[PATCH] enca_1.9.bb: remove unused configure test which is bound to fail.

[PATCH] enca_1.9.bb: remove unused configure test which is bound to fail.

[Graham Gower]

Enca's configure will fail on everything not arm or ix86.

Signed-off-by: Graham Gower <graham.gower@gmail.com>
---
 recipes/enca/enca_1.9.bb |    7 ++++++-
 site/arm-linux           |    3 ---
 site/ix86-common         |    3 ---
 3 files changed, 6 insertions(+), 7 deletions(-)

diff --git a/recipes/enca/enca_1.9.bb b/recipes/enca/enca_1.9.bb
index 548a824..dbffe1e 100644
--- a/recipes/enca/enca_1.9.bb
+++ b/recipes/enca/enca_1.9.bb
@@ -3,7 +3,7 @@ SECTION = "libs"
 PRIORITY = "optional"
 LICENSE = "GPLv2"
 HOMEPAGE = "http://trific.ath.cx/software/enca/"
-PR = "r3"
+PR = "r4"
 
 SRC_URI = "http://www.sourcefiles.org/Networking/Tools/Miscellanenous/enca-${PV}.tar.bz2 \
 	file://configure-hack.patch;patch=1 \
@@ -15,6 +15,11 @@ inherit autotools
 
 EXTRA_OECONF="--with-libiconv-prefix=${STAGING_DIR_HOST}${layout_exec_prefix}"
 
+do_configure_prepend() {
+	# remove failing test which checks for something that isn't even used
+	sed -i -e '/ye_FUNC_SCANF_MODIF_SIZE_T/d' configure.ac
+}
+
 do_configure_append() {
 	sed -i s:-I/usr/include::g Makefile
 	sed -i s:-I/usr/include::g */Makefile
diff --git a/site/arm-linux b/site/arm-linux
index 4212728..18dfb38 100644
--- a/site/arm-linux
+++ b/site/arm-linux
@@ -148,9 +148,6 @@ samba_cv_USE_SETREUID=${samba_cv_USE_SETREUID=yes}
 samba_cv_USE_SETUIDX=${samba_cv_USE_SETUIDX=yes}
 samba_cv_LINUX_LFS_SUPPORT=${samba_cv_LINUX_LFS_SUPPORT=yes}
 
-# enca
-yeti_cv_func_scanf_modif_size_t=yes
-
 # clamav
 clamav_av_func_working_snprintf_long=${clamav_av_func_working_snprintf_long=yes}
 clamav_av_have_in_port_t=${clamav_av_have_in_port_t=yes}
diff --git a/site/ix86-common b/site/ix86-common
index a5c4b60..b7a503f 100644
--- a/site/ix86-common
+++ b/site/ix86-common
@@ -104,9 +104,6 @@ ac_cv_func_posix_getpwnam_r=${ac_cv_func_posix_getpwnam_r=yes}
 dpkg_cv_va_copy=${ac_cv_va_copy=no}
 dpkg_cv___va_copy=${ac_cv___va_copy=yes}
 
-# enca
-yeti_cv_func_scanf_modif_size_t=yes
-
 # ettercap
 ettercap_cv_type_socklen_t=${ettercap_cv_type_socklen_t=yes}
 
-- 
1.6.0.3

Re: [PATCH] enca_1.9.bb: remove unused configure test which is bound to fail.

[Khem Raj]

On (26/08/09 16:48), Graham Gower wrote:
> Enca's configure will fail on everything not arm or ix86.
> 
> Signed-off-by: Graham Gower <graham.gower@gmail.com>
> ---
> recipes/enca/enca_1.9.bb |    7 ++++++-
> site/arm-linux           |    3 ---
> site/ix86-common         |    3 ---
> 3 files changed, 6 insertions(+), 7 deletions(-)
> 
> diff --git a/recipes/enca/enca_1.9.bb b/recipes/enca/enca_1.9.bb
> index 548a824..dbffe1e 100644
> --- a/recipes/enca/enca_1.9.bb
> +++ b/recipes/enca/enca_1.9.bb
> @@ -3,7 +3,7 @@ SECTION = "libs"
> PRIORITY = "optional"
> LICENSE = "GPLv2"
> HOMEPAGE = "http://trific.ath.cx/software/enca/"
> -PR = "r3"
> +PR = "r4"
> 
> SRC_URI = "http://www.sourcefiles.org/Networking/Tools/Miscellanenous/enca-${PV}.tar.bz2 \
> 	file://configure-hack.patch;patch=1 \
> @@ -15,6 +15,11 @@ inherit autotools
> 
> EXTRA_OECONF="--with-libiconv-prefix=${STAGING_DIR_HOST}${layout_exec_prefix}"
> 
> +do_configure_prepend() {
> +	# remove failing test which checks for something that isn't even used
> +	sed -i -e '/ye_FUNC_SCANF_MODIF_SIZE_T/d' configure.ac
> +}

you may not require this hunk. see below

> +
> do_configure_append() {
> 	sed -i s:-I/usr/include::g Makefile
> 	sed -i s:-I/usr/include::g */Makefile
> diff --git a/site/arm-linux b/site/arm-linux
> index 4212728..18dfb38 100644
> --- a/site/arm-linux
> +++ b/site/arm-linux
> @@ -148,9 +148,6 @@ samba_cv_USE_SETREUID=${samba_cv_USE_SETREUID=yes}
> samba_cv_USE_SETUIDX=${samba_cv_USE_SETUIDX=yes}
> samba_cv_LINUX_LFS_SUPPORT=${samba_cv_LINUX_LFS_SUPPORT=yes}
> 
> -# enca
> -yeti_cv_func_scanf_modif_size_t=yes
> -

set this to 'no'

