| 1 |
On Sun, Dec 25, 2005 at 09:39:45PM -0800, Robin H. Johnson wrote: |
| 2 |
> On Sun, Dec 25, 2005 at 07:41:01PM -0800, Brian Harring wrote: |
| 3 |
> > On Sun, Dec 25, 2005 at 06:48:02PM -0500, Mike Frysinger wrote: |
| 4 |
> > > On Sunday 25 December 2005 17:12, Brian Harring wrote: |
| 5 |
> > > > Can be defeated by a unpack ${DISTDIR}/file call, but that's invalid |
| 6 |
> > > > anyways. |
| 7 |
> > > |
| 8 |
> > > and what about things that do `cp ${DISTDIR}/aadsfasdf ${S}/` ? those are |
| 9 |
> > > going to fail to wont they ? |
| 10 |
> > Had originally thought about resetting DISTDIR to the symlink dir; it |
| 11 |
> > would address this concern. |
| 12 |
> Definetly change DISTDIR to point to the symlink dir. |
| 13 |
> |
| 14 |
> There were user-authored patches around more than year ago that |
| 15 |
> implemented this (a dir containing symlinks to the real files), as part |
| 16 |
> of somebodies multiple-distdir support. |
| 17 |
Haven't seen them, but the original implementation that got tagged |
| 18 |
into saviour was based on discussions/suggestions from you... |
| 19 |
|
| 20 |
Multiple distdir is a bit closer via this, but it still requires a lot |
| 21 |
of python side portage.fetch work to get there... |
| 22 |
|
| 23 |
Meanwhile, tagging this beast into trunk. |
| 24 |
~harring |
Archives