Re: [xml] [Proposal] How about write a book for LibXML2?

2008/11/19 Daniel Veillard <veillard redhat com>:
On Tue, Nov 11, 2008 at 03:43:13PM +0100, Elvis Stansvik wrote:
2008/11/8 Elvis Stansvik <elvstone gmail com>:
Just to throw something out there, I sketched out a preliminary layout
that could be used (and of course improved upon) as a skeleton for a
DocBook or Wiki:

I have now taken the liberty and set up a MediaWiki installation and
started adding some content. It's available at:

 Okay, that looks really nice ! Very good idea !
A couple of comments:
 - in the utility APIs, I think the URI parsing is generally useful


 - for the python bindings that should be split into the default
  bindings and lxml


I'm blocking search engine robots in robots.txt at the moment.

As you can see I've styled the wiki to look similar to If
you want the regular MonoBook skin, just change it in the preferences.

I have started to reconsider the structure of the main content (yet to
be written). Maybe it should be more task oriented and not sectioned
by the different APIs?

 well if you want to get content fast and avoid redundancy it's
good to be layered the way it is done now. Then later some reshuffling
could be done to be more user mindset (i.e. task) oriented.

That's true.

I also added created a tag extension so you can use e.g.
<api>xmlNode</api> and it will turn into a link to the libxml2 API
documentation. This extension uses information from the
libxml2-api.xml generated from libxml2 2.7.2 at the moment. I'm also
planning on patching the syntax highlighting extension I'm using to
automatically identify libxml2 symbols within <source /> tags and turn
them into API documentation links.


If anyone feel like helping out, just register and edit away! A simple
thing is to add installation instructions for your platform, as I've
only added it for the platforms I'm familar with.

Hope to get some response on this!

Daniel, what is running on? Would you consider again
making a for this? I'd be willing to maintain the
MediaWiki installation and handle security updates to it, as well as
combat any SPAM problems that might arise. The server I'm running this
on now is on a dynamic IP DSL line in Sweden, which is not really
optimal. is running on a RHEL4 box, not the best to handle things
like a Wiki, and also hosting a few other projects I care for. I would
prefer if the wWiki could be hosted somewhere else, but I'm fine adding
a DNS record pointing to the Wiki, and links.

I see. I've started looking into hosting it on our company web hotel,
and it looks like it should be able to run an up-to-date MediaWiki
installation. I've contacted our tech support there to see if they
would allow us to do Apache URL rewriting for pretty URLs they way I
have done at the current installation. They're usually very quick in

 Also I think the Wiki should be seen as the content editing side of
the project, it could regulary be imported into SVN doc and hence
available (statically) in That way the only traffic needed
for the wiki side should be for the people editing, the main traffic
would still be on the site.

A good idea. I'll look into the best export possibilities from
MediaWiki. A simple solution of the top of my head would be just
taking the HTML and make a small customization (if any at all) of the
default printable CSS. It looks pretty good.

 One of my reluctance on Wiki is that you only get transcient
informations, the state can change dramatically and in general it's hard
to ensure you won't loose good content. That's why I prefer a dual view
one for edit and regular imports (and checking at that stage) onto more
static content.

A valid concern. I'd like this setup too.

 Opinions ?

 thanks a lot !

No problem. I'll get back to you when I have a destination for a A record.



Daniel Veillard      | libxml Gnome XML XSLT toolkit
daniel veillard com  | Rpmfind RPM search engine | virtualization library

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]