| 1 |
-----BEGIN PGP SIGNED MESSAGE----- |
| 2 |
Hash: SHA1 |
| 3 |
|
| 4 |
On Monday 21 June 2004 21:35, Daniel wrote: |
| 5 |
> > > -* is the only way to add an ebuild for testing without forcing it on |
| 6 |
> > > ~arch users. i still have -* in keywords for the convenience of x86 |
| 7 |
> > > users who want to test and port x86 software to gcc 3.4 (and submit |
| 8 |
> > > patches on bugzilla) by simply adding "sys-devel/gcc -*" to |
| 9 |
> > > package.keywords. |
| 10 |
> > |
| 11 |
> > Adding 'sys-devel/gcc -*' to package.keywords is *INCORRECT*. This is |
| 12 |
> > undocumented (and IMO broken) behaviour that cannot be relied upon. |
| 13 |
> |
| 14 |
> man portage |
| 15 |
> |
| 16 |
> it's there and documented. |
| 17 |
|
| 18 |
Apologies for being harsh on that then. /me blames SpanKY |
| 19 |
|
| 20 |
That said, I motion for it to be removed. See bug #43670 for a long discussion |
| 21 |
on it. (http://bugs.gentoo.org/show_bug.cgi?id=43670). As well as all that is |
| 22 |
listed there, it ties the implementation to string matches. Consider the |
| 23 |
following: |
| 24 |
|
| 25 |
make.defaults:ARCH="amd64" |
| 26 |
package.keywords:foo/bar ~x86 |
| 27 |
foo/bar-1.0:KEYWORDS="~x86" |
| 28 |
foo/bar-1.5:KEYWORDS="x86" |
| 29 |
|
| 30 |
The current implementation will limit foo/bar to 1.0 which is clearly not what |
| 31 |
a user would intend the package.keywords entry to do. (NOTE: the above |
| 32 |
scenario has been addressed in a different way.) |
| 33 |
|
| 34 |
Regards, |
| 35 |
Jason Stubbs |
| 36 |
-----BEGIN PGP SIGNATURE----- |
| 37 |
Version: GnuPG v1.2.4 (GNU/Linux) |
| 38 |
|
| 39 |
iQCVAwUBQNbe71oikN4/5jfsAQIBtwP9ED+5UxLxchPB/nVm4uRbgvL5TxFlpWK+ |
| 40 |
tV+yYQDezYuSosUNO0Xj7CiptDDMv/iQo1dqmYi3UDv3/+k7hH+iOD4eRbnsqqi5 |
| 41 |
aVmtng29LZ1IgMMvr+oZSxIxX/9UjQ7IG5qUqQow4WRydcggVG6T05pah5ntd5+J |
| 42 |
iHCKTMRMf8I= |
| 43 |
=3g54 |
| 44 |
-----END PGP SIGNATURE----- |
| 45 |
|
| 46 |
-- |
| 47 |
gentoo-dev@g.o mailing list |
Archives