Perl bytecode on unix

[fpi]

1) How do I execute a perl file that I already converted to bytecode (perl -MO=Bytecode hello.pl)? Specifically, how do I use the byteload_fh () function mentioned in the docs? Even if you are not sure, I would appreciate a guess as to how to experiment, since I am not good with filehandles.

2)I found the ByteLoader.pm module, but I'm not sure if I'm using it correctly, and if I am, I can't get it to work all the time (even with a simple helloworld script, sometimes it works, sometimes not, beats me....). Basically I just open up the file containing the bytecode, add the following to the top of the file:

#!/usr/bin/perl
use ByteLoader 0.03;

after which the bytecode follows. I would rather use byteload_fh() if I only knew how.

3) will converting to bytecode include the perl modules that I have "used" in my script? or does that only happen with a full C (perlcc) compile?

[LD2]

fpi, to answer your first question.. you can use:
• 1. ByteLoader module
• 2. byteload_fh(FILEHANDLE) that is included with the B module
Which it seems like you're familiar with both of these ways...

From the looks of the second question, you're using ByteLoader correctly. Remember that it's still in experimental phase.. which would explain why it's a bit unstable. I believe there is an updated version coming out.

I believe you need to do a full compile(perlcc) to include other scripts that you used in your main script. If you run perlcc with the -b flag, it will generate bytecode instead of C code.. i.e. perlcc -b helloworld.pl

Here are a few good links:
• perlcompile
• Programming Perl, 3rd Edition - Chapter 18, Compiling
• B::Bytecode
• ByteLoader
• FileHandle
• Pass by Reference
• Passing Filehandles

[fpi]

LD2, thanks for the post. After reading through the links you posted and then further testing, here's the first little interesting thing I found:

perl -MO=Bytecode,-omyfile.bc myfile.pl produces the bytecode file, as before

perlcc -b -o myfile.bc myfile.pl produces the bytecode, and automatically sticks in the perl header and use Byteloader 0.03; statement. The only problem was the perl header pointed to /perl instead of /usr/bin/perl, so I had to go in and edit anyway. But at least now I know that I was using the ByteLoader module correctly, since that was the way that perlcc sets it up.

Thanks for the filehandle links too. I'm trying to experiment with byteloader_fh() again, but I'm getting "Undefined subroutine B::byteload_fh" errors. I reloaded the B modules....I just found out that you can't reload the B modules without reloading all of perl....