Skip to main content



      Home
Home » Modeling » TMF (Xtext) » [XText] Possibility to read the order of a node
[XText] Possibility to read the order of a node [message #44463] Tue, 19 May 2009 04:10 Go to next message
Eclipse UserFriend
Hi all,
is it possible to get the order (or position) where a node is located? I
need this for completion proposals.

Example:

RootModel: type name ( att1="s" cs="Attr"){
table tabname1 (attname=myatt1);
table tabname2 (attname=myatt2){ [1]
column col1 (att1 att2 . . att3 att4=value); [2]
};
};

If the cursor is at position [1] then the node name is "tabname2".
If the cursor is at position [2] then the node name is "col1". What I
want is that inside a table node (at every position) I know the right
container node name (i.e., "tabname2" if inside tabname2).
How can I achieve this?

Greets,
Ralf.
Re: [XText] Possibility to read the order of a node [message #44623 is a reply to message #44463] Wed, 20 May 2009 02:21 Go to previous messageGo to next message
Eclipse UserFriend
This is an issue which seems to have made it from oAW to TMF Xtext. See
http://www.openarchitectureware.org/forum/viewtopic.php?foru m=29&showtopic=10652&highlight=context
for an example grammar (oAW Xtext - here it is analogous).

The name of the function completeNode_children implies that there is a
safe way to access the container "Node" whose feature is to be completed.
Given the parameters (EObject model, Assignment assignment,
ContentAssistContext context, ICompletionProposalAcceptor acceptor) it is
not EObject model. In both cases model is of Type Node, but in the first
case (after the brace) model is the actual parent node (wa are after), in
the second cases model is the sibling node, not the container.

The question is: is there a unique and safe (best practice) way to access
the container node (without testing if the last parse node is a brace).
Maybe model is not the correct starting point, but assignment or context.
Re: [XText] Possibility to read the order of a node [message #44735 is a reply to message #44623] Thu, 21 May 2009 10:52 Go to previous message
Eclipse UserFriend
Hi Alex,

I think you are talking about this one:
https://bugs.eclipse.org/bugs/show_bug.cgi?id=276619 ? The bug will be
fixed asap.

Regards,
Sebastian

Am 20.05.2009 8:21 Uhr, schrieb Alexander Nittka:
> This is an issue which seems to have made it from oAW to TMF Xtext. See
> http://www.openarchitectureware.org/forum/viewtopic.php?foru m=29&showtopic=10652&highlight=context
> for an example grammar (oAW Xtext - here it is analogous).
>
> The name of the function completeNode_children implies that there is a
> safe way to access the container "Node" whose feature is to be
> completed. Given the parameters (EObject model, Assignment assignment,
> ContentAssistContext context, ICompletionProposalAcceptor acceptor) it
> is not EObject model. In both cases model is of Type Node, but in the
> first case (after the brace) model is the actual parent node (wa are
> after), in the second cases model is the sibling node, not the container.
>
> The question is: is there a unique and safe (best practice) way to
> access the container node (without testing if the last parse node is a
> brace). Maybe model is not the correct starting point, but assignment or
> context.
>
Previous Topic:Migrating from Oaw Xtext to TMF Xtext
Next Topic:[Xtext] Can any compare Xtext with TCS?
Goto Forum:
  


Current Time: Sat Jul 19 11:03:52 EDT 2025

Powered by FUDForum. Page generated in 0.05389 seconds
.:: Contact :: Home ::.

Powered by: FUDforum 3.0.2.
Copyright ©2001-2010 FUDforum Bulletin Board Software

Back to the top