[Seqinr-forum] Memory leak?

[Jean lobry]

At 17:41 +0100 5/03/10, Jean lobry wrote:
>At 10:22 +0000 5/03/10, Darren Obbard wrote:
>>Hi!
>>
>>Not sure if this is an R problem or a seqinr problem or merely stupidity
>>on my part, but I found it in a seqinr context using read.alignment()
>>
>
>I can reproduce the bug. Calling for garbage collection within the
>loop with gc() does not fix it. My first guess is that there is
>a memory leak problem at the C level in the read.alignment()
>function.
>
>Thanks for spotting the bug with a nice small reproducible example,
>I will investigate this further.
>
>Best,
>

I have just commited a fix that should be available out-of-the-box
in R-forge in a day or two (revision >= 1727). I have cleaned up memory
allocation at the C level in the read_fasta_align function. I didn't check
the remaining formats.

Best,

Z.

-- 
Pr. Jean R. Lobry (Dr. phil. hab.)
Head of Laboratoire de Police Scientifique de Lyon
Institut National de Police Scientifique
31 av. F. Roosevelt, F-69134 ECULLY CEDEX
Allo: +33 (0)4 72 86 89 60
e-mail: jean.lobry at interieur.gouv.fr