Antony Dovgal wrote: > On 04/17/2007 09:57 PM, Sean Coates wrote: >>> lemme try this again. You can in no way rely on the last digits of >>> a float >>> output. Those depend on your system, your CPU, your ini settings and any >>> operation you do with floats reduces the accuracy.... >> >> It's still a regression, and should be documented. > > I have a patch for it, but I want to decide how "compressed" floats > should look like: > 1) 1E+15 > or > 2) 1.0E+15 > > At that moment I'm unable to find any particular reasons to state that > one of these ways is correct (or more preferred) and the other one is > wrong (or less preferred).
They are both mathematically correct, and neither is really more correct. Given that, we should try to not change the behaviour across sub-versions. -Rasmus -- PHP Internals - PHP Runtime Development Mailing List To unsubscribe, visit: http://www.php.net/unsub.php
