Html Agility Pack

Solution files for .net 4.5 beta and vs11 beta.
I already use .net 4.5 for prototyping some aps and 4.5 .net version is necessary for me.

Just add files in their folders from the archive

I've compiled a version of HAPLight targeting Silverlight 5 for my usage. Sending it to you if you want to make use of it.

The changes:
1. Project now targets Silverlight 5.
2. Changed the Post-build event.
3. Removed the linked HtmlNodeNavigator.cs and create another copy instead. Silverlight 5 does not support StackFrame(int, bool) constructor.

Partial fix to parse "<p>foo<p>bar" as "<p>foo</p><p>bar", so that text "foo" is the correct content of the first <p> element.

The "resetter" element names were moved to a static HtmlNode dictionary (like ElementFlags) so that HAP clients can exert some control over how auto-closing happens. There is a special entry for <p> that causes it to *always* close any previous opened <p>... no other elements are considered a resetter for it.

A more-complete fix would be to give HAP an understanding of the "allowed children" and "allowed parents" concepts in SGML/HTML, but that would be pretty pervasive. This should catch a large number of <p> parsing cases with relatively little cost and/or risk.

Added Save overload which allows the saving of a HtmlDocument as a string to use if you want the output of the HtmlDocument as a string instead of writing to a string.

Would be great if this was added to HAP so I don't have to add it again to my code every time the project is updated.

Solution files for .net 4.5 ctp.
I already use .net 4.5 for prototyping some aps and .net version is necessary for me.

Just add files in their folders from the archive

Hi,

I'd like contribute to Html Agility Pack by providing an option for the end user to select HtmlNodes based on jQuery method syntax.

As a part of it I've created an extension method called Select for HtmlDocument, so that the end user can easily select the nodes using the jQuery syntax like htmlDocument.Select("#id, .class, div");

Attached patch consist of a solution called HtmlAgilityContrib in the branches. This new feature has been created using TDD.

Currently it supports basic selectors like id, tagName, className and multiple selectors. I've been currently working on Attribute based on selection and will update that also once i've done.

Also, I've found that there is less support for HTML 5 in Html Agility Pack (Correct me if i am wrong), If I get a chance, hope i would give my best efforts towards it.

Looking forward to add some more attractive features to Html Agility Pack. You can find more about me in my blog "http://sweettam.blogspot.com"

Regards,
Tamizhvendan

The HtmlDocument does not correctly flush the StreamWriter when writing to a stream. This patch fixes that.

public void Save(TextWriter writer)
{
if (writer == null)
{
throw new ArgumentNullException("writer");
}
DocumentNode.WriteTo(writer);
writer.Flush();
}

This fixes issue 28232, I have included a passing test. Option tags where assumed to be empty however they should really have the flags Empy | Closed as <option value=12>10.00</option> is entirely possible.

FIX: The Navigator.Name returned the element name even if it was positioned on one of its attributes. It is suppposed to return the attribute name (complete with namespace prefix, if any). Unit test included.

Fixes optional end tag support and the XPathNavigator Implementation.

In more detail, auto-closes elements with an optional end tag as per the spec, so e.g. "<dd>text<dt>more text" is parsed as "<dd>text</dd><dt>more text</dt>" instead of <dd></dd>text<dt></dt>more text"

Additionally, the HtmlNavigator implementation no longer crashes on documents with xml namespaces; instead these are stripped (not ideal, but HAP doesn't support namespaces, and adding support would be more complex - for HTML this generally works).

Finally HtmlNavigator avoids double-encoding entitites, by decoding entities before including them in XPathNavigator's tree representation.