On Fri, Apr 4, 2014 at 6:22 PM, Jim Jagielski <j...@jagunet.com> wrote:
> This seems to change some logic which appear only tangentially
> associated w/ the "save headers" issue... why is that?
Do you mean, ...
>
> On Apr 4, 2014, at 11:43 AM, Yann Ylavic <ylavic....@gmail.com> wrote:
>> Index: modules/proxy/mod_proxy_http.c
>> ===================================================================
>> --- modules/proxy/mod_proxy_http.c (revision 1584652)
>> +++ modules/proxy/mod_proxy_http.c (working copy)
>> @@ -717,24 +717,31 @@ int ap_proxy_http_request(apr_pool_t *p, request_r
>> apr_off_t bytes;
>> int force10, rv;
>> conn_rec *origin = p_conn->connection;
>> + apr_table_t *saved_headers_in;
>>
>> - if (apr_table_get(r->subprocess_env, "force-proxy-request-1.0")) {
>> - if (r->expecting_100) {
>> - return HTTP_EXPECTATION_FAILED;
>> - }
>> - force10 = 1;
>> - } else {
>> - force10 = 0;
>> - }
... this change?
Yes, sorry about it, it is part of another patch I'm working on
regarding 100-continue forwarding issue.
Please ignore this change, it is harmless (since
ap_proxy_create_hdrbrgd() already checks the expectation on HTTP/1.0),
but really not related to this thread.
>> +cleanup:
>> + /* Restore the original headers in (see comment above),
>> + * we won't modify them anymore.
>> + */
>> + r->headers_in = saved_headers_in;
>> + return rv;
>> }
>>
>
> Uggg... I hate gotos. We only use them sparingly and when
> breaking out of nasty, nasty code. Here, we are just doing so
> to safe some cut/paste... -0
>
I think I can avoid that, and don't like them either :p
This was to keep the patch simple for review (I can see gotos used not
far from here, and took the easy way).
