[scilab-Users] Trees in Scilab > 5.1.1

[Sylvestre Ledru]

Hello,

First, about the internal evolution of Scilab, you should send email on
the Dev Mailing list.

Le mardi 03 novembre 2009 à 21:21 +0100, Samuel Gougeon a écrit :
> Hi,
> 
> Trees of activable items (with callbacks) are a new and very nice
> item available for Scilab > 5.1.1.
> 4 functions are useful for building, displaying and printing trees:
> createNode(), createTree(), dumpTree(), and displaytree().
> 
> After trying these functions, some remarks might be done,
> as well as some questions remain. 
> IMO, there are too numerous to be posted as individual 
> bug+resquest reports, and since posting several questions
> in a single report is not correct, a post for discussion in the
> newsgroup is preferred (at least as a first step).
> 
> * About functions names : 
> IMO, capitalising the N in createNode() etc is a very good idea, 
> to make functions names more readable.
> Unfortunately, this is very exceptional in Scilab (so we have
> for instance getinstalledlookandfeels(), that looks not so pleasant
> than if it would have been getInstalledLookAndFeels() ).
> So, we should expect displayTree() instead of displaytree().
> Or, as a worse but more regular naming, createnode() instead of
> createNode(), etc.
It is true that we have a lack of consistency on the naming of the
function. We try as much as possible to avoid such issues for now.

> * About CreateTree() : This is the "branching function".
> It would be usefull to add a second argument (scalar boolean) 
> for specifying whether the branched node should by default 
> be developped (so displaying the downstream tree) or not.
Yes, that sounds reasonable (idem for foldTree). A SEP would be nice
here.

> * About displaytree() :
> returns an undocumented value, that looks simply the tree's
> handle passed in argument. Unfortunately, 
>   - no handle is returned for the window displaying the tree.
>   - winlist() does not list the new window
>   - it is possible neither to modify (naming, sizing, locating, 
>     foreground & background-colorizing etc) nor to delete the 
>     window with a script (or even by hand (except resizing and
>     positionning))
Could you submit a bug report on the undocumented value ?

> * a new foldTree(tree [, unfolded [, maxDepth] ) function could be 
>    usefull, to post-fold or post-develop a (sub)tree after displaying
> it.
>    The first argument could be either a single tree handle, or a list 
>    of handles. unfolded could be either a boolean (for assigning
> status),
>    or -1 (to switch trees' status)
> 
> Since Scilab 5.2 has not yet been released, these rems & ideas cannot 
> make an official Scilab Enhancement Proposal :-) 
> (moreover, SEP have to be posted on the dev-list, while all users may 
> discuss this here).
Please note that a SEP about the tree was be sent on February:
http://lists.scilab.org/cgi-bin/ezmlm-browse?list=dev&cmd=showmsg&msgnum=1228

Sylvestre