Mirrors/chatmail-core
1
0
Fork 0
mirror of https://github.com/chatmail/core.git synced 2026-04-27 18:36:30 +03:00
Code Issues Packages Projects Releases Wiki Activity

Fix #1804: remove <!doctype html> and accept invalid HTML

Browse Source 
This fixes #1804 in two ways: First, it removes a <!doctype html> from
the start of the mail, if there is any.

Then, it parses the html itself it quick-xml fails, just stripping
everything between < and >.

Both of these would have fixed this specific issue.

Also, add tests for both fixes.
This commit is contained in:
Hocuri
2020-08-19 12:20:01 +02:00
committed by holger krekel
parent f1ec1a0765
commit 1a736ca6c3
3 changed files with 185 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
src/dc_receive_imf.rs
View File

@@ -2623,4 +2623,26 @@ mod tests {
);
assert_eq!(last_msg.from_id, DC_CONTACT_ID_INFO);
}
#[async_std::test]
async fn test_html_only_mail() {
let t = TestContext::new_alice().await;
t.ctx
.set_config(Config::ShowEmails, Some("2"))
.await
.unwrap();
dc_receive_imf(
&t.ctx,
include_bytes!("../test-data/message/wrong-html.eml"),
"INBOX",
0,
false,
)
.await
.unwrap();
let chats = Chatlist::try_load(&t.ctx, 0, None, None).await.unwrap();
let msg_id = chats.get_msg_id(0).unwrap();
let msg = Message::load_from_db(&t.ctx, msg_id).await.unwrap();
assert_eq!(msg.text.unwrap(), " Guten Abend, \n\n Lots of text \n\n text with Umlaut ä... \n\n MfG [...]");
}
}

50
src/dehtml.rs
View File

@@ -25,7 +25,19 @@ enum AddText {
// dehtml() returns way too many newlines; however, an optimisation on this issue is not needed as
// the newlines are typically removed in further processing by the caller
pub fn dehtml(buf: &str) -> String {
let buf = buf.trim();
let s = dehtml_quick_xml(buf);
if !s.trim().is_empty() {
return s;
}
let s = dehtml_manually(buf);
if !s.trim().is_empty() {
return s;
}
buf.to_string()
}
pub fn dehtml_quick_xml(buf: &str) -> String {
let buf = buf.trim().trim_start_matches("<!doctype html>");
let mut dehtml = Dehtml {
strbuilder: String::with_capacity(buf.len()),
@@ -171,6 +183,24 @@ fn dehtml_starttag_cb<B: std::io::BufRead>(
}
}
pub fn dehtml_manually(buf: &str) -> String {
// Just strip out everything between "<" and ">"
let mut strbuilder = String::new();
let mut show_next_chars = true;
for c in buf.chars() {
match c {
'<' => show_next_chars = false,
'>' => show_next_chars = true,
_ => {
if show_next_chars {
strbuilder.push(c)
}
}
}
}
strbuilder
}
#[cfg(test)]
mod tests {
use super::*;
@@ -254,4 +284,22 @@ mod tests {
let txt = dehtml(input);
assert_eq!(txt.trim(), "lots of text");
}
#[test]
fn test_doctype_html() {
use crate::simplify::simplify;
let input = "<!doctype html>\n<b>fat text</b>";
let txt = simplify(dehtml(input), false).0;
assert_eq!(txt.trim(), "*fat text*");
let input = "<!some invalid html code>\n<b>some text</b>";
let txt = simplify(dehtml(input), false).0;
assert_eq!(txt.trim(), "some text");
// at least DC should show the text if the html is invalid
let input = "<This text is in brackets>";
let txt = simplify(dehtml(input), false).0;
assert_eq!(txt.trim(), "<This text is in brackets>");
}
}