git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jeff King <peff@peff.net>
To: Nelson Benitez Leon <nelsonjesus.benitez@seap.minhap.es>
Cc: git@vger.kernel.org, Junio C Hamano <gitster@pobox.com>
Subject: Re: [PATCH 3/6] http: fix http_proxy specified without protocol part
Date: Fri, 4 May 2012 03:22:20 -0400	[thread overview]
Message-ID: <20120504072220.GC21895@sigill.intra.peff.net> (raw)
In-Reply-To: <4FA2B4E6.3080009@seap.minhap.es>

On Thu, May 03, 2012 at 06:40:06PM +0200, Nelson Benitez Leon wrote:

> An earlier patch broke http_proxy specified as <host>:<port> by abusing
> credential_from_url().  Teach the function to parse that format, but the
> caller needs to be updated to handle the case where there is no protocol
> in the parse result.
> 
> Also allow keyring implementations to differentiate authentication
> material meant for http proxies and http destinations by using a
> different token "http-proxy" to consult them for the former.
> 
> Signed-off-by: Junio C Hamano <gitster@pobox.com>

Should this be:

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

?

Also, any time we read "an earlier patch broke..." and that earlier
patch is in this series, I have to wonder why the patches are not simply
reordered. Shouldn't the first half of this one come first, and just
explain the rationale as "teach credential_from_url to handle
protocol-less URLs, since those are used for proxy specifications, which
we will need to parse in a later patch".

> diff --git a/http.c b/http.c
> index 02f9fcd..22ffe0c 100644
> --- a/http.c
> +++ b/http.c
> @@ -366,17 +366,20 @@ static CURL *get_curl_handle(const char *url)
>  	}
>  	
>  	if (curl_http_proxy) {
> -		struct strbuf proxyhost = STRBUF_INIT;
> -
> -		if (!proxy_auth.host) /* check to parse only once */
> +		if (!proxy_auth.host) {
>  			credential_from_url(&proxy_auth, curl_http_proxy);
> +			if (!proxy_auth.protocol ||
> +			    !strcmp(proxy_auth.protocol, "http")) {
> +				free(proxy_auth.protocol);
> +				proxy_auth.protocol = xstrdup("http-proxy");
> +			}
> +		}

And then this hunk would just get squashed in in the first place.

>  		if (http_proactive_auth && proxy_auth.username && !proxy_auth.password)
>  			/* proxy string has username but no password, ask for password */
>  			credential_fill(&proxy_auth);
>  
> -		strbuf_addf(&proxyhost, "%s://%s", proxy_auth.protocol, proxy_auth.host);
> -		curl_easy_setopt(result, CURLOPT_PROXY, strbuf_detach(&proxyhost, NULL));
> +		curl_easy_setopt(result, CURLOPT_PROXY, curl_http_proxy);

And this too, which gets rid of my complaints about the previous patch.

-Peff

      reply	other threads:[~2012-05-04  7:22 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-05-03 16:40 [PATCH 3/6] http: fix http_proxy specified without protocol part Nelson Benitez Leon
2012-05-04  7:22 ` Jeff King [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20120504072220.GC21895@sigill.intra.peff.net \
    --to=peff@peff.net \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    --cc=nelsonjesus.benitez@seap.minhap.es \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).