Community
Participate
Working Groups
The code completion should ignore the case while typing. If I have a class MyClass and type "myc" and then STRG+Space the codecompletion should provide MyClass for selection as well.
I agree. Unfortunately our new DOM framework did not have that capability in CDT 3.0. We'll definitely take another look in 3.1.
Maybe not...
marking as FIXED
Re-opening - not all content assist cases related to this are passing. Bryan Wilkinson said: >I just had one concern regarding PDOMCPPNamespace.find(String name, boolean >prefixLookup). The visitor created in this method is set to always be case >sensitive. For content assist at least, case-sensitivity would need to depend >on whether or not it is a prefix lookup. So passing '!prefixLookup' instead of >'true' to the visitor's contructor would work for content assist, or else >another boolean parameter for the find(...) method would be needed. > >In fact, are there any situations in which we need to control case-sensitivity >independently of whether or not a prefix lookup is taking place? If this is >the case then it might be wise to remove any 'caseSensitive' booleans and >simply make case-sensitivity depend on the 'prefixLookup' booleans which >already existed. Bryan - if you're happy to fix up the remaining content assist cases I'll take a look at the patches. Otherwise, I'll try to take a look at this after my indexing work is complete. I've no strong feelings about whether caseSensitivity needs independent control (the OpenTypeDialog always does case-insensitive lookup now) thanks, Andrew
Fixing up the remaining content assist cases shouldn't be a problem. I'll hopefully have a patch ready soon.
Created attachment 59705 [details] proposed patch Patch ensures that all means used to search for bindings for content assist are case-insensitive.
verified and applied to HEAD, thanks!
I suppose this one can be closed now.
I've left it open as there are still some unit tests that reference this bug, which are marked as expected to fail (setExpectFailure) I've not looked into these though
searching for this bug's number in the tests shows these two are still marked to fail (?) org.eclipse.cdt.ui.tests.text.contentassist2.CompletionTest_NewTypeReference_Prefix org.eclipse.cdt.ui.tests.text.contentassist2.CompletionTest_SingleName_Method_Prefix
(In reply to comment #10) Thanks for the reminder, I forgot to update them. In fact there are three tests which actually pass: org.eclipse.cdt.ui.tests.text.contentassist2.CompletionTest_NewTypeReference_NoPrefix org.eclipse.cdt.ui.tests.text.contentassist2.CompletionTest_NewTypeReference_Prefix org.eclipse.cdt.ui.tests.text.contentassist2.CompletionTest_SingleName_Method_Prefix At least they are not (no longer) related to this bug.
assigning
done