> # clamav
> clamav_av_func_working_snprintf_long=${clamav_av_func_working_snprintf_long=yes}
> clamav_av_have_in_port_t=${clamav_av_have_in_port_t=yes}
> diff --git a/site/ix86-common b/site/ix86-common
> index a5c4b60..b7a503f 100644
> --- a/site/ix86-common
> +++ b/site/ix86-common
> @@ -104,9 +104,6 @@ ac_cv_func_posix_getpwnam_r=${ac_cv_func_posix_getpwnam_r=yes}
> dpkg_cv_va_copy=${ac_cv_va_copy=no}
> dpkg_cv___va_copy=${ac_cv___va_copy=yes}
> 
> -# enca
> -yeti_cv_func_scanf_modif_size_t=yes
> -

same here.

> # ettercap
> ettercap_cv_type_socklen_t=${ettercap_cv_type_socklen_t=yes}
> 
> -- 
> 1.6.0.3
> 
> 
> _______________________________________________
> Openembedded-devel mailing list
> Openembedded-devel@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel

Re: [PATCH] enca_1.9.bb: remove unused configure test which is bound to fail.

[Graham Gower]

2009/8/27 Khem Raj <raj.khem@gmail.com>:
>> +do_configure_prepend() {
>> +     # remove failing test which checks for something that isn't even used
>> +     sed -i -e '/ye_FUNC_SCANF_MODIF_SIZE_T/d' configure.ac
>> +}
>
> you may not require this hunk. see below
>

>> -# enca
>> -yeti_cv_func_scanf_modif_size_t=yes
>> -
>
> set this to 'no'
>

>> -# enca
>> -yeti_cv_func_scanf_modif_size_t=yes
>> -
>
> same here.


The two site files that specify yeti_cv_func_scanf_modif_size_t=yes
probably make things work on those two systems. But every other system
will fail to configure. Hence, a fix which is independent of the
architecture. And with that fix, I figured the arch dependent fixes
should be removed.

-Graham

Re: [PATCH] enca_1.9.bb: remove unused configure test which is bound to fail.

[Phil Blundell]

On Wed, 2009-08-26 at 16:48 +0930, Graham Gower wrote:
> Enca's configure will fail on everything not arm or ix86.
> 
> Signed-off-by: Graham Gower <graham.gower@gmail.com>
> ---
>  recipes/enca/enca_1.9.bb |    7 ++++++-
>  site/arm-linux           |    3 ---
>  site/ix86-common         |    3 ---
>  3 files changed, 6 insertions(+), 7 deletions(-)

Looks fine to me.  It does seem a bit weird that enca tests this scanf
thing and then doesn't use the information but, sure enough, I couldn't
find any reference to the result anywhere in the source code.

Do you have commit access or do you want me to check it in for you?

p.

Re: [PATCH] enca_1.9.bb: remove unused configure test which is bound to fail.

[Graham Gower]

2009/8/27 Phil Blundell <pb@reciva.com>:
>
> Do you have commit access or do you want me to check it in for you?

I don't have commit access, so if you could commit it that would be great.

-Graham

Re: [PATCH] enca_1.9.bb: remove unused configure test which is bound to fail.

[Phil Blundell]

On Thu, 2009-08-27 at 20:31 +0930, Graham Gower wrote:
> 2009/8/27 Phil Blundell <pb@reciva.com>:
> >
> > Do you have commit access or do you want me to check it in for you?
> 
> I don't have commit access, so if you could commit it that would be great.

Righto, done.

p.

Re: [PATCH] enca_1.9.bb: remove unused configure test which is bound to fail.

[Stanislav Brabec]

Phil Blundell wrote:
> On Wed, 2009-08-26 at 16:48 +0930, Graham Gower wrote:
> > Enca's configure will fail on everything not arm or ix86.
> > 
> > Signed-off-by: Graham Gower <graham.gower@gmail.com>
> > ---
> >  recipes/enca/enca_1.9.bb |    7 ++++++-
> >  site/arm-linux           |    3 ---
> >  site/ix86-common         |    3 ---
> >  3 files changed, 6 insertions(+), 7 deletions(-)
> 
> Looks fine to me.  It does seem a bit weird that enca tests this scanf
> thing and then doesn't use the information but, sure enough, I couldn't
> find any reference to the result anywhere in the source code.

It seems to be a relict from old versions.

Here is a new upstream, and the commit that fixes your problem:
http://gitorious.org/enca/enca/commit/163589d468dee45ab5495a8f4f8058d27da91a91


________________________________________________________________________
Stanislav Brabec
http://www.penguin.cz/~utx/zaurus