BIDI_EXTRACTION_AI/BiddingKG/dl/assembly/joint.py

# -*- coding: utf-8 -*-
"""联合体与多中标人金额装配（Phase 6 迁移）。

本模块由 ``interface/getAttributes.py`` 第 4804-5006 行迁移而来，负责：
    - ``get_win_joint``：从正文与实体列表中识别联合体信息并写入 prem
    - ``get_multi_winner_and_money``：装配多中标人（multi_winner）并收集正文/附件金额

迁移说明：原文件使用 ``from BiddingKG.dl.common.Utils import *``，本模块改为显式导入。
经核查，两个函数仅依赖标准库 ``re``，不依赖 common.Utils 中的任何名称。
"""
from __future__ import absolute_import

import re

__all__ = ["get_win_joint", "get_multi_winner_and_money"]


def get_win_joint(prem, list_entitys, list_sentences, list_articles):
    '''
    获取联合体信息， 添加到prem
    :param prem:
    :param list_entitys:
    :param list_sentences:
    :param list_articles:
    :return:
    '''
    try:
        if 'win_tenderer' in str(prem[0]['prem']) and re.search('联合(体|方|投标人)：|联合体?(成员|单位)[12345一二三四五]?：|(联合体?)?成员单位[12345一二三四五]?：|特殊普通合伙：|（联合(体|投标人)）|（联合体(成员|单位)方?[12345一二三四五]?）|（(联合体)?成员单位[12345一二三四五]?）|（特殊普通合伙|成员?）|[，；]成：|（成[），]|[与和][^，。]{6,100}联合体', list_articles[0].content):
            sentences = sorted(list_sentences[0], key=lambda x:x.sentence_index)
            for v in prem[0]['prem'].values():
                for d in v['roleList']:
                    if d.get('role_name', '') == 'win_tenderer':
                        winner = d.get('role_text')
                        join_l = [winner]
                        for list_entity in list_entitys:
                            for i in range(len(list_entity)-1):
                                _entity = list_entity[i]
                                b = _entity.wordOffset_begin
                                e = _entity.wordOffset_end
                                if _entity.entity_type in ['org', 'company'] and _entity.label==2 and _entity.values[2] > 0.7 and _entity.entity_text==winner: # 补充概率，修复多中标错误 746420740 ，推荐理由：中国移动通信集团辽宁有限公司大连分公司、大连天途有线电视网络股份有限公司、中国联合网络通信有限公司大连市分公司，参与了本项目投标，
                                    s = sentences[_entity.sentence_index].sentence_text
                                    find_joint = 0 # 是否包含联合体
                                    for j in range(i+1, len(list_entity)):
                                        behind_entity = list_entity[j]
                                        b2 = behind_entity.wordOffset_begin
                                        e2 = behind_entity.wordOffset_end
                                        if _entity.sentence_index == behind_entity.sentence_index and behind_entity.entity_type in ['org', 'company'] \
                                                and b2-e<13 and re.search('联合(体|方|投标人)：|联合体?(成员|单位)[12345一二三四五]?：|(联合体?)?成员单位[12345一二三四五]?：|特殊普通合伙：|[，；]成：|（成）$', s[e:b2]) or \
                                            re.search('（联合(体|方|投标人)）|（联合体?(成员|单位)方?[12345一二三四五]?）|（(联合体?)?成员单位[12345一二三四五]?）|（特殊普通合伙|成员?）|^（成[），]$', s[e2:e2+10]) and behind_entity.label in [2, 5]:
                                            join_l.append(behind_entity.entity_text)
                                            b = b2
                                            e = e2
                                            find_joint = 1
                                        elif (find_joint or re.search('[与和][^，。]{6,100}联合体', list_articles[0].content)) and behind_entity.entity_type in ['org', 'company'] and s[e:b2] in ['与', '和','；','、','&','，','/','//'] and (len(s)==e2 or s[e2] in ['；','、','&','，','/','//', '。', '）', '】'] or s[e2:e2+3]=='联合体'):
                                            join_l.append(behind_entity.entity_text)
                                            b = b2
                                            e = e2
                                        elif e == e2: # 修复重复实体导致中断情况
                                            continue
                                        else:
                                            break
                                    if len(join_l)>1:
                                        d['win_tenderer_joint'] = ','.join(set(join_l))



