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

Have return symbols in inline filtered html

Browse Source 
Before: "foo\nbar" was rendered as "foobar"
Now: "foo\nbar" is rendered as "foo ⏎ bar" (U+23CE unicode).
This commit is contained in:
miruka 2019-10-24 09:08:10 -04:00
parent 756edc90dd
commit 39b4706c75
1 changed files with 37 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

47
src/python/html_filter.py
View File

@ -26,6 +26,14 @@ class MarkdownInlineLexer(mistune.InlineLexer):
class HtmlFilter: class HtmlFilter:
inline_tags = {"font", "a", "sup", "sub", "b", "i", "s", "u", "code"}
block_tags = {
"h1", "h2", "h3", "h4", "h5", "h6","blockquote",
"p", "ul", "ol", "li", "hr", "br",
"table", "thead", "tbody", "tr", "th", "td", "pre",
}
link_regexes = [re.compile(r, re.IGNORECASE) for r in [ link_regexes = [re.compile(r, re.IGNORECASE) for r in [
(r"(?P<body>[a-zA-Z\d]+://(?P<host>[a-z\d._-]+(?:\:\d+)?)" (r"(?P<body>[a-zA-Z\d]+://(?P<host>[a-z\d._-]+(?:\:\d+)?)"
r"(?:/[/\-_.,a-z\d#%&?;=~]*)?(?:\([/\-_.,a-z\d#%&?;=~]*\))?)"), r"(?:/[/\-_.,a-z\d#%&?;=~]*)?(?:\([/\-_.,a-z\d#%&?;=~]*\))?)"),
@ -41,7 +49,7 @@ class HtmlFilter:
re.MULTILINE, re.MULTILINE,
) )
newlines_regex = re.compile(r"\n(\n*)") extra_newlines_regex = re.compile(r"\n(\n*)")
def __init__(self) -> None: def __init__(self) -> None:
@ -99,12 +107,8 @@ class HtmlFilter:
# https://matrix.org/docs/spec/client_server/latest#m-room-message-msgtypes # https://matrix.org/docs/spec/client_server/latest#m-room-message-msgtypes
# TODO: mx-reply and the new hidden thing # TODO: mx-reply and the new hidden thing
inline_tags = {"font", "a", "sup", "sub", "b", "i", "s", "u", "code"} inline_tags = self.inline_tags
tags = inline_tags | { all_tags = inline_tags | self.block_tags
"h1", "h2", "h3", "h4", "h5", "h6","blockquote",
"p", "ul", "ol", "li", "hr", "br",
"table", "thead", "tbody", "tr", "th", "td", "pre",
}
inlines_attributes = { inlines_attributes = {
"font": {"color"}, "font": {"color"},
@ -117,7 +121,7 @@ class HtmlFilter:
}} }}
return { return {
"tags": inline_tags if inline else tags, "tags": inline_tags if inline else all_tags,
"attributes": inlines_attributes if inline else attributes, "attributes": inlines_attributes if inline else attributes,
"empty": {} if inline else {"hr", "br"}, "empty": {} if inline else {"hr", "br"},
"separate": {"a"} if inline else { "separate": {"a"} if inline else {
@ -145,6 +149,7 @@ class HtmlFilter:
self._process_span_font, self._process_span_font,
self._img_to_a, self._img_to_a,
self._remove_extra_newlines, self._remove_extra_newlines,
self._newlines_to_return_symbol if inline else lambda el: el,
], ],
"element_postprocessors": [], "element_postprocessors": [],
"is_mergeable": lambda e1, e2: e1.attrib == e2.attrib, "is_mergeable": lambda e1, e2: e1.attrib == e2.attrib,
@ -182,9 +187,31 @@ class HtmlFilter:
if el.tag != "pre" and not pre_parent: if el.tag != "pre" and not pre_parent:
if el.text: if el.text:
el.text = self.newlines_regex.sub(r"\1", el.text) el.text = self.extra_newlines_regex.sub(r"\1", el.text)
if el.tail: if el.tail:
el.tail = self.newlines_regex.sub(r"\1", el.tail) el.tail = self.extra_newlines_regex.sub(r"\1", el.tail)
return el
def _newlines_to_return_symbol(self, el: HtmlElement) -> HtmlElement:
# Add a return unicode symbol (U+23CE) to blocks with siblings
# (e.g. a <p> followed by another <p>) or <br>.
# The <br> themselves will be removed by the inline sanitizer.
is_block_with_siblings = (el.tag in self.block_tags and
next(el.itersiblings(), None) is not None)
if el.tag == "br" or is_block_with_siblings:
el.tail = f" ⏎ {el.tail or ''}"
# Replace left \n in text/tail of <pre> content by the return symbol.
if el.text:
el.text = re.sub(r"\n", r" ⏎ ", el.text)
if el.tail:
el.tail = re.sub(r"\n", r" ⏎ ", el.tail)
return el return el