spider/Douyin_TikTok_Download_API
1
0
Fork 0
mirror of https://github.com/Evil0ctal/Douyin_TikTok_Download_API.git synced 2025-04-20 11:35:01 +08:00
Code Issues Packages Projects Releases Wiki Activity

1,添加了西瓜解析2，修改了readme/readme.en3,BUG convert_share_urls 这里有bug。 如果抖音的口令解析的出来其他的都是none，还没有好的解决方法，我想简单的用if判断，但这样会不会有其他新但平台加进来感觉这样处理有点太简单了

Browse Source
This commit is contained in:
‘Xuenew;git config --global user.email 18188108851@163.com’ 2023-09-25 23:12:48 +08:00
parent 0d75d270d9
commit 4581d6cb6a
10 changed files with 380 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
.idea/Douyin_TikTok_Download_API.iml generated Normal file
View File

@ -0,0 +1,12 @@
<?xml version="1.0" encoding="UTF-8"?>
<module type="PYTHON_MODULE" version="4">
<component name="NewModuleRootManager">
<content url="file://$MODULE_DIR$" />
<orderEntry type="inheritedJdk" />
<orderEntry type="sourceFolder" forTests="false" />
</component>
<component name="PyDocumentationSettings">
<option name="format" value="PLAIN" />
<option name="myDocStringFormat" value="Plain" />
</component>
</module>

37
.idea/inspectionProfiles/Project_Default.xml generated Normal file
View File

@ -0,0 +1,37 @@
<component name="InspectionProjectProfileManager">
<profile version="1.0">
<option name="myName" value="Project Default" />
<inspection_tool class="Eslint" enabled="true" level="WARNING" enabled_by_default="true" />
<inspection_tool class="PyPackageRequirementsInspection" enabled="true" level="WARNING" enabled_by_default="true">
<option name="ignoredPackages">
<value>
<list size="15">
<item index="0" class="java.lang.String" itemvalue="moviepy" />
<item index="1" class="java.lang.String" itemvalue="requests" />
<item index="2" class="java.lang.String" itemvalue="scipy" />
<item index="3" class="java.lang.String" itemvalue="matplotlib" />
<item index="4" class="java.lang.String" itemvalue="psycopg2" />
<item index="5" class="java.lang.String" itemvalue="pydub" />
<item index="6" class="java.lang.String" itemvalue="PyAudio" />
<item index="7" class="java.lang.String" itemvalue="numpy" />
<item index="8" class="java.lang.String" itemvalue="mysql-connector-python" />
<item index="9" class="java.lang.String" itemvalue="wincertstore" />
<item index="10" class="java.lang.String" itemvalue="retrying" />
<item index="11" class="java.lang.String" itemvalue="mitmproxy" />
<item index="12" class="java.lang.String" itemvalue="PyMySQL" />
<item index="13" class="java.lang.String" itemvalue="Requests" />
<item index="14" class="java.lang.String" itemvalue="redis" />
</list>
</value>
</option>
</inspection_tool>
<inspection_tool class="PyPep8NamingInspection" enabled="true" level="WEAK WARNING" enabled_by_default="true">
<option name="ignoredErrors">
<list>
<option value="N803" />
</list>
</option>
</inspection_tool>
<inspection_tool class="TsLint" enabled="true" level="WARNING" enabled_by_default="true" />
</profile>
</component>

6
.idea/inspectionProfiles/profiles_settings.xml generated Normal file
View File

@ -0,0 +1,6 @@
<component name="InspectionProjectProfileManager">
<settings>
<option name="USE_PROJECT_PROFILE" value="false" />
<version value="1.0" />
</settings>
</component>

7
.idea/misc.xml generated Normal file
View File

@ -0,0 +1,7 @@
<?xml version="1.0" encoding="UTF-8"?>
<project version="4">
<component name="JavaScriptSettings">
<option name="languageLevel" value="ES6" />
</component>
<component name="ProjectRootManager" version="2" project-jdk-name="Python 3.7 (Douyin_TikTok_Download_API)" project-jdk-type="Python SDK" />
</project>

8
.idea/modules.xml generated Normal file
View File

@ -0,0 +1,8 @@
<?xml version="1.0" encoding="UTF-8"?>
<project version="4">
<component name="ProjectModuleManager">
<modules>
<module fileurl="file://$PROJECT_DIR$/.idea/Douyin_TikTok_Download_API.iml" filepath="$PROJECT_DIR$/.idea/Douyin_TikTok_Download_API.iml" />
</modules>
</component>
</project>

6
.idea/vcs.xml generated Normal file
View File

@ -0,0 +1,6 @@
<?xml version="1.0" encoding="UTF-8"?>
<project version="4">
<component name="VcsDirectoryMappings">
<mapping directory="$PROJECT_DIR$" vcs="Git" />
</component>
</project>

147
.idea/workspace.xml generated Normal file
View File

@ -0,0 +1,147 @@
<?xml version="1.0" encoding="UTF-8"?>
<project version="4">
<component name="BranchesTreeState">
<expand>
<path>
<item name="ROOT" type="e8cecc67:BranchNodeDescriptor" />
<item name="LOCAL_ROOT" type="e8cecc67:BranchNodeDescriptor" />
</path>
<path>
<item name="ROOT" type="e8cecc67:BranchNodeDescriptor" />
<item name="REMOTE_ROOT" type="e8cecc67:BranchNodeDescriptor" />
</path>
<path>
<item name="ROOT" type="e8cecc67:BranchNodeDescriptor" />
<item name="REMOTE_ROOT" type="e8cecc67:BranchNodeDescriptor" />
<item name="GROUP_NODE:origin" type="e8cecc67:BranchNodeDescriptor" />
</path>
</expand>
<select />
</component>
<component name="ChangeListManager">
<list default="true" id="1927aa06-6b5f-4740-b2ae-70a32c544f86" name="Default Changelist" comment="" />
<option name="SHOW_DIALOG" value="false" />
<option name="HIGHLIGHT_CONFLICTS" value="true" />
<option name="HIGHLIGHT_NON_ACTIVE_CHANGELIST" value="false" />
<option name="LAST_RESOLUTION" value="IGNORE" />
</component>
<component name="Git.Settings">
<option name="RECENT_BRANCH_BY_REPOSITORY">
<map>
<entry key="$PROJECT_DIR$" value="xigua" />
</map>
</option>
<option name="RECENT_GIT_ROOT_PATH" value="$PROJECT_DIR$" />
</component>
<component name="ProjectId" id="2VqkuxviM0uPte5GQr4yI8Qa8JC" />
<component name="ProjectViewState">
<option name="hideEmptyMiddlePackages" value="true" />
<option name="showLibraryContents" value="true" />
</component>
<component name="PropertiesComponent">
<property name="RunOnceActivity.OpenProjectViewOnStart" value="true" />
<property name="RunOnceActivity.ShowReadmeOnStart" value="true" />
<property name="WebServerToolWindowFactoryState" value="false" />
</component>
<component name="RunManager">
<configuration name="scraper" type="PythonConfigurationType" factoryName="Python" temporary="true" nameIsGenerated="true">
<module name="Douyin_TikTok_Download_API" />
<option name="INTERPRETER_OPTIONS" value="" />
<option name="PARENT_ENVS" value="true" />
<envs>
<env name="PYTHONUNBUFFERED" value="1" />
</envs>
<option name="SDK_HOME" value="" />
<option name="WORKING_DIRECTORY" value="$PROJECT_DIR$" />
<option name="IS_MODULE_SDK" value="true" />
<option name="ADD_CONTENT_ROOTS" value="true" />
<option name="ADD_SOURCE_ROOTS" value="true" />
<EXTENSION ID="PythonCoverageRunConfigurationExtension" runner="coverage.py" />
<option name="SCRIPT_NAME" value="$PROJECT_DIR$/scraper.py" />
<option name="PARAMETERS" value="" />
<option name="SHOW_COMMAND_LINE" value="false" />
<option name="EMULATE_TERMINAL" value="false" />
<option name="MODULE_MODE" value="false" />
<option name="REDIRECT_INPUT" value="false" />
<option name="INPUT_FILE" value="" />
<method v="2" />
</configuration>
<recent_temporary>
<list>
<item itemvalue="Python.scraper" />
</list>
</recent_temporary>
</component>
<component name="SvnConfiguration">
<configuration />
</component>
<component name="TaskManager">
<task active="true" id="Default" summary="Default task">
<changelist id="1927aa06-6b5f-4740-b2ae-70a32c544f86" name="Default Changelist" comment="" />
<created>1695571223199</created>
<option name="number" value="Default" />
