2009/3/5 Parrot <[email protected]>:
> #401: pbc tools have inadequate test coverage
> --------------------+-------------------------------------------------------
>  Reporter:  cotto   |       Owner:
>     Type:  cage    |      Status:  new
>  Priority:  normal  |   Milestone:
> Component:  none    |     Version:
>  Severity:  medium  |    Keywords:
>     Lang:          |       Patch:
>  Platform:          |
> --------------------+-------------------------------------------------------
>  pbc_dump and pbc_disassemble were broken for the time between r37063 and
>  r37113.  They're working (afaict), but it disturbs me that breaking these
>  tools didn't cause any failures in make test (or make fulltest).
>
>  If these tools are worth maintaining (and I suspect so), they need to get
>  some increased test coverage.  I was able to segfault both tools by
>  running them on pbc_to_exe.pbc in Parrot's root dir, though they both
>  worked fine against a the pbc generated from a pir hello world.  It does
>  appear that both tools have some basic coverage, but we could really do
>  better, especially given all the pbc that's generated as part of Parrot's
>  build process.
>
>  Unfortunately I'm not sure how to make this into a closeable ticket.  A
>  good start would be that make test should fail if any of the following
>  lines are changed to something goofy:
>  * the GETATTR_Key_next_key(inteprp, key, key) lines in src/packdump.c
>  * the GETATTR_Key_next_key(interp, k, k); in src/debug.c
>
>  Anything too far beyond that will require someone with more knowledge of
>  these tools than me.

My patches to add tests for these tools were rejected so I'm pissed
off adding them again.
Someone else, sorry.
-- 
Reini
_______________________________________________
http://lists.parrot.org/mailman/listinfo/parrot-dev

Reply via email to