| 1 |
On Tue, 24 Apr 2012 00:14:15 -0400 |
| 2 |
Alexandre Rostovtsev <tetromino@g.o> wrote: |
| 3 |
|
| 4 |
> On Tue, 2012-04-24 at 06:05 +0200, Michał Górny wrote: |
| 5 |
> > On Mon, 23 Apr 2012 22:45:36 -0500 |
| 6 |
> > Doug Goldstein <cardoe@g.o> wrote: |
| 7 |
> > |
| 8 |
> > > So I've just had one reservation when using epatch_user for |
| 9 |
> > > allowing users to apply patches. And that's figuring out when to |
| 10 |
> > > run eautoreconf. I don't necessarily want to run it |
| 11 |
> > > unconditionally but sometimes users have patches which touch |
| 12 |
> > > autoconf files but my existing patch set doesn't so I'm not |
| 13 |
> > > calling eautoreconf. Does anyone have a suggested way to handle |
| 14 |
> > > this? |
| 15 |
> > |
| 16 |
> > inherit autotools-utils |
| 17 |
> |
| 18 |
> That doesn't help the numerous packages that don't ship some of the |
| 19 |
> autoconf macros they use in their source tarball, and therefore |
| 20 |
> require additional build-time dependencies (gnome-common, gtk-doc-am, |
| 21 |
> and so forth) to provide the macros needed for eautoreconf. |
| 22 |
|
| 23 |
We can't help them unless you want the whole tree (including those |
| 24 |
packages) to forcedly depend on them. |
| 25 |
|
| 26 |
It's simply better just to assume: if user wants user patches, he/she |
| 27 |
needs to have necessary deps installed. |
| 28 |
|
| 29 |
-- |
| 30 |
Best regards, |
| 31 |
Michał Górny |
Archives