If the random results are using the pseudorandom generator in a state of sin [st-and.ac.uk], shoudn't one set of tests jam the seed to produce a known sequence?
-- Bill
# I had a sig when sigs were cool
use Sig;
Well, whenever you load the module, a seed is set and printed as diagnosis, and thus, if the test fails, you know the seed and can reproduce it... Now, the way the seed is set nowadays, *that* is kind of stupid O:-)
If you're suggesting the user being able to pass the module his own seed, well... kind of makes sense... although I think the results could be different from machine to machine, but the user would have to be aware of that himself... but I'm not sure, as I don't know much about sinner generators O:
Seed whacking? (Score:1)
Bill
# I had a sig when sigs were cool
use Sig;
Re:Seed whacking? (Score:1)
Now, the way the seed is set nowadays, *that* is kind of stupid O:-)
If you're suggesting the user being able to pass the module his own seed, well... kind of makes sense... although I think the results could be different from machine to machine, but the user would have to be aware of that himself... but I'm not sure, as I don't know much about sinner generators O: