From da8f1c62b826b59e975b0fe327e780622547d54f Mon Sep 17 00:00:00 2001 From: Relakkes Date: Wed, 16 Oct 2024 21:02:27 +0800 Subject: [PATCH] =?UTF-8?q?feat:=20=E7=9F=A5=E4=B9=8E=E6=94=AF=E6=8C=81?= =?UTF-8?q?=E5=88=9B=E4=BD=9C=E8=80=85=E4=B8=BB=E9=A1=B5=E6=95=B0=E6=8D=AE?= =?UTF-8?q?=E7=88=AC=E5=8F=96(=E5=9B=9E=E7=AD=94=E3=80=81=E6=96=87?= =?UTF-8?q?=E7=AB=A0=E3=80=81=E8=A7=86=E9=A2=91)?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- README.md | 102 ++++++++++++++++--- config/base_config.py | 13 ++- media_platform/zhihu/client.py | 180 +++++++++++++++++++++++++++++++-- media_platform/zhihu/core.py | 57 ++++++++++- media_platform/zhihu/help.py | 166 ++++++++++++++++++++++++------ model/m_zhihu.py | 13 ++- schema/tables.sql | 13 ++- store/zhihu/__init__.py | 33 +++++- 8 files changed, 511 insertions(+), 66 deletions(-) diff --git a/README.md b/README.md index e849f9f..d7cd461 100644 --- a/README.md +++ b/README.md @@ -13,17 +13,8 @@ 原理：利用[playwright](https://playwright.dev/)搭桥，保留登录成功后的上下文浏览器环境，通过执行JS表达式获取一些加密参数 通过使用此方式，免去了复现核心加密JS代码，逆向难度大大降低 -[MediaCrawlerPro](https://github.com/MediaCrawlerPro) 版本已经迭代出来了，相较于开源版本的优势： -- 多账号+IP代理支持（重点！） -- 去除Playwright依赖，使用更加简单 -- 支持linux部署（Docker docker-compose） -- 代码重构优化，更加易读易维护（解耦JS签名逻辑） -- 完美的架构设计，更加易扩展，源码学习的价值更大 - - MediaCrawler仓库白金赞助商: ⚡️【IPCola全球独家海外IP代理】⚡️新鲜的原生住宅代理，超高性价比，超多稀缺国家 -> 【IPCola全球独家海外IP代理】使用此处阿江专属推荐码注册：atxtupzfjhpbdbl ，获得10%金额补贴。 ## 功能列表 | 平台 | 关键词搜索 | 指定帖子ID爬取 | 二级评论 | 指定创作者主页 | 登录态缓存 | IP代理池 | 生成评论词云图 | @@ -36,8 +27,80 @@ MediaCrawler仓库白金赞助商: | 贴吧 | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | | 知乎 | ✅ | ❌ | ✅ | ❌ | ✅ | ✅ | ✅ | + +## 创建并激活 python 虚拟环境 +> 如果是爬取抖音和知乎，需要提前安装nodejs环境，版本大于等于：`16`即可
+ ```shell + # 进入项目根目录 + cd MediaCrawler + + # 创建虚拟环境 + # 我的python版本是：3.9.6，requirements.txt中的库是基于这个版本的，如果是其他python版本，可能requirements.txt中的库不兼容，自行解决一下。 + python -m venv venv + + # macos & linux 激活虚拟环境 + source venv/bin/activate + + # windows 激活虚拟环境 + venv\Scripts\activate + + ``` + +## 安装依赖库 + + ```shell + pip install -r requirements.txt + ``` + +## 安装 playwright浏览器驱动 + + ```shell + playwright install + ``` + +## 运行爬虫程序 + + ```shell + ### 项目默认是没有开启评论爬取模式，如需评论请在config/base_config.py中的 ENABLE_GET_COMMENTS 变量修改 + ### 一些其他支持项，也可以在config/base_config.py查看功能，写的有中文注释 + + # 从配置文件中读取关键词搜索相关的帖子并爬取帖子信息与评论 + python main.py --platform xhs --lt qrcode --type search + + # 从配置文件中读取指定的帖子ID列表获取指定帖子的信息与评论信息 + python main.py --platform xhs --lt qrcode --type detail + + # 打开对应APP扫二维码登录 + + # 其他平台爬虫使用示例，执行下面的命令查看 + python main.py --help + ``` + +## 数据保存 +- 支持关系型数据库Mysql中保存（需要提前创建数据库） + - 执行 `python db.py` 初始化数据库数据库表结构（只在首次执行） +- 支持保存到csv中（data/目录下） +- 支持保存到json中（data/目录下） + +## MediaCrawlerPro +[MediaCrawlerPro](https://github.com/MediaCrawlerPro) 版本已经重构出来了，相较于开源版本的优势： +- 多账号+IP代理支持（重点！） +- 去除Playwright依赖，使用更加简单 +- 支持linux部署（Docker docker-compose） +- 代码重构优化，更加易读易维护（解耦JS签名逻辑） +- 代码质量更高，对于构建更大型的爬虫项目更加友好 +- 完美的架构设计，更加易扩展，源码学习的价值更大 + + +## 其他常见问题可以查看在线文档 +> +> 在线文档包含使用方法、常见问题、加入项目交流群等。 +> [MediaCrawler在线文档](https://nanmicoder.github.io/MediaCrawler/) +> + ## 开发者服务 -> 开源不易，希望大家可以Star一下MediaCrawler仓库、支持下我的课程、星球，十分感谢！！！
+> 开源不易，希望大家可以Star一下MediaCrawler仓库！！！！十分感谢！！！
+> 如果你对知识付费认可，可以看下下面我提供的付费服务，如果你是学生，请一定提前告知，会有优惠💰
- MediaCrawler源码剖析课程： 如果你想很快入门这个项目，或者想了具体实现原理，我推荐你看看这个我录制的视频课程，从设计出发一步步带你如何使用，门槛大大降低 @@ -65,12 +128,6 @@ MediaCrawler仓库白金赞助商: - [Python协程在并发场景下的幂等性问题](https://articles.zsxq.com/id_wocdwsfmfcmp.html) - [错误使用 Python 可变类型带来的隐藏 Bug](https://articles.zsxq.com/id_f7vn89l1d303.html) -## 使用教程文档 - -> MediaCrawler文档使用vitepress构建，包含使用方法、常见问题、加入项目交流群等。 -> -[MediaCrawler在线文档](https://nanmicoder.github.io/MediaCrawler/) - ## 感谢下列Sponsors对本仓库赞助 > 【IPCola全球独家海外IP代理】使用此处阿江专属推荐码注册：atxtupzfjhpbdbl ，获得10%金额补贴。 @@ -80,6 +137,19 @@ MediaCrawler仓库白金赞助商: 成为赞助者，可以将您产品展示在这里，每天获得大量曝光，联系作者微信：yzglan 或 email：relakkes@gmail.com +## MediaCrawler项目微信交流群 + +👏👏👏 汇聚爬虫技术爱好者，共同学习，共同进步。 + +❗️❗️❗️群内禁止广告，禁止发各类违规和MediaCrawler不相关的问题 + +### 加群方式 +> 备注：github，会有拉群小助手自动拉你进群。 +> +> 如果图片展示不出来或过期，可以直接添加我的微信号：yzglan，并备注github，会有拉群小助手自动拉你进群 + +![relakkes_wechat](docs/static/images/relakkes_weichat.jpg) + ## 打赏 如果觉得项目不错的话可以打赏哦。您的支持就是我最大的动力！ diff --git a/config/base_config.py b/config/base_config.py index 2cb0b6c..efccb96 100644 --- a/config/base_config.py +++ b/config/base_config.py @@ -1,6 +1,6 @@ # 基础配置 PLATFORM = "xhs" -KEYWORDS = "编程副业,编程兼职" +KEYWORDS = "编程副业,编程兼职" # 关键词搜索配置，以英文逗号分隔 LOGIN_TYPE = "qrcode" # qrcode or phone or cookie COOKIES = "" # 具体值参见media_platform.xxx.field下的枚举值，暂时只支持小红书 @@ -45,8 +45,8 @@ MAX_CONCURRENCY_NUM = 1 # 是否开启爬图片模式, 默认不开启爬图片 ENABLE_GET_IMAGES = False -# 是否开启爬评论模式, 默认不开启爬评论 -ENABLE_GET_COMMENTS = False +# 是否开启爬评论模式, 默认开启爬评论 +ENABLE_GET_COMMENTS = True # 是否开启爬二级评论模式, 默认不开启爬二级评论 # 老版本项目使用了 db, 则需参考 schema/tables.sql line 287 增加表字段 @@ -130,6 +130,13 @@ KS_CREATOR_ID_LIST = [ # ........................ ] + +# 指定知乎创作者主页url列表 +ZHIHU_CREATOR_URL_LIST = [ + "https://www.zhihu.com/people/yd1234567", + # ........................ +] + # 词云相关 # 是否开启生成评论词云图 ENABLE_GET_WORDCLOUD = False diff --git a/media_platform/zhihu/client.py b/media_platform/zhihu/client.py index ba29fbe..a1d28d8 100644 --- a/media_platform/zhihu/client.py +++ b/media_platform/zhihu/client.py @@ -5,18 +5,19 @@ from typing import Any, Callable, Dict, List, Optional, Union from urllib.parse import urlencode import httpx +from httpx import Response from playwright.async_api import BrowserContext, Page from tenacity import retry, stop_after_attempt, wait_fixed import config from base.base_crawler import AbstractApiClient from constant import zhihu as zhihu_constant -from model.m_zhihu import ZhihuComment, ZhihuContent +from model.m_zhihu import ZhihuComment, ZhihuContent, ZhihuCreator from tools import utils from .exception import DataFetchError, ForbiddenError from .field import SearchSort, SearchTime, SearchType -from .help import ZhiHuJsonExtractor, sign +from .help import ZhihuExtractor, sign class ZhiHuClient(AbstractApiClient): @@ -33,7 +34,7 @@ class ZhiHuClient(AbstractApiClient): self.timeout = timeout self.default_headers = headers self.cookie_dict = cookie_dict - self._extractor = ZhiHuJsonExtractor() + self._extractor = ZhihuExtractor() async def _pre_headers(self, url: str) -> Dict: """ @@ -95,7 +96,7 @@ class ZhiHuClient(AbstractApiClient): raise DataFetchError(response.text) - async def get(self, uri: str, params=None) -> Dict: + async def get(self, uri: str, params=None, **kwargs) -> Union[Response, Dict, str]: """ GET请求，对请求头签名 Args: @@ -109,7 +110,7 @@ class ZhiHuClient(AbstractApiClient): if isinstance(params, dict): final_uri += '?' + urlencode(params) headers = await self._pre_headers(final_uri) - return await self.request(method="GET", url=zhihu_constant.ZHIHU_URL + final_uri, headers=headers) + return await self.request(method="GET", url=zhihu_constant.ZHIHU_URL + final_uri, headers=headers, **kwargs) async def pong(self) -> bool: """ @@ -194,7 +195,7 @@ class ZhiHuClient(AbstractApiClient): } search_res = await self.get(uri, params) utils.logger.info(f"[ZhiHuClient.get_note_by_keyword] Search result: {search_res}") - return self._extractor.extract_contents(search_res) + return self._extractor.extract_contents_from_search(search_res) async def get_root_comments(self, content_id: str, content_type: str, offset: str = "", limit: int = 10, order_by: str = "sort") -> Dict: @@ -317,3 +318,170 @@ class ZhiHuClient(AbstractApiClient): all_sub_comments.extend(sub_comments) await asyncio.sleep(crawl_interval) return all_sub_comments + + async def get_creator_info(self, url_token: str) -> Optional[ZhihuCreator]: + """ + 获取创作者信息 + Args: + url_token: + + Returns: + + """ + uri = f"/people/{url_token}" + html_content: str = await self.get(uri, return_response=True) + return self._extractor.extract_creator(url_token, html_content) + + async