<option name="presentableId" value="Default" />
<updated>1695571223199</updated>
<workItem from="1695571224589" duration="834000" />
<workItem from="1695638899275" duration="6328000" />
<workItem from="1695654217519" duration="164000" />
</task>
<task id="LOCAL-00001" summary="1，添加西瓜视频解析&#10;2，添加了readme/readme.en里关于西瓜样例&#10;3, BUG convert_share_urls 这里有bug，如果抖音的口令解析的出来其他的都是none,这里我看你的逻辑我没动。我想就简单的判断是不是哔哩还是西瓜但这样会不会有点不好，将来有其他平台这里还得改，我在想有没有更简单的方案。">
<created>1695652986133</created>
<option name="number" value="00001" />
<option name="presentableId" value="LOCAL-00001" />
<option name="project" value="LOCAL" />
<updated>1695652986133</updated>
</task>
<option name="localTasksCounter" value="2" />
<servers />
</component>
<component name="TypeScriptGeneratedFilesManager">
<option name="version" value="2" />
</component>
<component name="Vcs.Log.Tabs.Properties">
<option name="TAB_STATES">
<map>
<entry key="MAIN">
<value>
<State />
</value>
</entry>
</map>
</option>
<option name="oldMeFiltersMigrated" value="true" />
</component>
<component name="VcsManagerConfiguration">
<MESSAGE value="1，添加西瓜视频解析&#10;2，添加了readme/readme.en里关于西瓜样例&#10;3, BUG convert_share_urls 这里有bug，如果抖音的口令解析的出来其他的都是none,这里我看你的逻辑我没动。我想就简单的判断是不是哔哩还是西瓜但这样会不会有点不好，将来有其他平台这里还得改，我在想有没有更简单的方案。" />
<option name="LAST_COMMIT_MESSAGE" value="1，添加西瓜视频解析&#10;2，添加了readme/readme.en里关于西瓜样例&#10;3, BUG convert_share_urls 这里有bug，如果抖音的口令解析的出来其他的都是none,这里我看你的逻辑我没动。我想就简单的判断是不是哔哩还是西瓜但这样会不会有点不好，将来有其他平台这里还得改，我在想有没有更简单的方案。" />
</component>
<component name="WindowStateProjectService">
<state width="1750" height="405" key="GridCell.Tab.0.bottom" timestamp="1695654184489">
<screen x="0" y="0" width="1792" height="1120" />
</state>
<state width="1750" height="405" key="GridCell.Tab.0.bottom/0.0.1792.1120@0.0.1792.1120" timestamp="1695654184489" />
<state width="1750" height="405" key="GridCell.Tab.0.center" timestamp="1695654184487">
<screen x="0" y="0" width="1792" height="1120" />
</state>
<state width="1750" height="405" key="GridCell.Tab.0.center/0.0.1792.1120@0.0.1792.1120" timestamp="1695654184487" />
<state width="1750" height="405" key="GridCell.Tab.0.left" timestamp="1695654184487">
<screen x="0" y="0" width="1792" height="1120" />
</state>
<state width="1750" height="405" key="GridCell.Tab.0.left/0.0.1792.1120@0.0.1792.1120" timestamp="1695654184487" />
<state width="1750" height="405" key="GridCell.Tab.0.right" timestamp="1695654184488">
<screen x="0" y="0" width="1792" height="1120" />
</state>
<state width="1750" height="405" key="GridCell.Tab.0.right/0.0.1792.1120@0.0.1792.1120" timestamp="1695654184488" />
<state x="581" y="280" key="RollbackChangesDialog" timestamp="1695654162351">
<screen x="0" y="0" width="1792" height="1120" />
</state>
<state x="581" y="280" key="RollbackChangesDialog/0.0.1792.1120@0.0.1792.1120" timestamp="1695654162351" />
<state x="496" y="297" key="Vcs.Push.Dialog.v2" timestamp="1695653575453">
<screen x="0" y="0" width="1792" height="1120" />
</state>
<state x="496" y="297" key="Vcs.Push.Dialog.v2/0.0.1792.1120@0.0.1792.1120" timestamp="1695653575453" />
</component>
<component name="com.intellij.coverage.CoverageDataManagerImpl">
<SUITE FILE_PATH="coverage/Douyin_TikTok_Download_API$scraper.coverage" NAME="scraper Coverage Results" MODIFIED="1695652653592" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="true" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$" />
</component>
</project>

