Matrix-Mirrors/anonymousland-synapse
1
0
Fork 0
mirror of https://git.anonymousland.org/anonymousland/synapse.git synced 2025-06-01 13:34:18 -04:00
Code Issues Projects Releases Packages Wiki Activity

Clean-up logic for rebasing URLs during URL preview. (#12219)

Browse source 
By using urljoin from the standard library and reducing the number
of places URLs are rebased.
This commit is contained in:
Patrick Cloke 2022-03-16 07:21:36 -04:00 • committed by GitHub
parent dda9b7fc4d
commit 4587b35929
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
4 changed files with 26 additions and 91 deletions
Download patch file Download diff file Expand all files Collapse all files

39
synapse/rest/media/v1/preview_html.py
View file

@ -16,7 +16,6 @@ import itertools
import logging
import re
from typing import TYPE_CHECKING, Dict, Generator, Iterable, Optional, Set, Union
from urllib import parse as urlparse
if TYPE_CHECKING:
from lxml import etree
@ -144,9 +143,7 @@ def decode_body(
return etree.fromstring(body, parser)
def parse_html_to_open_graph(
tree: "etree.Element", media_uri: str
) -> Dict[str, Optional[str]]:
def parse_html_to_open_graph(tree: "etree.Element") -> Dict[str, Optional[str]]:
"""
Parse the HTML document into an Open Graph response.
@ -155,7 +152,6 @@ def parse_html_to_open_graph(
Args:
tree: The parsed HTML document.
media_url: The URI used to download the body.
Returns:
The Open Graph response as a dictionary.
@ -209,7 +205,7 @@ def parse_html_to_open_graph(
"//*/meta[translate(@itemprop, 'IMAGE', 'image')='image']/@content"
)
if meta_image:
og["og:image"] = rebase_url(meta_image[0], media_uri)
og["og:image"] = meta_image[0]
else:
# TODO: consider inlined CSS styles as well as width & height attribs
images = tree.xpath("//img[@src][number(@width)>10][number(@height)>10]")
@ -320,37 +316,6 @@ def _iterate_over_text(
)
def rebase_url(url: str, base: str) -> str:
"""
Resolves a potentially relative `url` against an absolute `base` URL.
For example:
>>> rebase_url("subpage", "https://example.com/foo/")
'https://example.com/foo/subpage'
>>> rebase_url("sibling", "https://example.com/foo")
'https://example.com/sibling'
>>> rebase_url("/bar", "https://example.com/foo/")
'https://example.com/bar'
>>> rebase_url("https://alice.com/a/", "https://example.com/foo/")
'https://alice.com/a'
"""
base_parts = urlparse.urlparse(base)
# Convert the parsed URL to a list for (potential) modification.
url_parts = list(urlparse.urlparse(url))
# Add a scheme, if one does not exist.
if not url_parts[0]:
url_parts[0] = base_parts.scheme or "http"
# Fix up the hostname, if this is not a data URL.
if url_parts[0] != "data" and not url_parts[1]:
url_parts[1] = base_parts.netloc
# If the path does not start with a /, nest it under the base path's last
# directory.
if not url_parts[2].startswith("/"):
url_parts[2] = re.sub(r"/[^/]+$", "/", base_parts.path) + url_parts[2]
return urlparse.urlunparse(url_parts)
def summarize_paragraphs(
text_nodes: Iterable[str], min_size: int = 200, max_size: int = 500
) -> Optional[str]: