On 3 Nov., 05:35, leif <not.rea...@online.de> wrote: > On 2 Nov., 21:55, François Bissey <f.r.bis...@massey.ac.nz> wrote: > > > > > > This is a topic which I already touched in the "Merging tickets into > > > sagenb" thread, but I believe the discussion was not finished. > > > > Currently, every new Sage version has a new version of the following > > > spkgs: sage, sage_scripts, extcode, examples. However, it is likely > > > that not all those spkgs change in every version. > > > > So my proposal would be: > > > * Only update these spkgs whenever the actual code changes. Use the > > > Sage version as version number. This means that sage-4.6.1.alpha1 may > > > contain extcode-4.6.1.alpha0.spkg if extcode is not updated in > > > sage-4.6.1.alpha1. > > > * Also add sagenb to the list of spkgs handled this way. > > > > How to implement: > > > * Simplify sage-sdist such that it does not package sage_scripts, > > > extcode, examples (or make a sage-sdist-light script which does this). > > > * Merge sage_scripts, extcode, examples in the merger script. > > > As far as I am concerned it would nice not to bump packages more than > > necessary. > > +N > > > In the case of sage_script I believe the sage banner is updated > > to reflect the version each time. Otherwise there are little changes > > usually. > > We're currently [hopefully] addressing this at #9433, i.e. having the > version number outside of the Sage scripts repo. (Completely > incidentally, #9434 is related. Both are still a bit work in progress > though, slightly dragging.) > > > If you want to go that route, I would suggest that the versioning of these > > packages should be completely separate from the sage spkg, same as the > > notebook. > > +1 > > > My own opinion of course. > > No, mine too.
P.S.: I must admit I haven't yet reasoned much about the impact this would have on trac (tickets); perhaps both more structure *and* confusion... -Leif -- To post to this group, send an email to sage-devel@googlegroups.com To unsubscribe from this group, send an email to sage-devel+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URL: http://www.sagemath.org
