Arrgh, that should be $counts->minimum_ind since
the 0 values are less than all the positive counts.

On Thu, Feb 20, 2014 at 1:11 PM, Chris Marshall <[email protected]> wrote:
> It would probably help if we made the signature more
> mnemonic as to what the values are being returned,
> counts and values.  Since counts are always non-zero
> for the valid elements,
>
>   $counts->minimum_ind
>
> gives you the number of elements in the rle() encoded
> data.
>
> --Chris
>
>
> On Thu, Feb 20, 2014 at 12:57 PM, Chris Marshall <[email protected]> 
> wrote:
>> The current implementation seems to be the way that it is
>> in order to support threading of rle() operations.  rld() also
>> chooses an output size that is compatible with threaded
>> computations.  Maybe the signature could be changed and
>> the $a/$b outputs could be truncated to the minimum number
>> of elements.  In the single vector case, that would give the
>> desired result.
>>
>> --Chris
>>
>> On Tue, Feb 11, 2014 at 12:57 AM, Craig DeForest
>> <[email protected]> wrote:
>>> To be fair, rle could and should generate a truncated output PDL for 
>>> itself...
>>>
>>>
>>> On Feb 10, 2014, at 9:45 PM, Derek Lamb <[email protected]> wrote:
>>>
>>>> In your $idx below, you don't want the second occurrence of 0 in $b, you 
>>>> want
>>>> the first occurrence of 0 in $a.  Because if your data for some reason has 
>>>> no 0
>>>> in it, then your method fails.  All those zeroes in $rv are to be 
>>>> expected--the $a
>>>> and $b arrays have the same size as your input data, so you have to 
>>>> truncate.
>>>> Re-read the rle() docs.

_______________________________________________
Perldl mailing list
[email protected]
http://mailman.jach.hawaii.edu/mailman/listinfo/perldl

Reply via email to