Boost logo

Boost Testing :

From: Victor A. Wagner Jr. (vawjr_at_[hidden])
Date: 2005-05-01 18:17:50

At Sunday 2005-05-01 15:19, you wrote:
>Martin Wille wrote:
>>What we need is some indication about
>> 1. when a test was run
>I don't see how that helps?

but the person on whose machine _may_ know (maybe the cat ran across the
keyboard at that time and the tests needed to be restarted...and perhaps
something glitched).
There seems to be an presumption by many that the regression tests just
sort of automatically take care of themselves and no human is actually
involved with their running.
"They just magically appear, for free, doesn't cost anyone any time,
effort, or money"
Just because it's obvious that my regression tests are run by a scheduler,
any assumption that I don't take time to seee that they're running properly
or that the results look ok would be completely unwarranted. I run the
regressions on a machine that I use to do development. I use the new
libraries created in my work for my clients (yeah, sometimes risky but we
get a lot more testing that way)

I cannot speak for any of the other regression runners, but I have chatted
extensively (IRC) with Martin Wille about how the test are running, what
problems the tests themselves seem to have and is there perhaps something
wrong with the way _WE_ (the testers) have perhaps set up our testing

>> 3. How many CPU time, RAM, etc did the run take
>Yep.. And also handle displaying stack traces of aborts, time bounds, be a
>debugger, etc.. It's been mentioned before. And there was someone, can't
>remember now, who was willing to work on such a monitor.

I'd settle for WALL CLOCK time for the element to run.

>-- Grafik - Don't Assume Anything
>-- Redshift Software, Inc. -
>-- rrivera/ - grafik/
>-- 102708583/icq - grafikrobot/aim - Grafik/
>Boost-Testing mailing list

Victor A. Wagner Jr.
The five most dangerous words in the English language:
               "There oughta be a law"

Boost-testing list run by mbergal at