Rob Cozens wrote:
Richard, et al:
Do any of you do automated testing? If so, what do you test and how?
The SDB Test Client stack that ships with Serendipity Library has an
automated testing mode, where SDB database commands are called at
random and a log of the last 25 calls and their results is maintained.
Before releasing a major change to SDB, I run automated testing in
single user mode for 24 hours, server & one client for 24 hours, and
then server and multiple clients for at least another 24 hours.
Cool stuff. At Mark Wieder's suggestion I've also been looking at
Eggplant, which is driven by SenseTalk, an xTalk originally developed
for NeXT.
Interesting possibilities....
I have mentioned before that automated testing led me to a bug that
would have been very hard to find otherwise:
Command A could be called all day with no problem
Command B could be called all day with no problem -- so long has
Command A had not been called.
Once Command A was called, the next time Command B was called the
system crashed.
The SDB system or the operating system?
If the latter do you have a recipe posted at Bugzilla?
I try to stay up on crashes so I can avoid them.
I was at a Microsoft Developer Seminar once and the VB product manager
said something very respectable:
"When working in VB if you have a script error that's your issue, but
anytime you have a crash that's our issue. It shouldn't be possible in
a high-level tool to crash the OS, and we prioritize those issues highly."
--
Richard Gaskin
Managing Editor, revJournal
_______________________________________________________
Rev tips, tutorials and more: http://www.revJournal.com
_______________________________________________
use-revolution mailing list
use-revolution@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription
preferences:
http://lists.runrev.com/mailman/listinfo/use-revolution
