Bug 291419 - CCompletionProposal should include the ICElement so integrations can easily use the results from the DOMCompletionProposalComputer
Summary: CCompletionProposal should include the ICElement so integrations can easily u...
Status: NEW
Alias: None
Product: CDT
Classification: Tools
Component: cdt-editor (show other bugs)
Version: 6.0   Edit
Hardware: PC Windows Vista
: P3 normal (vote)
Target Milestone: ---   Edit
Assignee: Project Inbox CLA
QA Contact: Jonah Graham CLA
URL:
Whiteboard:
Keywords:
Depends on:
Blocks: 291421
  Show dependency tree
 
Reported: 2009-10-05 16:30 EDT by Shawn Minto CLA
Modified: 2020-09-04 15:26 EDT (History)
1 user (show)

See Also:


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Shawn Minto CLA 2009-10-05 16:30:25 EDT
When trying to create a new completion proposal processor to handle ranking the content assist by interest for the Mylyn integration, the ICElement was needed, but I was unable to get it from the proposal.  This element should be included as it is in JDT so integrations can take advantage of the easily accessible information on a ICElement when computing the proposals that they want to display.
Comment 1 Anton Leherbauer CLA 2009-10-09 07:48:43 EDT
Computing the ICElement for a proposal is expensive.  Even JDT computes this information only lazily when a proposal is selected in the proposal popup (LazyJavaCompletionProposal).
I.e. Filtering proposals based on associated ICElement is at least sub-optimal performance-wise.
Comment 2 Shawn Minto CLA 2009-10-09 11:54:30 EDT
If this is sub-optimal, is there a good way to compare a handle from and ICElement to and IBinding?
Comment 3 Nathan Ridge CLA 2017-01-10 03:00:02 EST
(In reply to Shawn Minto from comment #2)
> is there a good way to compare a handle from and ICElement to and IBinding?

I know the implementation of some CDT features such as type hierarchy and call hierarchy uses IndexUI.elementToBinding() to convert an ICElement to an IBinding.

I don't really have a good handle on whether this method is performant enough to run on each proposal during content assist.

If you're still interested in this, could you describe your use case a bit more?