Re: [PATCH 2/2] http, imap-send: stop using CURLOPT_VERBOSE

From: Jeff King <peff@peff.net>
Date: May 12, 2020, 19:27 UTC
Message-ID: <20200512192706.GD54565@coredump.intra.peff.net>
In-Reply-To: <20200512192300.203201-1-jonathantanmy@google.com>

On Tue, May 12, 2020 at 12:23:00PM -0700, Jonathan Tan wrote:

Show 14 quoted lines

> > PS I sometimes find the normal trace a bit verbose, but I do still want
> >    to see data. Do others feel the same? Particularly I find the "SSL"
> >    lines totally worthless (I guess maybe you could be debugging ssl
> >    stuff, but that would be the exception, I'd think). Ditto the split
> >    of data into two lines: one with the size and one with the actual
> >    data.
> > 
> >    I dunno. I haven't been debugging any git-over-http stuff lately, so
> >    it hasn't been bothering me. But I definitely have written perl
> >    scripts to extract the data to a more readable format. Maybe it would
> >    be easier if it had a few more knobs.
> 
> Data can be turned on using GIT_TRACE_CURL=1 and refraining from setting
> GIT_TRACE_CURL_NO_DATA. What knobs were you thinking of?

I still want to see data, but less cruft. I.e., something like "GIT_TRACE_CURL_SSL" (which I'd default to "off"), and probably just reducing:

  15:24:01.169101 [pid=55191] http.c:702            <= Recv data, 0000000004 bytes (0x00000004)
  15:24:01.169104 [pid=55191] http.c:717            <= Recv data: 3e..

to just the second line. Actually, we might not need a knob at all for SSL data. I was thinking that people might actually be debugging SSL problems with it, but since all of the non-printable characters are munged to "." anyway, it's basically useless (you can often pick out a few strings from the cert during handshake, but you'd be much better off to just connect with "openssl s_client" and ask it to dump the cert).

-Peff