python为火车头写插件

火车头的官方现在已支持python写插件，最开始按照官方文档安装了一个python3.8.8,调用插件总是报错，后面咨询客服说是版本太高，后面删除后python重新安装了一个python3.6,重新测试发现完美解决

贴一个写好的插件

# -*- coding: utf-8 -*-
# @Author: kaka
# @Date:   2022-03-03 09:45:11
# @Last Modified by:   kaka
# @Last Modified time: 2022-03-17 11:08:08
# @Email: zhckaka@sina.com
import os
import sys
import re
import datetime
# from w3lib import html
import html
from scrapy.selector import Selector
import importlib
from urllib.parse import urljoin
from urllib import parse
import json
import requests
import emoji
import re

if len(sys.argv) != 5:
    print(len(sys.argv))
    print("命令行参数长度不为5")
    sys.exit()
else:
    LabelCookie = parse.unquote(sys.argv[1])
    LabelUrl = parse.unquote(sys.argv[2])
    # PageType为List,Content,Pages分别代表列表页，内容页，多页http请求处理，Save代表内容处理
    PageType = sys.argv[3]
    SerializerStr = parse.unquote(sys.argv[4])
    if (SerializerStr[0:2] != '''{"'''):
        file_object = open(SerializerStr)
        try:
            SerializerStr = file_object.read()
            SerializerStr = parse.unquote(SerializerStr)
        finally:
            file_object.close()
    LabelArray = json.loads(SerializerStr)
    # 以下是用户编写代码区域
    if(PageType == "Save"):
        if(LabelArray['content_comments']):
            # 提取comment-copy中的内容
            comments = LabelArray['content_comments']
            text_str = r'''{0}'''.format(comments)
            json_data = json.loads(text_str)
            model_list = json_data["features"]["comments"]["models"]
            if model_list and len(model_list) > 0:
                level_one = {}
                # 先把一级的数据保存起来
                for ml_key, ml_val in model_list.items():
                    user_id = ml_key
                    parentId = ml_val.get("parentId", "")
                    media_list = ml_val.get("media", {}).get(
                        "richtextContent", {}).get("document", [])
                    if not parentId:
                        for md in media_list:
                            md_c = md.get("c", [])
                            # 保存评论
                            mc_list = []
                            for mc in md_c:
                                mt_text = mc.get("t")
                                mc_list.append(mt_text)
                                # print(mt_text)
                            level_one[user_id] = " ".join(mc_list)

                for ml_key, ml_val in model_list.items():
                    # print(ml_key)
                    user_id = ml_key
                    # print(user_id)
                    parentId = ml_val.get("parentId", "")
                    # print("pid",parentId)
                    media_list = ml_val.get("media", {}).get(
                        "richtextContent", {}).get("document", [])
                    for md in media_list:
                        md_c = md.get("c", [])
                        # 保存评论
                        if parentId and parentId in level_one.keys():
                            mc_list = []
                            for mc in md_c:
                                mt_text = mc.get("t")
                                mc_list.append(mt_text)
                            lv_one = level_one[parentId]
                            # new_one = ""
                            if "$$$" not in lv_one:
                                new_one = "{0}$$${1}".format(
                                    lv_one, " ".join(mc_list))
                                level_one[parentId] = new_one
                            else:
                                new_one = "{0}@@@{1}".format(
                                    lv_one, " ".join(mc_list))
                                level_one[parentId] = new_one
                level_one_list = []
                for k, v in level_one.items():
                    level_one_list.append(v)
                comments = "&&&".join(level_one_list)
                emoji_str = emoji.demojize(comments)
                # comments_new = re.sub(r':(.*?):', '', emoji_str).strip()
                LabelArray['content_comments'] = emoji_str
            else:
                LabelArray['content_comments'] = "errors"
    else:
        LabelArray['Html'] = '当前页面的网址为:' + LabelUrl + "\r\n页面类型为:" + PageType + \
            "\r\nCookies数据为:"+LabelCookie+"\r\n接收到的数据是:" + LabelArray['Html']
    # 以上是用户编写代码区域
    LabelArray = json.dumps(LabelArray)
    print(LabelArray)

　　这种方式清洗数据真是太舒服了，特此记录，python的路是越来越广了