On Wed, Nov 16, 2011 at 11:00 AM, Balachandran Sivakumar
<[email protected]> wrote:
> Hi,
>
> On Wed, Nov 16, 2011 at 7:54 AM, Girish Venkatachalam
> <[email protected]> wrote:
>>
>> #define HTTP_GET_CMD "GET /%s HTTP 1.1\r\n\r\n"
>>
>
>       Not nitpicking, but I guess that's an RFC violation ;) HTTP 1.1
> says Host header is mandatory. From RFC 2616, :

Okay. I only get irritated when people don't talk technical. You do and it is a
 genuine flaw. No problem. I am only glad.

Only that I dunno how to fix it. Are you supposed to say something like:

"HOST foobar \r\n\r\n"

or something?

I am sure it is before the GET right?

I am aware of many other flaws too. Like this won't work with HTTP 301
redirects as is the
 case with tinyurls which is what Shrinivasan is fond of.

And I am definitely not RFC compliant since I said that URL encoding
is not handled. And I don't
 handle all the header responses. So this is not a surprise.

Can you throw some light on your point please?

Moreover my point was only a broad example to illustrate the concept.

One has to build on top of it. It has educational value and gives a
real life kick.

For something professional more work is required.

But it is also a truth of life that majority of professional
commercial grade code is not
standards compliant.

>
> "A client MUST include a Host header field in all HTTP/1.1 request messages ."
>

Good. I am glad there is someone talking of RFC standards. Good. Keep it up.

> And nice initiative. People on the list, new to socket programming can
> possibly start with this. Thanks
>

Doesn't anybody want a code walk through?

Unless someone asks I am not gonna do it.

-Girish
_______________________________________________
ILUGC Mailing List:
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Reply via email to