michael/pyfedi
1
0
Fork 0
mirror of https://codeberg.org/rimu/pyfedi synced 2025-02-02 16:21:32 -08:00
Code Issues Projects Releases Packages Wiki Activity Actions

remove tracking from links fixes #34

Browse source
This commit is contained in:
rimu 2024-03-08 22:01:46 +13:00
parent 47bf0ddedb
commit 6914a47206
3 changed files with 10 additions and 7 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

4
app/activitypub/util.py
View file

@ -24,7 +24,7 @@ import pytesseract
from app.utils import get_request, allowlist_html, html_to_markdown, get_setting, ap_datetime, markdown_to_html, \
is_image_url, domain_from_url, gibberish, ensure_directory_exists, markdown_to_text, head_request, post_ranking, \
shorten_string, reply_already_exists, reply_is_just_link_to_gif_reaction, confidence, clean_link
shorten_string, reply_already_exists, reply_is_just_link_to_gif_reaction, confidence, remove_tracking_from_link
def public_key():
@ -1227,7 +1227,7 @@ def create_post(activity_log: ActivityPubLog, community: Community, request_json
post.image = image
else:
post.type = POST_TYPE_LINK
post.url = clean_link(post.url)
post.url = remove_tracking_from_link(post.url)
domain = domain_from_url(post.url)
# notify about links to banned websites.
already_notified = set() # often admins and mods are the same people - avoid notifying them twice

4
app/community/util.py
View file

@ -15,7 +15,7 @@ from app.constants import POST_TYPE_ARTICLE, POST_TYPE_LINK, POST_TYPE_IMAGE
from app.models import Community, File, BannedInstances, PostReply, PostVote, Post, utcnow, CommunityMember, Site, \
Instance, Notification, User
from app.utils import get_request, gibberish, markdown_to_html, domain_from_url, allowlist_html, \
html_to_markdown, is_image_url, ensure_directory_exists, inbox_domain, post_ranking, shorten_string, parse_page, clean_link
html_to_markdown, is_image_url, ensure_directory_exists, inbox_domain, post_ranking, shorten_string, parse_page, remove_tracking_from_link
from sqlalchemy import func
import os
@ -177,7 +177,7 @@ def save_post(form, post: Post):
post.body = form.link_body.data
post.body_html = markdown_to_html(post.body)
url_changed = post.id is None or form.link_url.data != post.url
post.url = clean_link(form.link_url.data)
post.url = remove_tracking_from_link(form.link_url.data)
post.type = POST_TYPE_LINK
domain = domain_from_url(form.link_url.data)
domain.post_count += 1

9
app/utils.py
View file

@ -260,9 +260,12 @@ def markdown_to_text(markdown_text) -> str:
def domain_from_url(url: str, create=True) -> Domain:
parsed_url = urlparse(url.lower().replace('www.', ''))
if parsed_url and parsed_url.hostname:
domain = Domain.query.filter_by(name=parsed_url.hostname.lower()).first()
find_this = parsed_url.hostname.lower()
if find_this == 'youtu.be':
find_this = 'youtube.com'
domain = Domain.query.filter_by(name=find_this).first()
if create and domain is None:
domain = Domain(name=parsed_url.hostname.lower())
domain = Domain(name=find_this)
db.session.add(domain)
db.session.commit()
return domain
@ -747,7 +750,7 @@ def sha256_digest(input_string):
return sha256_hash.hexdigest()
def clean_link(url):
def remove_tracking_from_link(url):
# strip ?si=abcDEFgh from youtu.be links
clean = re.search(r"(https://youtu.be/\w+)", url)