From: Ross Burton <Ross.Burton@arm.com>
To: "shinucha@cisco.com" <shinucha@cisco.com>
Cc: "openembedded-core@lists.openembedded.org"
<openembedded-core@lists.openembedded.org>,
"xe-linux-external@cisco.com" <xe-linux-external@cisco.com>
Subject: Re: [OE-core][PATCH v2] boost: make Boost.Locale ICU backend optional
Date: Tue, 4 Aug 2026 10:44:21 +0000 [thread overview]
Message-ID: <EBE6CBA0-EB72-47D6-AF42-32D22691D4AA@arm.com> (raw)
In-Reply-To: <20260729061232.3547458-1-shinucha@cisco.com>
On 29 Jul 2026, at 07:12, Shinu Chandran via lists.openembedded.org <shinucha=cisco.com@lists.openembedded.org> wrote:
> +BOOST_LOCALE_ICU_DEP = "${@bb.utils.contains('PACKAGECONFIG', 'locale', 'icu', '', d)}"
> +PACKAGECONFIG[locale] = ",,"
> +PACKAGECONFIG[locale-icu] = ",,${BOOST_LOCALE_ICU_DEP}"
The use of BOOST_LOCALE_ICU_DEP is overcomplicating things: if the distro turns on locale-icu but not locale then locale not enabling is entirely user error. Just set the dependency directly.
> PACKAGECONFIG[graph_parallel] = ",,,boost-mpi mpich"
> PACKAGECONFIG[mpi] = ",,mpich"
> PACKAGECONFIG[python] = ",,python3"
> @@ -145,10 +147,13 @@ BJAM_TOOLS = "--ignore-site-config \
>
> # use PARALLEL_MAKE to speed up the build
> BOOST_PARALLEL_MAKE = "${@oe.utils.parallel_make_argument(d, '-j%d')}"
> +# --disable-icu below controls Boost.Regex ICU support. Boost.Locale ICU is
> +# handled independently with boost.locale.icu.
So there’s an option to enable/disable the use of ICU in boost.locale, which you’re exposing.
There’s also an option to enable/disable the use of ICU in boost.regex, which is not exposed.
What’s the rationale behind enabling it in some parts of boost but not others? It seems like there should be an “icu” option which flips the toggles on all parts of boost that have an optional dependency on ICU.
> +BOOST_LOCALE_ICU_OPTS = "${@bb.utils.contains('PACKAGECONFIG', 'locale locale-icu', 'boost.locale.icu=on -sICU_PATH=${STAGING_EXECPREFIXDIR}', 'boost.locale.icu=off', d)}"
Instead of doing this, inline the argument into the PACKAGECONFIG[] statements and pass ${PACKAGECONFIG_CONFARGS} to BJAM_OPTS. This variable is generated at build time and contains the relevant enable/disable statements based on the selected configuration.
Ross
prev parent reply other threads:[~2026-08-04 10:45 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-07-28 5:16 [OE-core] [master] [PATCH] boost: make Boost.Locale ICU backend optional Shinu Chandran
2026-07-28 10:21 ` Alexander Kanavin
2026-07-29 6:12 ` [OE-core][PATCH v2] " Shinu Chandran
2026-08-04 10:44 ` Ross Burton [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=EBE6CBA0-EB72-47D6-AF42-32D22691D4AA@arm.com \
--to=ross.burton@arm.com \
--cc=openembedded-core@lists.openembedded.org \
--cc=shinucha@cisco.com \
--cc=xe-linux-external@cisco.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox