| 1 |
On Sat, 5 Nov 2016 08:38:22 +0100 |
| 2 |
"Paweł Hajdan, Jr." <phajdan.jr@g.o> wrote: |
| 3 |
|
| 4 |
> On 05/11/2016 00:54, Jonas Stein wrote: |
| 5 |
> > Today we have still 524 ebuilds with SRC_URI=*googlecode* in the tree |
| 6 |
> > [2] and should get these fixed before end of 2016. |
| 7 |
> > |
| 8 |
> > [3] https://wiki.gentoo.org/wiki/Shutdown_of_google_code |
| 9 |
> |
| 10 |
> The wiki page seems to indicate some sense of urgency. I'm not sure how |
| 11 |
> much of that is needed. |
| 12 |
> |
| 13 |
> Brainstorming mentions just cloning the URLs, but don't existing Gentoo |
| 14 |
> Mirrors have the same effect? My understanding is fetching will not |
| 15 |
> break even if original URLs go down. |
| 16 |
|
| 17 |
Thou shalt not rely on the mirrors! Though mirrors as a temporary |
| 18 |
solution are acceptable. As long as you don't turn it into 'wrong |
| 19 |
SRC_URI is fine, the file is on the mirrors'. |
| 20 |
|
| 21 |
> I'm also not fully convinced by the manpower argument. It seems the same |
| 22 |
> work needs to be done, one way or another. |
| 23 |
> |
| 24 |
> For the parts that can be automated and done in batch, this is indeed |
| 25 |
> more effective than 500 individual bugs. |
| 26 |
> |
| 27 |
> However, I'm not sure if finding the new homepage and download URL can |
| 28 |
> be that easily automated. In fact, users may do some work for us and |
| 29 |
> help point to the right page. |
| 30 |
|
| 31 |
Also, it would be nice not to have the berlios -> sourceforge crap |
| 32 |
repeated. If software is dead-dead, we should lastrite it or revive it, |
| 33 |
not point to some dead copy by malware hosting provider. |
| 34 |
|
| 35 |
-- |
| 36 |
Best regards, |
| 37 |
Michał Górny |
| 38 |
<http://dev.gentoo.org/~mgorny/> |
Archives