Re: [PATCH v2 0/2] Accept error packets in any context

From: Jonathan Nieder <jrnieder@gmail.com>
Date: Jan 15, 2019, 01:49 UTC
Message-ID: <20190115014950.GN162110@google.com>
In-Reply-To: <20190115014348.GM162110@google.com>

Jonathan Nieder wrote:

> Junio C Hamano wrote:

Show 14 quoted lines

>> In short, are you shooting js/smart-http-detect-remote-error topic
>> down and replacing it with this one?
>>
>> As that topic is not yet in 'next', I am perfectly fine doing that.
>> I just want to make sure that is what you meant, as my reading of
>> [4] was a bit fuzzy.
>
> Josh, looking at that branch, I see:
>
>  remote-curl: die on server-side errors
>
> 	Includes a test illustrating error handling in the ref
> 	advertisement.  Should that be revived as a standalone patch,
> 	without the remote-curl.c change?

In fact, it appears I have that locally:

 commit 7fe6abcd775dbffd919891d5838f8d125e41f160
 Author: Josh Steadmon <steadmon@google.com>
 Date:   Tue Dec 11 16:25:18 2018 -0800

    lib-httpd, t5551: check server-side HTTP errors

    Add an HTTP path to the test server config that returns an ERR pkt-line
    unconditionally. Verify in t5551 that remote-curl properly detects this
    ERR message and aborts.

    Signed-off-by: Josh Steadmon <steadmon@google.com>
    Signed-off-by: Jeff King <peff@peff.net>
    Signed-off-by: Junio C Hamano <gitster@pobox.com>

It's handled fine by the merge 7be333a6362882e8ffceef3de830dbbfafe99995 (Merge branch 'js/smart-http-detect-remote-error' into pu, 2019-01-11), though. So I think what is in "pu" is okay, without shooting any series down. (Alternatively we can combine them into a single five-patch series, if the maintainer prefers.)

Thanks, Jonathan