pyfpdb/IdentifySite.py

   1 #!/usr/bin/env python
   2 # -*- coding: utf-8 -*-
   3
   4 #Copyright 2010-2011 Chaz Littlejohn
   5 #This program is free software: you can redistribute it and/or modify
   6 #it under the terms of the GNU Affero General Public License as published by
   7 #the Free Software Foundation, version 3 of the License.
   8 #
   9 #This program is distributed in the hope that it will be useful,
  10 #but WITHOUT ANY WARRANTY; without even the implied warranty of
  11 #MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  12 #GNU General Public License for more details.
  13 #
  14 #You should have received a copy of the GNU Affero General Public License
  15 #along with this program. If not, see <http://www.gnu.org/licenses/>.
  16 #In the "official" distribution you can find the license in agpl-3.0.txt.
  17
  18 import L10n
  19 _ = L10n.get_translation()
  20
  21 import re
  22 import sys
  23 import os
  24 from optparse import OptionParser
  25 import codecs
  26 import Configuration
  27 import Database
  28
  29 __ARCHIVE_PRE_HEADER_REGEX='^Hand #(\d+)\s*$|\*{20}\s#\s\d+\s\*{20,25}\s+'
  30 re_SplitArchive = re.compile(__ARCHIVE_PRE_HEADER_REGEX, re.MULTILINE)
  31
  32 class IdentifySite:
  33     def __init__(self, config, in_path = '-', list = []):
  34         self.in_path = in_path
  35         self.config = config
  36         self.db = Database.Database(self.config)
  37         self.sitelist = {}
  38         self.filelist = {}
  39         self.generateSiteList()
  40         if list:
  41             for file in list:
  42                 self.idSite(file, self.sitelist)
  43         else:
  44             if os.path.isdir(self.in_path):
  45                 self.walkDirectory(self.in_path, self.sitelist)
  46             else:
  47                 self.idSite(self.in_path, self.sitelist)
  48
  49     def get_filelist(self):
  50         return self.filelist
  51
  52     def generateSiteList(self):
  53         """Generates a ordered dictionary of site, filter and filter name for each site in hhcs"""
  54         for site, hhc in self.config.hhcs.iteritems():
  55             filter = hhc.converter
  56             filter_name = filter.replace("ToFpdb", "")
  57             summary = hhc.summaryImporter
  58             result = self.db.get_site_id(site)
  59             if len(result) == 1:
  60                 self.sitelist[result[0][0]] = (site, filter, filter_name, summary)
  61             else:
  62                 pass
  63
  64     def walkDirectory(self, dir, sitelist):
  65         """Walks a directory, and executes a callback on each file"""
  66         dir = os.path.abspath(dir)
  67         for file in [file for file in os.listdir(dir) if not file in [".",".."]]:
  68             nfile = os.path.join(dir,file)
  69             if os.path.isdir(nfile):
  70                 self.walkDirectory(nfile, sitelist)
  71             else:
  72                 self.idSite(nfile, sitelist)
  73
  74     def __listof(self, x):
  75         if isinstance(x, list) or isinstance(x, tuple):
  76             return x
  77         else:
  78             return [x]
  79
  80     def idSite(self, file, sitelist):
  81         """Identifies the site the hh file originated from"""
  82         if file.endswith('.txt') or file.endswith('.xml'):
  83             self.filelist[file] = ''
  84             archive = False
  85             for id, info in sitelist.iteritems():
  86                 site = info[0]
  87                 filter = info[1]
  88                 filter_name = info[2]
  89                 summary = info[3]
  90                 mod = __import__(filter)
  91                 obj = getattr(mod, filter_name, None)
  92                 if summary:
  93                     smod = __import__(summary)
  94                     sobj = getattr(smod, summary, None)
  95
  96                 for kodec in self.__listof(obj.codepage):
  97                     try:
  98                         in_fh = codecs.open(file, 'r', kodec)
  99                         whole_file = in_fh.read(2000)
 100                         in_fh.close()
 101                         if filter_name in ('OnGame', 'Winamax'):
 102                             m = obj.re_HandInfo.search(whole_file)
 103                         elif filter_name in ('PartyPoker'):
 104                             m = obj.re_GameInfoRing.search(whole_file)
 105                             if not m:
 106                                 m = obj.re_GameInfoTrny.search(whole_file)
 107                         else:
 108                             m = obj.re_GameInfo.search(whole_file)
 109                             if m and re_SplitArchive.search(whole_file):
 110                                 archive = True
 111                             if not m and summary:
 112                                 m = sobj.re_TourneyInfo.search(whole_file)
 113                                 if m:
 114                                     filter = summary
 115                         if m:
 116                             self.filelist[file] = [site] + [filter] + [kodec] + [archive]
 117                             break
 118
 119                     except:
 120                         pass
 121
 122 def main(argv=None):
 123     if argv is None:
 124         argv = sys.argv[1:]
 125
 126     config = Configuration.Config(file = "HUD_config.test.xml")
 127     in_path = os.path.abspath('regression-test-files')
 128     IdSite = IdentifySite(config, in_path)
 129
 130     print "\n----------- SITE LIST -----------"
 131     for site, info in IdSite.sitelist.iteritems():
 132         print site, info
 133     print "----------- END SITE LIST -----------"
 134
 135     print "\n----------- ID REGRESSION FILES -----------"
 136     for file, site in IdSite.filelist.iteritems():
 137         print file, site
 138     print "----------- END ID REGRESSION FILES -----------"
 139
 140
 141 if __name__ == '__main__':
 142     sys.exit(main())