Re: [PATCH 4/4 v2] dpkg-base: Update git alternates

[Henning Schild <henning.schild@siemens.com>]

Am Mon, 27 Nov 2017 11:03:47 +0300
schrieb Alexander Smirnov <asmirnov@ilbers.de>:

> On 11/27/2017 10:44 AM, Henning Schild wrote:
> > Am Fri, 24 Nov 2017 18:12:43 +0300
> > schrieb Alexander Smirnov <asmirnov@ilbers.de>:
> >   
> >> On 11/24/2017 05:57 PM, Henning Schild wrote:  
> >>> Am Thu, 23 Nov 2017 13:25:32 +0300
> >>> schrieb Alexander Smirnov <asmirnov@ilbers.de>:
> >>>      
> >>>> Update git alternates to work in buildchroot.
> >>>>
> >>>> Signed-off-by: Alexander Smirnov <asmirnov@ilbers.de>
> >>>> ---
> >>>>    meta/classes/dpkg-base.bbclass | 7 +++++++
> >>>>    1 file changed, 7 insertions(+)
> >>>>
> >>>> diff --git a/meta/classes/dpkg-base.bbclass
> >>>> b/meta/classes/dpkg-base.bbclass index 21ae88d..d51a709 100644
> >>>> --- a/meta/classes/dpkg-base.bbclass
> >>>> +++ b/meta/classes/dpkg-base.bbclass
> >>>> @@ -18,6 +18,13 @@ dpkg_runbuild() {
> >>>>    
> >>>>    # Wrap the function dpkg_runbuild with the bind mount for
> >>>> buildroot do_build() {
> >>>> +    if [ -d ${WORKDIR}/git/.git ]; then
> >>>> +        OBJ_PATH=$(cat
> >>>> ${WORKDIR}/git/.git/objects/info/alternates)
> >>>> +        REPO_PATH=$(dirname $OBJ_PATH)
> >>>> +        REPO_NAME=$(basename $REPO_PATH)
> >>>> +        echo "/git/$REPO_NAME/objects" >
> >>>> ${WORKDIR}/git/.git/objects/info/alternates
> >>>> +    fi  
> >>>
> >>> Does that not overwrite a file so that the repo can now work only
> >>> in buildchroot and will be broken on the host?
> >>>      
> >>
> >> It does, but in general there is no use-case for having working
> >> repo clone on the host. I could restore original alternates in
> >> cleanup hook, but it really makes no sense.  
> > 
> > I do not like breaking that clone, even for a short while and
> > repairing it later.
> > The host-clone might be looked at while debugging.
> > And one could imagine the fetcher/unpacker supporting deduplication
> > if multiple recipes use the same repo. Even if such a feature is not
> > present in bitbake today, if it is implemented Isar will break.
> > Think one recipe being in do_build and another one in do_fetch at
> > the same time.  
> 
> Don't understand what you mean, but probably the following items
> could better describe what happens in this patch:
> 
> 1. There is Isar downloads folder 'tmp/downloads', which contains git 
> clones from origins.
> 2. There is workdir folder per recipe, which contains git clone from 
> Isar downloads folder.
> 3. The patch modifies clone in workdir, so the only current recipe is 
> user for this repo.
> 4. The downloads folder content stays unmodified for all the time.

I see, in that case i have no objections.


Thanks,
Henning

> Alex