From mboxrd@z Thu Jan  1 00:00:00 1970
From: Yann E. MORIN <yann.morin.1998@free.fr>
Date: Sat, 5 Dec 2015 18:29:32 +0100
Subject: [Buildroot] Overriding Buildroot packages
In-Reply-To: <565B859B.10400@mind.be>
References: <CAPER3A0P1Qyd_=ktyHt=udawndUfHue8SE_zDQxMHjz+r+pPtg@mail.gmail.com>
 <20151124164107.3570084a@free-electrons.com>
 <CAPER3A0tE2_edxtE65MaVUVUh3BOdqywb4jUANOe_4Jm+0fihw@mail.gmail.com>
 <5654E4F2.2020405@mind.be>
 <CAPER3A3yd6N6VWhM4uYJTq7pjBdiQ2=zQ-Mzu_aH+LBSNoaDSw@mail.gmail.com>
 <565B859B.10400@mind.be>
Message-ID: <20151205172932.GB3666@free.fr>
List-Id: <buildroot.busybox.net>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
To: buildroot@busybox.net

On 2015-11-30 00:09 +0100, Arnout Vandecappelle spake thusly:
> On 25-11-15 10:48, Hajime Branko Yamasaki Vukelic wrote:
> > On Tue, Nov 24, 2015 at 11:30 PM, Arnout Vandecappelle <arnout@mind.be> wrote:
> > 
> > [SNIP]
> > 
> >>> Ideally, ability to override some aspect of an existing package (e.g.,
> >>> only its version, or only the source URL, or something along those
> >>> lines) would be great to have. If not, then, a mechanism to make two
> >>> packages equivalent (something like virtual packages but applicaable
> >>> to packages that are not explicitly virtual) would also work (e.g.,
> >>> `POSTGRES_95_PROVIDES = postgres`, that causes having postgres-95
> >>> selected to satisfy dependency requirement for a package that
> >>> selects/depends on postgres).
> >>
> >>  There is a fairly easy way to make it possible to override the version and
> >> dependencies, but I think it's a bit ugly. In inner-generic-package, instead of
> >> defining all the rules directly, we could add yet another level of direction and
> >> delay the definition of the rules until after BR2_EXTERNAL has been included.
> > 
> > What if a package has <pkgname>.hash file? If I simply override the
> > version, wouldn't that still break when checking the hash?
> 
>  That's right, we'd need to find a workaround of that as well.

We already have a mechanism to avoid checking hashes;

    BR_NO_CHECK_HASH_FOR += tarball-name

Regards,
Yann E. MORIN.

-- 
.-----------------.--------------------.------------------.--------------------.
|  Yann E. MORIN  | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: |
| +33 662 376 056 | Software  Designer | \ / CAMPAIGN     |  ___               |
| +33 223 225 172 `------------.-------:  X  AGAINST      |  \e/  There is no  |
| http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL    |   v   conspiracy.  |
'------------------------------^-------^------------------^--------------------'