From: Michael G Schwern <schwern@pobox.com>
Date: Fri, 27 Jul 2012 08:19:37 GMT
Subject: Re: [PATCH 1/4] Extract some utilities from git-svn to allow extracting Git::SVN.
Message-ID: <50124F19.1050502@pobox.com>
In-Reply-To: <7vobn13hps.fsf@alter.siamese.dyndns.org>

On 2012.7.26 10:18 PM, Junio C Hamano wrote:
>> +# Note: not certain why this is in use instead of die.  Probably because
>> +# the exit code of die is 255?  Doesn't appear to be used consistently.
>> +sub fatal (@) { print STDERR "@_\n"; exit 1 }
> 
> Very true.  Also I do not think the line-noise prototype buys us
> anything (other than making the code look mysterious to non Perl
> programmers); we are not emulating any Perl's builtin with this
> function, and I do not see a reason why we want to force list
> context to its arguments, either.  But removal of it is not part of
> this step anyway, so I wouldn't complain.

The prototype does absolutely nothing since @ is the default prototype.  But
yes, I'm doing a very rote refactoring here.


>> +sub can_compress {
>> +    return $can_compress if defined $can_compress;
>> +
>> +    return $can_compress = eval { require Compress::Zlib; } ? 1 : 0;
>> +}
> 
> The original said "eval { require Compress::Zlib; 1; }"; presumably,
> when require does succeed, the value inside is the "1;" that has to
> be at the end of Compress::Zlib, so the difference should not matter.

Yes.  In other situations where you cannot guarantee that the statement in the
eval will return true it makes sense, but here it's redundant.


-- 
Being faith-based doesn't trump reality.
	-- Bruce Sterling