On 07/19/2012 12:13 AM, Nguyen Anh Kien wrote:
The Navigation and Node API:
To get a node, I must use: *navigation.getNode("foo", "bar",
"foobar")*
IMO, we should support get node by URI string param like
*navigation.getNode("foo/bar/foobar")*. Because we often has an URI of
node by node.getURI();.
The node.getURI() method is slightly different now and is used to
construct the URL to the navigation node (relative). So it would render
the site name as well as the "node uri". However I don't like the method
name, nor am I sure it's exactly what we want.
However I do think it's good to remove any internal separators even tho
it's very common for tree based structures to use the '/' character. You
have the String[] value of the node by doing node.getId().getPath(). I
might add getPath on the node object itself so it's more clear. I would
like to remove the dependency of the "/foo/bar" notation but I'll think
about it and if it's that important we can support it.
Also, *removeNode* and *addNode* methods should be the same
behaviour.
How do you mean ? Same return signature ? I'm not sure what the
implications are if I return the node that was removed since the
underlying impl is connected.
For a Node, We often has cloneable tasks for node, so it should
support clone() method in Node api
regards,
=============
Kien Nguyen
M&E: (+84) 933.975.960 - kienna(a)exoplatform.com
<mailto:kien.nguyen@exoplatform.com>
Skype & yahoo: nakien2a
Portal Team - eXo Platform South East Asia (
http://www.exoplatform.com
<
http://www.exoplatform.com/>)