cURL / Mailing Lists / curl-library / Single Mail


Re: Use read/write with same curl handle.

From: Daniel Stenberg <>
Date: Thu, 2 Mar 2006 07:54:40 +0100 (CET)

On Thu, 2 Mar 2006, Nilesh wrote:

>> Why? My example didn't cause any Expect: header to get sent...
> No idea, but in my case it tried to send Expect header. So I reset that
> because my server HTTP 1.0 based.

Please try my example. If that sends an Expect: header you're not using any
recent libcurl version.

Besides, your server being 1.0 or not shouldn't really matter as it is
supposed to ignore the Expect: header if it doesn't understand it. For cases
where libcurl uses the Expect header, it would just cause a minor delay.

> Yes, even I tried on ARM architecture which is BIG Endian and it worked
> fine. Some problem seems to be with PPC.

ARM cores are mostly used little endian, even if they like PPC features
endianess switch capabilities.

>> Now this puzzles me. You said you use 7.15.1, and when I look at line 1211
>> in transfer.c as it was back in 7.15.1 it concerns _writing_ data (as in
>> sending it off to the peer) and not receiving. Did you really add the
>> function invoke there?
> First of all why should lie.

"lie" is a strong word and I certainly won't claim you lie. I would rather say
"mix up", "confuse" or "make a mistake".

> Here is code where I tried putting hack.
> /# ifdef CODE_HACK
> result = Curl_client_write(data, CLIENTWRITE_BODY, k->str, nread);
> if(result)
> return result;
> #endif
> if(conn->bits.chunk) {
> /*

This "if"- line starts on line 1084 in transfer.c as it looked in 7.15.1 and
it is still in line 1084 in 7.15.2.

I did some quick searching around, but I couldn't find any libcurl version
that matches your claims. (I didn't try too many versions as it got tedious
quite fast! ;-)) Are you using a patched version, either by you or someone

  Commercial curl and libcurl Technical Support:
Received on 2006-03-02