automatic_crawler/application/pipeline/pipeline.py

import re
import sys
import os
import time

sys.path.append(os.getcwd())

from application.common import MysqlHelper, AliyunLogger


class PiaoQuanPipeline:
    """
    爬虫管道——爬虫规则判断
    """

    def __init__(self, platform, mode, rule_dict, env, item, trace_id):
        self.platform = platform
        self.mode = mode
        self.item = item
        self.rule_dict = rule_dict
        self.env = env
        self.trace_id = trace_id
        self.mysql = MysqlHelper(env=env, mode=mode, platform=platform)
        self.aliyun_log = AliyunLogger(platform=platform, mode=mode, env=env)

    def publish_time_flag(self):
        """
        判断发布时间是否过期
        :return: True or False
        """
        # 判断发布时间
        publish_time_stamp = self.item["publish_time_stamp"]
        update_time_stamp = self.item["update_time_stamp"]
        max_d = self.rule_dict.get("period", {}).get("max", 1000)
        min_d = self.rule_dict.get("period", {}).get("min", 1000)
        days = max_d if max_d > min_d else min_d
        if self.platform == "gongzhonghao":
            if (
                    int(time.time()) - publish_time_stamp
                    > 3600 * 24 * days
            ) and (
                    int(time.time()) - update_time_stamp
                    > 3600 * 24 * days
            ):
                self.aliyun_log.logging(
                    code="2004",
                    trace_id=self.trace_id,
                    data=self.item,
                    message="发布时间超过{}天".format(days),
                )
                return False
        else:
            if (
                    int(time.time()) - publish_time_stamp
                    > 3600 * 24 * days
            ):
                self.aliyun_log.logging(
                    code="2004",
                    trace_id=self.trace_id,
                    data=self.item,
                    message="发布时间超过{}天".format(days),
                )
                return False
        return True

    def title_flag(self):
        """
        视频标题是否满足需求
        :return:
        """
        title = self.item["video_title"]
        cleaned_title = re.sub(r"[^\w]", " ", title)
        # 敏感词
        # 获取敏感词列表
        sensitive_words = []
        if any(word in cleaned_title for word in sensitive_words):
            self.aliyun_log.logging(
                code="2003",
                trace_id=self.trace_id,
                message="标题中包含敏感词",
                data=self.item,
            )
            return False
        return True

    def download_rule_flag(self):
        """
        视频基础下载规则
        :return:
        """
        for key in self.item:
            if self.rule_dict.get(key):
                max_value = (
                    int(self.rule_dict[key]["max"])
                    if int(self.rule_dict[key]["max"]) > 0
                    else 999999999999999
                )
                if key == "peroid":  # peroid是抓取周期天数
                    continue
                else:
                    flag = int(self.rule_dict[key]["min"]) <= int(self.item[key]) <= max_value
                    if not flag:
                        self.aliyun_log.logging(
                            code="2004",
                            trace_id=self.trace_id,
                            data=self.item,
                            message="{}: {} <= {} <= {}, {}".format(
                                key,
                                self.rule_dict[key]["min"],
                                self.item[key],
                                max_value,
                                flag,
                            ),
                        )
                        return flag
            else:
                continue
        return True

    # 按照某个具体平台来去重
    def repeat_video(self):
        """
        视频是否重复
        :return:
        """
        # sql = f""" select * from crawler_video where platform="公众号" and out_video_id="{video_id}"; """
        out_id = self.item["out_video_id"]
        sql = f""" select 1 from crawler_video where platform = "{self.platform}" and out_video_id="{out_id}"; """
        repeat_video = self.mysql.select(sql=sql)
        # print(repeat_video)
        if repeat_video:
            self.aliyun_log.logging(
                code="2002",
                trace_id=self.trace_id,
                message="重复的视频",
                data=self.item,
            )
            return False
        return True

    def process_item(self):
        """
        全规则判断，符合规则的数据则return True
        :return:
        """
        if not self.publish_time_flag():
            # 记录相关日志
            return False
        if not self.title_flag():
            # 记录相关日志
            return False
        if not self.repeat_video():
            # 记录相关日志
            return False
        if not self.download_rule_flag():
            # 记录相关日志
            return False
        return True