]
yield {'packagename': line[0],
'filename': line[1],
'useragent': line[2],
'count': int(line[3])}
#reader.close()
def read_stats_dict(self, stats_file):
res = {}
for r in self.read_stats(stats_file):
key = (r['filename'], r['useragent'], r['packagename'])
value = r['count']
res[key] = value
return res
def build_local_stats(self, year, month, day, logfile, directory=None):
"""builds local stats with default values"""
filename = '%d-%.2d-%.2d.bz2' % (year, month, day)
if directory is not None:
filename = os.path.join(directory, filename)
self.build_daily_stats(year, month, day, logfile, filename,
compression='bz2')
def integrate_stats(self, targetdir, year, month, day, fd):
new = self.read_stats_dict(fd)
oldpath = "%s/days/%s-%.2s-%.2s.bz2" % (targetdir, year, month, day)
if os.path.exists(oldpath):
old = self.read_stats_dict(oldpath)
for k, v in new.items():
old[k] = old.get(k, 0) + v
else:
old = new
self._write_stats(oldpath, old, 'bz2')
monthpath = "%s/months/%s-%.2s.bz2" % (targetdir, year, month)
if os.path.exists(monthpath):
old = self.read_stats_dict(monthpath)
for k, v in new.items():
old[k] = old.get(k, 0) + v
else:
old = new
self._write_stats(monthpath, old, 'bz2')
return new
class ApacheLocalStats(LocalStats):
"""concrete class that uses the ApacheLogReader"""
def _get_logs(self, logfile, files_url):
return ApacheLogReader(logfile, files_url)
class ApacheDistantLocalStats(ApacheLocalStats):
"""Concrete class that gets the data from a distant file"""
is_url = re.compile(r'^http://')
def __init__(self, cache_folder='', timeout=5):
self.cache_folder = cache_folder
if not os.path.exists(cache_folder):
os.makedirs(cache_folder)
self.timeout = timeout
def get_and_cache(self, url):
"""retrieve the distant file and add it in the local
cache"""
basename = url.split('/')[-1]
filename = os.path.join(self.cache_folder, basename)
if os.path.exists(filename):
# in cache, let's return it
return filename, open(filename)
# not in cache, we need to retrieve it
# and store it
oldtimeout = socket.getdefaulttimeout()
socket.setdefaulttimeout(self.timeout)
try:
try:
content = urllib2.urlopen(url).read()
except (urllib2.URLError, socket.timeout):
return '', None
finally:
socket.setdefaulttimeout(oldtimeout)
f = open(filename, 'w')
try:
f.write(content)
finally:
f.close()
return filename, open(filename)
def read_stats(self, stats_file):
"""retrieve a distant file and works with it"""
if self.is_url.search(stats_file) is not None:
path, fileobj = self.get_and_cache(stats_file)
if path == '':
return iter([])
return ApacheLocalStats.read_stats(self, path)
|
