From jamatos at fc.up.pt Wed Sep 13 20:52:25 2006 From: jamatos at fc.up.pt (=?iso-8859-1?q?Jos=E9_Matos?=) Date: Wed, 13 Sep 2006 21:52:25 +0100 Subject: [Fedora-r-devel-list] rebuild for fc6 Message-ID: <200609132152.25229.jamatos@fc.up.pt> Hi Spot, to you have any estimated time to rebuild R for FC6? I am holding the update of my packages related with R to after your update. rpy R-mAr R-waveslim R-wavethresh Although the reason is probably a bit silly, since I am not sure how the .so's of those packages (those where it applies) depends on how R is compiled. The reason for the mass recompile is related with the linker, so I decided to play safe. Better to be sorrow than to be sorry. :-) -- Jos? Ab?lio From tcallawa at redhat.com Wed Sep 13 22:22:08 2006 From: tcallawa at redhat.com (Tom 'spot' Callaway) Date: Wed, 13 Sep 2006 17:22:08 -0500 Subject: [Fedora-r-devel-list] rebuild for fc6 In-Reply-To: <200609132152.25229.jamatos@fc.up.pt> References: <200609132152.25229.jamatos@fc.up.pt> Message-ID: <1158186128.28319.62.camel@localhost.localdomain> On Wed, 2006-09-13 at 21:52 +0100, Jos? Matos wrote: > Hi Spot, > to you have any estimated time to rebuild R for FC6? > > I am holding the update of my packages related with R to after your update. > rpy > R-mAr > R-waveslim > R-wavethresh > > Although the reason is probably a bit silly, since I am not sure how > the .so's of those packages (those where it applies) depends on how R is > compiled. The reason for the mass recompile is related with the linker, so I > decided to play safe. > > Better to be sorrow than to be sorry. :-) It just finished building for FC-6. No changes, just a rebuild. ~spot -- Tom "spot" Callaway || Red Hat || Fedora || Aurora || GPG ID: 93054260 "We must not confuse dissent with disloyalty. We must remember always that accusation is not proof and that conviction depends upon evidence and due process of law. We will not walk in fear, one of another. We will not be driven by fear into an age of unreason, if we dig deep in our history and our doctrine, and remember that we are not descended from fearful men -- not from men who feared to write, to speak, to associate and to defend causes that were, for the moment, unpopular." -- Edward R. Murrow, March 9, 1954 From jamatos at fc.up.pt Thu Sep 14 20:02:47 2006 From: jamatos at fc.up.pt (=?utf-8?q?Jos=C3=A9_Matos?=) Date: Thu, 14 Sep 2006 21:02:47 +0100 Subject: [Fedora-r-devel-list] rebuild for fc6 In-Reply-To: <1158186128.28319.62.camel@localhost.localdomain> References: <200609132152.25229.jamatos@fc.up.pt> <1158186128.28319.62.camel@localhost.localdomain> Message-ID: <200609142102.47601.jamatos@fc.up.pt> On Wednesday 13 September 2006 23:22, Tom 'spot' Callaway wrote: > > ??????Better to be sorrow than to be sorry. :-) > > It just finished building for FC-6. No changes, just a rebuild. I noticed. :-) The other packages already built correctly. Thanks. > ~spot -- Jos? Ab?lio From jamatos at fc.up.pt Thu Sep 14 20:18:11 2006 From: jamatos at fc.up.pt (=?iso-8859-1?q?Jos=E9_Matos?=) Date: Thu, 14 Sep 2006 21:18:11 +0100 Subject: [Fedora-r-devel-list] Automatic rebuild of R derived packages on every new major version. Message-ID: <200609142118.11345.jamatos@fc.up.pt> Hi, A new version of R is approaching, R 2.4.0 is scheduled for the first days of October. When version 2.4 is recompiled we need to recompile all the packages that depend on R. rpy needs to be compile for every new 2.4.x release, but R packages only need to be release for each x.y.0 release. That is exactly what happens with python, every package needs to be rebuild because pyc files depend on a major version. Python does not present any major problem since the core of the language is in Core, and every new version only appears once in the Core release and only minor updates are allowed during the distribution life time. R core language is in Extras and the correct way to build a new version is to rebuild the core language and then the different packages according to its dependencies. This procedure is tedious and error prone, especially as more and more packages enter Extras. Would it be possible to automate to some extent this procedure. I guess that this procedure could be of interest for other languages that don't go in Core. Should I bring this issue to Extras or Packagers mailing list, since this problem seems general enough and not only limited to R. What do other think? -- Jos? Ab?lio