Zergling_man/moment
Zergling_man/moment
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Improve HTML autolinking

Browse Source 
Allow example.com and 127.0.0.1, handle tel: and magnet:
This commit is contained in:
miruka 2019-04-17 12:06:09 -04:00
parent aa105e0c2c
commit a59c0d1572
2 changed files with 24 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
TODO.md
View File

@ -26,3 +26,5 @@
- Set Qt parents for all QObject - Set Qt parents for all QObject
- `<pre>` scrollbar on overflow - `<pre>` scrollbar on overflow
- Make links in room subtitle clickable, formatting?

23
harmonyqml/backend/html_filter.py
View File

@ -1,12 +1,23 @@
# Copyright 2019 miruka # Copyright 2019 miruka
# This file is part of harmonyqml, licensed under GPLv3. # This file is part of harmonyqml, licensed under GPLv3.
import re
import html_sanitizer.sanitizer as sanitizer import html_sanitizer.sanitizer as sanitizer
from lxml.html import HtmlElement from lxml.html import HtmlElement
from PyQt5.QtCore import QObject, pyqtProperty, pyqtSlot from PyQt5.QtCore import QObject, pyqtProperty, pyqtSlot
class HtmlFilter(QObject): class HtmlFilter(QObject):
link_regexes = [re.compile(r, re.IGNORECASE) for r in [
(r"(?P<body>.+://(?P<host>[a-z0-9._-]+)(?:/[/\-_.,a-z0-9%&?;=~]*)?"
r"(?:\([/\-_.,a-z0-9%&?;=~]*\))?)"),
r"mailto:(?P<body>[a-z0-9._-]+@(?P<host>[a-z0-9_.-]+[a-z]))",
r"tel:(?P<body>[0-9+-]+)(?P<host>)",
r"(?P<body>magnet:\?xt=urn:[a-z0-9]+:.+)(?P<host>)",
]]
def __init__(self) -> None: def __init__(self) -> None:
super().__init__() super().__init__()
self._sanitizer = sanitizer.Sanitizer(self.sanitizer_settings) self._sanitizer = sanitizer.Sanitizer(self.sanitizer_settings)
@ -15,6 +26,13 @@ class HtmlFilter(QObject):
sanitizer.normalize_overall_whitespace = lambda html: html sanitizer.normalize_overall_whitespace = lambda html: html
sanitizer.normalize_whitespace_in_text_or_tail = lambda el: el sanitizer.normalize_whitespace_in_text_or_tail = lambda el: el
# See FIXME note in sanitizer_settings
autolink_func = sanitizer.lxml.html.clean.autolink
sanitizer.lxml.html.clean.autolink = \
lambda el, **kw: autolink_func(
el, **self.sanitizer_settings["autolink"]
)
# Prevent custom attributes from being removed # Prevent custom attributes from being removed
sanitizer.lxml.html.clean.Cleaner.safe_attrs |= \ sanitizer.lxml.html.clean.Cleaner.safe_attrs |= \
self.sanitizer_settings["attributes"]["font"] self.sanitizer_settings["attributes"]["font"]
@ -51,7 +69,10 @@ class HtmlFilter(QObject):
}, },
"whitespace": {}, "whitespace": {},
"add_nofollow": False, "add_nofollow": False,
"autolink": True, "autolink": { # FIXME: arg dict not working
"link_regexes": self.link_regexes,
"avoid_hosts": [],
},
"sanitize_href": sanitizer.sanitize_href, "sanitize_href": sanitizer.sanitize_href,
"element_preprocessors": [ "element_preprocessors": [
sanitizer.bold_span_to_strong, sanitizer.bold_span_to_strong,