Re: [Emc-developers] Bug#1025433: Copyright issue

2022-12-07 Thread andy pugh
On Thu, 8 Dec 2022 at 00:42, Adam Ant  wrote:

>
> The correct course of action is to ask Paolo Mantegazza rather than
> debating symantics.
> You can not just pull substantial chunks of code from one source and then
> claim that you wrote it.
>

But Paulo didn't write it _either_

If you go back in time to 2003:
https://github.com/LinuxCNC/linuxcnc/blob/07bc1e161f834d8b192fe279819261294e5fe150/src/rtapi/procfs_macros.h
and compare that to a 1999 version of the RTAI file, the only line in
common is "#include "

At some point _both_ files included a chunk of work from Erwin Rol:

// proc print macros - Contributed by: Erwin Rol (er...@muffin.org)

_Most_ of both files was written by Erwin Rol. But both files existed
previously, and separately, as far as I can tell looking back in both
archives.

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is designed
for the especial use of mechanical geniuses, daredevils and lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1912
#ifndef PROCFS_MACROS_H
#define PROCFS_MACROS_H
/***
*PROC_FS MACROS*
/
#ifdef CONFIG_PROC_FS
#include 

/* proc print macros - Contributed by: Erwin Rol (er...@muffin.org)
   and shamelessly ripped from rtai_proc_fs.h, part of the RTAI
   project.  See http://www.rtai.org for more details.

   macro that holds the local variables that
   we use in the PROC_PRINT_* macros. We have
   this macro so we can add variables with out
   changing the users of this macro, of course
   only when the names don't colide!
*/

#define PROC_PRINT_VARS \
off_t pos = 0;  \
off_t begin = 0;\
int len = 0 /* no ";" */

/* macro that prints in the procfs read buffer.
   this macro expects the function arguments to be
   named as follows.
   static int FOO(char *page, char **start,
off_t off, int count, int *eof, void *data)  */

#define PROC_PRINT(fmt,args...) \
len += sprintf(page + len , fmt, ##args);   \
pos += len; \
if(pos < off) { \
len = 0;\
begin = pos;\
}   \
if(pos > off + count)   \
goto done;

/* macro to leave the read function from another
   place than at the end.   */
#define PROC_PRINT_RETURN  \
*eof = 1;  \
goto done   // no ";"

/* macro that should only used once at the end of the
   read function, to return from another place in the
   read function use the PROC_PRINT_RETURN macro.   */
#define PROC_PRINT_DONE \
*eof = 1;   \
done:   \
*start = page + (off - begin);  \
len -= (off - begin);   \
if(len > count) \
len = count;\
if(len < 0) \
len = 0;\
return len  // no ";"

#endif /* PROC_FS */
#endif /* PROCFS_MACROS_H */


rtai_proc_fs.h_RTAI_1999
Description: Binary data
___
Emc-developers mailing list
Emc-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-developers


Re: [Emc-developers] Bug#1025433: Copyright issue

2022-12-05 Thread andy pugh
On Mon, 5 Dec 2022 at 14:14, Adam Ant  wrote:

> Source: linuxcnc
>
> src/rtapi/procfs_macros.h incorrectly attributes copyright.
>
> The original file can be found here:
>
> http://svn.savannah.gnu.org/viewvc/rtai/magma/base/include/rtai_proc_fs.h?view=markup


At what point does a file become a new file? The file in LinuxCNC appears
to be a sub-set of an original file contributed to RTAI by Erwin Roll.

So who would be the correct copyright holder?  It is possibly not Paulo
either,

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is designed
for the especial use of mechanical geniuses, daredevils and lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1912

___
Emc-developers mailing list
Emc-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-developers