Re: [PATCH v2 1/4] streaming: allow to call close_istream(NULL);

From: Junio C Hamano <gitster@pobox.com>
To: "Nguyễn Thái Ngọc Duy" <pclouds@gmail.com>
Cc: git@vger.kernel.org, Nicolas Pitre <nico@fluxnic.net>
Subject: Re: [PATCH v2 1/4] streaming: allow to call close_istream(NULL);
Date: Fri, 18 May 2012 14:02:38 -0700	[thread overview]
Message-ID: <xmqqobpl2oep.fsf@junio.mtv.corp.google.com> (raw)
In-Reply-To: <1337169731-23416-1-git-send-email-pclouds@gmail.com> (=?utf-8?B?Ik5ndXnhu4VuCVRow6FpIE5n4buNYw==?= Duy"'s message of "Wed, 16 May 2012 19:02:08 +0700")

Nguyễn Thái Ngọc Duy  <pclouds@gmail.com> writes:

> This makes it more convenient in cleanup code, like free(NULL).

If the justification were "like fclose(NULL)", it have made more sense,
but fclose(NULL) does not silently turn itself into a successful no-op.
You are expected to check the returned value from the matching function
(i.e. fopen()) and do not call it.  So taking this patch alone, I smell
a bad design taste.

> Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
> ---
>  streaming.c |    2 +-
>  1 files changed, 1 insertions(+), 1 deletions(-)
>
> diff --git a/streaming.c b/streaming.c
> index 3a3cd12..38b39cd 100644
> --- a/streaming.c
> +++ b/streaming.c
> @@ -94,7 +94,7 @@ struct git_istream {
>  
>  int close_istream(struct git_istream *st)
>  {
> -	int r = st->vtbl->close(st);
> +	int r = st ? st->vtbl->close(st) : 0;
>  	free(st);
>  	return r;
>  }