From: Junio C Hamano <gitster@pobox.com>
Date: Wed, 16 Jan 2019 04:12:20 GMT
Subject: Re: [PATCH v2 0/4] Sideband the whole fetch v2 response
Message-ID: <xmqqy37lnsuz.fsf@gitster-ct.c.googlers.com>
In-Reply-To: <20190116003828.64889-1-jonathantanmy@google.com>

Jonathan Tan <jonathantanmy@google.com> writes:

> We could make the caller of demultiplex_sideband() store the outbuf, but
> at this point, it might be better to refactor packet_reader into its own
> file and have it depend on both pkt-line.h and sideband.h.

I do not know or too deeply care about the pkt-line / sideband
distinction, as they are logically one thing.  With a packet-reader
abstraction in place, doesn't this "here is the remainder of the
packet from the  previous round" belong to each reader instance?