8
README.en.md
View File

@ -81,6 +81,7 @@ _Download prohibited videos, perform data analysis, download without watermark o
- Douyin (overseas version of Douyin: TikTok) video/picture analysis - Douyin (overseas version of Douyin: TikTok) video/picture analysis
- Bilibili video analysis - Bilibili video analysis
- Xigua video analysis
- Batch analysis on the web page (supports Douyin/TikTok mixed submission) - Batch analysis on the web page (supports Douyin/TikTok mixed submission)
- Batch download of non-watermarked videos from the web parsing result page (removed for V3.X and above versions, please deploy V2.X version by yourself) - Batch download of non-watermarked videos from the web parsing result page (removed for V3.X and above versions, please deploy V2.X version by yourself)
- API call to get link data - API call to get link data
@ -126,6 +127,13 @@ asyncio.run(hybrid_parsing(url=input("Paste Douyin/TikTok/Bilibili share URL her
> 💡Tip: Including but not limited to the following examples, if you encounter link parsing failure, please open a new one[issue](https://github.com/Evil0ctal/Douyin_TikTok_Download_API/issues) > 💡Tip: Including but not limited to the following examples, if you encounter link parsing failure, please open a new one[issue](https://github.com/Evil0ctal/Douyin_TikTok_Download_API/issues)
- Xigua video link
```text
https://www.ixigua.com/7270448082586698281/
https://m.ixigua.com/video/7274710134306112054/
```
- Bilibili video link - Bilibili video link
```text ```text

8
README.md
View File

@ -102,6 +102,7 @@
- 抖音（抖音海外版: TikTok）视频/图片解析 - 抖音（抖音海外版: TikTok）视频/图片解析
- Bilibili视频解析 - Bilibili视频解析
- 西瓜视频解析
- 网页端批量解析(支持抖音/TikTok混合提交) - 网页端批量解析(支持抖音/TikTok混合提交)
- 网页端解析结果页批量下载无水印视频(V3.X以上版本移除，请自行部署V2.X版本) - 网页端解析结果页批量下载无水印视频(V3.X以上版本移除，请自行部署V2.X版本)
- API调用获取链接数据 - API调用获取链接数据
@ -147,6 +148,13 @@ asyncio.run(hybrid_parsing(url=input("Paste Douyin/TikTok/Bilibili share URL her
> 💡提示：包含但不仅限于以下例子，如果遇到链接解析失败请开启一个新 [issue](https://github.com/Evil0ctal/Douyin_TikTok_Download_API/issues) > 💡提示：包含但不仅限于以下例子，如果遇到链接解析失败请开启一个新 [issue](https://github.com/Evil0ctal/Douyin_TikTok_Download_API/issues)
- 西瓜视频链接
```text
https://www.ixigua.com/7270448082586698281/
https://m.ixigua.com/video/7274710134306112054/
```
- Bilibili视频链接 - Bilibili视频链接
```text ```text

