Kazu Yamamoto ( 山本和彦 )
kazu at iij.ad.jp
Mon Aug 22 10:59:05 CEST 2011
I have implemented data-thread-less version:
Since its performance is almost the same as the original, I have
merged it into master. Tommorrow I will try using C's localtime() and
strftime() as Snap does.
>> Actually, the idea we'd been passing around was having an IORef with
>> both the rendered time and the integral time. Whenever the rendered
>> time is requested, we first compare the stored integral time with the
>> current time. Only if there's a mismatch would we generate a new
>> rendered time.
> Thank you for explanation. Now I understand the idea passed around and
> what snap is doing. OK. I will implement this idea in wai-logger.
> web-devel mailing list
> web-devel at haskell.org
More information about the web-devel