import pandas as pd
import json
import random
import pandas as pd
import tqdm
import numpy as np
import os

DATA_DIR = "./data"
DATA_ASS_DIR = "./"
DATA_SAVE_DIR = "./data"


df = pd.read_csv(DATA_DIR+"/raw.csv", dtype=object, encoding="utf-8")

#del no_answer
df=df.drop(index = df[(df.status == "NO_ANSWER")].index.tolist())
#process answer format
answer=[]

for row in df.itertuples(index=True, name='hxd'):
    if row.status=="ACCEPTED":
        answer.append(1)
    else:
        answer.append(0)

df['status']=answer

#del less than 5
_users = df.groupby(["user_id"], as_index=True)
delete_users = []
for j in tqdm.tqdm(_users):
  if len(j[1]) < 5:
    delete_users.append(j[0])
print(f"del less than 5, del{len(delete_users)}totally")
df = df[~df['user_id'].isin(delete_users)]

#process pid
psp = json.load(open(DATA_ASS_DIR + "/psp_problem.json", encoding="utf-8"))
problemsetproblemid=df["problem_set_problem_id"]
problem_id_list=[]

for psp_id in problemsetproblemid:
    try:
        problem_id_list.append(psp[str(psp_id)])
    except:
        problem_id_list.append(0)
        continue



df["problem_id"]=problem_id_list
df=df.drop(index = df[(df.problem_id == 0)].index.tolist())

#process difficulty,skill,refcount
dif = json.load(open(DATA_ASS_DIR + "/problem_difficulty.json", encoding="utf-8"))
skill = json.load(open(DATA_ASS_DIR + "/problem_knowledge.json", encoding="utf-8"))
ref = json.load(open(DATA_ASS_DIR + "/problem_reference_count.json", encoding="utf-8"))

pid=df["problem_id"]

dif_list=[]
skill_list=[]
ref_list=[]
for psp_id in tqdm.tqdm(pid):
    try:
        skill_list.append(skill[str(psp_id)])
        dif_list.append(dif[str(psp_id)])
        ref_list.append(ref[str(psp_id)])
    except:
        continue

df["reference_count"]=ref_list
df["skill_id"]=skill_list
df["difficulty"]=dif_list

df.to_csv(DATA_SAVE_DIR + "/jisuan.csv",index=0)