/
pywik.py
executable file
·208 lines (186 loc) · 7.43 KB
/
pywik.py
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
#!/usr/bin/env python
# -*- coding: utf-8 -*-
# todo, enable and handle: $http_HEADER, $cookie_COOKIE in nginx logformat def
import sys, datetime, json
import pymongo
from operator import itemgetter
from plugins import init_plugins
from hashlib import sha1
# what i want to see
# good
# graph of hits/day
# popular pages
# referers
# search queries
# viewers (see historical visits if any, distinguish by agent?)
# bots
# non-404 error requests, hosts
# bad
# tor visits
# non-typo 404 pages
# top non-browser-user-agents (cross-ref with classes of other requests)
# masquerading crawlers
queries=[ ('pages', {'country': {'$ne': 'UA'},
'$or': [{'bad': { '$size': 0 }},{'bad': {'$exists': False}}],
'$and': [ { 'tags': 'page' }, { 'tags': { '$nin': ['bot'] } } ]}, ['path']),
('searchqueries', {'search_query': {'$ne': ""}},['search_query']),
('errors', {'$or': [{'status': {'$lt': 200}},
{'status': {'$gt': 400}}],
'status': {'$ne': 404}},['status', 'path']),
('unknown', {'tags': 'unknown'},['path']),
('notfound', {'tags': 'notfound'},['path']),
('bots', {'tags': 'bot'},['http_user_agent']),
('404s', {'status': 404},['path']),
]
spans={'today': 1,
'yesterday': 2,
'recently': 3,
'week': 7,
'month': 30,
'quarter': 121,
}
def spark(data):
if not data: return ('',0,0)
blocks = u' ▁▂▃▄▅▆▇█'
lo = float(min(data))
hi = float(max(data))
incr = (hi - lo)/(len(blocks)-1) or 1
return (''.join([(blocks[int((float(n) - lo)/incr)]
if n else
' ')
for n in data]),
int(lo),
int(hi))
def sparks(host, key, q):
db=pymongo.Connection().pywik
return spark([x['count']
for x
in db.__getitem__(host).group([key],
q,
{'count': 0},
'function(doc, out){ out.count++;}')])
def tostr(x):
try:
return str(x).encode('raw_unicode_escape').decode('utf-8')
except:
return repr(x)
def displayQ(host, title, q, fields):
db=pymongo.Connection().pywik
res=[]
cur=db.__getitem__(host).find(q, fields)
res.append('Total %s %s' % (title, cur.count()))
res.append(u'\n'.join(
[u"%s\t%s" % (int(x['count']),
u' '.join([tostr(x[f]) for f in fields]))
for x in sorted(db.__getitem__(host).group(fields,
q,
{'count': 0},
'function(doc, out){ out.count++;}'),
key=itemgetter('count'),
reverse=True)]).encode('utf8'))
return res
def pywik(site, span=1):
db=pymongo.Connection().pywik
now=datetime.datetime.now()
qspan = now - datetime.timedelta(days=(7 if span<8 else span))
span = now - datetime.timedelta(days=span)
res=[]
reports=queries[:]
reports.extend(init_plugins(site)[1])
for title, q, fields in reports:
#print >>sys.stderr, title, q, fields
qsparks=dict([x for x in q.items()])
qsparks["timestamp"]={'$gt': qspan, '$lt': now}
q["timestamp"]={'$gt': span, '$lt': now}
res.append(
{'total': db.__getitem__(site).find(q, fields).count(),
'fields': fields,
'title': title,
'sparks': tod3(db.__getitem__(site).group(fields+['year','month','day'],
qsparks,
{'count': 0},
'function(doc, out){ out.count++;}'),fields),
'lines': [ (int(x['count']),
dict([(f,x[f]) for f in fields]) )
for x in sorted(db.__getitem__(site).group(fields,
q,
{'count': 0},
'function(doc, out){ out.count++;}'),
key=itemgetter('count'),
reverse=True ) ] } )
return res
def ascii():
now=datetime.datetime.now()
span=None
for k, v in spans.items():
if k in sys.argv:
span = now - datetime.timedelta(days=v)
del sys.argv[sys.argv.index(k)]
if not span:
span = now - datetime.timedelta(days=1)
reports=[]
queries.extend(init_plugins(sys.argv[1])[1])
for title, q, fields in queries:
if title in sys.argv:
reports.append((title, q, fields))
del sys.argv[sys.argv.index(title)]
if not reports: reports=queries
res=[]
for title, q, fields in reports:
q["timestamp"]={'$gt': span, '$lt': now}
gfx, lo, hi = sparks(sys.argv[1], 'day', q)
res.append((u"%30s %8s %s %s" % (title, lo, gfx, hi)).encode('utf8'))
res.extend(displayQ(sys.argv[1], title, q, fields))
res.append('='*80)
return res
def getentries(site,key,val, span):
db=pymongo.Connection().pywik
now=datetime.datetime.now()
span = now - datetime.timedelta(days=span)
return db.__getitem__(site).find({key:val,
"timestamp": {'$gt': span,
'$lt': now}})
def tod3(entries,fields):
acc={}
mx = None
mn = None
for item in entries:
date=(int(item['year']),int(item['month']),int(item['day']))
title=u' '.join([unicode(item[k]) for k in fields])
tmp={'x': date, 'y': int(item['count']), 'title': title}
try:
acc[title].append(tmp)
except KeyError:
acc[title]=[tmp]
if not mx or mx < date:
mx=date
if not mn or mn > date:
mn=date[:3]
ret = []
for k, sparse in acc.items():
res=[]
i=0
if k in ['-','']: continue
cur=datetime.date(*mn)- datetime.timedelta(days=1)
for item in sorted(sparse, key=itemgetter('x')):
diff=(datetime.date(*item['x'])-cur).days
if diff>1: res.extend([{'x': len(res)+j, 'y': 0, 'text': item['title']} for j in xrange(diff-1)])
res.append({'x': len(res), 'y': item['y'], 'text': item['title']})
cur=datetime.date(*item['x'])
i=i+1
diff=(datetime.date(*mx) - cur).days
res.extend([{'x': len(res) + j, 'y': 0} for j in xrange(diff)])
ret.append(res)
return json.dumps(ret)
def timeline():
db=pymongo.Connection().pywik
for item in db.__getitem__(sys.argv[1]).find().sort([('time_local', pymongo.DESCENDING)]):
print item['timestamp'], \
item['status'], \
sha1(''.join([item['remote_addr'],item['http_user_agent']])).hexdigest()[:8], \
', '.join(item['tags']), \
item['path'].encode('utf8') if 'rss' not in item['tags'] else '', \
item['http_referer'].encode('utf8') if 'extref' in item['tags'] else ''
if __name__ == "__main__":
print '\n'.join(ascii())
timeline()