git-lfs validation

git-lfs validation

[Mauro Queirós]

[-- Attachment #1: Type: text/plain, Size: 522 bytes --]

Hi,



I noticed two things related to git-lfs on
poky/bitbake/lib/bb/fetch2/git.py, which I think may me improved:



First, git.py seems to always check the master branch when looking for lfs
objects (_contains_lfs function), even when the branch is specified in
SRC_URI;



Second, the warning message "Repository %s has LFS content but it is not
being fetched" is printed even when git-lfs is available and "lfs=1" is set.



I could also provide a patch if you think these two issues make sense.

[-- Attachment #2: Type: text/html, Size: 1280 bytes --]

Re: git-lfs validation

[Ross Burton]

Please do send a patch, or ideally two separate patches.

Ross

On Mon, 10 Feb 2020 at 14:14, Mauro Queirós <maurofrqueiros@gmail.com> wrote:
>
> Hi,
>
>
>
> I noticed two things related to git-lfs on poky/bitbake/lib/bb/fetch2/git.py, which I think may me improved:
>
>
>
> First, git.py seems to always check the master branch when looking for lfs objects (_contains_lfs function), even when the branch is specified in SRC_URI;
>
>
>
> Second, the warning message "Repository %s has LFS content but it is not being fetched" is printed even when git-lfs is available and "lfs=1" is set.
>
>
>
> I could also provide a patch if you think these two issues make sense.
>
> --
> _______________________________________________
> bitbake-devel mailing list
> bitbake-devel@lists.openembedded.org
> http://lists.openembedded.org/mailman/listinfo/bitbake-devel

[bitbake-devel] git-lfs validation

[maurofrqueiros]

[-- Attachment #1.1: Type: text/plain, Size: 1776 bytes --]

---------- Forwarded message ---------
De: Mauro Queirós <maurofrqueiros@gmail.com>
Date: quinta, 28/05/2020 à(s) 10:57
Subject: Re: [bitbake-devel] git-lfs validation
To: Ross Burton <ross@burtonini.com>
Cc: <bitbake-devel@lists.openembedded.org>


Hi,

Sorry for the long delay, but only now I managed to get back to this issue.

Actually, I found out another issue. When "lfs=0" is set, the LFS-tracked
files are still downloaded. The flag only allows ignoring errors if Git-LFS
is not installed.

I think that something like the patch attached should work to skip LFS
smudging if "lfs=0" is set. This way, only the pointers are downloaded,
which I think was the intent of creating this flag.

Regards!

Ross Burton <ross@burtonini.com> escreveu no dia segunda, 10/02/2020 à(s)
15:57:

> Please do send a patch, or ideally two separate patches.
>
> Ross
>
> On Mon, 10 Feb 2020 at 14:14, Mauro Queirós <maurofrqueiros@gmail.com>
> wrote:
> >
> > Hi,
> >
> >
> >
> > I noticed two things related to git-lfs on
> poky/bitbake/lib/bb/fetch2/git.py, which I think may me improved:
> >
> >
> >
> > First, git.py seems to always check the master branch when looking for
> lfs objects (_contains_lfs function), even when the branch is specified in
> SRC_URI;
> >
> >
> >
> > Second, the warning message "Repository %s has LFS content but it is not
> being fetched" is printed even when git-lfs is available and "lfs=1" is set.
> >
> >
> >
> > I could also provide a patch if you think these two issues make sense.
> >
> > --
> > _______________________________________________
> > bitbake-devel mailing list
> > bitbake-devel@lists.openembedded.org
> > http://lists.openembedded.org/mailman/listinfo/bitbake-devel
>

[-- Attachment #1.2: Type: text/html, Size: 2925 bytes --]

[-- Attachment #2: 0001-git.py-skip-smudging-if-lfs-0-is-set.patch --]
[-- Type: application/x-patch, Size: 810 bytes --]

Re: git-lfs validation

[Mauro Queirós]

[-- Attachment #1: Type: text/plain, Size: 199 bytes --]

Patches related to all issues described before:
https://patchwork.openembedded.org/patch/172928/
https://patchwork.openembedded.org/patch/172929/
https://patchwork.openembedded.org/patch/172930/

[-- Attachment #2: Type: text/html, Size: 211 bytes --]