Archaic wrote:
jhalfs was discussed in dozens of threads a long time ago. What you are
arguing against is a method to test the book directly from it's XML.
That is extremely powerful. And if it is the MD5s you don't like, I
cannot understand that, either. It is common (and recommended) practice
to verify checksums. I would hope before any md5 is added to the book
that the dev who is adding it has checked it against the upstream
checksum file (if it exists). But many upstream packages don't have
md5's. For that reason alone, I think the book should have them. The
rest are for convenience.
I can understand that, but we should just provide a link to the sources
md5 instead of creating our own. It helps prevent any contamination of
the file by us.
If I'm missing the point of what's going on, somebody give me a clue. I
know I haven't done a lot of work in LFS lately, but today since I had
some free time, I ventured out and looked at the outstanding tickets,
and this is how I felt after looking at them.
Perhaps you should visit more often. ;) There have been a few udev rules
threads that I'd appreciate your comments on, too.
I've never seen a ticket assigned to me to answer questions. After all
that's why we have the ticket system in place.
--
http://linuxfromscratch.org/mailman/listinfo/lfs-dev
FAQ: http://www.linuxfromscratch.org/faq/
Unsubscribe: See the above information page
