NOTE: use Perl; is on undef hiatus. You can read content, but you can't post it. More info will be forthcoming forthcomingly.
All the Perl that's Practical to Extract and Report
subconsciously, maybe there are helpful little gnomes running around
in the repository and fixing bugs while we sleep, I don't know..
-- Jarkko Hietaniemi
Stories, comments, journals, and other submissions on use Perl; are Copyright 1998-2006, their respective owners.
Pure Perl (Score:2)
The advantages of doing this are twofold. First, you eliminate compiler and binary compatibility issues, which are a major hassle on Windows. Second, you improve the likelihood of receiving patches, since most of your user base doesn't understand C or have a compiler with which to debug a C extension (especially Perl extensions - oof).
Speed is a non-issue. With W
Re:Pure Perl (Score:1)
Based on my experiments with FFI systems and Perl, the speed issue is mostly negligible. If Win32::API builds up thunks once, then lets you call the foreign functions through the thunks, then speed is nearly the same as that of XS. The real expense seems to be parameter marshalling and unmarshalling and the double-dispatch (first to the thunk, then to the actual function pointer). You have to do that with both systems.
... or you could use
pack/unpackand Perl'ssyscall, but I've only ever seen one person do that.Reply to This
Parent