# -*- coding: utf-8 -*-
import jieba
import jieba.posseg as pseg
import jieba.analyse
import codecs
import math
import functools
from gensim import corpora
from gensim.models import LdaModel
from gensim import models
from gensim.corpora import Dictionary
#测试使用
#2021.03.14
string = u'图书评论是近代报刊业兴起后，在世界各国得到长足发展的一种新型评论体裁。而不论是书评理论还是书评实践都有一个不小的疏漏，即忽视了图书的形式因素。'


def cut_word(string):  # 精确模式-结巴分词
    seg_list = jieba.cut(string, cut_all=False)
    line = " ".join(seg_list)  # 精确模式
    return line


def cut_stop_word(string):  # 去停用词版本
    stopwords = codecs.open('stopwords.txt', 'r', encoding='utf8').readlines()
    stopwords = [w.strip() for w in stopwords]
    sentence_depart = jieba.cut(string.strip(), cut_all=False)
    line = []
    for word in sentence_depart:
        if word not in stopwords:
            if word != '\t':
                if word != ' ':
                    if word != '\n':
                        line.append(word)
    return line


def word_mark(string):  # 词性标注
    sentence = pseg.cut(string)
    line = ""
    for x in sentence:
        line = line + x.word + x.flag + "/"
    return line


def key_word_extract(string):  # 基于tf-idf提取6个关键词
    tags = jieba.analyse.extract_tags(string, topK=6, withWeight=False, allowPOS=())
    line = " ".join(tags)
    return line

def predictkeyword(file_path):
    f = open(file_path, 'r', encoding='utf8', errors='ignore')
    texts = f.read().replace(' ', '')
    texts = texts.replace('\n', '')
    return key_word_extract(texts)


def Recall_lcs_Gram(str1, str2):
    lstr1 = len(str1)
    lstr2 = len(str2)
    record = [[0 for i in range(lstr2 + 1)] for j in range(lstr1 + 1)]  # 多一位
    maxNum = 0  # 最长匹配长度
    p = 0  # 匹配的起始位
    for i in range(lstr1):
        for j in range(lstr2):
            if str1[i] == str2[j]:
                # 相同则累加
                record[i + 1][j + 1] = record[i][j] + 1
                if record[i + 1][j + 1] > maxNum:
                    # 获取最大匹配长度
                    maxNum = record[i + 1][j + 1]
                    # 记录最大匹配长度的终止位置
                    p = i + 1
    Recall_lcs=maxNum/lstr2
    return Recall_lcs

def Precision_lcs_Gram(str1, str2):
    lstr1 = len(str1)
    lstr2 = len(str2)
    record = [[0 for i in range(lstr2 + 1)] for j in range(lstr1 + 1)]  # 多一位
    maxNum = 0  # 最长匹配长度
    p = 0  # 匹配的起始位
    for i in range(lstr1):
        for j in range(lstr2):
            if str1[i] == str2[j]:
                # 相同则累加
                record[i + 1][j + 1] = record[i][j] + 1
                if record[i + 1][j + 1] > maxNum:
                    # 获取最大匹配长度
                    maxNum = record[i + 1][j + 1]
                    # 记录最大匹配长度的终止位置
                    p = i + 1
    Precision_lcs=maxNum/lstr1
    return Precision_lcs

def acc(file_path,keyword_path):
    f = open(file_path, 'r', encoding='utf8', errors='ignore')
    k = open(keyword_path, 'r', encoding='utf8', errors='ignore')
    texts = f.read()
    key = k.read()
    keys = key.split(' ')
    prekey = key_word_extract(texts).split(' ')
    len_keys = len(keys)
    len_prekey = len(prekey)
    acckey = Precision_lcs_Gram(key.replace(' ', ''),key_word_extract(texts).replace(' ', ''))
    return acckey

def rec(file_path,keyword_path):
    f = open(file_path, 'r', encoding='utf8', errors='ignore')
    k = open(keyword_path, 'r', encoding='utf8', errors='ignore')
    texts = f.read()
    key = k.read()
    keys = key.split(' ')
    prekey = key_word_extract(texts).split(' ')
    len_keys = len(keys)
    len_prekey = len(prekey)
    reckey = Recall_lcs_Gram(key.replace(' ', ''),key_word_extract(texts).replace(' ', ''))
    return reckey