Re: [PATCH v3 3/4] http: handle proxy proactive authentication

[Nelson Benitez Leon <nelsonjesus.benitez@xxxxxxxxxxxxxx>]

On 03/06/2012 09:30 AM, Jeff King wrote:
> On Mon, Mar 05, 2012 at 04:19:40PM +0100, Nelson Benitez Leon wrote:
> 
>> diff --git a/http.c b/http.c
>> index 8932da5..b0b4362 100644
>> --- a/http.c
>> +++ b/http.c
>> @@ -43,6 +43,7 @@ static int curl_ftp_no_epsv;
>>  static const char *curl_http_proxy;
>>  static const char *curl_cookie_file;
>>  static struct credential http_auth = CREDENTIAL_INIT;
>> +static struct credential proxy_auth = CREDENTIAL_INIT;
>>  static int http_proactive_auth;
>>  static const char *user_agent;
>>  
>> @@ -303,6 +304,17 @@ static CURL *get_curl_handle(void)
>>  		}
>>  	}
>>  	if (curl_http_proxy) {
>> +		credential_from_url(&proxy_auth, curl_http_proxy);
>> +		if (http_proactive_auth && proxy_auth.username && !proxy_auth.password) {
>> +			/* proxy string has username but no password, ask for password */
>> +			struct strbuf pbuf = STRBUF_INIT;
>> +			credential_fill(&proxy_auth);
>> +			strbuf_addf(&pbuf, "%s://%s:%s@%s",proxy_auth.protocol,
>> +			    	proxy_auth.username, proxy_auth.password,
>> +			    	proxy_auth.host);
> 
> Can we pull this out into a helper function, since the next patch will
> need to do the exact same thing in the 407 case?

Ok.

> Also, when turning it back into a URL to hand to curl, should we be
> percent-encoding the items we put in? If my password has an "@" in it,
> wouldn't we generate a bogus URL? Although looking at how the http auth
> code handles this, we set CURLOPT_USERPWD directly. Should you be
> setting CURLOPT_PROXYUSERPWD instead of munging the proxy URL?

Ok, but it seems is CURLOPT_PROXYUSERNAME and CURLOPT_PROXYPASSWORD what
we need here as per documentation[1]

[1] http://curl.haxx.se/libcurl/c/curl_easy_setopt.html#CURLOPTPROXYUSERNAME

>> +			free ((void *)curl_http_proxy);
> 
> Please don't cast to void. This is C, not C++, and casts to void
> pointers are implicit.  They can never help, and might cover up an
> actual type error (e.g., casting a non-pointer type).

Ok, will remove it, I copy/paste it from the http code and I must admit
I didn't understand why this was needed.

--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html