def get_creator_answers(self, url_token: str, offset: int = 0, limit: int = 20) -> Dict: + """ + 获取创作者的回答 + Args: + url_token: + offset: + limit: + + Returns: + + + """ + uri = f"/api/v4/members/{url_token}/answers" + params = { + "include":"data[*].is_normal,admin_closed_comment,reward_info,is_collapsed,annotation_action,annotation_detail,collapse_reason,collapsed_by,suggest_edit,comment_count,can_comment,content,editable_content,attachment,voteup_count,reshipment_settings,comment_permission,created_time,updated_time,review_info,excerpt,paid_info,reaction_instruction,is_labeled,label_info,relationship.is_authorized,voting,is_author,is_thanked,is_nothelp;data[*].vessay_info;data[*].author.badge[?(type=best_answerer)].topics;data[*].author.vip_info;data[*].question.has_publishing_draft,relationship", + "offset": offset, + "limit": limit, + "order_by": "created" + } + return await self.get(uri, params) + + async def get_creator_articles(self, url_token: str, offset: int = 0, limit: int = 20) -> Dict: + """ + 获取创作者的文章 + Args: + url_token: + offset: + limit: + + Returns: + + """ + uri = f"/api/v4/members/{url_token}/articles" + params = { + "include":"data[*].comment_count,suggest_edit,is_normal,thumbnail_extra_info,thumbnail,can_comment,comment_permission,admin_closed_comment,content,voteup_count,created,updated,upvoted_followees,voting,review_info,reaction_instruction,is_labeled,label_info;data[*].vessay_info;data[*].author.badge[?(type=best_answerer)].topics;data[*].author.vip_info;", + "offset": offset, + "limit": limit, + "order_by": "created" + } + return await self.get(uri, params) + + async def get_creator_videos(self, url_token: str, offset: int = 0, limit: int = 20) -> Dict: + """ + 获取创作者的视频 + Args: + url_token: + offset: + limit: + + Returns: + + """ + uri = f"/api/v4/members/{url_token}/zvideos" + params = { + "include":"similar_zvideo,creation_relationship,reaction_instruction", + "offset": offset, + "limit": limit, + "similar_aggregation": "true" + } + return await self.get(uri, params) + + async def get_all_anwser_by_creator(self, creator: ZhihuCreator, crawl_interval: float = 1.0, + callback: Optional[Callable] = None) -> List[ZhihuContent]: + """ + 获取创作者的所有回答 + Args: + creator: 创作者信息 + crawl_interval: 爬取一次笔记的延迟单位（秒） + callback: 一次笔记爬取结束后 + + Returns: + + """ + all_contents: List[ZhihuContent] = [] + is_end: bool = False + offset: int = 0 + limit: int = 20 + while not is_end: + res = await self.get_creator_answers(creator.url_token, offset, limit) + if not res: + break + utils.logger.info(f"[ZhiHuClient.get_all_anwser_by_creator] Get creator {creator.url_token} answers: {res}") + paging_info = res.get("paging", {}) + is_end = paging_info.get("is_end") + contents = self._extractor.extract_content_list_from_creator(res.get("data")) + if callback: + await callback(contents) + all_contents.extend(contents) + offset += limit + await asyncio.sleep(crawl_interval) + return all_contents + + + async def get_all_articles_by_creator(self, creator: ZhihuCreator, crawl_interval: float = 1.0, + callback: Optional[Callable] = None) -> List[ZhihuContent]: + """ + 获取创作者的所有文章 + Args: + creator: + crawl_interval: + callback: + + Returns: + + """ + all_contents: List[ZhihuContent] = [] + is_end: bool = False + offset: int = 0 + limit: int = 20 + while not is_end: + res = await self.get_creator_articles(creator.url_token, offset, limit) + if not res: + break + paging_info = res.get("paging", {}) + is_end = paging_info.get("is_end") + contents = self._extractor.extract_content_list_from_creator(res.get("data")) + if callback: + await callback(contents) + all_contents.extend(contents) + offset += limit + await asyncio.sleep(crawl_interval) + return all_contents + + + async def get_all_videos_by_creator(self, creator: ZhihuCreator, crawl_interval: float = 1.0, + callback: Optional[Callable] = None) -> List[ZhihuContent]: + """ + 获取创作者的所有视频 + Args: + creator: + crawl_interval: + callback: + + Returns: + + """ + all_contents: List[ZhihuContent] = [] + is_end: bool = False + offset: int = 0 + limit: int = 20 + while not is_end: + res = await self.get_creator_videos(creator.url_token, offset, limit) + if not res: + break + paging_info = res.get("paging", {}) + is_end = paging_info.get("is_end") + contents = self._extractor.extract_content_list_from_creator(res.get("data")) + if callback: + await callback(contents) + all_contents.extend(contents) + offset += limit + await asyncio.sleep(crawl_interval) + return all_contents diff --git a/media_platform/zhihu/core.py b/media_platform/zhihu/core.py index 6a3de2a..9898104 100644 --- a/media_platform/zhihu/core.py +++ b/media_platform/zhihu/core.py @@ -10,7 +10,7 @@ from playwright.async_api import (BrowserContext, BrowserType, Page, import config from base.base_crawler import AbstractCrawler -from model.m_zhihu import ZhihuContent +from model.m_zhihu import ZhihuContent, ZhihuCreator from proxy.proxy_ip_pool import IpInfoModel, create_ip_pool from store import zhihu as zhihu_store from tools import utils @@ -18,7 +18,7 @@ from var import crawler_type_var, source_keyword_var from .client import ZhiHuClient from .exception import DataFetchError -from .help import ZhiHuJsonExtractor +from .help import ZhihuExtractor from .login import ZhiHuLogin @@ -31,7 +31,7 @@ class ZhihuCrawler(AbstractCrawler): self.index_url = "https://www.zhihu.com" # self.user_agent = utils.get_user_agent() self.user_agent = "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/128.0.0.0 Safari/537.36" - self._extractor = ZhiHuJsonExtractor() + self._extractor = ZhihuExtractor() async def start(self) -> None: """ @@ -74,7 +74,7 @@ class ZhihuCrawler(AbstractCrawler): await self.zhihu_client.update_cookies(browser_context=self.browser_context) # 知乎的搜索接口需要打开搜索页面之后cookies才能访问API，单独的首页不行 - utils.logger.info("[ZhihuCrawler.start] Zhihu跳转到搜索页面获取搜索页面的Cookies，改过程需要5秒左右") + utils.logger.info("[ZhihuCrawler.start] Zhihu跳转到搜索页面获取搜索页面的Cookies，该过程需要5秒左右") await self.context_page.goto(f"{self.index_url}/search?q=python&search_source=Guess&utm_content=search_hot&type=content") await asyncio.sleep(5) await self.zhihu_client.update_cookies(browser_context=self.browser_context) @@ -88,7 +88,7 @@ class ZhihuCrawler(AbstractCrawler): raise NotImplementedError elif config.CRAWLER_TYPE == "creator": # Get creator's information and their notes and comments - raise NotImplementedError + await self.get_creators_and_notes() else: pass @@ -169,6 +169,53 @@ class ZhihuCrawler(AbstractCrawler): callback=zhihu_store.batch_update_zhihu_note_comments ) + async def get_creators_and_notes(self) -> None: + """ + Get creator's information and their notes and comments + Returns: + + """ + utils.logger.info("[ZhihuCrawler.get_creators_and_notes] Begin get xiaohongshu creators") + for user_link in config.ZHIHU_CREATOR_URL_LIST: + utils.logger.info(f"[ZhihuCrawler.get_creators_and_notes] Begin get creator {user_link}") + user_url_token = user_link.split("/")[-1] + # get creator detail info from web html content + createor_info: ZhihuCreator = await self.zhihu_client.get_creator_info(url_token=user_url_token) + if not createor_info: + utils.logger.info(f"[ZhihuCrawler.get_creators_and_notes] Creator {user_url_token} not found") + continue + + utils.logger.info(f"[ZhihuCrawler.get_creators_and_notes] Creator info: {createor_info}") + await zhihu_store.save_creator(creator=createor_info) + + # 默认只提取回答信息，如果需要文章和视频，把下面的注释打开即可 + + # Get all anwser information of the creator + all_content_list = await self.zhihu_client.get_all_anwser_by_creator( + creator=createor_info, + crawl_interval=random.random(), + callback=zhihu_store.batch_update_zhihu_contents + ) + + + # Get all articles of the creator's contents + # all_content_list = await self.zhihu_client.get_all_articles_by_creator( + # creator=createor_info, + # crawl_interval=random.random(), + # callback=zhihu_store.batch_update_zhihu_contents + # ) + + # Get all videos of the creator's contents + # all_content_list = await self.zhihu_client.get_all_videos_by_creator( + # creator=createor_info, + # crawl_interval=random.random(), + # callback=zhihu_store.batch_update_zhihu_contents + # ) + + # Get all comments of the creator's contents + await self.batch_get_content_comments(all_content_list) + + @staticmethod def format_proxy_info(ip_proxy_info: IpInfoModel) -> Tuple[Optional[Dict], Optional[Dict]]: """format proxy info for playwright and httpx""" diff --git a/media_platform/zhihu/help.py b/media_platform/zhihu/help.py index e9e929a..edd16a3 100644 --- a/media_platform/zhihu/help.py +++ b/media_platform/zhihu/help.py @@ -1,8 +1,10 @@ # -*- coding: utf-8 -*- -from typing import Dict, List +import json +from typing import Dict, List, Optional from urllib.parse import parse_qs, urlparse import execjs +from parsel import Selector from constant import zhihu as zhihu_constant from model.m_zhihu import ZhihuComment, ZhihuContent, ZhihuCreator @@ -29,11 +31,11 @@ def sign(url: str, cookies: str) -> Dict: return ZHIHU_SGIN_JS.call("get_sign", url, cookies) -class ZhiHuJsonExtractor: +class ZhihuExtractor: def __init__(self): pass - def extract_contents(self, json_data: Dict) -> List[ZhihuContent]: + def extract_contents_from_search(self, json_data: Dict) -> List[ZhihuContent]: """ extract zhihu contents Args: @@ -45,21 +47,34 @@ class ZhiHuJsonExtractor: if not json_data: return [] - result: List[ZhihuContent] = [] search_result: List[Dict] = json_data.get("data", []) search_result = [s_item for s_item in search_result if s_item.get("type") in ['search_result', 'zvideo']] - for sr_item in search_result: - sr_object: Dict = sr_item.get("object", {}) - if sr_object.get("type") == zhihu_constant.ANSWER_NAME: - result.append(self._extract_answer_content(sr_object)) - elif sr_object.get("type") == zhihu_constant.ARTICLE_NAME: - result.append(self._extract_article_content(sr_object)) - elif sr_object.get("type") == zhihu_constant.VIDEO_NAME: - result.append(self._extract_zvideo_content(sr_object)) + return self._extract_content_list([sr_item.get("object") for sr_item in search_result if sr_item.get("object")]) + + + def _extract_content_list(self, content_list: List[Dict]) -> List[ZhihuContent]: + """ + extract zhihu content list + Args: + content_list: + + Returns: + + """ + if not content_list: + return [] + + res: List[ZhihuContent] = [] + for content in content_list: + if content.get("type") == zhihu_constant.ANSWER_NAME: + res.append(self._extract_answer_content(content)) + elif content.get("type") == zhihu_constant.ARTICLE_NAME: + res.append(self._extract_article_content(content)) + elif content.get("type") == zhihu_constant.VIDEO_NAME: + res.append(self._extract_zvideo_content(content)) else: continue - - return result + return res def _extract_answer_content(self, answer: Dict) -> ZhihuContent: """ @@ -72,22 +87,23 @@ class ZhiHuJsonExtractor: res = ZhihuContent() res.content_id = answer.get("id") res.content_type = answer.get("type") - res.content_text = extract_text_from_html(answer.get("content")) + res.content_text = extract_text_from_html(answer.get("content", "")) res.question_id = answer.get("question").get("id") res.content_url = f"{zhihu_constant.ZHIHU_URL}/question/{res.question_id}/answer/{res.content_id}" - res.title = extract_text_from_html(answer.get("title")) - res.desc = extract_text_from_html(answer.get("description")) + res.title = extract_text_from_html(answer.get("title", "")) + res.desc = extract_text_from_html(answer.get("description", "") or answer.get("excerpt", "")) res.created_time = answer.get("created_time") res.updated_time = answer.get("updated_time") - res.voteup_count = answer.get("voteup_count") - res.comment_count = answer.get("comment_count") + res.voteup_count = answer.get("voteup_count", 0) + res.comment_count = answer.get("comment_count", 0) # extract author info - author_info = self._extract_author(answer.get("author")) + author_info = self._extract_content_or_comment_author(answer.get("author")) res.user_id = author_info.user_id res.user_link = author_info.user_link res.user_nickname = author_info.user_nickname res.user_avatar = author_info.user_avatar + res.user_url_token = author_info.url_token return res def _extract_article_content(self, article: Dict) -> ZhihuContent: @@ -106,17 +122,18 @@ class ZhiHuJsonExtractor: res.content_url = f"{zhihu_constant.ZHIHU_URL}/p/{res.content_id}" res.title = extract_text_from_html(article.get("title")) res.desc = extract_text_from_html(article.get("excerpt")) - res.created_time = article.get("created_time") - res.updated_time = article.get("updated_time") - res.voteup_count = article.get("voteup_count") - res.comment_count = article.get("comment_count") + res.created_time = article.get("created_time", 0) or article.get("created", 0) + res.updated_time = article.get("updated_time", 0) or article.get("updated", 0) + res.voteup_count = article.get("voteup_count", 0) + res.comment_count = article.get("comment_count", 0) # extract author info - author_info = self._extract_author(article.get("author")) + author_info = self._extract_content_or_comment_author(article.get("author")) res.user_id = author_info.user_id res.user_link = author_info.user_link res.user_nickname = author_info.user_nickname res.user_avatar = author_info.user_avatar + res.user_url_token = author_info.url_token return res def _extract_zvideo_content(self, zvideo: Dict) -> ZhihuContent: @@ -129,25 +146,34 @@ class ZhiHuJsonExtractor: """ res = ZhihuContent() - res.content_id = zvideo.get("zvideo_id") + + if "video" in zvideo and isinstance(zvideo.get("video"), dict): # 说明是从创作者主页的视频列表接口来的 + res.content_id = zvideo.get("video").get("video_id") + res.content_url = f"{zhihu_constant.ZHIHU_URL}/zvideo/{res.content_id}" + res.created_time = zvideo.get("published_at") + res.updated_time = zvideo.get("updated_at") + else: + res.content_id = zvideo.get("zvideo_id") + res.content_url = zvideo.get("video_url") + res.created_time = zvideo.get("created_at") + res.content_type = zvideo.get("type") - res.content_url = zvideo.get("video_url") res.title = extract_text_from_html(zvideo.get("title")) res.desc = extract_text_from_html(zvideo.get("description")) - res.created_time = zvideo.get("created_at") res.voteup_count = zvideo.get("voteup_count") res.comment_count = zvideo.get("comment_count") # extract author info - author_info = self._extract_author(zvideo.get("author")) + author_info = self._extract_content_or_comment_author(zvideo.get("author")) res.user_id = author_info.user_id res.user_link = author_info.user_link res.user_nickname = author_info.user_nickname res.user_avatar = author_info.user_avatar + res.user_url_token = author_info.url_token return res @staticmethod - def _extract_author(author: Dict) -> ZhihuCreator: + def _extract_content_or_comment_author(author: Dict) -> ZhihuCreator: """ extract zhihu author Args: @@ -165,6 +191,7 @@ class ZhiHuJsonExtractor: res.user_link = f"{zhihu_constant.ZHIHU_URL}/people/{author.get('url_token')}" res.user_nickname = author.get("name") res.user_avatar = author.get("avatar_url") + res.url_token = author.get("url_token") return res def extract_comments(self, page_content: ZhihuContent, comments: List[Dict]) -> List[ZhihuComment]: @@ -209,7 +236,7 @@ class ZhiHuJsonExtractor: res.content_type = page_content.content_type # extract author info - author_info = self._extract_author(comment.get("author")) + author_info = self._extract_content_or_comment_author(comment.get("author")) res.user_id = author_info.user_id res.user_link = author_info.user_link res.user_nickname = author_info.user_nickname @@ -254,3 +281,80 @@ class ZhiHuJsonExtractor: query_params = parse_qs(parsed_url.query) offset = query_params.get('offset', [""])[0] return offset + + @staticmethod + def _foramt_gender_text(gender: int) -> str: + """ + format gender text + Args: + gender: + + Returns: + + """ + if gender == 1: + return "男" + elif gender == 0: + return "女" + else: + return "未知" + + + def extract_creator(self, user_url_token: str, html_content: str) -> Optional[ZhihuCreator]: + """ + extract zhihu creator + Args: + user_url_token : zhihu creator url token + html_content: zhihu creator html content + + Returns: + + """ + if not html_content: + return None + + js_init_data = Selector(text=html_content).xpath("//script[@id='js-initialData']/text()").get(default="").strip() + if not js_init_data: + return None + + js_init_data_dict: Dict = json.loads(js_init_data) + users_info: Dict = js_init_data_dict.get("initialState", {}).get("entities", {}).get("users", {}) + if not users_info: + return None + + creator_info: Dict = users_info.get(user_url_token) + if not creator_info: + return None + + res = ZhihuCreator() + res.user_id = creator_info.get("id") + res.user_link = f"{zhihu_constant.ZHIHU_URL}/people/{user_url_token}" + res.user_nickname = creator_info.get("name") + res.user_avatar = creator_info.get("avatarUrl") + res.url_token = creator_info.get("urlToken") or user_url_token + res.gender = self._foramt_gender_text(creator_info.get("gender")) + res.ip_location = creator_info.get("ipInfo") + res.follows = creator_info.get("followingCount") + res.fans = creator_info.get("followerCount") + res.anwser_count = creator_info.get("answerCount") + res.video_count = creator_info.get("zvideoCount") + res.question_count = creator_info.get("questionCount") + res.article_count = creator_info.get("articlesCount") + res.column_count = creator_info.get("columnsCount") + res.get_voteup_count = creator_info.get("voteupCount") + return res + + + def extract_content_list_from_creator(self, anwser_list: List[Dict]) -> List[ZhihuContent]: + """ + extract content list from creator + Args: + anwser_list: + + Returns: + + """ + if not anwser_list: + return [] + + return self._extract_content_list(anwser_list) diff --git a/model/m_zhihu.py b/model/m_zhihu.py index 0dc6b6a..08c5142 100644 --- a/model/m_zhihu.py +++ b/model/m_zhihu.py @@ -15,8 +15,8 @@ class ZhihuContent(BaseModel): question_id: str = Field(default="", description="问题ID, type为answer时有值") title: str = Field(default="", description="内容标题") desc: str = Field(default="", description="内容描述") - created_time: int = Field(default="", description="创建时间") - updated_time: int = Field(default="", description="更新时间") + created_time: int = Field(default=0, description="创建时间") + updated_time: int = Field(default=0, description="更新时间") voteup_count: int = Field(default=0, description="赞同人数") comment_count: int = Field(default=0, description="评论数量") source_keyword: str = Field(default="", description="来源关键词") @@ -25,6 +25,7 @@ class ZhihuContent(BaseModel): user_link: str = Field(default="", description="用户主页链接") user_nickname: str = Field(default="", description="用户昵称") user_avatar: str = Field(default="", description="用户头像地址") + user_url_token: str = Field(default="", description="用户url_token") class ZhihuComment(BaseModel): @@ -57,7 +58,15 @@ class ZhihuCreator(BaseModel): user_link: str = Field(default="", description="用户主页链接") user_nickname: str = Field(default="", description="用户昵称") user_avatar: str = Field(default="", description="用户头像地址") + url_token: str = Field(default="", description="用户url_token") gender: str = Field(default="", description="用户性别") ip_location: Optional[str] = Field(default="", description="IP地理位置") follows: int = Field(default=0, description="关注数") fans: int = Field(default=0, description="粉丝数") + anwser_count: int = Field(default=0, description="回答数") + video_count: int = Field(default=0, description="视频数") + question_count: int = Field(default=0, description="提问数") + article_count: int = Field(default=0, description="文章数") + column_count: int = Field(default=0, description="专栏数") + get_voteup_count: int = Field(default=0, description="获得的赞同数") + diff --git a/schema/tables.sql b/schema/tables.sql index 6e304f5..33e02d3 100644 --- a/schema/tables.sql +++ b/schema/tables.sql @@ -474,6 +474,7 @@ CREATE TABLE `zhihu_content` ( `user_link` varchar(255) NOT NULL COMMENT '用户主页链接', `user_nickname` varchar(64) NOT NULL COMMENT '用户昵称', `user_avatar` varchar(255) NOT NULL COMMENT '用户头像地址', + `user_url_token` varchar(255) NOT NULL COMMENT '用户url_token', `add_ts` bigint NOT NULL COMMENT '记录添加时间戳', `last_modify_ts` bigint NOT NULL COMMENT '记录最后修改时间戳', PRIMARY KEY (`id`), @@ -482,6 +483,7 @@ CREATE TABLE `zhihu_content` ( ) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_0900_ai_ci COMMENT='知乎内容（回答、文章、视频）'; + CREATE TABLE `zhihu_comment` ( `id` int NOT NULL AUTO_INCREMENT COMMENT '自增ID', `comment_id` varchar(64) NOT NULL COMMENT '评论ID', @@ -513,10 +515,17 @@ CREATE TABLE `zhihu_creator` ( `user_link` varchar(255) NOT NULL COMMENT '用户主页链接', `user_nickname` varchar(64) NOT NULL COMMENT '用户昵称', `user_avatar` varchar(255) NOT NULL COMMENT '用户头像地址', + `url_token` varchar(64) NOT NULL COMMENT '用户URL Token', `gender` varchar(16) DEFAULT NULL COMMENT '用户性别', `ip_location` varchar(64) DEFAULT NULL COMMENT 'IP地理位置', - `follows` int NOT NULL DEFAULT '0' COMMENT '关注数', - `fans` int NOT NULL DEFAULT '0' COMMENT '粉丝数', + `follows` int NOT NULL DEFAULT 0 COMMENT '关注数', + `fans` int NOT NULL DEFAULT 0 COMMENT '粉丝数', + `anwser_count` int NOT NULL DEFAULT 0 COMMENT '回答数', + `video_count` int NOT NULL DEFAULT 0 COMMENT '视频数', + `question_count` int NOT NULL DEFAULT 0 COMMENT '问题数', + `article_count` int NOT NULL DEFAULT 0 COMMENT '文章数', + `column_count` int NOT NULL DEFAULT 0 COMMENT '专栏数', + `get_voteup_count` int NOT NULL DEFAULT 0 COMMENT '获得的赞同数', `add_ts` bigint NOT NULL COMMENT '记录添加时间戳', `last_modify_ts` bigint NOT NULL COMMENT '记录最后修改时间戳', PRIMARY KEY (`id`), diff --git a/store/zhihu/__init__.py b/store/zhihu/__init__.py index fc944d7..703d9cb 100644 --- a/store/zhihu/__init__.py +++ b/store/zhihu/__init__.py @@ -3,7 +3,7 @@ from typing import List import config from base.base_crawler import AbstractStore -from model.m_zhihu import ZhihuComment, ZhihuContent +from model.m_zhihu import ZhihuComment, ZhihuContent, ZhihuCreator from store.zhihu.zhihu_store_impl import (ZhihuCsvStoreImplement, ZhihuDbStoreImplement, ZhihuJsonStoreImplement) @@ -25,6 +25,21 @@ class ZhihuStoreFactory: raise ValueError("[ZhihuStoreFactory.create_store] Invalid save option only supported csv or db or json ...") return store_class() +async def batch_update_zhihu_contents(contents: List[ZhihuContent]): + """ + 批量更新知乎内容 + Args: + contents: + + Returns: + + """ + if not contents: + return + + for content_item in contents: + await update_zhihu_content(content_item) + async def update_zhihu_content(content_item: ZhihuContent): """ 更新知乎内容 @@ -71,3 +86,19 @@ async def update_zhihu_content_comment(comment_item: ZhihuComment): local_db_item.update({"last_modify_ts": utils.get_current_timestamp()}) utils.logger.info(f"[store.zhihu.update_zhihu_note_comment] zhihu content comment:{local_db_item}") await ZhihuStoreFactory.create_store().store_comment(local_db_item) + + +async def save_creator(creator: ZhihuCreator): + """ + 保存知乎创作者信息 + Args: + creator: + + Returns: + + """ + if not creator: + return + local_db_item = creator.model_dump() + local_db_item.update({"last_modify_ts": utils.get_current_timestamp()}) + await ZhihuStoreFactory.create_store().store_creator(local_db_item) \ No newline at end of file