# -*- coding: utf-8 -*-
"""
Created on Sun Apr  7 17:19:35 2019

@author: quliang
"""

import pymysql
from pymongo import MongoClient
import logging
import re
import json

def get_text(s):
    
    pattern=re.compile(r'\'text\': \'(.*?)\'')
    res_list = re.findall(pattern,str(s))
    res_str = ' '.join(res_list)
    
    return res_str

def get_court_consider(s): 
    
    pattern=re.compile(r'\'text\': \"(.*?)\"')
    res_list = re.findall(pattern,str(s))
    res_str = ' '.join(res_list)
    
    return res_str


def data_to_mysql(data,keys,test=False):

    db_data='test' if test else 'renren_data'
    db = pymysql.connect(host='193.112.56.204', user='renrendata', passwd='data@2018',
         db=db_data, port=3306, charset='utf8') 
    cursor = db.cursor()
    cols = ", ".join('`{}`'.format(k) for k in keys)
    val_cols = ', '.join('%({})s'.format(k) for k in keys)
    sql = "insert into judgement_doc(%s) values(%s)"
    res_sql = sql % (cols, val_cols)
    #print(res_sql)
    cursor.executemany(res_sql,data)
    db.commit()

logger = logging.getLogger()
logging.basicConfig(format='%(asctime)s - %(filename)s[line:%(lineno)d] - %(levelname)s: %(message)s')
logger.setLevel(logging.ERROR)

client = MongoClient('mongodb://admin:admin@193.112.23.12:27017')
db=client.law_nlp

#collist = db.list_collection_names()

collist=['zhejiang', 'fujian', 'hebei', 'neimenggu', 
 'hainan', 'hubei','heilongjiang', 'jiangsu', 'anhui','guizhou','liaoning', 
 'qinghai','gansu','chongqing','henan','hunan', 'guangxi']

#collist1=['zhejiang']

keys=['province', 'save_id', 'judgementId', 'judgementType', 'reason_name', 
 'title', 'regulation', 'trial_procedure', 'original_told', 'was_told', 
 'court_get', 'court_consider', 'court_result','frist_court_get','appellor_told',
 'by_appellor_told','frist_court_consider','second_court_result']




for col in collist:
    w=[]
    collection=db[col]
    print('finding data ing')
    result = collection.find({"content.judgementType":"判决","content.title":{'$regex':'婚'}})
    if not result:
        continue
    print('{} data collected finished'.format(col))
    result = list(result)
    lens=len(list(result))
    print('{}有{}条数据'.format(col,lens))
    for inx,res in enumerate(result):
        logger.info('res detail',res)
        panjue_data={k:'' for k in keys}
        panjue_data['province']=col
        panjue_data['save_id']=inx
        panjue_data['judgementType']=res['content']['judgementType']
        panjue_data['title']=res['content']['title']
        
        if res.get('judgementId'):
            panjue_data['judgementId']=res['judgementId']
            
        if res['content'].get('reason'):
            if res['content']['reason'].get('name'):
                panjue_data['reason_name']=res['content']['reason']['name']
        
        #解析法条
        if res['content'].get('regulationGroupByTrialRoundInfos'):
            for trialRoundText in res['content']['regulationGroupByTrialRoundInfos']:
                #一审 ， 二审 分隔符&&
                if trialRoundText.get('trialRoundText'):
                    panjue_data['regulation']+='&'+trialRoundText['trialRoundText']+'&'+'\n'
                    for sectionParagraphs in trialRoundText['regulations']:
                        logger.info('sectionParagraphs detail',sectionParagraphs)
                        panjue_data['regulation']+=sectionParagraphs['text']
                        panjue_data['regulation']+=get_text(sectionParagraphs)
                    #多个法条分割符'######'
                    panjue_data['regulation']+='\n'
        else:
            logger.info('judgementId',res['judgementId'],col,'is not exsixt keywords:regulationGroupByTrialRoundInfos')
        
        #解析各个分块       
        for sub in res['content']['paragraphs']:
            if sub['typeText']=='审理经过':
                panjue_data['trial_procedure']=get_text(sub['subParagraphs'])
                continue
            if sub['typeText']=='一审法院查明':
                panjue_data['frist_court_get']=get_text(sub['subParagraphs'])
                continue
            if sub['typeText']=='原告诉称':
                panjue_data['original_told']=get_text(sub['subParagraphs'])
                continue
            if sub['typeText']=='上诉人诉称':
                panjue_data['appellor_told']=get_text(sub['subParagraphs'])
                continue
            if sub['typeText']=='被告辩称':
                panjue_data['was_told']=get_text(sub['subParagraphs'])
                continue
            if sub['typeText']=='被上诉人辩称':
                panjue_data['by_appellor_told']=get_text(sub['subParagraphs'])
                continue
            if sub['typeText']=='本院查明':
                panjue_data['court_get']=get_text(sub['subParagraphs'])
                continue
            if sub['typeText']=='一审法院认为':
                panjue_data['frist_court_consider']=get_court_consider(sub['subParagraphs'])
                continue
            if sub['typeText']=='本院认为':
                panjue_data['court_consider']=get_court_consider(sub['subParagraphs'])
                continue
            if sub['typeText']=='裁判结果':
                panjue_data['court_result']=get_text(sub['subParagraphs'])
                continue
            if sub['typeText']=='二审裁判结果':
                panjue_data['second_court_result']=get_text(sub['subParagraphs'])
                continue
        #str_json=json.dumps(panjue_data,indent=2, ensure_ascii=False)
        #with open('data.txt','a') as f:
            #f.write(str_json+'\n')
        w.append(panjue_data)
        if inx%1000==0:
            print('=======>  {}有{}条数据已保存'.format(col,inx))

data_to_mysql(w,keys,test=False)