XML-RPC client module for Node.js (uses libxml.js)

github.com

27 points·caludio··7 comments

I notice this client is based on libxml.js.

Does anyone know if anyone is working on a "flexible" HTML/XHTML parser a la BeautifulSoup / Nokogiri / TagSoup, etc? Node could become very useful as a base for building scrapers if this existed.

libxml2 has an excellent tag soup parser (the Python lxml module makes heavy use of it) - but I don't think libxml.js exposes it yet.

As an alternative to libxml2 one could use tagsoup as a separate process and a JS only parser like XML for script.

I've been trying to model libxml.js after Nokogiri. I wanted to get something built and working first. The next step is to expose libxml2's html parser.

Someone else has started working on find-by-CSS a la Nokogiri. I'll merge that into libxml.js when it's ready.

BTW, I'm looking for more help on this project. A new job has diminished the amount of time I can spend on OSS projects.

Hi, thanks for working on this.

I went ahead and wrote a very simple test for HTML parsing and added an parseHTML() binding. I've sent you a pull request.

You can check out the commit here:

http://github.com/Maciek416/libxmljs/commit/87f28186cd4d2f2f...

Why use libxml when JavaScript already has a standard XML API, E4X (ECMAScript for XML), as specified by ECMA 357? At least libxml should use the faster native XML support behind the scenes if available.

Here's your answer:

http://code.google.com/p/v8/issues/detail?id=235

"There are currently no plans for implementing E4X in V8."

This doesn't seem to have moved forward, and the thread ends in a link to libxmljs :)

Reply on news.ycombinator.com