Improve duplicate handling

This commit is contained in:
Klaus-Uwe Mitterer 2015-03-19 20:56:57 +01:00
parent 5ad1fac2cc
commit b7b619ba0b

View file

@ -70,16 +70,18 @@ while True:
print "Error in " + str(twid) print "Error in " + str(twid)
print e print e
continue continue
words = text.split(" ") words = text.split(" ")
for word in words: for word in words:
if word[0] == "@" and (not user == word[1:]) and (not sender == word[1:]): wordlist += word
wordlist += word
wordlist = list(set(wordlist))
wordlist = list(set(wordlist)) try:
for word in wordlist: wordlist.remove(user)
comment += word + " " wordlist.remove(sender)
except ValueError:
pass
cur.execute("INSERT INTO tweets VALUES(%i,'%s','%s','%s',0)" % (twid,date.strftime("%Y-%m-%dT%H:%M:%S"),sender,comment.strip())) cur.execute("INSERT INTO tweets VALUES(%i,'%s','%s','%s',0)" % (twid,date.strftime("%Y-%m-%dT%H:%M:%S"),sender,comment.strip()))
sql_conn.commit() sql_conn.commit()