@@ -8,7 +8,7 @@ # the nodes are the persons # the edges have a type to/cc and a weight (number of mails) -import sys, os, datetime, getopt, ConfigParser, platform +import sys, os, cStringIO, codecs, datetime, getopt, platform, csv #TODO , ConfigParser if(platform.machine()=='i686'): import psyco from sqlobject import * @@ -17,8 +17,8 @@ from lib.utils import decode_header CONFIG = {'database': os.path.abspath('../db/messages.db'), 'personmapfile' : '../db/persons.map', - 'format': "dot", # dot|graphxml|log - 'stats': True, # display overall stats + 'format': "csv", # dot|graphxml|log|csv + 'stats': False, # display overall stats 'egg': False, # filter out entity: 'egger': 'Marsiske Stefan', } @@ -32,7 +32,7 @@ def usage(): print "\t-h This Help" print "\t-s|--stats Display top ten stats on in/out contacts." # print "\t-f |--format== [dot,log,graphxml] Output format." - print "\t-f |--format== [dot,log] Output format." + print "\t-f |--format== [dot,log,csv] Output format." def counter(start=0): while True: @@ -41,6 +41,31 @@ def counter(start=0): nodeIdGenerator=iter(counter(0)) edgeIdGenerator=iter(counter(0)) +class UnicodeWriter: + """ + A CSV writer which will write rows to CSV file "f", + which is encoded in the given encoding. + """ + def __init__(self, f, dialect=csv.excel, encoding="utf-8", **kwds): + # Redirect output to a queue + self.queue = cStringIO.StringIO() + self.writer = csv.writer(self.queue, dialect=dialect, **kwds) + self.stream = f + self.encoder = codecs.getincrementalencoder(encoding)() + + def writerow(self, row): + #ORIG:self.writer.writerow([s.encode("utf-8") for s in row]) + self.writer.writerow(row) + # Fetch UTF-8 output from the queue ... + data = self.queue.getvalue() + data = data.decode("utf-8") + # ... and reencode it into the target encoding + data = self.encoder.encode(data) + # write to the target stream + self.stream.write(data) + # empty queue + self.queue.truncate(0) + class Node: def __init__(self,name,receivers = []): self.__dict__['id'] = "n"+str(nodeIdGenerator.next()) @@ -205,30 +230,35 @@ def loadPersonMap(): personmap[email] = name.strip() def buildGraph(): - graph = Graph() - q = Role.select() - for edge in q: - sender = edge.msg - if(edge.msg.sender and edge.msg.sender.owner): - sender = edge.msg.sender.owner.fullname - elif(edge.msg.sender): - sender = edge.msg.sender.username+"@"+edge.msg.sender.mailserver - if(sender and personmap.has_key(sender)): - sender = personmap[sender] - - receiver = edge.email - if(edge.email and edge.email.owner): - receiver = edge.email.owner.fullname - elif(edge.email): - receiver = edge.email.username+"@"+edge.email.mailserver - if(receiver and personmap.has_key(receiver)): - receiver = personmap[receiver] - - if(CONFIG['format']=="log"): - print edge.msg.delivered,sender, edge.header.name, receiver - - graph.addToGraph(edge.msg.delivered,sender,edge.header.name, receiver) - return graph + graph = Graph() + q = Role.select() + csvcoder=None + if(CONFIG['format']=="csv"): + csvcoder=UnicodeWriter(sys.stdout) + for edge in q: + sender = edge.msg + if(edge.msg.sender and edge.msg.sender.owner): + sender = edge.msg.sender.owner.fullname + elif(edge.msg.sender): + sender = edge.msg.sender.username+"@"+edge.msg.sender.mailserver + if(sender and personmap.has_key(sender)): + sender = personmap[sender] + + receiver = edge.email + if(edge.email and edge.email.owner): + receiver = edge.email.owner.fullname + elif(edge.email): + receiver = edge.email.username+"@"+edge.email.mailserver + if(receiver and personmap.has_key(receiver)): + receiver = personmap[receiver] + + if(CONFIG['format']=="log"): + print edge.msg.delivered,sender, edge.header.name, receiver + elif(CONFIG['format']=="csv"): + csvcoder.writerow(map(lambda x: str(x),(edge.msg.delivered,sender, edge.header.name, receiver))) + + graph.addToGraph(edge.msg.delivered,sender,edge.header.name, receiver) + return graph def dumpResults(graph): #print graph @@ -266,7 +296,7 @@ if __name__=='__main__': elif o in ("-s", "--stats"): CONFIG['stats'] = True elif o in ("-f", "--format"): - if(a and a in ("dot","log","graphxml")): + if(a and a in ("dot", "log", "graphxml", "csv")): CONFIG['format'] = a else: usage() utils/getgraph 6ef7d3e6e984fa49642cba9c5b7f7578efdb656a Stefan Marsiske stefan.marsiske@gmail.com http://github.com/stef/maelstrom/commit/898a3bf88dd7e5c3e723c7cf495824481caaf6b3 898a3bf88dd7e5c3e723c7cf495824481caaf6b3 2009-03-07T14:59:11-08:00 2009-03-07T14:59:11-08:00 [enh] added csv export of log style output 77f5d9df95d88a0f8953332217481c2d6b1a76ca Stefan Marsiske stefan.marsiske@gmail.com