LongArticleTaskServer/applications/utils/common.py

"""
@author: luojunhui
"""

import random
import string
import hashlib
import math
import statistics
from scipy.stats import t
from odps import ODPS

from datetime import datetime, timezone, date, timedelta
from typing import List

from requests import RequestException
from urllib.parse import urlparse, parse_qs
from tenacity import (
    stop_after_attempt,
    wait_exponential,
    retry_if_exception_type,
)
from applications.config import name_map


def str_to_md5(strings):
    """
    字符串转化为 md5 值
    :param strings:
    :return:
    """
    # 将字符串转换为字节
    original_bytes = strings.encode("utf-8")
    # 创建一个md5 hash对象
    md5_hash = hashlib.md5()
    # 更新hash对象，传入原始字节
    md5_hash.update(original_bytes)
    # 获取16进制形式的MD5哈希值
    md5_value = md5_hash.hexdigest()
    return md5_value


def proxy():
    """
    快代理
    """
    # 隧道域名:端口号
    tunnel = "j685.kdltps.com:15818"

    # 用户名密码方式
    username = "t14070979713487"
    password = "hqwanfvy"
    proxies = {
        "http": "http://%(user)s:%(pwd)s@%(proxy)s/"
        % {"user": username, "pwd": password, "proxy": tunnel},
        "https": "http://%(user)s:%(pwd)s@%(proxy)s/"
        % {"user": username, "pwd": password, "proxy": tunnel},
    }
    return proxies


def async_proxy():
    return {
        "url": "http://j685.kdltps.com:15818",
        "username": "t14070979713487",
        "password": "hqwanfvy",
    }


def request_retry(retry_times, min_retry_delay, max_retry_delay):
    """
    :param retry_times:
    :param min_retry_delay:
    :param max_retry_delay:
    """
    common_retry = dict(
        stop=stop_after_attempt(retry_times),
        wait=wait_exponential(min=min_retry_delay, max=max_retry_delay),
        retry=retry_if_exception_type((RequestException, TimeoutError)),
        reraise=True,  # 重试耗尽后重新抛出异常
    )
    return common_retry


def yield_batch(data, batch_size):
    """
    生成批次数据
    :param data:
    :param batch_size:
    :return:
    """
    for i in range(0, len(data), batch_size):
        yield data[i : i + batch_size]


def extract_root_source_id(path: str) -> dict:
    """
    提取path参数
    :param path:
    :return:
    """
    params = parse_qs(urlparse(path).query)
    jump_page = params.get("jumpPage", [None])[0]
    if jump_page:
        params2 = parse_qs(jump_page)
        res = {
            "video_id": params2["pages/user-videos?id"][0],
            "root_source_id": params2["rootSourceId"][0],
        }
        return res
    else:
        return {}


def show_desc_to_sta(show_desc):
    def decode_show_v(show_v):
        """

        :param show_v:
        :return:
        """
        foo = show_v.replace("千", "e3").replace("万", "e4").replace("亿", "e8")
        foo = eval(foo)
        return int(foo)

    def decode_show_k(show_k):
        """

        :param show_k:
        :return:
        """
        this_dict = {
            "阅读": "show_view_count",  # 文章
            "看过": "show_view_count",  # 图文
            "观看": "show_view_count",  # 视频
            "赞": "show_like_count",
            "付费": "show_pay_count",
            "赞赏": "show_zs_count",
        }
        if show_k not in this_dict:
            print(f"error from decode_show_k, show_k not found: {show_k}")
        return this_dict.get(show_k, "show_unknown")

    show_desc = show_desc.replace("+", "")
    sta = {}
    for show_kv in show_desc.split("\u2004\u2005"):
        if not show_kv:
            continue
        show_k, show_v = show_kv.split("\u2006")
        k = decode_show_k(show_k)
        v = decode_show_v(show_v)
        sta[k] = v
    res = {
        "show_view_count": sta.get("show_view_count", 0),
        "show_like_count": sta.get("show_like_count", 0),
        "show_pay_count": sta.get("show_pay_count", 0),
        "show_zs_count": sta.get("show_zs_count", 0),
    }
    return res


