On Fri, Sep 9, 2011 at 11:06 AM, Sisyphus <[email protected]> wrote:

>
> ----- Original Message ----- From: "David Myers" <[email protected]>
> To: <[email protected]>
> Sent: Friday, September 09, 2011 11:46 PM
> Subject: Re: [Perldl] PDL compilation issue(s)
>
>
>
>  Thanks Sisyphus.
>>
>> CPAN was too much work, trying to make the edit.
>>
>
> I'm no fan of CPAN.pm and generally use it only if I'm wanting to install
> one of those abominations that think it's really cool to have a dependency
> list 5 miles long.
> Incidentally, I don't think there was any need for you to start over from
> scratch again - you could just have gone to the current (failing) build
> directory and then, after having amended the Basic/Core/typemap, run 'make
> test' followed by 'make install'.
> With the Basic/Core/typemap "fixed", the 'make test' should have picked up
> from where the failure occurred, and carried on as if nothing had happened.
>
>
>  Downloaded PDL-2.4.9_008 and installed.
>>
>
> That's probably the best solution, anyway :-)
> Glad you got there !!
>
> Did the Slatec stuff get built ? (It requires a fortran compiler and
> ExtUtils::F77.)
> Only reason I ask is that I believe that certain aspects of PDL::Stats work
> far more efficiently if the Slatec stuff was built. You'll find that
> PDL/Stats/GLM.pm checks to see whether Slatec is available.
>
>
Yes indeed. In fact, the logistic regression function won't run at all
without PDL::Slatec. gfortran with the stock Ubuntu repo works fine, so it
shouldn't be a problem. If you had installed PDL and PDL-Stats without
gfortran, just add it and you should not have to reinstall PDL or PDL-Stats.


>
>
>

>  Though to note, after the 3rd pgplot.t test, make test hung. Installed it
>> anyway.
>>
>
I think you need to press q or something to exit the test. Maybe that's the
problem.



>  I've no experience regarding this - though I see that Chris has offered
> some advice.
>
> Cheers,
> Rob
>
> ______________________________**_________________
> Perldl mailing list
> [email protected]
> http://mailman.jach.hawaii.**edu/mailman/listinfo/perldl<http://mailman.jach.hawaii.edu/mailman/listinfo/perldl>
>


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

Reply via email to