ai_qback/gpt/get_article2.py

# -*- coding: utf-8 -*-

from gpt.chatgpt import get_answer_from_gpt
from gpt.gpt_check import CheckGptAnswer, CheckArticleResult
from tools.new_mysql import MySQLUploader
from tools.loglog import logger, log_err_e
from tools.thread_pool_manager import pool_executor
from common.common_data import all_exchange_words
from common.split_text import split_text_to_word

from pydantic import BaseModel
from cachetools import TTLCache
from concurrent.futures import wait
from random import randint, shuffle
import json
import requests
from openpyxl import load_workbook
from tenacity import retry, stop_after_attempt, wait_fixed


def get_article_difficulty(article) -> int:
    """获取文章的难度值"""
    url = "http://qbank.yunzhixue.cn/api/article/analysis"

    data = {"body": article, "question": ""}
    try:
        response = requests.post(url, json=data)
    except Exception as e:
        log_err_e(e, msg="获取文章难度值；")
        return 0

    if response.status_code == 200:
        difficult_value = response.json()['data']['difficult']
        return difficult_value
    else:
        logger.error(f"错误状态码{response.status_code}")


def find_interval(number) -> int:
    """
    判断一个数字属于哪个难度等级区间。31级是例外情况，需要排查

    :param number: 要检查的数字。
    :return: 返回包含该数字的区间，如果没有找到，则返回 None。
    """
    intervals = [(1, 200), (201, 250), (251, 300), (301, 350), (351, 400), (401, 450), (451, 550), (551, 650), (651, 750), (751, 850), (851, 950),
                 (951, 1100),
                 (1101, 1250), (1251, 1400), (1401, 1550), (1551, 1700), (1701, 1900), (1901, 2100), (2101, 2300), (2301, 2600), (2601, 2900),
                 (2901, 3200),
                 (3201, 3500), (3501, 3900), (3901, 4300), (4301, 4700), (4701, 5100), (5101, 5500), (5501, 5900), (5901, 6500), (6501, 99999)]
    for index, (start, end) in enumerate(intervals, start=1):
        if start <= number <= end:
            return index
    logger.error(f"文章难度判断不对:{number}")
    return 0


def merge_and_split(list1, list2):
    combined = list1 + list2
    import random
    random.shuffle(combined)

   
    two_thirds = []
    one_third = []

   
    total_length = len(combined)
    if total_length > 15:
        two_thirds = combined[:15]
        one_third = combined[15:]
    else:
        two_thirds = combined
        one_third = []

    return two_thirds, one_third


class GetArticle:
    def __init__(self):
        self.m = MySQLUploader()

        self.callback_url_dict = {}
        self.real_ip_dict = {} 
        self.demo_name = {}

        self.article_result = {} 

       
        self.punctuation = [",", ".", "!", "?", ":", ";", '"', "–", "_", "-", "...", "......"]
        all_exchange_words.update(self.punctuation)

       
        self.exchange_data: dict[str, list] = {} 
        self.read_spring_bamboo_exchange_table()

   
    def read_spring_bamboo_exchange_table(self):
        """变形是键，原型是值"""
        wb = load_workbook(r"data/春笋单词对照变形.xlsx", read_only=True, data_only=True)
        ws = wb.active
        for row in ws.values:
            prototype = row[0] 
            exchange = row[1] 
            if prototype not in self.exchange_data:
                self.exchange_data[prototype] = [exchange]
            else:
                self.exchange_data[prototype].append(exchange)
        wb.close()

   
    def parser_insert_to_mysql(self, resp_result):
        for single_article in resp_result['articles']:
           
            article = single_article['body']
            article_json = json.dumps(single_article)
            difficult_value = find_interval(get_article_difficulty(article)) 
            if not difficult_value:
                logger.error("文章难度等级为0；")
            sql = "INSERT INTO spring_bamboo_article (article_json,difficult_level) VALUES (%s,%s)"
            self.m.execute_(sql, (article_json, difficult_value))

   
    def submit_task(self, core_words: list, extend_words: list, take_count: int, student_stage: int, real_ip: str,
                    demo_name: str, article_difficulty: int):
        """
        words_meaning_list: 词义id 包含词义ID的数组集合，用于生成文章。- 示例：[110, 111, 112, 113, 114]
        take_count: 取文章数量 （int类型，正常是2篇，最大8篇）
        student_stage: 学段（int类型：1.小学;2.初中;3.高中;）
        demo_name: 项目名称
        article_difficulty:文章难度值1-4200模糊范围
        """
        task_id = randint(10000000, 99999999)
       
        logger.info(f"生成文章id。task_id:{task_id}")