                                        # behind_entity = list_entity[i + 1]
                                # if _entity.sentence_index== behind_entity.sentence_index and _entity.entity_type in ['org', 'company'] and _entity.label==2\
                                #         and _entity.entity_text==winner and behind_entity.entity_type in ['org', 'company'] and behind_entity.label==5:
                                #     s = sentences[_entity.sentence_index].sentence_text
                                #     b = _entity.wordOffset_begin
                                #     e = _entity.wordOffset_end
                                #     b2 = behind_entity.wordOffset_begin
                                #     e2 = behind_entity.wordOffset_end
                                    # if re.search('（联合体）', s[e2:e2+6]) and b2-e<3:
                                    #     print('联合体：', s[max(0, b-10):e2+10])
                                    #     d['win_tenderer_joint'] = '%s，%s'%(_entity.entity_text, behind_entity.entity_text)
                                    #     break
                                    # elif re.search('（联合体((牵头|主办)(人|方|单位)|主体)|牵头(人|方|单位)）|(联合体)?成员：|特殊普通合伙：', s[e:b2]) and b2-e<10:
                                    #     d['win_tenderer_joint'] = '%s，%s' % (_entity.entity_text, behind_entity.entity_text)
                                    #     print('联合体：', s[max(0, b - 10):e2 + 10])
                                    #     break
    except Exception as e:
        print('获取联合体抛出异常', e)

def get_multi_winner_and_money(channel_dic, prem, list_entitys,list_sentences, all_winner=False):
    '''
    获取多中标人及正文、附件所有金额，多中标人multi_winner写入prem，返回金额列表
    :param channel_dic:
    :param prem:
    :param list_entitys:
    :param list_sentences:
    :return:
    '''

    def add_multi_winner(pack_l, winner_l):
        if len(prem[0]['prem']) > 1 and len(set([it[0] for it in pack_l])) > 1:  # 多标段多中标人处理
            pk_dic = {}
            for ent in winner_l:
                for i in range(len(pack_l)):
                    pk, s1, b1, _ = pack_l[i]
                    if ent[1] < s1 or ent[1] == s1 and ent[2] < b1:
                        break
                    elif (ent[1] > s1 or ent[1] == s1 and ent[2] > b1):
                        if i < len(pack_l) - 1:
                            pk2, s2, b2, _ = pack_l[i + 1]
                            if (ent[1] < s2 or ent[1] == s2 and ent[2] < b2):
                                if pk not in pk_dic:
                                    pk_dic[pk] = set()
                                pk_dic[pk].add(ent[0])
                            else:
                                continue
                        else:
                            if pk not in pk_dic:
                                pk_dic[pk] = set()
                            pk_dic[pk].add(ent[0])
                    else:
                        continue
            for pk, multi_winner in pk_dic.items():
                multi_winner = multi_winner - tenderee_or_agency
                if len(multi_winner) < 2:
                    continue
                for k, v in prem[0]['prem'].items():
                    if pk == k:
                        for d in v['roleList']:
                            if d.get('role_name', '') == 'win_tenderer':
                                if d.get('role_text', '') in multi_winner and 'multi_winner' not in d:
                                    d['multi_winner'] = ','.join(set(multi_winner))
        elif 0 < len(prem[0]['prem']) < 3: # 修复 单包多中标人 例：285780273
            multi_winner = set([it[0] for it in winner_l]) - tenderee_or_agency
            if len(multi_winner) > 1:
                for v in prem[0]['prem'].values():
                    for d in v['roleList']:
                        if d.get('role_name', '') == 'win_tenderer':
                            if d.get('role_text', '') in multi_winner and 'multi_winner' not in d:
                                d['multi_winner'] = ','.join(set(multi_winner))
                            break

