From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <alexander.turenko@tarantool.org>
Received: from smtp20.mail.ru (smtp20.mail.ru [94.100.179.251])
 (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits))
 (No client certificate requested)
 by dev.tarantool.org (Postfix) with ESMTPS id 955BC430407
 for <tarantool-patches@dev.tarantool.org>;
 Tue, 25 Aug 2020 17:51:10 +0300 (MSK)
Date: Tue, 25 Aug 2020 17:51:03 +0300
From: Alexander Turenko <alexander.turenko@tarantool.org>
Message-ID: <20200825145103.hiqmjkimuk3al3rt@tkn_work_nb>
References: <20200825140108.52090-1-huston.mavr@gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
In-Reply-To: <20200825140108.52090-1-huston.mavr@gmail.com>
Subject: Re: [Tarantool-patches] [PATCH v2] build: refactor static build
	process
List-Id: Tarantool development patches <tarantool-patches.dev.tarantool.org>
List-Unsubscribe: <https://lists.tarantool.org/mailman/options/tarantool-patches>, 
 <mailto:tarantool-patches-request@dev.tarantool.org?subject=unsubscribe>
List-Archive: <https://lists.tarantool.org/pipermail/tarantool-patches/>
List-Post: <mailto:tarantool-patches@dev.tarantool.org>
List-Help: <mailto:tarantool-patches-request@dev.tarantool.org?subject=help>
List-Subscribe: <https://lists.tarantool.org/mailman/listinfo/tarantool-patches>, 
 <mailto:tarantool-patches-request@dev.tarantool.org?subject=subscribe>
To: HustonMmmavr <huston.mavr@gmail.com>
Cc: tarantool-patches@dev.tarantool.org, yaroslav.dynnikov@gmail.com

I looked very briefly (not thoroughly at all) on this iteration.

There is nothing that confuses me (except few tiny comments below).

I hope Igor will do thorough review.

WBR, Alexander Turenko.

> +if (APPLE)
> +    find_program(C_COMPILER clang)
> +    find_program(CXX_COMPILER clang++)
> +else()
> +    find_program(C_COMPILER gcc)
> +    find_program(CXX_COMPILER g++)
> +endif()

Can we just leave it default?

In offline discussion Alexandr B. said that tarantool builds with gcc,
but icu with clang that gives some problem.

Possible solution is to pass ${CMAKE_C_COMPILER} (and CXX too where
necessary) to a subproject as we do for c-ares and curl. It seems it is
already done, so maybe it worth to re-check whether it solves the
problem.

Anyway, if we really need to set a compiler here explicitly, I don't
mind. Just noted that this way is somewhat unusual as I see.

> diff --git a/static-build/test/static-build/box.lua b/static-build/test/static-build/box.lua
> new file mode 100755
> index 000000000..bad8a9055
> --- /dev/null
> +++ b/static-build/test/static-build/box.lua
> @@ -0,0 +1,3 @@
> +#!/usr/bin/env tarantool
> +
> +require('console').listen(os.getenv('ADMIN'))

Is looks redundant, see the comment below.

> diff --git a/static-build/test/static-build/suite.ini b/static-build/test/static-build/suite.ini
> new file mode 100644
> index 000000000..92e349466
> --- /dev/null
> +++ b/static-build/test/static-build/suite.ini
> @@ -0,0 +1,5 @@
> +[default]
> +core = app
> +description = Static build tests
> +script = box.lua
> +is_parallel = True

'script' does not have sense for 'core = app' test, it is for 'core =
tarantool' tests.