Re: [OE-core] [PATCH] classes: decode output data to text

2023-01-16 Thread Pawel Zalewski
That is a good point, looking at the codebase I think that the go-to for dealing with it is using the call to ".decode('utf8')". Thanks, Pawel On Mon, 16 Jan 2023 at 11:01, Richard Purdie < richard.pur...@linuxfoundation.org> wrote: > On Thu, 2022-12-22 at 10:38 +, Pawel Zalewski wrote: > >

Re: [OE-core] [PATCH] classes: decode output data to text

2023-01-16 Thread Richard Purdie
On Thu, 2022-12-22 at 10:38 +, Pawel Zalewski wrote: > The default return value from subprocess.check_output is an encoded byte. > The applied fix will decode the value to a string. > > Signed-off-by: Pawel Zalewski > --- > meta/classes/fs-uuid.bbclass | 2 +- > 1 file changed, 1 insertion(+

Re: [OE-core] [PATCH] classes: decode output data to text

2023-01-16 Thread Richard Purdie
On Fri, 2023-01-06 at 14:32 +0100, Alexander Kanavin wrote: > Now I see what's going on. The fs-uuid class contains two functions: > get_rootfs_uuid, and replace_rootfs_uuid. The former is not used > anywhere, but the latter is. Can you introduce get_rootfs_uuid() into > something in oe-core, perha

Re: [OE-core] [PATCH] classes: decode output data to text

2023-01-06 Thread Alexander Kanavin
Now I see what's going on. The fs-uuid class contains two functions: get_rootfs_uuid, and replace_rootfs_uuid. The former is not used anywhere, but the latter is. Can you introduce get_rootfs_uuid() into something in oe-core, perhaps it could enhance existing code somewhere (maybe complement replac

Re: [OE-core] [PATCH] classes: decode output data to text

2023-01-06 Thread Pawel Zalewski
1. inherit fs-uuid in some-recipe 2. set ROOTFS to some-existing-rootfs 3. make call to get_rootfs_uuid somewhere in that recipe ie. "${@get_rootfs_uuid(d)}" 3. bitbake some-recipe Pawel On Fri, 6 Jan 2023 at 12:43, Alexander Kanavin wrote: > > Can you please describe steps to reproduce? > > Ale

Re: [OE-core] [PATCH] classes: decode output data to text

2023-01-06 Thread Alexander Kanavin
Can you please describe steps to reproduce? Alex On Fri 6. Jan 2023 at 13.40, Pawel Zalewski wrote: > It is not actually being used by default in Yocto ? > This was found in kirkstone. > But regardless, it is wrong and will drop an error. > > Kind regards, > Pawel > > On Fri, 6 Jan 2023 at 12:0

Re: [OE-core] [PATCH] classes: decode output data to text

2023-01-06 Thread Pawel Zalewski
It is not actually being used by default in Yocto ? This was found in kirkstone. But regardless, it is wrong and will drop an error. Kind regards, Pawel On Fri, 6 Jan 2023 at 12:06, Richard Purdie wrote: > > On Thu, 2022-12-22 at 10:38 +, Pawel Zalewski wrote: > > The default return value fr

Re: [OE-core] [PATCH] classes: decode output data to text

2023-01-06 Thread Richard Purdie
On Thu, 2022-12-22 at 10:38 +, Pawel Zalewski wrote: > The default return value from subprocess.check_output is an encoded byte. > The applied fix will decode the value to a string. > > Signed-off-by: Pawel Zalewski > --- > meta/classes/fs-uuid.bbclass | 2 +- > 1 file changed, 1 insertion(+

[OE-core] [PATCH] classes: decode output data to text

2022-12-22 Thread Pawel Zalewski
The default return value from subprocess.check_output is an encoded byte. The applied fix will decode the value to a string. Signed-off-by: Pawel Zalewski --- meta/classes/fs-uuid.bbclass | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/meta/classes/fs-uuid.bbclass b/meta/clas