> > + finish my YAPC talk > > + create a short (perl) script that shows me the unit > > out of the Configure diff > > l1:/pro/3gl/CPAN 127 > diff2unit.pl 17011 17041 17097 17156 17239 17278 17339 17489 >17522 17671 17698 17715 17731 17739 17827 17866 17879 > Reading Units 528 > 17011/ 1 > 17011/ 2 > 17041/ 1 > 17097/ 1 > 17156/ 1 > 17239/ 1 > 17278/ 1 > 17278/ 2 > 17339/ 1 > 17489/ 1 > 17489/ 2 > 17489/ 3 > 17522/ 1 > 17671/ 1 > 17671/ 2 > 17671/ 3 > 17671/ 4 > 17671/ 5 > 17671/ 6 > 17671/ 7 > 17698/ 1 Extensions.U > 17715/ 1 d_modfl.U > 17715/ 2 d_modfl.U > 17715/ 3 > 17731/ 1 > 17731/ 2 > 17739/ 1 prefix.U > 17827/ 1 usenm.U > 17827/ 2 > 17866/ 1 > 17879/ 1 gccvers.U > l1:/pro/3gl/CPAN 128 > > > A good start. Andy patched in his meta-doc-patch, the .U files in > metaconfig/dist-3.0at70b [1] > > Which files should *I* edit. the files there, or the files in metaconfig/U [2] > and how do the files in [1] promote to [2], and how does p4 know?
You should edit [2], the metaconfig/U, since that's where Perl's metaconfig picks up the units to regenerate Configure and config_h.SH. The transfer from [1] to [2] has to be done manually. The difference between [1] and [2] is that [1] is the "official metaconfig tree", and the [2] is the modified units used to build the Perl Configure. (Yes, it would perhaps make sense to create, say, dist-3.1, that would merge back in the changes from the Perl units.) > I guess that the promotion for [1] to [2] is like rebuilding metaconfig, and I > am to p4 edit the files in 3.0at70b > > right? No. For now, edit files in [2], in metaconfig/U. If you take a look at metaconfig/U, you will find a lot of files that live dual lives, in metaconfig/U, and in 3.0at70b. The default is to prefer the metaconfig/U versions over the 3.0at70b versions. > > + backport the Configure changes > > + modify makedepend to `tag' the units > > + modify makedepend to incorporate Andy's CBU wishes > > + find out how those changes can actually be used :) -- Jarkko Hietaniemi <[EMAIL PROTECTED]> http://www.iki.fi/jhi/ "There is this special biologist word we use for 'stable'. It is 'dead'." -- Jack Cohen
