southfox/garden
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

[fix] do not get external link for node
All checks were successful
/ deploy (push) Successful in 42s
Details

Browse source
This commit is contained in:
SouthFox 2024-10-09 00:57:49 +08:00
parent 0d40891025
commit 9c4411eaa2
1 changed files with 14 additions and 4 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

18
soup.py
View file

@ -7,9 +7,19 @@ from bs4 import BeautifulSoup
def re_link(link): def get_node_link(node_contend: str):
urls = re.findall(r'href=[\'"]?([^\'" >]+)', link) node_soup = BeautifulSoup(node_contend, "lxml")
return urls
temp_link = node_soup.find_all(name='a',class_='internal-link')
node_link = set()
for i in temp_link:
i = i.get("href")
if i == "":
continue
node_link.add(i)
return node_link
def new_section(soup): def new_section(soup):
@ -59,7 +69,7 @@ def soup_link():
break break
node_id_link = '/' + file_name + '/#' +node_element["id"] node_id_link = '/' + file_name + '/#' +node_element["id"]
node_dict["links"] = re_link(node_content) node_dict["links"] = get_node_link(node_content)
node_dict["id-link"] = node_id_link node_dict["id-link"] = node_id_link
node_dict["backlinks"] = [] node_dict["backlinks"] = []
nodes_dict[node_element["id"]] = node_dict nodes_dict[node_element["id"]] = node_dict