Source Code for Module Ganga.Lib.LCG.Utility

  1  #!/usr/bin/env python 
  2  import re 
  3  import time 
  4  import random 
  5  import gzip 
  6  from Ganga.Utility.logging import getLogger 
  7  from Ganga.Lib.LCG.ElapsedTimeProfiler import ElapsedTimeProfiler 
  8  from Ganga.Lib.LCG.Compatibility import * 
  9   
 10 -def get_uuid(*args): 
 11      ''' Generates a universally unique ID. ''' 
 12      t = long( time.time() * 1000 ) 
 13      r = long( random.random()*100000000000000000L ) 
 14      try: 
 15          a = socket.gethostbyname( socket.gethostname() ) 
 16      except: 
 17          # if we can't get a network address, just imagine one 
 18          a = random.random()*100000000000000000L 
 19      data = str(t)+' '+str(r)+' '+str(a)+' '+str(args) 
 20   
 21      md5_obj = get_md5_obj() 
 22      md5_obj.update( data ) 
 23      data = md5_obj.hexdigest() 
 24           
 25      return data 
 26   
 27 -def urisplit(uri): 
 28     """ 
 29     Basic URI Parser according to STD66 aka RFC3986 
 30   
 31     >>> urisplit("scheme://authority/path?query#fragment") 
 32     ('scheme', 'authority', 'path', 'query', 'fragment')  
 33   
 34     """ 
 35     # regex straight from STD 66 section B 
 36     regex = '^(([^:/?#]+):)?(//([^/?#]*))?([^?#]*)(\?([^#]*))?(#(.*))?' 
 37     p = re.match(regex, uri).groups() 
 38     scheme, authority, path, query, fragment = p[1], p[3], p[4], p[6], p[8] 
 39     #if not path: path = None 
 40     return (scheme, authority, path, query, fragment)  
 41   
 42 -def readStrippedLines(fileName): 
 43      '''reads in a list of strings from a file''' 
 44   
 45      lines = [] 
 46      f = open(fileName, 'r') 
 47      for l in f.readlines(): 
 48          lines.append(l.strip()) 
 49      f.close() 
 50      return lines 
 51   
 52 -def filter_string_list(allList, filterList, type=0): 
 53      '''picks a list of strings from allList (mis-)matching the elementes in the filterList 
 54           - type = 0 : including lists given by filterLists 
 55           - type = 1 : excluding lists given by filterLists 
 56      ''' 
 57   
 58      matchedDict = {} 
 59      allDict     = {} 
 60   
 61      for item in allList: 
 62          allDict[item] = True 
 63   
 64      if type == 1: 
 65          matchedDict = allDict 
 66   
 67      for filter in filterList: 
 68          if filter.find('*') > 0: 
 69              wc = ".*".join(filter.split('*')) 
 70              for item in allDict.keys(): 
 71                  if re.match(wc, item) != None: 
 72                      if type == 0: matchedDict[item] = True 
 73                      if type == 1: del matchedDict[item] 
 74          else: 
 75              if allDict.has_key(filter): 
 76                  if type == 0: matchedDict[filter] = True 
 77                  if type == 1: del matchedDict[filter] 
 78   
 79      return matchedDict.keys() 
 80   
 81 -def get_md5sum(fname, ignoreGzipTimestamp=False): 
 82      ''' Calculates the MD5 checksum of a file ''' 
 83   
 84      profiler = ElapsedTimeProfiler(getLogger(name='Profile.LCG')) 
 85      profiler.start() 
 86   
 87   
 88      ## if the file is a zipped format (determined by extension), 
 89      ## try to get checksum from it's content. The reason is that 
 90      ## gzip file contains a timestamp in the header, which causes 
 91      ## different md5sum value even the contents are the same. 
 92      #re_gzipfile = re.compile('.*[\.tgz|\.gz].*$') 
 93   
 94      f = None 
 95   
 96      if ignoreGzipTimestamp and (fname.find('.tgz') > 0 or fname.find('.gz') > 0): 
 97          f = gzip.open(fname,'rb') 
 98      else: 
 99          f = open(fname, 'rb') 
100   
101      m = get_md5_obj() 
102   
103      while True: 
104          d = f.read(8096) 
105          if not d: 
106              break 
107          m.update(d) 
108      f.close() 
109   
110      md5sum = m.hexdigest() 
111   
112      profiler.check('md5sum calculation time') 
113   
114      return md5sum 
115