Something similar to what you have described I included into the following
test:
https://github.com/kmx/pdl-io-sereal/blob/master/t/1-basic.t
Anyway your use case can be more tricky.
Enjoy holiday!
--
kmx
On 11.9.2015 17:54, Ingo Schmid wrote:
Hi kmx,
awesome you took up on that! That will really help me.
sorry, I'm on holidays for the next two weeks, so no promise I'll access
work.
Maybe I'll do a quick check later tonight.
The test is to store a piddle in the header of a piddle in the header of
a piddle in the ...
And successfully restore that.
Ingo
On 09/10/2015 08:36 PM, kmx wrote:
Hi Ingo,
I have just released my new experiment - PDL::IO::Sereal -
https://metacpan.org/pod/PDL::IO::Sereal
If you have some spare time could you please test whether it handles
well your scenario?
--
kmx
On 3.9.2015 18:05, Ingo Schmid wrote:
Hi again,
it's not working quite as well, though. The header of the piddles
stored in the header of a piddle are lost. ;-(
Ingo
On 09/03/2015 05:42 PM, Ingo Schmid wrote:
to answer my own question:
I need to both
use Storable;
use PDL::IO::Storable;
then store (or better nstore) both header and piddle separately. The
same on the retrieve side. But at least it's only one extra call per
piddle.
Ingo
On 09/03/2015 05:25 PM, Ingo Schmid wrote:
Hi,
I've a piddle that has a header which contains piddles.I want to
store that on disk and retrieve later from other programs.
- Storable has problems with piddles, otherwise readflex+ nstore
(header) would do it.
- PDL::IO::Storable seems to lose the header.Atleast, retrieve
restores the piddle without it's header.
I'd hate to manually walk the header for piddles to store
individually, if can be avoided. Am I missing something?
Ingo
------------------------------------------------------------------------------
_______________________________________________
pdl-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/pdl-general
