Re: [PATCH 3/7] pack-protocol.txt: Mark all LFs in push-cert as required

From: Junio C Hamano <gitster@pobox.com>
Date: Jul 1, 2015, 20:39 UTC
Message-ID: <xmqq4mln8ve2.fsf@gitster.dls.corp.google.com>
In-Reply-To: <xmqq8uaz8vjb.fsf@gitster.dls.corp.google.com>

Junio C Hamano <gitster@pobox.com> writes:

Show 6 quoted lines

>> I am moderately negative about this; wouldn't it make the end result
>> cleaner to fix the implementation?
>
> I think that something like this should be sufficient.  As the
> receiving end, we must not complain if there is no terminator.
> ...

And the change we are *not* going to make, but I made temporarily only for testing, on the sending side to violate our "sender SHOULD terminate with LF" rule would look like this:

There is a slight complication on sending an empty line without any termination, though ;-) The reader that calls packet_read() cannot tell such a payload from a flush packet, I think.

*That* may be something we want to document.

diff --git a/send-pack.c b/send-pack.c
index 2a64fec..1a743db 100644
--- a/send-pack.c
+++ b/send-pack.c
@@ -273,9 +273,11 @@ static int generate_push_cert(struct strbuf *req_buf,
 
 	packet_buf_write(req_buf, "push-cert%c%s", 0, cap_string);
 	for (cp = cert.buf; cp < cert.buf + cert.len; cp = np) {
+		int len;
 		np = next_line(cp, cert.buf + cert.len - cp);
+		len = (np <= cp + 1) ? 1 : (np - cp - 1);
 		packet_buf_write(req_buf,
-				 "%.*s", (int)(np - cp), cp);
+				 "%.*s", len, cp);
 	}
 	packet_buf_write(req_buf, "push-cert-end\n");