From: Jonathan Nieder <jrnieder@gmail.com>
Date: Thu, 13 Jan 2011 05:38:49 GMT
Subject: Re: [PATCH 1/2] unpack-trees: handle lstat failure for existing directory
Message-ID: <20110113053849.GA10371@burratino>
In-Reply-To: <buobp3ls9l1.fsf@dhlpc061.dev.necel.com>

Miles Bader wrote:
> Jonathan Nieder <jrnieder@gmail.com> writes:

>> @@ -1382,7 +1382,9 @@ static int verify_absent_1(struct cache_entry *ce,
>>  		return check_ok_to_remove(ce->name, ce_namelen(ce),
>>  				ce_to_dtype(ce), ce, &st,
>>  				error_type, o);
>> -
>> +	if (errno != ENOENT)
>> +		return error("cannot stat '%s': %s", ce->name,
>> +				strerror(errno));
>>  	return 0;
>
> Is errno guaranteed to be set to something relevant at this point in the
> code...?

Yes, because lstat failed.  But perhaps that is a hint that the code
should be restructured as

	} else if (lstat(... )) {
		if (errno != ENOENT)
			return error(...
		return 0;
	} else {
		return check_ok_to_remove(...
	}