| 1 |
On Sat, Mar 21, 2020 at 11:22:40AM -0700, Alec Warner wrote: |
| 2 |
> On Sat, Mar 21, 2020 at 1:03 AM Alexander Tsoy <alexander@××××.me> wrote: |
| 3 |
> |
| 4 |
> > В Сб, 21/03/2020 в 00:53 -0700, Matt Turner пишет: |
| 5 |
> > > On Fri, Mar 20, 2020 at 9:55 PM Kent Fredric <kentnl@g.o> |
| 6 |
> > > wrote: |
| 7 |
> > > > If X is "noarch" and its dependency Y is "amd64", then a user on |
| 8 |
> > > > "sparc" |
| 9 |
> > > > will be able to install "X", but not its dependency "Y". |
| 10 |
> > > |
| 11 |
> > > Thank you. This is a good explanation of the problem. |
| 12 |
> > > |
| 13 |
> > > How do other distributions handle this? Arch, Fedora, and Debian have |
| 14 |
> > > "noarch" packages. Surely they've found a reasonable way to make this |
| 15 |
> > > work. |
| 16 |
> > |
| 17 |
> > Binary distros usually have separate repositories for each |
| 18 |
> > architecture. |
| 19 |
> > |
| 20 |
> |
| 21 |
> Pretty much this. There is not 1 repository, there are N. This means that |
| 22 |
> if leaf package A "noarch" depends on package B (only stable on x86) then |
| 23 |
> in the x86 tree, A and B will be available. In the sparc tree, B is not |
| 24 |
> available and so A is uninstallable and also not available. |
| 25 |
> |
| 26 |
> We had discussed doing this in the past but in practice we use a bunch of |
| 27 |
> files to compute these boundaries on the fly and this is not particularly |
| 28 |
> cheap in the current implementation. |
| 29 |
|
| 30 |
This part of the thread has been the best explanation I've seen for why |
| 31 |
this couldn't work for us, so I'm cool with this. |
| 32 |
|
| 33 |
Thanks, |
| 34 |
|
| 35 |
William |
Archives