build_cascade.py

#coding=utf-8
import datetime
import codecs

if __name__ == "__main__":
    result = []
    l4 = []
    f = codecs.open("3638347916812769_retweets_id_text_time.txt",encoding="utf-8")
    rootId = "3638347916812769_retweets"
    l = f.readlines()
    l2 = [tuple(x.strip().split("\t")) for x in l]
    l3 = [(x[0],x[1],datetime.datetime.strptime(x[2],"%Y-%m-%d %H:%M:%S")) for x in l2] # (id,text,time)
    l3.sort(key=lambda x:x[2])
    for item in l3:
        if not "//@" in item[1]:
            result.append((rootId,item[0]))
        else:
            l4.append(item)