Conflict between BBCLASSEXTEND and named recipes.

Conflict between BBCLASSEXTEND and named recipes.

[Jeremy Puhlman]

There seems to be a bug with available providers when a recipe file
provides the same name as an extended recipe. The extended version of
the recipe does not seem to be available at all.

Example:

If zlib_1.2.6.bb exists in a layer, with BBCLASSEXTEND="native
zlib-nativesdk" defined, and zlib-native_1.2.5.bb exists, the version
1.2.6 of zlib-native is not selectable at all.


bitbake -s | grep zlib
zlib                                                :1.2.6-r0
zlib-native                                         :1.2.5-r0
zlib-nativesdk                                      :1.2.6-r0

Even with PREFERRED_PROVIDER_zlib-naitve = 1.2.6 the results are the
same. I have been poking around the parser, but really haven't made any
progress, figured I would poke the list.


-- 
Jeremy Puhlman
Montavista Software, LLC.

Re: Conflict between BBCLASSEXTEND and named recipes.

[Denys Dmytriyenko]

On Fri, Mar 16, 2012 at 10:09:18AM -0700, Jeremy Puhlman wrote:
> There seems to be a bug with available providers when a recipe file
> provides the same name as an extended recipe. The extended version of
> the recipe does not seem to be available at all.
> 
> Example:
> 
> If zlib_1.2.6.bb exists in a layer, with BBCLASSEXTEND="native
> zlib-nativesdk" defined, and zlib-native_1.2.5.bb exists, the version
> 1.2.6 of zlib-native is not selectable at all.
> 
> 
> bitbake -s | grep zlib
> zlib                                                :1.2.6-r0
> zlib-native                                         :1.2.5-r0
> zlib-nativesdk                                      :1.2.6-r0
> 
> Even with PREFERRED_PROVIDER_zlib-naitve = 1.2.6 the results are the
> same. I have been poking around the parser, but really haven't made any
> progress, figured I would poke the list.

I don't have a solution, but I can confirm that this was a known issue all the 
way back when BBCLASSEXTEND got introduced in Classic OE. The workaround was 
to remove or mask the actual *-native.bb recipes.

-- 
Denys

Re: Conflict between BBCLASSEXTEND and named recipes.

[Jeremy Puhlman]

> 
> I don't have a solution, but I can confirm that this was a known issue all the 
> way back when BBCLASSEXTEND got introduced in Classic OE. The workaround was 
> to remove or mask the actual *-native.bb recipes.
> 

Thanks. I though I searched the archives, but apparently not well
enough. My first suggestion was BBMASK and that seemed to work as well.

-- 
Jeremy Puhlman
Montavista Software, LLC.

Re: Conflict between BBCLASSEXTEND and named recipes.

[Martin Jansa]

[-- Attachment #1: Type: text/plain, Size: 1442 bytes --]

On Fri, Mar 16, 2012 at 10:09:18AM -0700, Jeremy Puhlman wrote:
> There seems to be a bug with available providers when a recipe file
> provides the same name as an extended recipe. The extended version of
> the recipe does not seem to be available at all.
> 
> Example:
> 
> If zlib_1.2.6.bb exists in a layer, with BBCLASSEXTEND="native
> zlib-nativesdk" defined, and zlib-native_1.2.5.bb exists, the version
> 1.2.6 of zlib-native is not selectable at all.
> 
> 
> bitbake -s | grep zlib
> zlib                                                :1.2.6-r0
> zlib-native                                         :1.2.5-r0
> zlib-nativesdk                                      :1.2.6-r0
> 
> Even with PREFERRED_PROVIDER_zlib-naitve = 1.2.6 the results are the
> same. I have been poking around the parser, but really haven't made any
> progress, figured I would poke the list.

I though that setting PREFERRED_VERSION should be enough in this case.

notice that it's not PREFERRED_PROVIDER and you have "naitve"/"native"
typo there. Can you retest with correct form if you haven't already?

Cheers,
> 
> 
> -- 
> Jeremy Puhlman
> Montavista Software, LLC.
> 
> _______________________________________________
> bitbake-devel mailing list
> bitbake-devel@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/bitbake-devel

-- 
Martin 'JaMa' Jansa     jabber: Martin.Jansa@gmail.com

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 205 bytes --]