From: Ivan Todoroski <grnch_lists@gmx.net>
Date: Sun, 25 Mar 2012 09:52:35 GMT
Subject: Re: [PATCH/RFC 2/2] remote-curl: send the refs to fetch-pack on stdin
Message-ID: <4F6EEAE3.60204@gmx.net>
In-Reply-To: <20120325012432.GD27651@sigill.intra.peff.net>

On 25.03.2012 03:24, Jeff King wrote:
> On Sat, Mar 24, 2012 at 09:54:16PM +0100, Ivan Todoroski wrote:
>> -static int rpc_service(struct rpc_state *rpc, struct discovery *heads)
>> +static int rpc_service(struct rpc_state *rpc, struct discovery *heads,
>> +		int nr_fetch, struct ref **to_fetch)
> 
> I was curious why you needed to add new arguments for this, since we
> surely must be passing the information into rpc_service already (since
> it has to put them on the command line). And the answer is that they are
> already in the argv member of the struct rpc_state. I wonder if it would
> fit the existing style of the code to pass the arguments through a
> member in the rpc_state in the same way. I don't feel strongly about it,
> though.

Yeah, I realized this was ugly after I sent it.... not only are new 
arguments added to rpc_service(), but they seem completely arbitrary and 
specific to fetch-pack. There is no reason other callers of 
rpc_service() would want args in the exact same format.

On the other hand I can't reuse rpc_state.argv because that is already 
passed to client.argv for start_command().

Would it be OK if I add a new memeber in rpc_state, e.g. "struct strbuf 
*stdin_preamble"? If non-NULL, it would contain any data the caller of 
rpc_service() wants shoved into the stdin of the sub-command before 
anything else. That way the caller is free to format this data in 
whatever format they need, not only in the pkt-line format needed by 
fetch-pack.