read_pdf_plt/parsePDF.py

import pdfplumber
import pandas as pd
import numpy as np;
basePath = "D:/";
# pdf_path = basePath+"闻海雁532329198801060347(2).pdf"
basePath = "D:/mydocument/myproject/git/busscredit/Crerdai/";
pdf_path = basePath+"雷雨晴130630199006130027.pdf"
# pdf_path=basePath+"杨安140402197102111236.pdf"
# pdf_path=basePath+"刘盼兰130133198912261210.pdf"
# pdf_path=basePath+"马维强130521198604045272.pdf"
# pdf_path=basePath+"郑晨晨130681199008205811.pdf"
reportTime=None;
reportTimeCn="报告时间："
queryreq = {}
identity = {}

queryreqDf=pd.DataFrame;
identityDf=pd.DataFrame;
addressDf=pd.DataFrame;
mateDf=pd.DataFrame;
CREDITSUMMARYCUE_df = pd.DataFrame
FIRSTLOANOPENMONTH_df=pd.DataFrame
FIRSTLOANCARDOPENMONTH_df=pd.DataFrame;

FIRSTLOANOPENMONTH = "";#首笔贷款账龄
FIRSTLOANCARDOPENMONTH=""#首笔贷记卡账龄
OVERDUESUM_df=pd.DataFrame;
def parseQueryreq(reportTime):
    queryreq["reportTime"]=reportTime.split("：")[1];

headers_queryreq = ["被查询者姓名","被查询者证件类型","被查询者证件号码","查询机构","查询原因"];
header_identity=['性别', None, '出生日期', '婚姻状况', '学历', '学位', '就业状况', '国籍', '电子邮箱']
header_address=['通讯地址', None, None, None, None, '户籍地址', None, None, None]
header_mate = ['姓名', '证件类型', '证件号码', '工作单位', '联系电话']
header_CREDITSUMMARYCUE=['业务类型', None, '账户数', '首笔业务发放月份']
header_OVERDUESUM=['账户类型', '账户数', '月份数', '单月最高逾期/透支总额', '最长逾期/透支月数']
header_SHAREANDDEBT=['管理机构数', '账户数', '授信总额', '余额', '最近6个月平均应还款']


headers = [0] * 100
ignoreText=["异议信息提示","信息主体"]

def isIgnore(text):
    for txt in ignoreText:
        if text.find(txt)>=0:
            return True;
    return False;

#根据列确定表头
#判断条件需要修改
def hasHeader(headers,rows):
    text = rows[0]
    for header in headers:
        if header == text:
            return True
    return False;

def headerOf(headers,entityHeader):
    for header in headers:
        if str(entityHeader).find(header)>=0:
            return True
    return False;
def getMinMonth(FIRSTLOANOPENMONTH_df):
    FIRSTLOANOPENMONTH_values = [];
    FIRSTLOANOPENMONTH = ""
    for value in FIRSTLOANOPENMONTH_df.values:
        if value == '--':
            continue
        FIRSTLOANOPENMONTH_values.append(value)
    if len(FIRSTLOANOPENMONTH_values)>0:
        far = np.array(FIRSTLOANOPENMONTH_values);
        findex = np.argmax(far)
        FIRSTLOANOPENMONTH = FIRSTLOANOPENMONTH_values[findex][0];
    return FIRSTLOANOPENMONTH;

pd.Series()
with pdfplumber.open(pdf_path) as pdf:
    for page in pdf.pages:
        # first_page = pdf.pages[1]
        for table in page.extract_tables():
            df = pd.DataFrame(table)
            # 第一列当成表头：
            # df = pd.DataFrame(table[1:],columns=table[0])
            tmp=df.loc[0,:].tolist()
            tmp2=['业务类型',None,'账户数','首笔业务发放月份']
            #判断表头
            if tmp==tmp2 :
                #要处理的一段数据
                dfData=df[df[3]!='--'] #过滤
                print(dfData[3].min())
                print("$$$$$$$$$$$$$$$")

            #@print(df)