On Thu, Aug 30, 2001 at 07:07:42PM +0200, Tels wrote:
> Attached is a script to test all modules in a dist,
Attachment error #1
> I think a "make
> pod-cover" and "make test-cover" would be cool, but I am not able to do such
> a beast.
Me also. ExtUtils::MakeMaker is still one hell of a blackbox to me
right now, so I'll handwave over that for now.
> Having a covered() function giving a list of functions covered would be
> needed to sum all covered/uncovered up and calculate the real ratio.
Umm, the value returned by coverage is a real ratio, or is this more
for your 'cover all these' script? Actually for that you could add
them all up, and it'd be the total percentage for all packages tested,
so I'm confused.
> (OT: whats wrong with covered()/uncovered()? naked() is funny, but I rather
> have a working pair than a funny name and no counterpart because nobody
> could find a cute name :-|
It's more not there for tuits. Just after I waved 0.01 past people
they said "Okay, now I know I have 30% coverage, so what do I have to
add docs to". Other parts have been more exciting/in need of
attention. 0.05 has covered, naked stays as naked though.
> Beside that, it seems not be able to find doc paragraphs for multiply
> functions like this:
>
> =head2 get_foo/get_blah/get_baz
>
> These functions get you an item back of the number and type you specified:
>
> $object->get_foo(5); # return 5th foo
>
> Neither does getfoo, get_blah, get_baz work. I think
>
> get_foo(),get_blah()
> get_foo() and get_blah()
> getfoo()/get_blah()
>
> are all valid variants, too.
True, though I haven't personally seen those in the wild before. Will
add those cases and/or make the Pod parsing more pluggable, probably for
0.06.
> A quick hack would be to exclude any sub starting with "(" from the
> analysis, but this might not the right fix.
That's achievable now with also_private => qr/^\(/, though I'll look
and see if that can be coped with more sensibly when I get a chance.
Sorry to be deferring on so many things this time round. I'm helluva
busy at the moment and won't have a real chance to look at anything
until tomorrow night when I get back home.
--
Richard Clamp <[EMAIL PROTECTED]>