146
scraper.py
View File

@ -10,7 +10,7 @@
# If this project is helpful to you, please give me a star, thank you! # If this project is helpful to you, please give me a star, thank you!
# @备注: # @备注:
# 核心代码，估值1块(๑•̀ㅂ•́)و✧ # 核心代码，估值1块(๑•̀ㅂ•́)و✧
# 用于爬取Douyin/TikTok/Bilibili的数据并以字典形式返回。 # 用于爬取Douyin/TikTok/Bilibili/xigua的数据并以字典形式返回。
# 如果本项目对您有帮助，请给我一个star，谢谢！ # 如果本项目对您有帮助，请给我一个star，谢谢！
import re import re
@ -23,7 +23,9 @@ import asyncio
import traceback import traceback
import configparser import configparser
import urllib.parse import urllib.parse
import random
from zlib import crc32
from typing import Union from typing import Union
from tenacity import * from tenacity import *
@ -47,6 +49,22 @@ class Scraper:
self.bilibili_api_headers = { self.bilibili_api_headers = {
'User-Agent': 'com.ss.android.ugc.trill/494+Mozilla/5.0+(Linux;+Android+12;+2112123G+Build/SKQ1.211006.001;+wv)+AppleWebKit/537.36+(KHTML,+like+Gecko)+Version/4.0+Chrome/107.0.5304.105+Mobile+Safari/537.36' 'User-Agent': 'com.ss.android.ugc.trill/494+Mozilla/5.0+(Linux;+Android+12;+2112123G+Build/SKQ1.211006.001;+wv)+AppleWebKit/537.36+(KHTML,+like+Gecko)+Version/4.0+Chrome/107.0.5304.105+Mobile+Safari/537.36'
} }
self.ixigua_api_headers = {
'authority': 'ib.365yg.com',
'accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.7',
'accept-language': 'zh-CN,zh;q=0.9',
'cache-control': 'no-cache',
'pragma': 'no-cache',
'sec-ch-ua': '"Chromium";v="116", "Not)A;Brand";v="24", "Google Chrome";v="116"',
'sec-ch-ua-mobile': '?0',
'sec-ch-ua-platform': '"macOS"',
'sec-fetch-dest': 'document',
'sec-fetch-mode': 'navigate',
'sec-fetch-site': 'none',
'sec-fetch-user': '?1',
'upgrade-insecure-requests': '1',
'user-agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/116.0.0.0 Safari/537.36'
}
# 判断配置文件是否存在/Check if the configuration file exists # 判断配置文件是否存在/Check if the configuration file exists
if os.path.exists('config.ini'): if os.path.exists('config.ini'):
self.config = configparser.ConfigParser() self.config = configparser.ConfigParser()
@ -210,6 +228,37 @@ class Scraper:
else: else:
print('该链接为原始链接,无需转换,原始链接为: {}'.format(url)) print('该链接为原始链接,无需转换,原始链接为: {}'.format(url))
return url return url
elif 'ixigua.com' in url:
"""
西瓜视频链接类型(不全)：
1. https://v.ixigua.com/ienrQ5bR/
2. https://www.ixigua.com/7270448082586698281
3. https://m.ixigua.com/video/7270448082586698281
西瓜用户链接类型(不全)：
1. https://www.ixigua.com/home/3189050062678823
西瓜直播链接类型(不全)：
"""
if 'v.ixigua.com' in url:
print('正在通过西瓜分享链接获取原始链接...')
try:
async with aiohttp.ClientSession() as session:
async with session.get(url, headers=self.ixigua_api_headers, proxy=self.proxies, allow_redirects=False,
timeout=10) as response:
print("asdfasdf",response.headers)
if response.status == 302:
url = response.headers['Location'].split('?')[0] if '?' in response.headers[
'Location'] else \
response.headers['Location']
print('获取原始链接成功, 原始链接为: {}'.format(url))
return url
except Exception as e:
print('获取原始链接失败！')
print(e)
# return None
raise e
else:
print('该链接为原始链接,无需转换,原始链接为: {}'.format(url))
return url
"""__________________________________________⬇️Douyin methods(抖音方法)⬇️______________________________________""" """__________________________________________⬇️Douyin methods(抖音方法)⬇️______________________________________"""
@ -365,7 +414,7 @@ class Scraper:
"""__________________________________________⬇️bilibili methods(Bilibili方法)⬇️______________________________________""" """__________________________________________⬇️bilibili methods(Bilibili方法)⬇️______________________________________"""
# 获取TikTok视频ID/Get TikTok video ID # 获取bilibili视频ID/Get BiliBili video ID
async def get_bilibili_video_id(self, original_url: str) -> Union[str, None]: async def get_bilibili_video_id(self, original_url: str) -> Union[str, None]:
""" """
获取视频id 获取视频id
@ -427,6 +476,73 @@ class Scraper:
except Exception as e: except Exception as e:
raise ValueError(f'获取BiliBili视频数据出错了:{e}') raise ValueError(f'获取BiliBili视频数据出错了:{e}')
"""__________________________________________⬇️xigua methods(xigua方法)⬇️______________________________________"""
# 获取西瓜拿播放地址的接口
def get_xigua_json_url(self,video_id):
# 获取json文件的地址
r = str(random.random())[2:]
url_part = "/video/urls/v/1/toutiao/mp4/{}?r={}".format(video_id, r)
s = crc32(url_part.encode())
json_url = "https://ib.365yg.com{}&s={}&nobase64=true".format(url_part, s)
return json_url
# 获取西瓜视频ID/Get xigua video ID
async def get_ixigua_video_id(self, original_url: str) -> Union[str, None]:
"""
获取视频id
:param original_url: 视频链接
:return: 视频id
"""
try:
# 转换链接/Convert link
original_url = await self.convert_share_urls(original_url)
# 获取视频ID/Get video ID
if 'www.ixigua.com/' in original_url:
video_id = re.findall('ixigua\.com/(\d+)', original_url)[0]
elif 'm.ixigua.com/video' in original_url:
video_id = re.findall('/video/(\d+)', original_url)[0]
# 返回视频ID/Return video ID
return video_id
except Exception as e:
raise ValueError(f'获取西瓜视频ID出错了:{e}')
@retry(stop=stop_after_attempt(4), wait=wait_fixed(7))
async def get_ixigua_video_data(self, video_id: str) -> Union[dict, None]:
"""
获取单个视频信息
:param video_id: 视频id
:return: 视频信息
"""
print('正在获取西瓜视频数据...')
try:
# 构造访问链接/Construct the access link
video_url = f'https://m.ixigua.com/video/{video_id}?wid_try=1'
print("video_url",video_url)
async with aiohttp.ClientSession() as session:
async with session.get(video_url, headers=self.ixigua_api_headers, proxy=self.proxies,
timeout=10) as response:
response = await response.text()
search = re.search("\"vid\":\"([^\"]+)\",", response)
vid = search.group(1)
print('获取视频vid信息成功！')
play_url_api = self.get_xigua_json_url(vid)
print(f"正在获取视频数据API: {play_url_api}")
async with aiohttp.ClientSession() as session:
async with session.get(play_url_api, headers=self.ixigua_api_headers, proxy=self.proxies,
timeout=10) as response:
response = await response.json()
video_data = response.get("data",{}).get("video_list",{}).get("video_3",{}).get("main_url","")
video_data = {
'status': 'success',
'message': "更多接口请查看(More API see): https://api.tikhub.io/",
'type': 'video',
'platform': '西瓜',
'video_url': video_data,
}
return video_data
except Exception as e:
raise ValueError(f'获取西瓜视频数据出错了:{e}')
"""__________________________________________⬇️Hybrid methods(混合方法)⬇️______________________________________""" """__________________________________________⬇️Hybrid methods(混合方法)⬇️______________________________________"""
# 判断链接平台/Judge link platform # 判断链接平台/Judge link platform
@ -435,6 +551,8 @@ class Scraper:
url_platform = 'douyin' url_platform = 'douyin'
elif 'bilibili' in video_url: elif 'bilibili' in video_url:
url_platform = 'bilibili' url_platform = 'bilibili'
elif 'xigua' in video_url:
url_platform = 'xigua'
elif 'tiktok' in video_url: elif 'tiktok' in video_url:
url_platform = 'tiktok' url_platform = 'tiktok'
else: else:
@ -456,6 +574,7 @@ class Scraper:
video_id = await self.get_douyin_video_id(video_url) if url_platform == 'douyin' \ video_id = await self.get_douyin_video_id(video_url) if url_platform == 'douyin' \
else await self.get_tiktok_video_id(video_url) if url_platform == 'tiktok' \ else await self.get_tiktok_video_id(video_url) if url_platform == 'tiktok' \
else await self.get_bilibili_video_id(video_url) if url_platform == 'bilibili' \ else await self.get_bilibili_video_id(video_url) if url_platform == 'bilibili' \
else await self.get_ixigua_video_id(video_url) if url_platform == 'xigua' \
else None else None
# 如果获取不到视频ID抛出异常/If the video ID cannot be obtained, an exception is thrown # 如果获取不到视频ID抛出异常/If the video ID cannot be obtained, an exception is thrown
@ -467,6 +586,7 @@ class Scraper:
data = await self.get_douyin_video_data(video_id) if url_platform == 'douyin' \ data = await self.get_douyin_video_data(video_id) if url_platform == 'douyin' \
else await self.get_tiktok_video_data(video_id) if url_platform == 'tiktok' \ else await self.get_tiktok_video_data(video_id) if url_platform == 'tiktok' \
else await self.get_bilibili_video_data(video_id) if url_platform == 'bilibili' \ else await self.get_bilibili_video_data(video_id) if url_platform == 'bilibili' \
else await self.get_ixigua_video_data(video_id) if url_platform == 'xigua' \
else None else None
if data: if data:
@ -475,6 +595,10 @@ class Scraper:
if url_platform == 'bilibili': if url_platform == 'bilibili':
print("获取Bilibili视频数据成功！") print("获取Bilibili视频数据成功！")
return data return data
# 如果是西瓜平台则返回视频数据/If it is a ixigua platform, return video data
if url_platform == 'xigua':
print("获取西瓜视频数据成功！")
return data
# 如果是抖音/TikTok平台则继续进行数据解析/If it is a Douyin/TikTok platform, continue to parse the data # 如果是抖音/TikTok平台则继续进行数据解析/If it is a Douyin/TikTok platform, continue to parse the data
print(f"获取**{url_platform}**视频数据成功，正在判断数据类型...") print(f"获取**{url_platform}**视频数据成功，正在判断数据类型...")
@ -654,11 +778,19 @@ class Scraper:
"""__________________________________________⬇️Test methods(测试方法)⬇️______________________________________""" """__________________________________________⬇️Test methods(测试方法)⬇️______________________________________"""
async def async_test(_douyin_url: str = None, _tiktok_url: str = None, _bilibili_url: str = None) -> None: async def async_test(_douyin_url: str = None, _tiktok_url: str = None, _bilibili_url: str = None, _ixigua_url: str = None) -> None:
# 异步测试/Async test # 异步测试/Async test
start_time = time.time() start_time = time.time()
print("<异步测试/Async test>") print("<异步测试/Async test>")
print('\n--------------------------------------------------')
print("正在测试异步获取西瓜视频ID方法...")
ixigua_id = await api.get_ixigua_video_id(_ixigua_url)
print(f"西瓜视频ID: {ixigua_id}")
print("正在测试异步获取西瓜视频数据方法...")
ixigua_data = await api.get_ixigua_video_data(ixigua_id)
print(f"西瓜视频数据: {str(ixigua_data)[:100]}")
print('\n--------------------------------------------------') print('\n--------------------------------------------------')
print("正在测试异步获取哔哩哔哩视频ID方法...") print("正在测试异步获取哔哩哔哩视频ID方法...")
bilibili_id = await api.get_bilibili_video_id(_bilibili_url) bilibili_id = await api.get_bilibili_video_id(_bilibili_url)
@ -688,7 +820,8 @@ async def async_test(_douyin_url: str = None, _tiktok_url: str = None, _bilibili
douyin_hybrid_data = await api.hybrid_parsing(_douyin_url) douyin_hybrid_data = await api.hybrid_parsing(_douyin_url)
tiktok_hybrid_data = await api.hybrid_parsing(_tiktok_url) tiktok_hybrid_data = await api.hybrid_parsing(_tiktok_url)
bilibili_hybrid_data = await api.hybrid_parsing(_bilibili_url) bilibili_hybrid_data = await api.hybrid_parsing(_bilibili_url)
print(f"抖音、TikTok、哔哩哔哩混合解析全部成功！") xigua_hybrid_data = await api.hybrid_parsing(_ixigua_url)
print(f"抖音、TikTok、哔哩哔哩、西瓜混合解析全部成功！")
print('\n--------------------------------------------------') print('\n--------------------------------------------------')
# 总耗时/Total time # 总耗时/Total time
@ -704,4 +837,7 @@ if __name__ == '__main__':
douyin_url = 'https://v.douyin.com/rLyrQxA/6.66' douyin_url = 'https://v.douyin.com/rLyrQxA/6.66'
tiktok_url = 'https://www.tiktok.com/@evil0ctal/video/7217027383390555438' tiktok_url = 'https://www.tiktok.com/@evil0ctal/video/7217027383390555438'
bilibili_url = "https://www.bilibili.com/video/BV1Th411x7ii/" bilibili_url = "https://www.bilibili.com/video/BV1Th411x7ii/"
asyncio.run(async_test(_douyin_url=douyin_url, _tiktok_url=tiktok_url, _bilibili_url=bilibili_url)) ixigua_url = "https://www.ixigua.com/7270448082586698281"
# ixigua_url = "ttps://v.ixigua.com/ienrQ5bR/" # convert_share_urls 这里有bug 如果抖音的口令解析的出来其他的都是none
asyncio.run(async_test(_douyin_url=douyin_url, _tiktok_url=tiktok_url, _bilibili_url=bilibili_url, _ixigua_url=ixigua_url))