Jamie Bennett wrote:
Good to see your around again, long time no see.
alfs development seems to have stalled a bit again, but we'll see if we
can't get that hopping again. We had some good progress on the
communication protocol, we may want to see if we can work on refining
that some, as well as work on some means of authentication next.
Do we have a definitive answer documented on what the finished profile
syntax is going to be? Also is there a todo list somewhere? bugzilla?
Now that I'm back in the land of the connected it would be nice to get
my teeth into something.
Nope I don't believe a definitive answer was reached on the profile
syntax, I think there was definite agreement on moving away from xml
though. So far there's no todo list, though one should probably be
created at some point.
On a side note, did you get any more work done on the POC code you
posted about on your blog a while back?
I think along side the authentication work, some form of extensive
logging should be thought about. Logging, especially in LFS and the land
of ICA testing, is essential and the more detailed the logs are the more
likely the new tool will be adopted by the mainstream LFS builders.
Agreed.
I believe that with the work done on jhlfs it has shown that the correct
way forward from here should be to auto-generate the profiles. A jhlfs
Makefile->profile hack in perl/python would be all that is needed to
generate the book profiles using jhlfs to download and parse the book.
I've been thinking along the same lines about that too, I know
preliminary support is being worked in for BLFS, don't know what the
plans are for CLFS. But I do agree auto generation is starting to be
more along the lines of the way to go.
Thomas
--
http://linuxfromscratch.org/mailman/listinfo/alfs-discuss
FAQ: http://www.linuxfromscratch.org/faq/
Unsubscribe: See the above information page
