Re: [PATCH] Reason pharses into the HTTP response
Guillaume Cottenceau <[email protected]>
| Newsgroups | gmane.comp.mobile.kannel.devel |
|---|---|
| Message-ID | <[email protected]> |
Zoltán Dudás <zoltan.dudas 'at' sei.hu> writes:
> Hi All !
>
> We use the Kannel for our project and we made some little modifications on it.
> I am going to send these patches back to the community.
> The first one is a very simply modification, it changes the constant 'foo'
> word to real reason phrases
> (http://www.w3.org/Protocols/rfc2616/rfc2616-sec6.html) in the status-line of
> the HTTP response.
>
> Index: gateway/gwlib/http.c
> ===================================================================
> RCS file: /home/cvs/gateway/gwlib/http.c,v
> retrieving revision 1.217
> diff -r1.217 http.c
> 2350c2350
> < response = octstr_format("HTTP/1.0 %d Foo\r\n", status);
> ---
> > response = octstr_format("HTTP/1.0 %d %s\r\n", status,
FAQ reads:
"If you want to send us modifications, please send the output of
cvs diff -u (-u makes cvs use unified diffs, instead of plain
diffs; unified diffs are easier to read, since they provide more
context and can be applied to a later version of the same file)"
Unified diffs are the way to go.
[...]
> > static char *http_error_messages[] = {
> > "OK", /* 200 */
> > "Created", /* 201 */
> > "Accepted", /* 202 */
> > "No Content", /* 204 */
[...]
> > switch (status) {
> > case HTTP_OK:
> > result= http_error_messages[0];
> > break;
> > case HTTP_CREATED:
> > result= http_error_messages[1];
> > break;
> > case HTTP_ACCEPTED:
> > result= http_error_messages[2];
> > break;
> > case HTTP_NO_CONTENT:
> > result= http_error_messages[3];
> > break;
Hum you really don't have another solution that to hardcode the
row number of the char* array? This is inelegant IMHO since it
duplicates information. And I'm not sure Kannel people don't
prefer octstr over char* pretty much :).
--
Guillaume Cottenceau