| 1 |
On Fri, May 19, 2006 at 02:32:13PM -0700, Donnie Berkholz wrote: |
| 2 |
> Robin H. Johnson wrote: |
| 3 |
> > On Fri, May 19, 2006 at 09:08:08AM -0700, Greg KH wrote: |
| 4 |
> >> On Fri, May 19, 2006 at 01:45:30PM +0200, Fernando J. Pereda wrote: |
| 5 |
> >>> Also, git-sources *should* use this eclass once it is in the tree since |
| 6 |
> >>> people using it will save _lots_ of bandwidth and disk space. |
| 7 |
> >> Yes, I'll convert it over once you feel it is ready, just let me know. |
| 8 |
> > I'd like to voice an objection to converting git-sources using a |
| 9 |
> > git.eclass and performing live fetching. |
| 10 |
> > |
| 11 |
> > This takes away the ability to use to existing git snapshots when there |
| 12 |
> > is no or minimal network connectivity. |
| 13 |
> |
| 14 |
> I'm not sure I understand. If you have enough connectivity to get a |
| 15 |
> snapshot to the machine, why wouldn't you have enough to get a .git/ to it? |
| 16 |
Simple case - consider a disconnected machine, that you use sneakernet |
| 17 |
to get files to - I've had a few in the past where the hardware was new |
| 18 |
enough that networking was broken or not supported yet, and I had to try |
| 19 |
a few patches and snapshots before actually getting it to work. |
| 20 |
|
| 21 |
Another case might be a machine where HTTP/FTP access is allowed out via |
| 22 |
a filter proxy, but not anything else (I've had a few cases like this in |
| 23 |
doing Gentoo consulting). |
| 24 |
|
| 25 |
-- |
| 26 |
Robin Hugh Johnson |
| 27 |
E-Mail : robbat2@g.o |
| 28 |
GnuPG FP : 11AC BA4F 4778 E3F6 E4ED F38E B27B 944E 3488 4E85 |
Archives