Question

I have an HTML file that it isn’t syntactically correct, I’m parsing it with HTML Agility Pack.

But if I have a link like

<a href="http://google.com/!/!!!">Google</a>

it’s a problem, is there a possible way to detect broken links so that when an error is found (no page is available on that link) the application will store that link in a list and return it?

Same problem on tags, example:

<img hhh="jjj"/>

here the image tag is all wrong, this should be in the ‘errors for repair’ list too.

Thanks in advance.

Answer 1

You need to loop through Document.DocumentNode.Descendants("a") and check whether the href tag is bad.

Similarly, you can loop through Document.DocumentNode.Descendants("img") and check for src attributes.

EDIT:

To check for bad attributes, you can maintain a Dictionary<string, IEnumerable<string>> that maps tag names to valid attributes, then use LINQ to find missing attributes, like this:

from tag in Document.DocumentNode.Descendants()
let legalAttributes = allAttributes[tag.TagName]
from attribute in tag.Attributes
where !legalAttributes.Contains(attribute.Name, StringComparer.OrdinalIgnoreCase)
select new { Tag = tag.OuterHtml, Attribute = attribute.Name }