SilvioCVdeAlmeida <[EMAIL PROTECTED]> writes:
>> Are you calling sub as main::subroutine() or as
>> subroutine() ?
>>
>> XS code's "current package" is the current package
>> of the calling perl code, NOT the package in which
>> XS resides. (This is becuase current package is a
>> per Current Op thing and XS doesn't have OPs so last
>> executed perl statement's one gets used.)
>
>ThatÂs it! My project has a scheme based on the example of a Persistent
>Interpreter from perlembed.pod, where each script file defines its own
>unique package name, for the purpose of the cache system. So, I always
>call a subroutine from inside a package.
>
>What OP above means?
"Opcode" - the internal form perl source is compiled into.
>
>> Another thing is that XS code has no "lexical scope"
>> so my/our declarations can't be seen.
>
>I suppose the concept here is the same as from the current package you
>described above.
It can look the same, but is is NOT the same concept.
You cannot prefix "my $foo" with anything to get access to it
from another file.
>
>> >Perl docs tells nothing more than almost nothing
>> >about "eval_pv", so I would greatly appreciate some
>> >link/info about it.
>> >
>>
>> Essentially it is "as if" the code that called your
>> XS had done
>>
>> eval "Some string";
>
>Indeed,
>
> sub pmain { "PackageMAIN\n" }
>
> package Pack;
>
> sub ppack { "PackagePACK\n" }
>
> eval "print 1, pmain()";
> eval "print 2, ppack()";
>
> package main;
>
> eval "print 3, pmain()";
> eval "print 4, ppack()";
>
>prints:
>
> 2PackagePACK
> 3PackageMAIN
>
>
>Thank you, Nick.
>
>Silvio
