Re: [fossil-users] Compiling fossil under windows

2010-04-11 Thread D. Richard Hipp

On Apr 11, 2010, at 6:46 AM, Rene de Zwart wrote:


 Having had my way with fossil compiling under windows. I was looking  
 for
 new challenges by compiling with other free compilers.
 I tried digital mars c compiler. Which promptly said
  unistd.h is for unix systems (The smart little bugger :-)

 Do I correctly infer from this that native windows compilers are not
 tested/used/supported?


Correct.

I don't own a windows machine. The windows binaries on the website are  
generated by cross-compiling off of Linux.  I do have the capability  
of running windows under VMWare (for testing), but I don't bring  
VMWare up that often and do not have any compilers installed there.

D. Richard Hipp
d...@hwaci.com



___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users


Re: [fossil-users] Compiling fossil under windows

2010-04-11 Thread Jeremy Cowgar
On 4/11/2010 6:46 AM, Rene de Zwart wrote:
 Having had my way with fossil compiling under windows. I was looking for
 new challenges by compiling with other free compilers.
 I tried digital mars c compiler. Which promptly said
unistd.h is for unix systems (The smart little bugger :-)

 Do I correctly infer from this that native windows compilers are not
 tested/used/supported?


I compile Fossil with MinGW all the time. I have not tried any other 
Windows compiler.

Jeremy

___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users


Re: [fossil-users] Compiling fossil under windows

2010-04-11 Thread Rene de Zwart

Op Zo, 11 april, 2010 12:59, schreef D. Richard Hipp:

 On Apr 11, 2010, at 6:46 AM, Rene de Zwart wrote:


 Having had my way with fossil compiling under windows. I was looking
 for
 new challenges by compiling with other free compilers.
 I tried digital mars c compiler. Which promptly said
  unistd.h is for unix systems (The smart little bugger :-)

 Do I correctly infer from this that native windows compilers are not
 tested/used/supported?


 Correct.

 I don't own a windows machine. The windows binaries on the website are
 generated by cross-compiling off of Linux.  I do have the capability
 of running windows under VMWare (for testing), but I don't bring
 VMWare up that often and do not have any compilers installed there.

 D. Richard Hipp
 d...@hwaci.com



 ___
 fossil-users mailing list
 fossil-users@lists.fossil-scm.org
 http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users

Richard,

I had success compiling fossil with Digital Mars Compiler.
This compiler identifies itself with __DMC__
so basically everwhere
  if __MINGW32__ or if!__min...@__
stands has to change to
   if __MINGW32__ || __DMC__ or if !__MINGW32__  !__DMC__
there are a few quirks
dmc doesn't have

-) close-, open- and read-dir but a public header fixes that in construct,
add and vfile
-)  strncasecmp and strcasecmp a simple  -Dstrncasecmp=memicmp
-Dstrcasecmp=stricmp
-) winsock inclusing is different
I had to do
  #if defined(__MINGW32__)
  #  include windows.h   /* for Sleep once server works again */
  #  include winsock2.h  /* socket operations */
  #  define sleep Sleep/* windows does not have sleep, but   
Sleep */
  #  include ws2tcpip.h
  #elif defined(__DMC__)
typedef int socklen_t;
  #  include winsock2.h  /* socket operations */
  #else
in cgi.c http_socket.c
it seems that winsock2.h includes windows.h

if it is of importance to you i can send the diffs


-- 
Rene de Zwart

___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users


[fossil-users] Fossil import problem

2010-04-11 Thread verizon
Hi,

I'm running fossil version [2255e4e3ba] 2009-12-20 02:58:18 UTC on UNIX 
(and linux) and am trying to import about 1 Gig of files (it's a Linux 
distribution and tool set for a propriety system).  When I do this the import 
works but a subsequent pull fails.  There is one directory (target/include) 
that only contains links  which are pointers to .h files to elsewhere in the 
distribution.  

0 lrwxrwxrwx 1 xxx TG1   23 2010-04-11 17:29 cvmip.h - ../../executive/cvmip.h
:
:
0 lrwxrwxrwx 1 xxx TG11 2010-04-11 17:29 open-license - .

When I import this into Fossil it shows lines like this for each of the files 
in this directory:

Add ./xxx/target/include/open-license/open-license/open-license/open- 
license/open-license/open-license/open-license/open-license/open-license 
/open-license/open-license/open-license/open-license/open-license/cvmip. h

The import makes it past these files and completes the rest of the files in the 
set.  But when I try to make a working directory from the repository the open 
stops after pulling these files and showing similar lines when pulling the 
files.  The subsequent files in the set are not copied out but do show up in 
the Fossil Files list (in the UI).

I know my Fossil is old but I've been keeping consistent version of a bunch of 
machines and haven't had any troubles till now.  Is this a problem that would 
be fixed in a newer version of Fossil ?  Any other suggestions on what I can 
do. 

Don't know if this will help but the project is a mostly standard Debian MIPS 
distribution with proprietary extensions).  Others have probably put a Linux 
distro into Fossil and I was hoping for some suggestions if you encountered 
this problem.

--jim schimpf
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users