gnustep-dev
[Top][All Lists]
Advanced

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

Re: NSXML* classes changed to use libxml2 directly... Should we merge th


From: Ivan Vučica
Subject: Re: NSXML* classes changed to use libxml2 directly... Should we merge this into the trunk?
Date: Mon, 23 Jan 2012 18:32:43 +0100

I'm currently working on an app that makes heavy use of NSXML*, with
the final goal of porting to Linux so I'm eager to try this out, and
can't wait for the code to make its way into trunk. Probably
post-FOSDEM?

If you want to test with my code, it is available here:
 http://bitbucket.org/ivucica/opml
Example file to try this out with:
 http://static.curry.com/worldoutline/admin/2012/01/06/archive149.opml

Some details on the code:
Editor was written upon request from Adam Curry who makes heavy use of
a platform called Frontier, which currently doesn't exist for Linux
and doesn't seem portable. This editor is an outliner, and edits .opml
files. These files are basically a way to store text content in
outlines, as well as to "include" other outlines over HTTP. This
editor already contains some GNUstep-related hacks, and opens a blank
file. None of hacks are related to my use of NSXML*, since I never got
it to work with NSXML*; previous GNUstep trunk I tested with died in
NSDocument's initWithData:, apparently because that method wasn't
implemented. Note that GNUstep version of the code currently makes use
of .xib files instead of .gorm or .nib files.

There is no way to add new nodes, to change level of nodes in the
outline. There is also no way to change outline node type nor outline
node attributes, such as the included outline's URL.

Latest committed version also reportedly crashes on OS X upon saving,
but I haven't worked on it since mailing the last build to Adam Curry
so I haven't confirmed this.

23. 1. 2012., u 18:17, Gregory Casamento <address@hidden> napisao:

> Hey guys,
>
> I originally implemented the NSXML(Document,Element,Node,etc) classes
> using NSXMLParser and keeping the information in arrays and
> dictionaries.   This was okay, up to a point, but would never work for
> XPath and XQuery functions unless you were willing to re-implement
> XPath.   By leveraging libxml2 it allows us to use any and all other
> libraries which use libxml2 to do operations on XML content.
>
> For this reason it was decided to change the implementation to use
> libxml2.  Currently the branch I created to develop these changes
> (nsxml_libxml2_changes) is much more mature and functional than the
> code that is currently on the trunk.
>
> I am soliciting the opinion of the group regarding adding this code
> into the trunk once it is ready so that people don't look at the
> already obsolete code on the trunk and begin making changes that are
> not relavant to the current direction it is going.
>
> Please let me know what all of you think.
>
> Thanks, GC
> --
> Gregory Casamento
> Open Logic Corporation, Principal Consultant
> yahoo/skype: greg_casamento, aol: gjcasa
> (240)274-9630 (Cell)
> http://www.gnustep.org
> http://heronsperch.blogspot.com
>
> _______________________________________________
> Gnustep-dev mailing list
> address@hidden
> https://lists.gnu.org/mailman/listinfo/gnustep-dev



reply via email to

[Prev in Thread] Current Thread [Next in Thread]