paulmataruso/zulip
1
0
Fork 0
mirror of https://github.com/zulip/zulip.git synced 2025-11-03 13:33:24 +00:00
Code Issues Packages Projects Releases Wiki Activity

markdown: Rewrite YouTube URL parser without regex spaghetti.

Browse Source 
This also adds support for the new YouTube Shorts URLs.

Signed-off-by: Anders Kaseorg <anders@zulip.com>
This commit is contained in:
Anders Kaseorg
2023-02-09 15:45:11 -08:00
committed by Tim Abbott
parent 53aa3f6c71
commit 0a1904a6a7
2 changed files with 27 additions and 27 deletions
Download Patch File Download Diff File Expand all files Collapse all files

48
zerver/lib/markdown/__init__.py
View File

@@ -26,7 +26,7 @@ from typing import (
TypeVar,
Union,
)
from urllib.parse import urlencode, urljoin, urlsplit
from urllib.parse import parse_qs, urlencode, urljoin, urlsplit
from xml.etree.ElementTree import Element, SubElement
import ahocorasick
@@ -811,28 +811,30 @@ class InlineInterestingLinkProcessor(markdown.treeprocessors.Treeprocessor):
def youtube_id(self, url: str) -> Optional[str]:
if not self.zmd.image_preview_enabled:
return None
# YouTube video id extraction regular expression from https://pastebin.com/KyKAFv1s
# Slightly modified to support URLs of the forms
# - youtu.be/<id>
# - youtube.com/playlist?v=<id>&list=<list-id>
# - youtube.com/watch_videos?video_ids=<id1>,<id2>,<id3>
# If it matches, match.group(2) is the video id.
schema_re = r"(?:https?://)"
host_re = r"(?:youtu\.be/|(?:\w+\.)?youtube(?:-nocookie)?\.com/)"
param_re = (
r"(?:(?:(?:v|embed)/)"
r"|(?:(?:(?:watch|playlist)(?:_popup|_videos)?(?:\.php)?)?(?:\?|#!?)(?:.+&)?v(?:ideo_ids)?=))"
)
id_re = r"([0-9A-Za-z_-]+)"
youtube_re = r"^({schema_re}?{host_re}{param_re}?)?{id_re}(?(1).+)?$"
youtube_re = youtube_re.format(
schema_re=schema_re, host_re=host_re, id_re=id_re, param_re=param_re
)
match = re.match(youtube_re, url)
# URLs of the form youtube.com/playlist?list=<list-id> are incorrectly matched
if match is None or match.group(2) == "playlist":
return None
return match.group(2)
id = None
split_url = urlsplit(url)
if split_url.scheme in ("http", "https"):
if split_url.hostname in (
"m.youtube.com",
"www.youtube.com",
"www.youtube-nocookie.com",
"youtube.com",
"youtube-nocookie.com",
):
query = parse_qs(split_url.query)
if split_url.path in ("/watch", "/watch_popup") and "v" in query:
id = query["v"][0]
elif split_url.path == "/watch_videos" and "video_ids" in query:
id = query["video_ids"][0].split(",", 1)[0]
elif split_url.path.startswith(("/embed/", "/shorts/", "/v/")):
id = split_url.path.split("/", 3)[2]
elif split_url.hostname == "youtu.be" and split_url.path.startswith("/"):
id = split_url.path[len("/") :]
if id is not None and re.fullmatch(r"[0-9A-Za-z_-]+", id):
return id
return None
def youtube_title(self, extracted_data: UrlEmbedData) -> Optional[str]:
if extracted_data.title is not None: