def main(pool):
emailDatabase = email_database.buildEmailDatabase()
start_time = time.time()
mostEmails = stat_mostEmails(pool,emailDatabase.values())
countDatabase = {}
print "--------------------"
print "Most emails sent:"
for k in mostEmails:
print str(k[0]) + " sent " + str(k[1]) + " emails"
#keep a database of email counts to use later for the reading level
countDatabase[k[0]] = k[1]
print "-------------------"
print "Most replied to emails"
mostRepliedTo = stat_mostRepliedTo(pool,emailDatabase.values())
for k in mostRepliedTo[:20]:
print str(k[0]) + " had " + str(k[1]) + " email replies to his/her email threads"
mostThreads = stat_mostThreads(pool,emailDatabase.values())
print "--------------------"
print "Most likely to reply to an email thread"
for k in mostThreads[:20]:
print str(k[0]) + " has a thread reply frequency score of " + str(k[1])
print "--------------------"
readingLevel = stat_readingDifficulty(pool,emailDatabase.values())
avgReadingLevel = sorted([(k[0],k[1]/countDatabase[k[0]]) for k in readingLevel], key=lambda obj : (-obj[1],obj[0]))
print "Highest grade reading level of emails"
for k in avgReadingLevel[:10]:
print str(k[0]) + " averaged a score of " + str(k[1])
print "-------------------"
print "Lowest grade reading level of emails"
for k in avgReadingLevel[::-1][:10]:
print str(k[0]) + " averaged a score of " + str(k[1])
print "-------------------"
print "finished in "+str(time.time() - start_time)+"s"
def main():
f = open("emails.js","w")
emailDatabase = email_database.buildEmailDatabase()
# complete list of threads
emailList = []
threads = {}
for k in emailDatabase:
del emailDatabase[k]["content"]
parsedEmail = re.search(EMAIL_PATTERN,emailDatabase[k]["from"]).group(0)
parsedDate = email.utils.parsedate_tz(emailDatabase[k]["date"])
timestamp = time.mktime(parsedDate[:-1])+parsedDate[-1]
emailDatabase[k]["from"] = parsedEmail
emailList.append(emailDatabase[k])
try:
threads[emailDatabase[k]["threadid"]].append((parsedEmail, timestamp, emailDatabase[k]["subject"]))
except KeyError:
threads[emailDatabase[k]["threadid"]] = [(parsedEmail, timestamp, emailDatabase[k]["subject"])]
threadLinks = []
#thread mappings
for thread in threads:
#sort thread groups based on timestamp to get the first email
sortedThreadEmails = sorted(threads[thread], key=lambda obj: obj[1])
for em in range(1,len(sortedThreadEmails)):
if sortedThreadEmails[em][0] == sortedThreadEmails[0][0]:
continue
threadLinks.append({"source":sortedThreadEmails[em][0], "target":sortedThreadEmails[0][0], "subject":sortedThreadEmails[0][2]})
print("var threadlinks = "+json.dumps(threadLinks)+";",file=f)
#emails by date
dateList = {}
for em in emailList:
date = time.mktime(time.strptime(" ".join(em["date"].split()[1:4]), "%d %b %Y"))
date = int(date)*1000
if date in dateList:
dateList[date].append(em)
else:
dateList[date] = [em]
print("var emailsbydate = "+json.dumps(dateList)+";",file=f)
#email authors
authors = [re.search(EMAIL_PATTERN,em["from"]).group(0) for em in emailList]
print("var authors = "+json.dumps(list(set(authors)))+";",file=f)
