BIDI_ML_INFO_EXTRACTION/BiddingKG/dl/time/re_servicetime.py

#coding:UTF-8
import re
import pandas as pd
import numpy as np
from bs4 import BeautifulSoup
# from sqlalchemy import create_engine

TEST_MODE = False


# before = '(?P<before>' \
#          '合同期限|工期/交货期/服务期|工期，\(日历天\)|工期\(交货期\)|合格工期\(天\)|服务期限\(年\)|工期\(天\)' \
#          '|工期要求|项目周期|工期\(交货期\)|计划工期\(服务期限\)|服务时限|履行期限|服务周期|供货期' \
#          '|合格工期|计划工期\(服务期\)|服务期\(日历天\)|服务，期|交货\(完工\)时间|交付\(服务、完工\)时间' \
#          '|交货时间|工期\(日历天\)' \
#          '|服务期限为|计划工期|工期要求|服务期限|服务期' \
#          '|投标工期|设计工期|合格服务周期|总工期|服务时间|流转期限|维护期限|服务时限|交货期|服务要求' \
#          '|完成时间|服务期限|中标工期|项目周期|期限要求|周期|工期|供货期|合同履行日期|计划周期|工期' \
#          ')'

before = '(?P<before>' \
         '合同期限|工期/交货期/服务期|工期，|工期\(交货期\)|合格工期|服务期限|工期' \
         '|工期要求|项目周期|工期\(交货期\)|计划工期\(服务期限\)|服务时限|履行期限|服务周期|供货期' \
         '|合格工期|计划工期\(服务期\)|服务期|服务，期|交货\(完工\)时间|交付\(服务、完工\)时间' \
         '|交货时间|工期|质保期' \
         '|服务期限为|计划工期|工期要求|服务期限|服务期' \
         '|投标工期|设计工期|合格服务周期|总工期|服务时间(范围)?|流转期限|维护期限|服务时限|交货期' \
         '|完成时间|服务期限|中标工期|项目周期|期限要求|周期|供货期|合同履行日期|计划周期' \
         '|履约期限|合同约定完成时限|合同完成日期|承诺完成日期' \
         '|合同起始日起|合同履约期|履约截止日期|承包期限|合同完成日期|特许经营期限' \
         '|服务期间|服务履行期|委托(管理)?期限|经营期限' \
         ')'


# ^(?!.*abc).*$ 排除abc字符串
before_wuye = '(?P<before>' \
              '(履约期限、地点等简要信息[：:]((履约|时间|期限){1,2}[：:])?)' \
              ')'
# '|(履约期限、地点等简要信息[^\d半一二三四五六七八九十壹两叁贰肆伍陆柒捌玖拾]{0,25})' \
# (履约期限、地点等简要信息.{0,25}(?= [\d半一二三四五六七八九十壹两叁贰肆伍陆柒捌玖拾]+([年月日]|个月)|20[21]))

before2 = '(?P<before2>' \
          '自合同签订之日起至|合同签订之日起|自合同签订之日起|签订合同后|系统开发' \
          '|合同签订之日起至|自合同签订之日|合同签定后|自签订合同之日起|自合同签订起' \
          '|[自从]?合同签[订定]生效之日起|自合同签订后不超过|合同签订日至' \
          '|合同签订生效之日起|本项目招标有效期' \
          '|[自从于]?签[订定署字](合同|协议书|协议)并?(期|开始履行|生效|有效期|约定|验收合格|期限|开始服务){0,2}(之[日后]|日期?[后起]|后|起|算|为)+[^。\d半一二三四五六七八九十壹两叁贰肆伍陆柒捌玖拾]{0,4}' \
          '|[自从于]?(采购)?(合同|协议书|协议)(正式)?签[订定署字](完[成毕])?并?(期|开始履行|生效|有效期|约定|验收合格|期限|开始服务){0,2}(之[日后]|日期?[后起]|后|起|算|为)+[^。\d半一二三四五六七八九十壹两叁贰肆伍陆柒捌玖拾]{0,5}' \
          '|签订合同起' \
          '|项目的有效期限为|项目服务为|签订合同期为' \
          '|(合同|协议书)签[订定署字]生效(之[日后]|后|起)[^。\d半一二三四五六七八九十壹两叁贰肆伍陆柒捌玖拾]{1,4}' \
          '|[自从于]服务(合同|协议书|协议)生效(之[日后]|后|起)[^。\d半一二三四五六七八九十壹两叁贰肆伍陆柒捌玖拾]{1,4}' \
          '|(本次)?采购周期' \
          '|(项目招标)?履行期|[自从于]?(合同|协议书|协议)生效(之[日后]|后|起)[^。\d半一二三四五六七八九十壹两叁贰肆伍陆柒捌玖拾]{1,3}' \
          '|服务(有效期|年限)|本?合同有效期|(正式)?入驻(之[日后]|后|起|算)+' \
          '|(合同|协议书|协议)生效(之[日后]|后|起|算)+' \
          '|自?(提供服务|采购人指定|合同约定)(之[日后]|后|起|算)+' \
          '|本?项目合同期(为|是)*' \
          '|交付使用(之[日后]|后|起|算)+|' \
          ')'
        # '|[自从于].{2,15}之日[^。\d半一二三四五六七八九十壹两叁贰肆伍陆柒捌玖拾]{1,4}' \

