Dne 9.8.2012 19:07, Ken Dreyer napsal(a):
I have contributed a couple patches to upstream already, and I understand that this is going to have to involve some developer legwork. I figured I would try it as an experiment to see what is possible.
That is the word of man ;) Good to hear that.
I'm slowly publishing my interim work on fedorapeople.org. I intend to document FPG compromises very clearly on the wiki page, and then slowly chip away at them. At the same time, I can be submitting the "easier" requisite gems for review into Fedora proper. If I can keep a running dashboard of the issues, hopefully it will not be insurmountable :) I understand that Ruby's culture is all about bundling, and trying to counteract that as a distro can be a bit like pushing a boulder uphill. But at the same time, it's not like the question of bundling is a new one. So I figured I might as well take a crack at it with Gitorious and see what it's like. One of the areas I could use the most help with from the Ruby SIG is understanding what dependencies are a "bad idea". For example, in the Redmine review, Jan pointed out that one of the Gem dependencies was abandoned upstream[3]. In my wiki page[4], does anyone see Gems in the list that stand out as similarly problematic?
On first look, I'm curious about hoe, echoe, gemcutter, rubyforge. They seems to be development or better to say deployment dependencies. So I believe it should be possible to get rid of them. They are probably used in some Rakefile for some management tasks.
Vit
- Ken [1] https://groups.google.com/d/msg/gitorious/gpyb3Qh1Hb8/J2M-9rrpIw8J [2] https://lists.fedoraproject.org/pipermail/ruby-sig/2012-July/001074.html [3] https://bugzilla.redhat.com/825495#c1 [4] https://fedoraproject.org/wiki/User:Ktdreyer/Gitorious _______________________________________________ ruby-sig mailing list [email protected] https://admin.fedoraproject.org/mailman/listinfo/ruby-sig
_______________________________________________ ruby-sig mailing list [email protected] https://admin.fedoraproject.org/mailman/listinfo/ruby-sig
