#751: test failures related to src/pmc/handle.pmc
--------------------+-------------------------------------------------------
Reporter: mikehh | Owner: whiteknight
Type: cage | Status: assigned
Priority: major | Milestone: 1.3
Component: core | Version: trunk
Severity: medium | Keywords:
Lang: | Patch:
Platform: |
--------------------+-------------------------------------------------------
Comment(by jkeenan):
Replying to [comment:3 whiteknight]:
> I've been thinking about this issue since yesterday, and I think we're
going about it all wrong. Instead of blindly following the test and
creating a new t/pmc/handle.t test file, I say we fix the test to not
require it.
>
> Handle is intended to be an abstract type, and should not be
instantiated and used directly. Instead, it is inherited by other IO-
related PMC types to provide a common base. As such, it's already tested
as well as it is going to be
Whiteknight:
At the time I added ''t/pmc/handle.t'' I was unaware of the special nature
of the Handle PMC. And since it was a recent addition, I figured that
fixing the test was just another bit of cage cleaning such as is often
required after a branch is merged into trunk. I was not previously aware
of ''test_file_coverage.t'' or the coding standard policy which it implies
and enforces. On IRC, Infinoid (IIRC) noted that its original placement
in ''t/distro/'' meant that it would typically get run only once a month
as part of a release manager's pre-release run of `make fulltest`. So we
discussed moving it to ''t/codingstd/'' so that it would get run as part
of the more often run `make codetest`. And then I wrote and committed
''t/pmc/handle.t''.
Little did I know that that would prove to be the most controversial
commit I've made to Parrot in more than a year! (See
[http://tinyurl.com/nzb38j discussion on list].)
Now, one response would be to revert my commit of ''t/pmc/handle.t'' so
that ''test_file_coverge.t'' once again FAILs and so that the issue is not
papered over. I don't think that would be a good idea three days before a
release. Instead, I have opened a new TT, TT #759, which calls for a
reconsideration of the 'each PMC must have its own test file' policy. I
will leave to the release manager, ''i.e.,'' you, whether to move on that
reconsideration in the days before the release.
Thank you very much.[[BR]]
kid51
--
Ticket URL: <https://trac.parrot.org/parrot/ticket/751#comment:4>
Parrot <https://trac.parrot.org/parrot/>
Parrot Development
_______________________________________________
parrot-tickets mailing list
[email protected]
http://lists.parrot.org/mailman/listinfo/parrot-tickets
