On 12-apr-2010, at 10:20, Derick Rethans wrote: > On Mon, 22 Mar 2010, Felix De Vliegher wrote: > >> On 17-mrt-2010, at 19:09, Derick Rethans wrote: >> >>> On Wed, 17 Mar 2010, Felix De Vliegher wrote: >>> >>>> On 17-mrt-2010, at 17:52, Derick Rethans wrote: >>>> >>>>> On Wed, 17 Mar 2010, Felix De Vliegher wrote: >>>>> >>>>>> On 17-mrt-2010, at 17:27, Frederic Hardy wrote: >>>>>>> >>>>>>> Why not use arrayIterator::seek() ? >>>>>> >>>>>> Because the functionality isn't exactly the same. >>>>>> ArrayIterator::seek() only sets the array pointer, array_seek would >>>>>> also return the value + have fseek()-like functionality with the >>>>>> SEEK_* consts and optional negative offsets. >>>>> >>>>> To be honest, I'd rather have the proposed array_seek() return a status >>>>> whether the seek worked or not. Notices are uncool and you can already >>>>> retrieve data/key with key() and current(). >>>>> >>>> >>>> Update: http://phpbenelux.eu/array_seek-return.patch.txt >>>> I've kept the fseek()-style return values (0 when fine, -1 when seek fails) >>> >>> Any reason why you picked that over the (IMO more logical) true/false >>> approach? >> >> No, it makes more sense to use the boolean return values, I was just >> using your fseek() analogy. Although I still find it useful to return >> the seeked value, and false when seek fails (basically how next(), >> reset() and friends behave). > > Has this been added to trunk now? Or not yet?
No, still have it lying around. Can I commit this? Cheers, Felix -- PHP Internals - PHP Runtime Development Mailing List To unsubscribe, visit: http://www.php.net/unsub.php
