Community
Participate
Working Groups
Created attachment 210520 [details] example code showing a subwords bug I attached an example class with four markers where subwords completion should be triggered. $1+2 are working cases, $3+4 fail. I would expect the same result as $1 on all markers. Additionally, it is strange, that $2 works but $3 doesn't.
we are victims of JDT's way how/when/where which proposals are proposed. For parameter completions (on empty prefix), JDT only proposes the method signatures (more than one if overloaded) but no potential arguments for these methods. They are show after the first letter is typed. Here the hacky approach of Subwords completion fails because JDT per default does not make any usable completions. In order to solve this, we have to create fall back proposals for each local, method or field accessible in the current context and check their prefixes accordingly. This is definitely something we want to have some time during M6 or M7. Tagging this bug with helpwanted. If anyone is interested to fix this, contributions are welcome.
Test case for this is here: http://goo.gl/qvzyf --> def void testBug370572_2() (annotated with @Ignore ATM)
joining this wit aggregator bug for subwords *** This bug has been marked as a duplicate of bug 377373 ***