On Sun, 2016-10-02 at 17:43 -0700, Stephano Cetola wrote:
> If attempting to patch a git repo without a proper git config setup,
> an error will occur saying user.name/user.email are needed by git
> am/apply. After some code was removed from kernel-yocto, it was
> simple enough to reproduce this error by creating a kernel patch and
> using a container to build.
> 
> This patch abstracts out functionality that existed in buildhistory
> for use in other classes. It also adds a call to this functionality
> to the kernel-yocto class.
> 
> Fixes [YOCTO #10346]
> 
> introduced in
> OE-core revision 0f698dfd1c8bbc0d53ae7977e26685a7a3df52a3
> 
> Signed-off-by: Stephano Cetola <[email protected]>
> ---
>  meta/classes/buildhistory.bbclass | 13 ++++---------
>  meta/classes/kernel-yocto.bbclass |  2 ++
>  meta/classes/sanity.bbclass       | 15 +++++++++++++++
>  3 files changed, 21 insertions(+), 9 deletions(-)
> 
> diff --git a/meta/classes/buildhistory.bbclass
> b/meta/classes/buildhistory.bbclass
> index 6e5de0e..ed79029 100644
> --- a/meta/classes/buildhistory.bbclass
> +++ b/meta/classes/buildhistory.bbclass
> @@ -688,6 +688,7 @@ END
>       rm $commitmsgfile
>  }
>  
> +inherit sanity
>  buildhistory_commit() {
>       if [ ! -d ${BUILDHISTORY_DIR} ] ; then
>               # Code above that creates this dir never executed,
> so there can't be anything to commit
> @@ -708,15 +709,9 @@ END
>                       git tag -f build-minus-2 build-minus-1 >
> /dev/null 2>&1 || true
>                       git tag -f build-minus-1 > /dev/null 2>&1 ||
> true
>               fi
> -             # If the user hasn't set up their name/email, set
> some defaults
> -             # just for this repo (otherwise the commit will fail
> with older
> -             # versions of git)
> -             if ! git config user.email > /dev/null ; then
> -                     git config --local user.email
> "buildhistory@${DISTRO}"
> -             fi
> -             if ! git config user.name > /dev/null ; then
> -                     git config --local user.name "buildhistory"
> -             fi
> +
> +             check_git_config "buildhistory"
> +
>               # Check if there are new/changed files to commit
> (other than metadata-revs)
>               repostatus=`git status --porcelain | grep -v "
> metadata-revs$"`
>               HOSTNAME=`hostname 2>/dev/null || echo unknown`
> diff --git a/meta/classes/kernel-yocto.bbclass b/meta/classes/kernel-
> yocto.bbclass
> index 53659f2..035edeb 100644
> --- a/meta/classes/kernel-yocto.bbclass
> +++ b/meta/classes/kernel-yocto.bbclass
> @@ -156,9 +156,11 @@ do_kernel_metadata() {
>       fi
>  }
>  
> +inherit sanity
>  do_patch() {
>       cd ${S}
>  
> +     check_git_config "kernel-yocto"
>       meta_dir=$(kgit --meta)
>       (cd ${meta_dir}; ln -sf patch.queue series)
>       if [ -f "${meta_dir}/series" ]; then
> diff --git a/meta/classes/sanity.bbclass
> b/meta/classes/sanity.bbclass
> index 7682ffb..40033ab 100644
> --- a/meta/classes/sanity.bbclass
> +++ b/meta/classes/sanity.bbclass
> @@ -1023,3 +1023,18 @@ python check_sanity_eventhandler() {
>  
>      return
>  }
> +
> +# If the user hasn't set up their name/email, set some defaults
> +check_git_config() {
> +     if [ -z "$1" ]; then
> +             name="OE"
> +     else
> +             name="$1"
> +     fi
> +     if ! git config user.email > /dev/null ; then
> +             git config --local user.email "${name}@${DISTRO}"
> +     fi
> +     if ! git config user.name > /dev/null ; then
> +             git config --local user.name "${name}"
> +     fi
> +}

I worry you're causing a load of the classes to end up cross linked
here which is bad from a usability standpoint. Nobody can easily
disable the sanity checks after this patch.

If we're keeping this as a shell function this function would probably
go to utils.bbclass. Adding something to lib/oe is probably ultimately
the way to go but I appreciate its harder to write the above in a
python lib, so perhaps utils.bbclass is good enough for now? We can
assume utils is inherited globally.

Cheers,

Richard


-- 
_______________________________________________
Openembedded-core mailing list
[email protected]
http://lists.openembedded.org/mailman/listinfo/openembedded-core

Reply via email to