On Mon, Oct 17, 2011 at 4:42 AM, Bert Huijben <b...@qqmail.nl> wrote:
>
>
>> -----Original Message-----
>> From: Mat Booth [mailto:mat.bo...@wandisco.com]
>> Sent: maandag 17 oktober 2011 12:26
>> To: Peter Samuelson
>> Cc: dev@subversion.apache.org
>> Subject: Re: Minor patch suggestion:
>>
>> On 14 October 2011 18:25, Peter Samuelson <pe...@p12n.org> wrote:
>> >
>> > [Mat Booth]
>> >> --- subversion/libsvn_subr/cmdline.c (revision 1183368)
>
> I don't know anybody that still does real Subversion work with the 2002/2003
> Visual Studio versions.
> Visual Studio 2005, 2008 and 2010 are the most common releases and then
> Visual C++ 6.0 (From 1998). But apache httpd and some other old projects
> still require Visual C++ 6.0 (from 1998).
>
> AnkhSVN has about 1000 users that still use Visual Studio 2002/2003. This
> number is measurable in our update process, but nothing compared to our
> total user base. (We have no numbers on VC 6.0 usage)
>
Not sure how relevant this as (as you point out VC6 is ancient), but
trunk won't build with VC6. Attached is a portion of a patch
(trivially modified) sent to me off list by David Darj that is needed
to get trunk to compile with VC6. David's original patch includes
changes to support building the Perl bindings, too.
It is not ready to be committed and I'm not suggesting it (I haven't
even tried to run the tests), this is just FYI.
Is the project going to support building with VC6? Or skip it because
it's too old?
--
Joe
Index: subversion/libsvn_subr/cache-membuffer.c
===================================================================
--- subversion/libsvn_subr/cache-membuffer.c (revision 1180755)
+++ subversion/libsvn_subr/cache-membuffer.c (working copy)
@@ -116,7 +116,7 @@
/* Don't create cache segments smaller than this value unless the total
* cache size itself is smaller.
*/
-#define MIN_SEGMENT_SIZE 0x2000000ull
+#define MIN_SEGMENT_SIZE 0x2000000ui64
/* We don't mark the initialization status for every group but initialize
* a number of groups at once. That will allow for a very small init flags
Index: subversion/libsvn_subr/cache.c
===================================================================
--- subversion/libsvn_subr/cache.c (revision 1180755)
+++ subversion/libsvn_subr/cache.c (working copy)
@@ -211,15 +211,15 @@
{
enum { _1MB = 1024 * 1024 };
- apr_uint64_t misses = info->gets - info->hits;
- double hit_rate = (100.0 * info->hits)
- / (info->gets ? info->gets : 1);
- double write_rate = (100.0 * info->sets)
+ apr_int64_t misses = info->gets - info->hits;
+ double hit_rate = (100.0 * (apr_int64_t)info->hits)
+ / (info->gets ? (apr_int64_t)info->gets : 1);
+ double write_rate = (100.0 * (apr_int64_t)info->sets)
/ (misses ? misses : 1);
- double data_usage_rate = (100.0 * info->used_size)
- / (info->data_size ? info->data_size : 1);
- double data_entry_rate = (100.0 * info->used_entries)
- / (info->total_entries ? info->total_entries : 1);
+ double data_usage_rate = (100.0 * (apr_int64_t)info->used_size)
+ / (info->data_size ? (apr_int64_t)info->data_size :
1);
+ double data_entry_rate = (100.0 * (apr_int64_t)info->used_entries)
+ / (info->total_entries ?
(apr_int64_t)info->total_entries : 1);
return svn_string_createf(result_pool,
Index: subversion/include/svn_types.h
===================================================================
--- subversion/include/svn_types.h (revision 1180755)
+++ subversion/include/svn_types.h (working copy)
@@ -29,6 +29,7 @@
/* ### this should go away, but it causes too much breakage right now */
#include <stdlib.h>
+#include <limits.h> /* for ULONG_MAX */
#include <apr.h> /* for apr_size_t, apr_int64_t, ... */
#include <apr_errno.h> /* for apr_status_t */
