[email protected] wrote:
On Mon, Jan 11, 2010 at 02:06:11PM -0800, Brock Pytlik wrote:
[email protected] wrote:
On Mon, Jan 11, 2010 at 01:40:01PM -0800, Brock Pytlik wrote:
[email protected] wrote:
On Mon, Jan 11, 2010 at 01:20:00PM -0800, Brock Pytlik wrote:
Webrev:
http://cr.opensolaris.org/~bpytlik/ips-13791-v1/
This looks okay. You're letting the fd for stdout remain the one that
was inherited from the parent process, though. Is it necessary to print
the command that failed if everything else is getting logged to stdout
for the test suite already?
-j
There's sample output in the bug. To me, it seemed appropriate, but
if you'd like something different, please let me know. I thought
printing the exact command would be helpful instead of trying to
reconstruct it from the test or the errors.
What I'm trying to ask is any of this output going to be duplicated if
the test framework's logging is already grabbing stdout?
-j
No, I don't think so. At least I didn't see any duplicated output
when I triggered this exception. I don't get why you think it would,
but that's ok.
This wasn't a rhetorical question, so perhaps that's why you're confused
about why I asked. I'm not entirely certain what output the test suite
saves, and where it saves it. That's why I asked.
-j
I'm not either. I don't see any possible way the stdout in this
situation could be duplicated. It's possible that it might be lost. If
the stdout could be useful for some reason, I'll spend time worrying
about it. The stderr, as far as I can tell, is saved exactly in one
place then immediately shown to the user via the exception. If this
doesn't clear things up, let's talk in person as this seems to be going
around in circles. :)
Brock
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