before3 = '(?P<before3>' \
          '([\(（]日历天[\)）]|[\(（]天[\)）]|[\(（]年[\)）]|[\(（]月[\)）])?' \
          ')'

before4 = '(?P<before4>' \
          '(履约|[本项目原则上]*一招|期限|(服务|合同)(期|)|合计|均为|开工后|不超过|中选后|计划|达到|本合同|)' \
          ')'

charac = '(?P<charac>' \
         '[:：,，【（）】]*' \
         ')'

# charac前后、center前、after1前 需加
before5 = '(?P<before5>' \
          '[自为约是起暂定的拟有效期限从共计至算是要求总服务到本项目]{0,5}' \
          ')'
before6 = '(?P<before6>' \
          '[自为约是起暂定的拟有效期限从共计至算是要求总服务到本项目]{0,5}' \
          ')'
before7 = '(?P<before7>' \
          '[自为约是起暂定的拟有效期限从共计至算是要求总服务到本项目]{0,5}' \
          ')'

center = '(?P<center>' \
         '(\d{2,4}[-.年/](\d{1,2}[-.月/])?(\d{0,2}[日号]?)?[-～~开始起至到—－]+(\d{2,4}[-.年/]\d{1,2}[-.月/]\d{0,2}[日号]?|\d{2,4}[-.年/]\d{1,2}[-.月/]?|\d{1,2}[-.月/]\d{1,2}[日号]?|\d{2,4}[-.年/]|\d{1,2}[-.月/]|\d{1,2}[日号]?)' \
         '|\d{2,4}[-.年/]\d{1,2}[-.月/](\d{1,2}[日号]?)?' \
         '|[+\d半一二三四五六七八九十壹两叁贰肆伍陆柒捌玖拾]+)(\)|)' \
         ')'

center2 = '(?P<center2>' \
          '[.\d]+个?[月年]' \
          ')'

number = '(?P<number>' \
         '[\d半一二三四五六七八九十壹两叁贰肆伍陆柒捌玖拾]+' \
         ')'

after = '(?P<after>' \
        '[个,，（\(]*(日历|工作|学|)([年月日天周]|周年|整年)(内|）|\)|)|' \
        ')'
        # '|周|号|天|个月|个年|(（|\(|)年(）|\)|)|(（|\(|)月(）|\)|)|(（|\(|)日(）|\)|)' \
        # '|个日历天|日历天|\(日历天\)|\(天\)|周内|，日历天|工作日|个工作日|' \

after1 = '(?P<after1>' \
         '\d{2,4}[-.年/](\d{1,2}[-.月/])?(\d{1,2}[日号])?[-～~开始起至到—]+(\d{2,4}[-.年/]\d{1,2}[-.月/]\d{0,2}[日号]?|\d{2,4}[-.年/]\d{1,2}[-.月/]?|\d{1,2}[-.月/]\d{1,2}[日号]?|\d{2,4}[-.年/]|\d{1,2}[-.月/]|\d{1,2}[日号]?)([】）]?)' \
         ')'

after2 = '(?P<after2>' \
         '\d+' \
         ')'

after3 = '(?P<after3>' \
         '(([\d半一二三四五六七八九十壹两叁贰肆伍陆柒捌玖拾][年月日])?)' \
         ')'

after4 = '(?P<after4>' \
         '[^\d半一二三四五六七八九十壹两叁贰肆伍陆柒捌玖拾]{0,25}止' \
         ')'


reg = re.compile(before + before3 + before7 + charac + before5 + before2 + before4 + before6 + center + after)

reg1 = re.compile(before + before3 + before7 + charac + before5 + after4 + after3)

reg2 = re.compile(before + before3 + before7 + charac + before5 + before2 + before6 + after1)

reg3 = re.compile(before + before3 + before7 + charac + before5 + before2 + after2)

reg4 = re.compile(before2[:-2]+before2[-1:] + before5 + center + after)

reg5 = re.compile(before + before3 + before7 + charac + before5 + before2 + before4 + before6 + center2 + after)

# reg4 = re.compile(before2[:-2]+before2[-1:] + number + after)
# print(before2[:-2]+before2[-1:])

reg_wuye = re.compile(before_wuye + before4 + before5 + center + after)

reg_not = re.compile(u'(工期延误|工期节点|工期管理'
                     u'|工期、|终止)'
                     u'|工期情况|划工期内|服务期内'
                     u'|(\d{1,2}:\d{1,2}(:\d{1,2})?)')

reg_not1 = re.compile(u'(履行日期：见|服务期限应按|签订合同前，'
                      u'|务期限：1、|签订日期|证金在合同签|服务期限截止'
                      u')')

# reg_not2 = re.compile(u'(截止|1\\.|1、)')
# reg_not2 = re.compile(u'')

reg_right_digit = re.compile(u'[\d半一二三四五六七八九十壹两叁贰肆伍陆柒捌玖拾]+')

reg_right_unit = re.compile(u'[-.年月日号天～~至到—/]')

reg_error = re.compile(u'公告|发布|中')


def re_service_time(text):
    if TEST_MODE:
        # print(chardet.detect(text))
        text = re.sub("\s*", "", text)

    text_list = []
    text_list.append(text)
    # 初始化
    all_output_list = []
    all_text_index_list = []

    for index in range(len(text_list)):
        # 初始化
        output_list = []
        input_str = text_list[index]

        # 替换混淆词
        for _reg_not in [reg_not, reg_not1]:
            match_iter = re.finditer(_reg_not, input_str)
            for match in match_iter:
                word_index = match.span()
                word = match.group()
                instead = "#" * len(word)
                print("word, instead, word_index", word, instead, word_index)
                input_str = input_str[:word_index[0]] + instead + input_str[word_index[1]:]

        if TEST_MODE:
            print("input_str", input_str)

        # 匹配
        output_list, text_index_list = re_find_all_result(reg_wuye, input_str)
        if TEST_MODE:
            print("output_str, text_index reg_wuye", output_list, text_index_list)
        output_list, text_index_list = filter_service_time(output_list, text_index_list)
        prob = 0.9

        if len(output_list) == 0:
            output_list, text_index_list = re_find_all_result(reg2, input_str)
            if TEST_MODE:
                print("output_str, text_index reg2", output_list, text_index_list)
            output_list, text_index_list = filter_service_time(output_list, text_index_list)
            prob = 0.8

        if len(output_list) == 0:
            output_list, text_index_list = re_find_all_result(reg, input_str)
            if TEST_MODE:
                print("output_str, text_index reg", output_list, text_index_list)
            output_list, text_index_list = filter_service_time(output_list, text_index_list)
            prob = 0.8

        if len(output_list) == 0:
            output_list, text_index_list = re_find_all_result(reg1, input_str)
            if TEST_MODE:
                print("output_str, text_index reg1", output_list, text_index_list)
            output_list, text_index_list = filter_service_time(output_list, text_index_list)
            prob = 0.8

        if len(output_list) == 0:
            output_list, text_index_list = re_find_all_result(reg3, input_str)
            if TEST_MODE:
                print("output_str, text_index reg3", output_list, text_index_list)
            output_list, text_index_list = filter_service_time(output_list, text_index_list)
            prob = 0.8

        if len(output_list) == 0:
            output_list, text_index_list = re_find_all_result(reg4, input_str)
            if TEST_MODE:
                print("output_str, text_index reg4", output_list, text_index_list)
            output_list, text_index_list = filter_service_time(output_list, text_index_list)
            prob = 0.5

        if len(output_list) == 0:
            output_list, text_index_list = re_find_all_result(reg5, input_str)
            if TEST_MODE:
                print("output_str, text_index reg5", output_list, text_index_list)
            output_list, text_index_list = filter_service_time(output_list, text_index_list)
            prob = 0.8

        # 添加
        all_output_list += output_list
        all_text_index_list += text_index_list

    index2word = []
    for i in range(len(all_text_index_list)):
        word = text[all_text_index_list[i][0]:all_text_index_list[i][1]]
        if i != len(all_text_index_list)-1:
            word = word + " "
        index2word.append(word)

    if TEST_MODE:
        print("index2word all_text_index_list", index2word, all_text_index_list)
    return index2word, all_text_index_list, prob


def filter_service_time(output_list, text_index_list):
    # 过滤
    delete_list = []
    for i in range(len(output_list)):
        output = output_list[i]

        # 日期影响
        if re.findall("日", output) and not re.findall(reg_right_unit, re.sub("日期", "", output)):
            delete_list.append([output, text_index_list[i]])
            print("delete output", output)
            continue
        # 不包含数字、单位的
        if not re.findall(reg_right_digit, output):
            delete_list.append([output, text_index_list[i]])
            continue
        if not re.findall(reg_right_unit, output):
            delete_list.append([output, text_index_list[i]])
            continue
        # 包含不要的字
        if re.findall(reg_error, output):
            delete_list.append([output, text_index_list[i]])
            continue
        # 类似2019年的
        if not re.findall("[-./月日天号]", output):
            if len(re.findall("年", output)) == 1:
                year_time = re.search("\d+", output)
                if year_time is not None and int(year_time.group()) >= 2000:
                    delete_list.append([output, text_index_list[i]])
    for output, text_index in delete_list:
        if output in output_list:
            output_list.remove(output)
        if text_index in text_index_list:
            text_index_list.remove(text_index)

    if TEST_MODE:
        print("delete_list", delete_list)
    return output_list, text_index_list


def re_find_all_result(reg, input, unit="", index=0):
    """

    :param reg: 正则表达式
    :param input: 待匹配句子
    :param unit: 需要加的单位
    :param index: 字符串拼接的开始位置
    :return: 正则后的字符串
    """
    # 全文下标
    text_index = []
    match1 = re.finditer(reg, input)
    output_list = []
    for i in match1:
        d = i.groupdict()

        if TEST_MODE:
            for key in d.keys():
                if d.get(key):
                    print('d.get("' + key + '")', d.get(key))

        front_len = 0
        for key in d.keys():
            if d.get(key) and key in ["before", "before2", "before4",
                                      "before5", "before6", "before7", "charac",
                                      "after4"]:
                front_len += len(d.get(key))
        # 特殊情况
        if d.get("before3"):
            front_len -= len(d.get("before7"))
            front_len -= len(d.get("charac"))

        text_index.append([i.start()+front_len, i.end()])
        output_list.append(input[i.start()+front_len: i.end()])
    return output_list, text_index


def calculateLen(ss, i):
    front_len = 0
    back_len = 0
    for index in range(i):
        front_len += len(ss[index])
    for index in range(i+1, len(ss)):
        back_len += len(ss[index])
    return front_len, back_len


def extract_servicetime(text):
    list_servicetime = []
    word_list, text_index_list, prob = re_service_time(text)
    # print(word, text_index_list)
    for i in range(len(text_index_list)):
        d = {"body": word_list[i], "begin_index": text_index_list[i][0], "end_index": text_index_list[i][1], "prob": prob}
        if len(word_list[i]) <= 35:
            list_servicetime.append(d)
    if TEST_MODE:
        print("list_servicetime", list_servicetime)
    return list_servicetime


def test_from_str():
    # s = """
    # """
    s = "5元/年 服务期：交付使用之日起三年； 承诺服务等级"
    print(extract_servicetime(s))
    print(re.findall('(\d{2,4}[-.年/]|\d{1,2}[-.月/]|\d{1,2}[日号]?)+[-～~起至到—]+\d{2,4}[-.年/]', s))


def test_from_csv():
    df = pd.read_csv("D:/BIDI_DOC/招标方式_服务期限_提取/serviceTime_text.csv")
    result_list = []
    for index, row in df.iterrows():
        result = extract_servicetime(row["text"])
        result_list.append(str(result))

    df["new_word"] = pd.DataFrame(result_list)
    df.to_csv("D:/BIDI_DOC/招标方式_服务期限_提取/serviceTime_text_new.csv")


def test_from_xlsx():
    df = pd.read_excel("D:/BIDI_DOC/比地_文档/service_time_error.xlsx")
    result_list = []
    for index, row in df.iterrows():
        text = row["dochtmlcon"]
        soup = BeautifulSoup(text, "lxml")
        text = soup.get_text(strip=True)
        result = extract_servicetime(text)
        result_list.append(str(result))

    df["new_word"] = pd.DataFrame(result_list)
    df.to_excel("D:/BIDI_DOC/比地_文档/service_time_error_new.xlsx", index=False)


# def test_from_db():
#     engine = create_engine("mysql+pymysql://root:pwdformysql0922@192.168.2.170:3306/"
#                            "exportdb?charset=utf8")
#     sql = 'SELECT docid, doctextcon, service_time_1 FROM `wuye_zhouqi_1` where service_time_1 <> "" and service_time_1 is not null;'
#     # 建立dataframe
#     df = pd.read_sql_query(sql, engine)
#     result_list = []
#     for index, row in df.iterrows():
#         result = extract_servicetime(row["doctextcon"])
#         if len(result) > 0:
#             temp = ""
#             for r in result:
#                 temp += r.get("body") + "##"
#             result_list.append(temp)
#         else:
#             result_list.append(np.nan)
#
#     df["new_service_time"] = pd.DataFrame(result_list)
#     df.to_excel("D:/BIDI_DOC/比地_文档/service_time_from_wuye_zhouqi.xlsx", index=False)


if __name__ == '__main__':
    test_from_str()