Em 24-09-2015 21:23, Bruce Dubbs escreveu: > Ken Moffat wrote: >> On Thu, Sep 24, 2015 at 04:40:09PM -0500, Bruce Dubbs wrote: >>> Fernando de Oliveira wrote: >>> >>>> I am trying to get back to normal working speed, still without success. >>>> Have not yet started LFS-7.8-rc1. >>> >>> OK, I think we can finish tagging without you doing it. It's better >>> to have >>> a 2nd set of eyes on the packages anyway. As of this message, there >>> are 60 >>> packages left to tag: 28 gnome, 22 lxqt, 3 networking, 6 server, and >>> lxdm. >>> >> >> I agree about a 2nd set of eyes. >> >>> The only thing I need from you is ticket #6972 (Fix starting X). >>> >>> The only other tickets are Plasma5 which I will be working on tonight >>> and >>> the bind security update. >> >> Can I mention the perl module ticket, used by bind's tessuite ? I >> think it was Net-DNS. I build that while testing the >> Mail-SpamAssassin *builds*, but I do not regard that as verification >> that the module works, and certainly not in the context of Bind.
IIRC, something, a test (?) of Bind depends on it. That's why I do both at the same time and they are normally released almost simultaneously. >>> >>>> I like not tagging the packages I usually update, because I may be >>>> reproducing the same mistake. Therefore, if somebody could tag lxqt and >>>> gnome, I really appreciate. >>>> >>>> About gnome, first i started thinking in updating to 3.18.0 for 7.8, >>>> but >>>> it depends on many other packages and could fail to build, without the >>>> new gtk.3,gdk-pixbuf, and many others that need to be updated but are >>>> not security nor at the end of the chain. >>> >>> Agree. Lets keep the freeze for gnome and lxqt. >>> >> For gnome, after merging the v3 packages into the book I rather wish >> I had not done that - in its early days, v3 was still somewhat usable, >> now it seems to be much less so. But anyway - >> >> Has anybody *recently* reviewed the runtime gnome packages ? >> >> I am thinking about trying to test what I can (I use evince and >> gucharmap, but without recommended deps such as gobject-introspection >> and nautilus, so for the moment I cannot tag them). Some things I >> cannot test, e.g. cheese (needs a supported movie camera, I think) > > I can test cheese. > >> and the networkmanager applet (no wifi on my machines running 7.8, >> and similarly I built NetworkManager itself for *latest* plasma but >> without trying to *use* it I do not regard that as a sufficient test >> of whether or not it works - after all, it's primary purpose is for >> wifi). > > I can look. I added a wifi card to my test system and find it > reasonably quick. I also do not have to worry about pulling ethernet > cable between rooms. No need of wifi to test nm-applet. It is for all connections. I never use NM, but to test it and nm-applet (which) are released simultaneously, just instal both, start nm-applet (after having started BN, and change from one conection to the other, that uses DHCPCD, then othe with fixed ip, then kill nm-applet, stop NM, and restart the network. > >> But then I looked at the gnome runtime deps - I see totem-pl-parser, >> and I remember it from when I used (or, at some times, *tried* to >> use) totem. But AFAICS nothing *directly* references it (totem, and >> its string of dependencies, have been archived). >> >> So, if I were to build everything from the gnome packages, how would >> I know if totem-pl-parser is working ? > > gnome/applications/brasero.xml: <xref linkend="totem-pl-parser"/> > > However I don't know what I'd need to do to actually *use* > totem-pl-parser. Generally for libraries I consider the build > successful if it is properly linked to parent apps. It's really > upstream's job to do detailed testing. > > -- Bruce > -- []s, Fernando, soon to be called Sisyphus -- http://lists.linuxfromscratch.org/listinfo/blfs-dev FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
