#!/usr/bin/python

# get-www-stats - Debian web site popularity statistics
# Copyright 2010 Marcin Owsiany <porridge@debian.org>
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program.  If not, see <http://www.gnu.org/licenses/>.


# This program is run from a crontab on a Debian website mirror like this:
#
# MAILTO="porridge@debian.org"
# # Atomically and concurrent-safely create a stats.tgz
# 18 3   *   *   *     cd "$HOME" && d=$(mktemp -d stats-wip-XXXXXXXXXX) && printf '{"hostname":"\%s"}' $(hostname -f) > "$d/stats.meta.json" && ./get-www-stats > "$d/stats.json" && tar zcf stats-wip.tgz "$d" && rm -rf "$d" && mv stats-wip.tgz stats.tgz
#
# And the output is transferred to dde.debian.net like this:
#
# MAILTO="porridge@debian.org"
# # Atomically transfer stats and replace them.
# 18 4   *   *   *     cd $HOME && { [ ! -e stats-old ] || please_cleanup_failed_run ; } && cp -al stats-new stats-old && ln -s stats-old stats-old.s && mv -T stats-old.s stats && { scp -q -i .ssh/stats-transfer-nopass senfl.debian.org:stats.tgz stats.tgz || scp_failed ; } && rm -rf stats-new && mkdir stats-new && tar zxf stats.tgz -C stats-new --strip-components=1 && rm stats.tgz && ln -s stats-new stats-new.s && mv -T stats-new.s stats && rm -rf stats-old
#
# The output is then exported via DDE (see http://wiki.debian.org/DDE) and used
# by the stattrans.pl script to sort the page lists in the Debian web site
# translation statistics pages.

try:
  import json
except ImportError:
  import simplejson as json

from gzip import open as gzopen
import logging
import os
import re
import sys

#logging.basicConfig(level=logging.INFO)

logs_dir = '/var/log/apache2'
logs_prefix = 'www.debian.org-access.log'
logs_count = 10

logs = []
for f in os.listdir(logs_dir):
  if not f.startswith(logs_prefix):
    continue
  parts = f.split('-')
  if len(parts) == 2:
    logs.append((99999999, f, False))
  elif len(parts) == 3:
    if f.endswith('.gz'):
      gzipped = True
      stamp = parts[2][:-3]
    else:
      gzipped = False
      stamp = parts[2]
    logs.append((int(stamp), f, gzipped))
  else:
    logging.warn('Skipping unexpected filename [%s].' % f)

counts = {}

for n, f, gzipped in sorted(logs)[-logs_count:]:
  logfile = os.path.join(logs_dir, f)
  logging.info('Reading %s.' % logfile)
  opener = gzipped and gzopen or open
  for line in opener(logfile):
    line = line.rstrip()
    tokens = line.split()
    url = tokens[6]
    url = re.sub(r'\...\.html$', '', url)
    url = re.sub(r'/$', '/index', url)
    if url in counts:
      counts[url] += 1
    else:
      counts[url] = 1
  
if '/index' not in counts:
  raise Exception('No data for /index')
elif counts['/index'] < 50000:
  logging.warn('Less than 50k hits for /index')
elif counts['/index'] < 10000:
  raise Exception('Less than 10k hits for /index')

json.dump(sorted([(v, k) for (k, v) in counts.iteritems() if v > 2], reverse=True),
          sys.stdout,
          indent=2)

# for v, k in sorted([(v, k) for (k, v) in counts.iteritems()], reverse=True):
#   print '%8d %s' % (v, k)
#   if v < 3:
#     break

# Perl original:
# @f=split;
# $s = $f[6];
# $s =~ s,\...\.html,,;
# $s =~ s,/$,/index,;
# $S{$s} += 1;
# END{
#   printf "%d normalized URLs\n", scalar keys %S;
#   foreach my $k (sort { $S{$b} <=> $S{$a} } keys %S) {
#     printf "%8d %s\n", $S{$k}, $k
#   }
# }