Open GoogleCodeExporter opened 9 years ago
The log shows that I used version 1.8.9. Turns out this is because I used an
old checkout of the code and performed only an `hg pull`, without an `hg
update`. I just *properly* updated the working directory and re-ran the test
using version 1.8.11. Can confirm the ticket is still valid.
Original comment by stephan...@gmail.com
on 4 Mar 2015 at 9:52
Thanks for the bug report! I verified that the issue still occurs in our
working repository.
Original comment by wdi...@gmail.com
on 17 Mar 2015 at 4:30
Original comment by Jonathan...@gmail.com
on 8 Apr 2015 at 9:55
Original comment by jtha...@cs.washington.edu
on 18 Apr 2015 at 6:42
This test case also illustrates a type argument inference problem with multiple
dependencies.
Re-opened the issue, as the test case is @skipped.
Original comment by wdi...@gmail.com
on 27 May 2015 at 7:34
It looks like this test previously passed before we fixed the JDK/Issue12 problems (as of the time of this writing the JDK/Issue12 fix is being passed around as a bundle).
When type argument inference fails on Issue404 (for the reasons stated in other
emails), it infers a wildcard for the type arguments that could not be
inferred. The bound annotations of this wildcard come from the type parameter
to which its an argument. The annotations on the type parameter in this case
come from the JDK and have changed (correctly) because of the Issue12/JDK fixes.
So, while the error currently emitted when using the Issue12 bundle should not
be emitted in the long run, it should be until we fix type argument inference
to handle these particular cases.
Original comment by jbu...@cs.washington.edu
on 27 May 2015 at 9:28
Original issue reported on code.google.com by
stephan...@gmail.com
on 28 Feb 2015 at 10:53Attachments: