From: Jonathan Nieder <jrnieder@gmail.com>
Date: Fri, 03 Dec 2010 09:40:20 GMT
Subject: Re: [PATCH 06/10] parse-options: never suppress arghelp if LITERAL_ARGHELP is set
Message-ID: <20101203094020.GA11910@burratino>
In-Reply-To: <4CF8B55F.8020203@gmail.com>

Stephen Boyd wrote:
> On 12/01/10 15:31, Jonathan Nieder wrote:

>> The motivation is to allow update-index to correctly advertise
>> 
>> 	--cacheinfo <mode> <object> <path>
>> 	                      add the specified entry to the index
>> 
>> while abusing PARSE_OPT_NOARG to disallow the "sticked form"
>> 
>> 	--cacheinfo=<mode> <object> <path>
[...]
> parse-options should accept both forms of --cacheinfo above if the
> option isn't marked PARSE_OPT_NOARG. Marking it NOARG to get rid of the
> equals sign in the usage seems wrong when both the equals sign and no
> equals sign can be accepted.

Just to clarify: the NOARG was not meant to affect the usage message
but the actual accepted usage.  The idea was that

	git update-index --cacheinfo=100644 87a8767c87b file.c

should be rejected, because if it is accepted that would tempt people
to try

	git update-index --cacheinfo=100644 -q 87a8767c87b file.c

which fails.  That is, the argument to --cacheinfo is not <mode>,
since --cacheinfo takes _three_ arguments and therefore the sticked
form sends a wrong message.

> I know this is a bit more code, but it also means that we don't have
> this approach bite someone else down the line when they make assumptions
> about options marked as NOARG not taking arguments.
>
> We should probably add another check like "if argh is set and
> PARSE_OPT_NOARG is true error out" so this can't be done.

I agree with your conclusion.  Let's drop this patch, and I'll look
into adding the check and a PARSE_OPT_NOSTICKED flag.