        self.real_ip_dict[task_id] = real_ip
        self.demo_name[task_id] = demo_name

        try:
            resp_result = self.run_task(core_words, extend_words, task_id, take_count, student_stage, article_difficulty)
            self.parser_insert_to_mysql(resp_result) 
            return resp_result
        except Exception as e:
            err_msg = f"GetArticle提交任务失败{type(e).__name__},{e}"
            log_err_e(e, msg="GetArticle提交任务失败;")

            return err_msg

   
    @retry(stop=stop_after_attempt(2), wait=wait_fixed(3), reraise=True)
    def get_article(self, core_words: list, extend_words: list, student_stage: int, task_id: int, take_count: int, article_difficulty) -> dict:
        article_grade = find_interval(article_difficulty)
        if 0 < article_grade <= 10:
            article_word_count = 50 + 10 * article_grade 
        elif 10 < article_grade <= 20:
            article_word_count = 150 + 30 * (article_grade - 10) 
        else:
            article_word_count = 450 + 20 * (article_grade - 20) 

       
        diffculty_control = {
            1: {"grade": "小学", "article_word_count": article_word_count, "desc_difficulty": "最简单最容易没有难度", "paragraph_count": "1-2",
                "desc2": "文章整体非常简洁，通俗易懂，适合初学者，刚入门，单词全是最常见的，语句通顺即可。",
                "choice_desc": "选择题难度尽可能简单，但是不要让所有选择题让其直接在文中找到答案，允许1-2个选择题很简单，参考中国小学生水平"},
            2: {"grade": "初中", "article_word_count": article_word_count, "desc_difficulty": "简单、常见、难度低", "paragraph_count": "2-3",
                "desc2": "文章整体难度适中，大约和中国初中生，中国CET-3，雅思4分这样的难度标准。",
                "choice_desc": "选择题难度适中，但是不要所有选择题让其直接在文中找到答案，参考中国初中生水平，中考标准。"},
            3: {"grade": "初中", "article_word_count": article_word_count, "desc_difficulty": "简单、常见、难度低", "paragraph_count": "2-3",
                "desc2": "文章整体难度适中，大约和中国初中生，中国CET-3，雅思4分这样的难度标准。",
                "choice_desc": "选择题难度适中，但是不要所有选择题让其直接在文中找到答案，参考中国初中生水平，中考标准。"},
            4: {"grade": "高中", "article_word_count": article_word_count, "desc_difficulty": "常见、高中难度的", "paragraph_count": "3-5",
                "desc2": "文章整体难度适中，大约和中国的高中生，中国CET-6，雅思6分这样的难度标准。",
                "choice_desc": "选择题难度偏难，要有迷惑性混淆性，答案不要出现直接在文中，4个选项要学生推理或逻辑判断，参考中国高中生水平，高考标准。"}
        }
        take_count_dict = {0: "", 1: "一", 2: "二", 3: "三", 4: "四", 5: "五", 6: "六", 7: "七", 8: "八", 9: "九"}
        different_cou = take_count_dict.get(take_count, "")

        grade = diffculty_control[student_stage]["grade"] 
        select_word_count = diffculty_control[student_stage]["article_word_count"] 
        select_diffculty = diffculty_control[student_stage]["desc_difficulty"] 
        select_paragraph_count = diffculty_control[student_stage]["paragraph_count"] 
        desc2 = diffculty_control[student_stage]["desc2"]
        choice_desc = diffculty_control[student_stage]["choice_desc"] 

       
        shuffle(core_words)
        core_words_meaning_str = ";".join([str(i['meaning_id']) + ' ' + i["spell"] + ":" + i["meaning"] for i in core_words])
        extend_words_meaning_str = ";".join([str(i['meaning_id']) + ' ' + i["spell"] + ":" + i["meaning"] for i in extend_words])

        no_escape_code = r"\\n\\n"
        json_model = r'{"difficultSentences":[{"english":"string","chinese":"string"}],"usedMeanIds":[0,0,0],"englishArticle":"string","chineseArticle":"string","questions":[{"trunk":"string","analysis":"string","candidates":[{"label":"string","text":"string","isRight":0}]}]}'
        sys_prompt = "你是一个专业的英语老师，擅长根据用户提供的词汇生成对应的英语文章和中文翻译和4个配套选择题。"
        q = f"""下面我会为你提供两组数据，[单词组1]和[单词组2]（里面包含词义id，英语单词，中文词义），优先使用[单词组1]内的单词，请根据这些单词的中文词义，\
生成一篇带中文翻译的考场英语文章，英语文章和中文翻译要有[标题]。注意这个单词有多个词义时，生成的英语文章一定要用提供的中文词义。并挑选一句复杂的句子和其中文翻译，放入difficultSentences。\
英语文章，放入"englishArticle"中。中文翻译，放入"chineseArticle"中。最终文中使用到的单词id放入"usedMeanIds"中。\
4个选择题，放入questions字段。questions结构下有4个选择题对象，其中trunk是[英语]问题文本，analysis是[中文]的问题分析，candidates是4个ABCD选项，内部有label是指选项序号A B C D ，text是[英语]选项文本，isRight是否正确答案1是正确0是错误。

