Something (JDK 7?) caused a type-mapping regression in DDR generation about pljava HOT 3 CLOSED

The line where isAssignable(tm, ktm) actually returns false for tm and ktm that both represent java.lang.String is here.

from pljava.

I did find some related reports in the Java bug database:

http://bugs.java.com/bugdatabase/view_bug.do?bug_id=8038455
http://bugs.java.com/bugdatabase/view_bug.do?bug_id=7026845
http://bugs.java.com/bugdatabase/view_bug.do?bug_id=6191665
http://bugs.java.com/bugdatabase/view_bug.do?bug_id=2206926

8038455 is written as an 'enhancement' (due for Java 9) and I didn't see anything clearly reflecting that they knew stuff had broken in 7 that worked in 6. (I also found a machine that still had 6 on it, and confirmed it's still not broken in 6.) So I made another report they will (I hope) at least check, to make sure that this issue is the same as what they're enhancing for 9....

I can't give that bug number yet ... their system never has given a number at submission time, it has to go through one review queue before they even decide it gets a number.

from pljava.

The workaround added in #42 is probably going to be the last word on this until at least Java 9. Closing.

from pljava.