Replies: 2 comments 3 replies
-
|
I would need a test case to investigate this. |
Beta Was this translation helpful? Give feedback.
-
|
Try this test, without the fix, it will throw an error: [Fact]
public void When_HtmlBody_Contains_InvalidTags()
{
var visitor = new HtmlPreviewVisitor();
var message = new MimeMessage();
var body = new TextPart(TextFormat.Html) { Text = "<html><body>Hello, World!<mailto:[email protected].< p=\"\"></p> </body></html>" };
message.Body = body;
message.Accept(visitor);
Assert.Contains("Hello, World!", visitor.HtmlBody);
} |
Beta Was this translation helpful? Give feedback.
-
|
@jstedfast did you check the test? |
Beta Was this translation helpful? Give feedback.
-
|
I haven't had a chance to look at this yet. |
Beta Was this translation helpful? Give feedback.
-
|
Turns out that '<' is a valid tag name character according to the HTML5 tokenizer specification, so the tokenizer was happily gobbling that character up into the tag name but HtmlWriter's logic for validating tag names assumed '<' was an invalid tag name character. shrug I've updated the code to allow that in a tag name, so the next (4.11.0) release will not throw an exception there. In the meantime, your workaround is probably as good as any. |
Beta Was this translation helpful? Give feedback.
-
Hi @jstedfast,
I encountered an issue recently with one of my clients regarding email parsing. Upon reviewing the email body, I discovered it contained invalid HTML tags, such as:
This invalid tag caused the HtmlPreviewVisitor to throw an error during the parsing of the HTML body:
To address this, I implemented a fix by adding a try...catch block at the end of the HtmlTagCallback method to ignore invalid tags:
Could you please share your thoughts on this fix? I appreciate your feedback.
Best regards,
Mahmood
Beta Was this translation helpful? Give feedback.
All reactions