gitalk自动初始化
使用jekyll-text-theme之后, 使用gitalk作为评论系统, 需要自己手动初始化.
参考利用 Github Action 自动初始化 Gitalk 评论之Python篇
改造初始化gitalk的id方式
由于本主题使用的是自定义的key作为gitalk的id即github/issues的label, 而不是hexo使用的md5方式, 所以需要手动获取一下gitalk的id, 然后进行初始化
import requests
from bs4 import BeautifulSoup
import re
def get_gitalk_id(url):
r = requests.get(url)
soup = BeautifulSoup(r.text, 'html.parser')
# title作为issue的标题
# title = soup.title.string
gitalk_container = soup.findAll(id='js-gitalk-container')[0].next_element
key = str(re.findall(r"id: '([^']+)'", gitalk_container.string)[0])
return key
# 将about, archive 和 首页排除
def filter_posts(urls):
return list(filter(lambda x: not x.endswith(("about.html", "archive.html", "/")), urls))
整体思路如下
- 通过sitemap获取网站url
- 过滤不需要初始化的url
- 抓取页面, 获取gitalk中的id, 和title
- 判断不存在issue的label, 然后初始化
其他内容参考init-gitalk
附上完整代码
gitalk_init.py文件
启动命令python gitalk_init.py https://blog.lautumn.cn https://blog.lautumn.cn/sitemap.xml xxxxxxxxxxxx lautumn1990 blog, 其中xxxxxxxxxxxx 为token
#!/usr/bin/env python3
# -*- coding:utf-8 _*-
"""
@author: lautumn
@license: Apache Licence
@file: gitalk_init.py
@time: 2021/8/20 18:45
"""
import requests
import json
import sys
import xml.etree.ElementTree as ET
from bs4 import BeautifulSoup
import re
if len(sys.argv) != 6:
print("Usage:")
print(sys.argv[0], "site_url sitemap_url token username repo_name")
site_url = sys.argv[1]
sitemap_url = sys.argv[2]
token = sys.argv[3]
username = sys.argv[4]
repo_name = sys.argv[5]
# 根据url找key
url_key_map = {}
# 根据url找title
url_title_map = {}
def get_comments(session, label=''):
issues = []
if label is None or label == '':
return issues
url = 'https://api.github.com/repos/' + username + '/' + repo_name + '/issues?q=is&labels=Gitalk,' + label
r = session.get(url=url)
data = json.loads(r.text)
for issue in data:
issues.append(issue['body'].split('?')[0])
return issues
def get_url_key_map(urls):
global url_key_map, url_title_map
for url in urls:
key = ""
title = ""
try:
r = requests.get(url)
soup = BeautifulSoup(r.text, 'html.parser')
title = soup.title.string
gitalk_container = soup.findAll(id='js-gitalk-container')[0].next_element
key = str(re.findall(r"id: '([^']+)'", gitalk_container.string)[0])
except:
print("pass url %s , error" % url)
pass
url_key_map[url] = key
url_title_map[url] = title
def get_posts():
global url_key_map
post_urls = []
r = requests.get(sitemap_url)
root = ET.fromstring(r.text)
for child in root:
post_urls.append(child[0].text)
post_urls = filter_posts(post_urls)
print(post_urls)
get_url_key_map(post_urls)
post_urls = list(filter(lambda x: url_key_map.get(x) != '' and url_key_map.get(x) is not None, post_urls))
print(url_key_map)
return post_urls
def filter_posts(urls):
return list(filter(lambda x: not x.endswith(("about.html", "archive.html", "/")), urls))
def get_post_title(url):
global url_title_map
return url_title_map[url]
def init_gitalk(session, not_initialized):
github_url = "https://api.github.com/repos/" + username + "/" + repo_name + "/issues"
for url in not_initialized:
print("\n\n" + url)
title = get_post_title(url=url)
# issuse lable 限制最大长度为50，使用自定义的key, jekyll模板定义的
gtalk_id = url_key_map.get(url)
issue = {
'title': title,
'body': url,
'labels': ['Gitalk', gtalk_id]
}
print('[{}] checking...'.format(title))
is_existed = get_comments(session=session, label=gtalk_id)
if is_existed:
print(f"issues [ {gtalk_id} ] already exist")
continue
print('[{}] initializing...'.format(title))
resp = session.post(url=github_url, data=json.dumps(issue))
if resp.status_code == 201:
print('Created')
else:
print(f'issuse: {issue}')
print(f'failed: {resp.text}' )
break
def main():
print("\n\n-------------------")
session = requests.Session()
session.auth = (username, token)
session.headers = {
'Accept': 'application/vnd.github.v3+json',
'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/85.0.4183.59 Safari/537.36 Edg/85.0.564.30'
}
post_urls = get_posts()
init_gitalk(session=session, not_initialized=post_urls)
if __name__ == '__main__':
main()