要求:
1.必须用提供的这个词义的单词，其他单词使用{select_diffculty}的单词。{desc2}{choice_desc}
2.优先保证文章语句通顺，意思不要太生硬。不要为了使用特定的单词，造成文章语义前后不搭，允许不使用个别词义。
3.文章中使用提供单词，一定要和提供单词的中文词义匹配，尤其是一词多义时，务必使用提供单词的词义。必须要用提供单词的词义。如果用到的词义与提供单词词义不一致，请不要使用这个单词。
4.生成的文章要求{select_word_count}词左右，可以用{no_escape_code}字符分段，一般{select_paragraph_count}个段落左右。第一段是文章标题。
5.生成文章优先使用[单词组1]的词义，其次可以挑选使用[单词组2]的词义。允许不使用[单词组1]的个别单词，优先保证文章整体意思通顺连贯和故事完整。
6.回复紧凑无空格的json数据，示例：{json_model}

提供[单词组1]：{core_words_meaning_str};
提供[单词组2]：{extend_words_meaning_str};
"""
        try:
            real_ip = self.real_ip_dict[task_id]
            demo_name = self.demo_name[task_id]
            r_json = json.loads(get_answer_from_gpt(q, temperature=1, json_resp=True, real_ip=real_ip, demo_name=demo_name, model='gpt-4.1',
                                                    check_fucn=CheckArticleResult.get_article_1, max_tokens=8000, sys_prompt=sys_prompt))

           
            allWordAmount = 0
            allWordAmount += len(split_text_to_word(r_json["englishArticle"]))
            for i in r_json["questions"]:
                count_trunk = len(split_text_to_word(i["trunk"]))
                count_candidates = sum([len(split_text_to_word(ii["text"])) for ii in i["candidates"]])
                allWordAmount += count_trunk
                allWordAmount += count_candidates

           
            usedMeanIds: list = r_json['usedMeanIds'] 
           
            article_words = split_text_to_word(r_json['englishArticle'])
           
            for i in core_words + extend_words:
                meaning_id = i.get('meaning_id', 0)
                if not meaning_id:
                    continue
                word = i["spell"]
                if meaning_id not in usedMeanIds and word in self.exchange_data: 
                    words_exchanges_list = self.exchange_data[word]
                    for exchange_word in words_exchanges_list:
                        if exchange_word in article_words:
                            usedMeanIds.append(meaning_id)
                            break

           
            r_json["body"] = r_json.pop("englishArticle")
            r_json["chinese"] = r_json.pop("chineseArticle")

            return {**r_json, "allWordAmount": allWordAmount}
        except json.decoder.JSONDecodeError:
            logger.error("gpt生成文章回复json格式化错误")
            raise
        except Exception as e:
            logger.error(f"gpt生成文章回复其他错误.{type(e).__name__} {e}")
            raise

   
    def run_get_article_task(self, core_words, extend_words, task_id, take_count, student_stage, article_difficulty) -> dict:
        """
        :param core_words: 核心单词数据，优先级1;可能为空
        :param extend_words: 扩展单词数据，优先级2;可能为空
        :param task_id: 任务id
        :param take_count: 文章数量
        :param student_stage: 学段标识，整型,123
        :param article_difficulty:文章难度1-4200模糊范围
        :return:
        """
        futures = []
        for i in range(take_count):
            futures.append(pool_executor.submit(self.get_article, core_words, extend_words, student_stage, task_id, take_count, article_difficulty))
        wait(futures)
        return_json = {"articles": []}
        for t in futures:
            return_json["articles"].append(t.result())
        return return_json

   
    def run_task(self, core_words, extend_words, task_id, take_count, student_stage, article_difficulty):
        try:
            outside_json = self.run_get_article_task(core_words, extend_words, task_id, take_count, student_stage, article_difficulty)
            logger.success(f"文章2任务完成。taskid：{task_id}\n{outside_json}")
            return outside_json
        except Exception as e:
            logger.error(f"{type(e).__name__} {e}")
        finally:
            self.real_ip_dict.pop(task_id)
            self.demo_name.pop(task_id)