Re: Clone fails on a repo with too many heads/tags

From: Ivan Todoroski <grnch_lists@gmx.net>
Date: Mar 21, 2012, 20:02 UTC
Message-ID: <4F6A33C5.2080909@gmx.net>
In-Reply-To: <20120321171423.GA13140@sigill.intra.peff.net>

On 21.03.2012 18:14, Jeff King wrote:

Show 12 quoted lines

> On Wed, Mar 21, 2012 at 07:28:24AM -0700, Shawn O. Pearce wrote:
> 
>>> Would it be OK for fetch-pack.c to use the packetized format (pkt-line.h)
>>> for reading the list of refs from stdin?
>> This is probably the easiest way to implement the sneak-into-stdin
>> patch. Use a pkt-line for each argument that should have been in the
>> argv array from the command line, and a flush pkt to terminate the
>> list.
> 
> Something in me feels slightly uncomfortable with that, just because
> simple newline-delimited formats make it easy for people to hack on the
> tool and feed input from unexpected sources.

I understand what you mean. How about this:

If both --stdin and --stateless-rpc are specified to fetch-pack, it will use pkt-line to read the refs from stdin before handing off stdin to get_remote_heads().

However, if only --stdin is specified, it will read refs from stdin in a script-friendly newline delimited format, one ref per line. This is okay because when --stateless-rpc is not specified get_remote_heads() reads from an fd different from stdin so there is no issue with residual buffers in this case.

This way you preserve scriptability for any other callers who don't use --stateless-rpc.

How does this sound?