Message158701
| Author |
mark.dickinson |
| Recipients |
Dave.Reid, mark.dickinson, rhettinger, serhiy.storchaka, vstinner |
| Date |
2012年04月19日.07:16:55 |
| SpamBayes Score |
-1.0 |
| Marked as misclassified |
Yes |
| Message-id |
<1334819816.84.0.26828775277.issue14591@psf.upfronthosting.co.za> |
| In-reply-to |
| Content |
The latest patch has the disadvantage that it'll often change the behaviour of jumpahead for people on 32-bit platforms, which may lead to unnecessary breakage.
Here's a better version that only fixes mt[0] in the unlikely (but possible) event that mt[1] through mt[623] are all zero. That should mean that users on 32-bit machines who are depending on jumpahead being reproducible won't notice (unless they're very unlucky indeed). |
|