import re
import glob
import copy

name_list = glob.glob(‘./*.txt‘)

date_totul = []
indictee_totul = []
court_totul = []
procuratorate_totul = []
with open(‘./result.txt‘,‘a‘,encoding=‘utf-8‘) as f_r:
    for name in name_list:
        f_r.write(‘<{0}>\n\n‘.format(name.split(‘\\‘)[-1]))
        with open(name,encoding=‘utf-8‘) as f:
            lines = f.read()

            # 时间匹配
            #xxxx年xx月xx日；同年xx月xx日；xxxx年xx月x旬；xxxx年xx月底；xxxx年xx月；xx月xxx日
            # |优先匹配前面的，无符合才匹配后面的
            pattern_t = re.compile(
                ‘[0-9〇一二三四五六七八九]{4}年.{1,2}月.{1,3}日‘
                ‘|同年.{1,2}月.{1,3}日‘
                ‘|[0-9〇一二三四五六七八九]{4}年.{1,2}月.{1}旬‘
                ‘|[0-9〇一二三四五六七八九]{4}年.{1,2}月底‘
                ‘|[0-9〇一二三四五六七八九]{4}年.{1,2}月‘
                ‘|[0-9〇一二三四五六七八九十]{1,2}月.{1,3}日‘)
            date_step = [date for date in pattern_t.findall(lines)]
            # print(date_step)
            for i in date_step:
                f_r.write(‘<time>{0}</time>\n\n‘.format(i))
            date_totul.extend(date_step)

            # 被告匹配
            pattern_i = re.compile(‘被告人(.{2,4}?)[,，]|被上诉人：(.+?)。|被执行人：(.+?)。‘)
            pattern_i2 = re.compile(‘被申诉人\（.+\）：(.+)。|被申请人\（.+\）：(.+)。‘)
            defendant = list(set([item[0] for item in re.findall(pattern_i,lines) + pattern_i2.findall(lines) if ‘死刑‘ not in item[0]]))

            if defendant != [] and defendant != [‘‘]:
                print(defendant)

                # indictee_totul.append(indictee_step)
                for item in defendant:
                    f_r.write(‘<defendant>{0}</defendant>\n\n‘.format(item))

            # 法院匹配
            pattern_c = re.compile(‘[\n。,，《；](.{,15}?人民法院)‘)
            _court_list =  [name.group(1) for name in pattern_c.finditer(lines)]
            _court_step = []
            # print(_court_list)
            for _court in _court_list:
                _court_step.append(_court.split(‘。‘)[-1]
                                   .split(‘，‘)[-1]
                                   .split(‘《‘)[-1]
                                   .split(‘、‘)[-1]
                                   .split(‘；‘)[-1])
                _court_step = list(set(_court_step))
                for court_name in _court_step:
                    if ‘由‘ in court_name or                                    ‘向‘ in court_name or                                     ‘受‘ in court_name or                                     ‘和‘ in court_name:
                        # print(name)
                        _court_step.append(court_name.split(‘由‘)[-1].
                                           split(‘向‘)[-1].
                                           split(‘受‘)[-1].
                                           split(‘和‘)[-1])
                        _court_step.remove(court_name)
                        # print(_court_step)
                    if ‘不服‘ in court_name or                                     ‘后被‘ in court_name or                                     ‘报请‘ in court_name or                                     ‘书证‘ in court_name or                                     ‘核准‘ in court_name or                                     ‘指令‘ in court_name or                                     ‘维持‘ in court_name or                                     ‘撤销‘ in court_name or                             ‘参照‘ in court_name:
                        _court_step.append(copy.deepcopy(court_name[2::]))
                        _court_step.remove(court_name)

                    _court_step = list(set(_court_step))
                    if ‘人民法院‘ in _court_step:
                        _court_step.remove(‘人民法院‘)

            for i in _court_step:
                f_r.write(‘<court>{0}</court>\n\n‘.format(i))
            # print(name,_court_step)

            # 检察院匹配
            pattern_p = re.compile(‘审理(.+)指控‘)
            procuratorate_step = list(set([name.group(1) for name in pattern_p.finditer(lines)]))
            procuratorate_totul.extend(procuratorate_step)
            # print(name,procuratorate_step)
            for i in procuratorate_step:
                f_r.write(‘<procuratorate>{0}</procuratorate>\n\n‘.format(i))

            # 地点匹配
            pattern_pl_1 = re.compile(‘(.{2}省.+?县)‘)
            pattern_pl_2 = re.compile(‘.{2}省.{2}市‘)
            pattern_pl_3 = re.compile(‘.{2}省.+?自治州‘)
            pattern_pl_4 = re.compile(‘.{2}省.+?乡‘)
            pattern_pl_5 = re.compile(‘.{2}市.{2}区‘)
            pattern_pl_6 = re.compile(‘.{2}市.{2}镇‘)
            pattern_pl_7 = re.compile(‘.{2}市.+?开发区‘)
            place_step = list(set([name.group(0) for name in pattern_pl_1.finditer(lines)]))
            place_step.extend(list(set([name.group(0) for name in pattern_pl_2.finditer(lines)])))
            place_step.extend(list(set([name.group(0) for name in pattern_pl_3.finditer(lines)])))
            place_step.extend(list(set([name.group(0) for name in pattern_pl_4.finditer(lines)])))
            place_step.extend(list(set([name.group(0) for name in pattern_pl_5.finditer(lines)])))
            place_step.extend(list(set([name.group(0) for name in pattern_pl_6.finditer(lines)])))
            place_step.extend(list(set([name.group(0) for name in pattern_pl_7.finditer(lines)])))
            place_step_n = []
            for place_name in place_step:
                if len(place_name)<=15:
                    # print(name,place_name)
                    place_step_n.append(place_name)
                if ‘×ב in place_name:
                    if place_name in place_step_n:
                        place_step_n.append(place_name.split(‘×ב)[0])
                        place_step_n.remove(place_name)
                if ‘XX‘ in place_name:
                    if place_name in place_step_n:
                        place_step_n.append(place_name.split(‘XX‘)[0])
                        place_step_n.remove(place_name)
        #     print(name, [(len(item)) for item in place_step_n])
        # print(name,place_step,len(place_step))
        # print(name,place_step_n,len(place_step_n))
        for i in place_step_n:
            f_r.write(‘<location>{0}</location>\n\n‘.format(i))
        f_r.write(‘</{0}>\n\n‘.format(name.split(‘\\‘)[-1]))