Re: [PATCH v2] Makefile: use sha1collisiondetection by default on OSX and Darwin

From: Junio C Hamano <gitster@pobox.com>
Date: Dec 15, 2022, 21:17 UTC
Message-ID: <xmqqlen8s6xv.fsf@gitster.g>
In-Reply-To: <patch-v2-1.1-3de7cdbd260-20221215T084129Z-avarab@gmail.com>

Ævar Arnfjörð Bjarmason  <avarab@gmail.com> writes:

> Let's also use sha1collisiondetection on OSX, to do so we'll need to
> split up the "APPLE_COMMON_CRYPTO" flag into that flag and a new
> "APPLE_COMMON_CRYPTO_SHA1".

That's well reasoned

This leaves these in Makefile:

        # Define NO_APPLE_COMMON_CRYPTO if you are building on Darwin/Mac OS X
        # and do not want to use Apple's CommonCrypto library.  This allows you
        # to provide your own OpenSSL library, for example from MacPorts.

        ifndef NO_APPLE_COMMON_CRYPTO
                NO_OPENSSL = YesPlease
                APPLE_COMMON_CRYPTO = YesPlease
                COMPAT_CFLAGS += -DAPPLE_COMMON_CRYPTO
        endif

Because we only mention what it does to use NO_APPLE_COMMON_CRYPTO here, without promising anything about what happens otherwise, we can do this change without breaking any promises ;-)

However there is this bit:

        ifdef APPLE_COMMON_CRYPTO
                LIB_4_CRYPTO += -framework Security -framework CoreFoundation
        endif

So, if one says

    make NO_APPLE_COMMON_CRYPTO=NoThanks APPLE_COMMON_CRYPTO_SHA1=YesPlease

presumably the end result will fail to link? I _think_ that falls into "if it hurts, don't do that", and

 (1) automatically disabling the latter when the former is set would
     be more confusing than helpful.

 (2) explicitly detecting the situation and issue an error message
     from the Makefile might appear nicer, but if the linker does
     the failing with a messaage fine, that would be sufficient.

so in the end, the posted patch as-is should be fine, I think.