Code Completion for Language Keywords
The keywords completion should be easy, right? ;-)
Not sure. ;-) The issue should be additionally investigated.
If a keyword has a related definite list of the code patterns that are implied by the language then these patterns
should be proposed instead of the stand-alone keyword, i.e. the code template feature should be involved instead of the
simple code completion.
E.g. code completion for the "extends" keyword should unconditionally propose input of either the super class name or a
list such names.
I see the following requirement:
Code Completion SHOULD NOT provoke creation of the syntactically incorrect code.
It was just a joke ... you can watch my slow progress in file
in mercurial. It is (very) far from complete now ... The status of this issue is "started" which is correct.
And yes I am aware that it is better to suggest nothing than something that will be syntactically incorrect.
moving opened issues from TM <= 6.1 to TM=Dev
restoring original TM, sorry for any inconvenience
Reopenning after jiriprox's action.
I am trying to hack this together before the J1. It will certainly violate the requirement
"Code Completion SHOULD NOT provoke creation of the syntactically incorrect code."
but with the current infrastructure this is the only thing I am able to do.
This has been implemented but may still contain bugs. I am closing this and will handle the bugs as separate reports -
when somebody spots them.
Perhaps this is a misunderstanding on my part but, if you type:
public class Test extends
and press ctrl+space
no completions classes are listed. This is the example listed in this issue so I am reopening. I was expecting to
see CustomNode, Frame, Application, etc...
Lark, we still have the issue that says: the source tree broken --> no completion, right? When you type the source is
broken AFAICT. Please for now try on non-broken source (green mark in the upper right hand corner).
I know that the bug with broken source is serious for for code completion feature but what can I do? You are on the Cc:
of the discussion about that one ...
I suggest to close this one for now. Thoughts?
Ok. I am going to add the Release note keyword as a placeholder. If the jira issue is fixed before release then we
don't need to releasenote it but I don't want to loose track of the limitations on this issue.
Closing - hope I understood the Ok correctly as Ok to close it as resolved (with the keyword added).