bug#67848: [PATCH] WIP: gnu: icedove: Update to 115.5.2.

bug#67848: [PATCH] WIP: gnu: icedove: Update to 115.5.2.

[Maxim Cournoyer]

Hi Jonathan,

Jonathan Brielmaier <jonathan.brielmaier@web.de> writes:

> * gnu/packages/gnuzilla.scm (all-mozilla-102-locales,
> %icecat-102-base-version, %icecat-102-version, icecat-102-source):
> Remove variables.
> (%icedove-build-id, %icedove-version, thunderbird-comm-source): Update for 115.5.2.
> (thunderbird-comm-l10n): Switch to provided tarball.
> (icedove-source)[origin]: Add `zstd` package. Use `icecat-source`.
> Provide l10n folder from tarball.
> (icedove-minimal)[arguments]: Use MACH_BUILD_PYTHON_NATIVE_PACKAGE_SOURCE instead
> of deprecated MACH_USE_SYSTEM_PYTHON. Use GUIX_PYTHONPATH and bundled
> NSS library.
> [inputs]: Use `icu4c-73-promise`. Add `nspr`. Remove `nss` for now.

Thanks for working on this!

I don't think it's acceptable to reintroduce bundling for NSS, unless we
have strong reasons to.  We should rather introduce a new NSS variant at
a sufficient version in a prior commit.  NSS takes time to build, but
otherwise shouldn't be too difficult, if I recall correctly my
experience doing it.

-- 
Thanks,
Maxim

bug#67848: [PATCH] WIP: gnu: icedove: Update to 115.5.2.

[Jonathan Brielmaier via Bug reports for GNU Emacs, the Swiss army knife of text editors]

Am 17.12.23 um 23:02 schrieb Maxim Cournoyer:> I don't think it's
acceptable to reintroduce bundling for NSS, unless we
> have strong reasons to.  We should rather introduce a new NSS variant at
> a sufficient version in a prior commit.  NSS takes time to build, but
> otherwise shouldn't be too difficult, if I recall correctly my
> experience doing it.

Oke, I'll try to prepare a newer NSS and use it also in Icecat (which
switched also to the bundled NSS).

bug#67848: [PATCH] WIP: gnu: icedove: Update to 115.5.2.

[Maxim Cournoyer]

Hi Jonathan,

Jonathan Brielmaier <jonathan.brielmaier@web.de> writes:

> Am 17.12.23 um 23:02 schrieb Maxim Cournoyer:> I don't think it's
> acceptable to reintroduce bundling for NSS, unless we
>> have strong reasons to.  We should rather introduce a new NSS variant at
>> a sufficient version in a prior commit.  NSS takes time to build, but
>> otherwise shouldn't be too difficult, if I recall correctly my
>> experience doing it.
>
> Oke, I'll try to prepare a newer NSS and use it also in Icecat (which
> switched also to the bundled NSS).

Sounds good, I'll be happy to review it!

-- 
Thanks,
Maxim