forked from tantalor/gogogon
/
ranks.py
executable file
·149 lines (127 loc) · 4.94 KB
/
ranks.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
#!/usr/bin/env python
import datetime
import os
import sys
import subprocess
import bitly
from domain import domain
import json
import csv
import optparse
from csv_unicode_writer import UnicodeWriter
import logging
GROUPSIZE = 10
LOG_INPUT_DIR = "/var/log/gogogon"
LOG_INPUT_PREFIX = os.path.join(LOG_INPUT_DIR, "consumer.log")
RANKS_OUTPUT_DIR = os.path.join(LOG_INPUT_DIR, "ranks")
def main():
# setup logger
formatter = logging.Formatter('%(process)d %(levelname)s %(asctime)s %(message)s', '%Y-%m-%d %H:%M:%S')
handler = logging.FileHandler("/var/log/gogogon/ranks.log")
handler.setFormatter(formatter)
logger = logging.getLogger()
logger.addHandler(handler)
logger.setLevel(logging.DEBUG)
logger.debug("starting up")
today = datetime.datetime.today()
one_day = datetime.timedelta(1)
yesterday = today - one_day
ymd = "%04d-%02d-%02d" % (yesterday.year, yesterday.month, yesterday.day)
# find yesterday's log
logfile = os.path.join(LOG_INPUT_DIR, "consumer.log.%s" % ymd)
# But allow this to be overridden
parser = optparse.OptionParser()
parser.add_option('-f', '--file', dest="logfile",
default=logfile)
parser.add_option('-o', '--output-directory', dest="output_dir",
default= RANKS_OUTPUT_DIR)
parser.add_option('-a', '--agency', dest="use_agency_domain",
default=False)
options, remainder = parser.parse_args()
logfile = options.logfile
output_dir = options.output_dir
if not os.path.exists(logfile):
raise RuntimeError('Log file does not exist: ' + logfile)
if not os.path.exists(output_dir):
raise RuntimeError('Output directory does not exist: ' + output_dir)
# sort and uniq the log
cmd = 'grep INFO %s | cut -f 5- -d " " | sort | uniq -c' % logfile
pipe = subprocess.Popen(cmd, shell=True, stdout=subprocess.PIPE)
# collect up global hashes and click counts
details = dict()
for line in pipe.stdout:
(count, global_hash, url) = line.strip().split(' ', 2)
details[global_hash] = dict(
u=url,
global_clicks=long(count),
agency=domain(url),
global_hash=global_hash,
)
# grab hashes in groups of GROUPSIZE size
for i in xrange(1+len(details)/GROUPSIZE):
hashes = details.keys()[i*GROUPSIZE:i*GROUPSIZE+GROUPSIZE]
# lookup titles
for item in bitly.info(hashes=hashes):
if 'title' not in item: continue
details[item['hash']]['title']=item['title']
# lookup yesterday's clicks
for item in bitly.clicks_by_day(hashes=hashes, days=2):
if 'clicks' not in item: continue
clicks = int(item['clicks'][1]['clicks'])
if clicks > details[item['hash']]['global_clicks']:
details[item['hash']]['global_clicks'] = clicks
# sort by global clicks descending
records = details.values()
records.sort(key=lambda x: x["global_clicks"], reverse=True)
logger.debug("writing output files")
write_output_files(records, ymd, output_dir)
if options.use_agency_domain:
write_agency_domain_files(records, output_dir, ymd)
logger.debug("shutting down")
def write_output_files(records, ymd, output_dir=RANKS_OUTPUT_DIR, latest=True):
# output files
json_file = os.path.join(output_dir, "%s.json" % ymd)
csv_file = os.path.join(output_dir, "%s.csv" % ymd)
json_latest_file = os.path.join(output_dir, "latest.json")
# write json
json.dump(records, file(json_file, 'w'))
if latest:
json.dump(records[:10], file(json_latest_file, 'w'))
# write csv
csv_writer = UnicodeWriter(file(csv_file, 'w'))
csv_writer.writerow(["Long URL", "Page Title", "Clicks", "Agency Domain", "Global hash"])
for record in records:
if not 'title' in record or not record['title']: continue
url = record['u'] if type(record['u']) == unicode else record['u'].decode('utf8')
csv_writer.writerow([
url,
record['title'],
str(record['global_clicks']),
record['agency'],
record['global_hash'],
])
def write_agency_domain_files(records, output_dir, ymd):
domains = dict()
for record in records:
domain = domains.setdefault(record['agency'],
dict(agency=record['agency'], global_clicks=0))
domain['global_clicks'] = domain['global_clicks'] + record['global_clicks']
domain_records = domains.values()
domain_records.sort(key=lambda x: x["global_clicks"], reverse=True)
json_file = os.path.join(output_dir, "domain-%s.json" % ymd)
csv_file = os.path.join(output_dir, "domain-%s.csv" % ymd)
json_latest_file = os.path.join(output_dir, "latest-domain.json")
# write json
json.dump(domain_records, file(json_file, 'w'))
json.dump(domain_records[:10], file(json_latest_file, 'w'))
# write csv
csv_writer = csv.writer(file(csv_file, 'w'))
csv_writer.writerow(["Agency Domain", "Clicks"])
for record in domain_records:
if not 'agency' in record: continue
csv_writer.writerow([
record['agency'],
record['global_clicks'],
])
if __name__ == '__main__':
main()