Tim Sneddon's VMS fix for huge HTTP POSTs

                                  Changelog

Daniel (5 November 2004)

- Tim Sneddon made libcurl send no more than 64K in a single first chunk when

  doing a huge POST on VMS, as this is a system limitation. Default on general

  systems is 100K.

Daniel (4 November 2004)

- Andres Garcia made it build on mingw againa, my --retry code broke the build.

This release includes the following bugfixes:

 o huge POSTs on VMS

 o configure no longer uses pkg-config on cross-compiles

 o potential gzip decompress memory leak

 o "-C - --fail" on a HTTP page already downloaded

 o pycurl 7.12.2: http://pycurl.sf.net/

 o TclCurl 0.12.2: http://personal1.iddeo.es/andresgarci/tclcurl/english/

 o libcurl.NET 1.1: http://www.seasideresearch.com/downloads.html

This release would not have looked like this without help, code, reports and

advice from friends like these:

 Peter Wullinger, Guillaume Arluison, Alexander Krasnostavsky, Mohun Biswas,

 Tomas Pospisek, Gisle Vanem, Dan Fandrich, Paul Nolan

 Tomas Pospisek, Gisle Vanem, Dan Fandrich, Paul Nolan, Andres Garcia,

 Tim Sneddon

        Thanks! (and sorry if I forgot to mention someone)

      if(data->set.postfields) {

        if((data->state.authhost.done || data->state.authproxy.done )

           && (postsize < (100*1024))) {

           && (postsize < MAX_INITIAL_POST_SIZE)) {

          /* If we're not done with the authentication phase, we don't expect

             to actually send off any data yet. Hence, we delay the sending of

             the body until we receive that friendly 100-continue response */

          /* The post data is less than 100K, then append it to the header.

             This limit is no magic limit but only set to prevent really huge

             POSTs to get the data duplicated with malloc() and family. */

          /* The post data is less than MAX_INITIAL_PORT_SIZE, then append it

             to the header. This limit is no magic limit but only set to

             prevent really huge POSTs to get the data duplicated with

             malloc() and family. */

          result = add_buffer(req_buffer, "\r\n", 2); /* end of headers! */

          if(result)

   public curl/curl.h header. */

#define CURLAUTH_PICKNONE (1<<30) /* don't use auth */

/* MAX_INITIAL_POST_SIZE indicates the number of kilobytes that will be sent

   in the initial part of a multi-part POST message. This is primarily for

   OpenVMS where the maximum number of bytes allowed per I/O is 64K.  For

   other systems that do not define this, the default is (as it was

   previously) 100K. */

#ifndef MAX_INITIAL_POST_SIZE

#define MAX_INITIAL_POST_SIZE (100*1024)

#endif

#endif

#endif

/* MSK, 03/09/04, Seems to work for all platforms I've built on so far.    */

/*      Added HAVE_SYS_IOCTL_H, IOCTL_3_ARGS and SIZEOF_CURL_OFF_T defines */

/* MSK, 06/04/04, Added HAVE_INET_NTOP                                     */

/* TES, 11/05/04, Added MAX_INITIAL_POST_SIZE, HAVE_BASENAME               */

/* Define cpu-machine-OS */

#ifdef __ALPHA

#endif

#endif

/* Define to set number of kilobytes in initial POST message. On VMS systems

   this is important as the default is 100 and the maximum amount of data

   transferable in a VMS $QIO is 64K. All VMS versions prior to Alpha/I64

   V8.2 and TCP/IP V5.5 are affected by this. */

#define MAX_INITIAL_POST_SIZE (60*1024)

/* Define if you have the ANSI C header files.  */

#define STDC_HEADERS 1

/* Define if you have the geteuid function.  */

#define HAVE_GETEUID 1

/* Define if you have the basename function. */

#define HAVE_BASENAME 1

/* Define if you have the gethostbyaddr function.  */

#define HAVE_GETHOSTBYADDR 1