def generate_gzh_id(url):
    biz = url.split("biz=")[1].split("&")[0]
    idx = url.split("&idx=")[1].split("&")[0]
    sn = url.split("&sn=")[1].split("&")[0]
    url_bit = "{}-{}-{}".format(biz, idx, sn).encode()
    md5_hash = hashlib.md5()
    md5_hash.update(url_bit)
    md5_value = md5_hash.hexdigest()
    return md5_value


def timestamp_to_str(timestamp, string_format="%Y-%m-%d %H:%M:%S") -> str:
    """
    :param string_format:
    :param timestamp:
    """
    dt_object = (
        datetime.utcfromtimestamp(timestamp).replace(tzinfo=timezone.utc).astimezone()
    )
    date_string = dt_object.strftime(string_format)
    return date_string


def days_remaining_in_month():
    # 获取当前日期
    today = date.today()

    # 获取下个月的第一天
    if today.month == 12:
        next_month = today.replace(year=today.year + 1, month=1, day=1)
    else:
        next_month = today.replace(month=today.month + 1, day=1)

    # 计算本月最后一天（下个月第一天减去1天）
    last_day_of_month = next_month - timedelta(days=1)

    # 计算剩余天数
    remaining_days = (last_day_of_month - today).days

    return remaining_days


def generate_task_trace_id():
    random_str = "".join(random.choices(string.ascii_lowercase + string.digits, k=16))
    return f"Task-{datetime.now().strftime('%Y%m%d%H%M%S')}-{random_str}"


def ci_lower(data: List[int], conf: float = 0.95) -> float:
    """
    计算data的置信区间下限
    """
    if len(data) < 2:
        raise ValueError("Sample length less than 2")

    n = len(data)
    mean = statistics.mean(data)
    std = statistics.stdev(data) / math.sqrt(n)
    # t 分位点（左侧）：ppf 返回负值
    t_left = t.ppf((1 - conf) / 2, df=n - 1)
    return mean + t_left * std


def get_task_chinese_name(data):
    """
    通过输入任务详情信息获取任务名称
    """
    task_name = data["task_name"]
    task_name_chinese = name_map.get(task_name, task_name)

    # account_method
    if task_name == "crawler_gzh_articles":
        account_method = data.get("account_method", "")
        account_method = account_method.replace(
            "account_association", "账号联想"
        ).replace("search", "")
        crawl_mode = data.get("crawl_mode", "")
        crawl_mode = crawl_mode.replace("search", "搜索").replace("account", "抓账号")
        strategy = data.get("strategy", "")
        return f"{task_name_chinese}\t{crawl_mode}\t{account_method}\t{strategy}"
    elif task_name == "article_pool_cold_start":
        platform = data.get("platform", "")
        platform = platform.replace("toutiao", "今日头条").replace("weixin", "微信")
        strategy = data.get("strategy", "")
        strategy = strategy.replace("strategy", "策略")
        category_list = data.get("category_list", [])
        category_list = "、".join(category_list)
        crawler_methods = data.get("crawler_methods", [])
        crawler_methods = "、".join(crawler_methods)
        return f"{task_name_chinese}\t{platform}\t{crawler_methods}\t{category_list}\t{strategy}"
    else:
        return task_name_chinese


def fetch_from_odps(query):
    client = ODPS(
        access_id="LTAIWYUujJAm7CbH",
        secret_access_key="RfSjdiWwED1sGFlsjXv0DlfTnZTG1P",
        endpoint="http://service.cn.maxcompute.aliyun.com/api",
        project="loghubods",
    )
    with client.execute_sql(query).open_reader() as reader:
        if reader:
            return [item for item in reader]
        else:
            return []