Re: [PATCH 6/9] kernel-2.eclass: port to esplit.eclass

--=-=-=
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable

Ulrich M=C3=BCller <[email protected]> writes:

>>>>>> On Thu, 09 Jul 2026, Sam James wrote:
>
>> --- a/eclass/kernel-2.eclass
>> +++ b/eclass/kernel-2.eclass
>> @@ -792,7 +792,7 @@ universal_unpack() {
>>  	debug-print "Inside universal_unpack"
>>=20=20
>>  	local OKV_ARRAY
>> -	local IFS=3D.; OKV_ARRAY=3D(${OKV}); unset IFS
>> +	mapfile -d '' OKV_ARRAY < <(IFS=3D. esplit "${OKV}")
>
> TBH, I find the replacement code harder to understand than the original.
> (Probably it is also less efficient?)

I could go either way in this instance, tbh. No real stake in it other
than needing the violation gone.

>
> Also, what about the other instances of this line, and shouldn't
> esplit.eclass be inherited?

I can change those too and I suppose I should to be consistent with
Mike's earlier change, I just didn't because in some other fixed
ebuilds, people only changed the offending instance, maybe because they
preferred the original one but can't use it. Anyway, I'll change it.

It's also why I missed the inherit: I changed the wrong instance :)

>
>>=20=20
>>  	cd "${WORKDIR}" || die
>>  	if [[ ${#OKV_ARRAY[@]} -ge 3 && ${KV_MAJOR} -ge 3 ]]; then

sam

--=-=-=
Content-Type: application/pgp-signature; name="signature.asc"

-----BEGIN PGP SIGNATURE-----

iQEBBAEWCgCpFiEEJaa7iN2bdkxrVUHCc4QJ9SDfkZAFAmpQzygbFIAAAAAABAAO
bWFudTIsMi41KzEuMTIsMiwyXxSAAAAAAC4AKGlzc3Vlci1mcHJAbm90YXRpb25z
Lm9wZW5wZ3AuZmlmdGhob3JzZW1hbi5uZXQyNUE2QkI4OEREOUI3NjRDNkI1NTQx
QzI3Mzg0MDlGNTIwREY5MTkwDxxzYW1AZ2VudG9vLm9yZwAKCRBzhAn1IN+RkLa4
AQCRpwBteH0GYeLx/JbaRjeBwa0TTB43f4vGB5uJpIonIQD/bss8CV7siWbmhDkk
VHcjRt9G+L5RqjXp9IeIepO62QE=
=Ol3E
-----END PGP SIGNATURE-----
--=-=-=--