patch to support doctype with test

[deanmao]

Hi, earlier I made an issue, but figured it would be more efficient if I just wrote the code. This is the issue:
#47

[tautologistics]

Thanks! I am going to merge your pull but expand it to generally support XML/HTML declarations of the type "<!" and "<?". Looking into that now...

[deanmao]

Also there was another small issue with attributes that had no values,
something like this:

. The "by-zero" part caused it
to think that it needed to wait for more data and would stop parsing the
rest of the output. The commit below fixes it:

deanmao@fdcdd0a

I've been using your new 2.x parser with my fixes on lots of webpages now
-- so far no issues!

On Fri, May 4, 2012 at 6:14 AM, Chris Winberry <
reply@reply.github.com

wrote:

Thanks! I am going to merge your pull but expand it to generally support
XML/HTML declarations of the type "<!" and "<?". Looking into that now...

---

Reply to this email directly or view it on GitHub:

#48 (comment)

[CWindatt]

This fix was merged with the following commit, but this pull request remains open.
6bcd609

There is an issue with this fix that it fails to update the location correctly. It will increase the col based on the doctype data, but doesn't account for the text. To fix, add the following to the end of _updateLocation line 624

} else if (node.type === Mode.Doctype) {
this._col += 10;
}