IRC channel logs

2025-08-11.log

back to list of logs

<lanodan>Ooh talk about "Reviewing live-bootstrap" in an hour or so https://program.why2025.org/why2025/talk/33HD7W/ (stream will be available on https://streaming.media.ccc.de/why2025/cassiopeia )
<lanodan>Was https://github.com/arsv/perl-cross tried to bootstrap perl?
<stikonas>lanodan: no, haven't seen it before
<stikonas>but I doubt it can skip pregenerated file issues that we were trying to sort out in live-bootstrap
<stikonas>ultimately, perl source uses some header files that are generated using perl
<stikonas>well, perhaps it could be used instead of our handwritten makefiles...
<stikonas>but to be honest it's not that complicated https://github.com/fosslinux/live-bootstrap/blob/master/steps/perl-5.000/mk/main.mk
<stikonas>probably most of the complexity is actually in https://github.com/fosslinux/live-bootstrap/blob/master/steps/perl-5.000/files/config.h
<lanodan>Ah I thought it was mostly due to how the big Configure script of perl requires perl to generate
<stikonas>lanodan: well, that's why we had makefiles, yes
<stikonas>but there are also perl scripts
<stikonas>I've ported them to awk for perl 5.000
<stikonas>actually 2 .pl scripts were ported to bash and one to awk
<stikonas>e.g. see https://github.com/fosslinux/live-bootstrap/blob/master/steps/perl-5.000/files/opcode.awk
<stikonas>instead of https://github.com/Perl/perl5/blob/perl-5.000/opcode.pl
<stikonas>ok, I seem to have forgotten how it works
<stikonas>after one more look it seems that one script was ported from perl to sh and the other from perl to a combination of bash and awk script
<stikonas>that's the sh part: https://github.com/fosslinux/live-bootstrap/blob/master/steps/perl-5.000/files/opcode.sh