    moneys = []
    moneys_attachment = []
    if channel_dic['docchannel']['life_docchannel'] in ['中标信息','候选人公示','合同公告'] and 'win_tenderer' in str(prem):
        sentences = sorted(list_sentences[0], key=lambda x: x.sentence_index)
        finalists = [] # 入围供应商
        multi_winner_l = [] # 保存中标人名称列表
        tenderee_or_agency = set()
        package_l = []
        i = 0
        while i < len(list_entitys[0])-1:
            ent = list_entitys[0][i]
            b_idx_fr = ent.wordOffset_begin
            e_idx_fr = ent.wordOffset_end
            i += 1
            if ent.entity_type in ['money']:
                money = float(ent.entity_text)
                if ent.in_attachment:
                    moneys_attachment.append(money)
                else:
                    moneys.append(money)
            elif ent.entity_type in ['package']:
                package_l.append((ent.entity_text, ent.sentence_index, ent.wordOffset_begin, ent.in_attachment))
            elif ent.entity_type in ['org', 'company']:
                sentence_text = sentences[ent.sentence_index].sentence_text
                pre_text = sentence_text[max(0, b_idx_fr - 10):b_idx_fr]
                if ent.label in [0,1] and ent.values[ent.label] > 0.8:
                    tenderee_or_agency.add(ent.entity_text)
                elif ent.label == 2 and (ent.values[ent.label] > 0.8 or all_winner):
                    multi_winner_l.append((ent.entity_text, ent.sentence_index, ent.wordOffset_begin, ent.in_attachment))
                    for j in range(i, len(list_entitys[0])):
                        ent_bh = list_entitys[0][j]
                        b_idx_bh = ent_bh.wordOffset_begin
                        e_idx_bh = ent_bh.wordOffset_end
                        if ent_bh.entity_type in ['org','company'] and ent_bh.label in [2,5] and ent_bh.sentence_index == ent.sentence_index and b_idx_bh - e_idx_fr in [1, 2]:
                            if sentence_text[e_idx_fr:b_idx_bh] in ['；', '、', '&', '，', '/', '//'] and (
                                    len(sentence_text) == e_idx_bh or sentence_text[e_idx_bh] in ['；', '、', '&', '，','/', '//','。']):  # 修复多中标人刚好在文末index超出报错，例子 407126558
                                multi_winner_l.append((ent_bh.entity_text, ent_bh.sentence_index, ent_bh.wordOffset_begin, ent_bh.in_attachment))
                                e_idx_fr = e_idx_bh
                                i = j + 1
                            else:
                                break
                        elif ent_bh.entity_type in ['org', 'company'] and ent_bh.label == 5 and ent_bh.sentence_index == ent.sentence_index and b_idx_bh == e_idx_fr: # 两实体间没符号分割情况
                            multi_winner_l.append((ent_bh.entity_text, ent_bh.sentence_index, ent_bh.wordOffset_begin, ent_bh.in_attachment))
                            e_idx_fr = e_idx_bh
                            i = j + 1
                        elif ent_bh.entity_type in ['org', 'company'] and ent_bh.label == 5 and e_idx_fr == e_idx_bh: # 处理 514603520 中国邮政储蓄银行股份有限公司淄博市临淄区支行 实体由于字典匹配重复两次情况
                            i = j + 1
                        else:
                            break
                    if re.search('入围', pre_text) and re.search('未入围', pre_text)==None:
                        finalists.append((ent.entity_text, ent.sentence_index, ent.wordOffset_begin, ent.in_attachment))
                elif all_winner==1 and ent.label in [3,4,5] and re.search('第[一二三四五六七八九十0-9]+名|候选(人|单位)|入围(单位|供应商)|投标银行', pre_text) and re.search('未', pre_text)==None:
                    multi_winner_l.append((ent.entity_text, ent.sentence_index, ent.wordOffset_begin, ent.in_attachment))

        if len(multi_winner_l)>=2:
            winner_main = [it for it in multi_winner_l if not it[3]]
            winner_attn = [it for it in multi_winner_l if it[3]]
            pack_main = [it for it in package_l if not it[3]]
            pack_attn = [it for it in package_l if it[3]]
            if len(set([it[0] for it in winner_main]))>=2: # 有两个及以上多中标人及多标段 例：441612746
                add_multi_winner(pack_main, winner_main)
            elif len(set([it[0] for it in winner_attn]))>=2:
                add_multi_winner(pack_attn, winner_attn)
        if len(finalists)>=2: # 多入围候选人
            winner_main = [it for it in finalists if not it[3]]
            winner_attn = [it for it in finalists if it[3]]
            pack_main = [it for it in package_l if not it[3]]
            pack_attn = [it for it in package_l if it[3]]
            if len(set([it[0] for it in winner_main]))>=2: # 有两个及以上多中标人及多标段 例：276326152
                add_multi_winner(pack_main, winner_main)
            elif len(set([it[0] for it in winner_attn]))>=2:
                add_multi_winner(pack_attn, winner_attn)
    else:
        for i in range(len(list_entitys[0])):
            ent = list_entitys[0][i]
            if ent.entity_type in ['money']:
                money = float(ent.entity_text)
                if ent.in_attachment:
                    moneys_attachment.append(money)
                else:
                    moneys.append(money)
    return {'moneys': list(set(moneys)), 'moneys_attachment': list(set(moneys_attachment))}