On Sat, 2007-01-06 at 16:28 +0100, Nils Larsch wrote:
> OpenSC wrote:
> > #128: Patch: PIV Card update patch
> > -------------------------+--------------------------------------------------
> >  Reporter:  harningt     |        Owner:  [email protected]
> >      Type:  enhancement  |       Status:  new                          
> >  Priority:  normal       |    Milestone:  0.11.2                       
> > Component:  opensc       |      Version:  0.11.1                       
> >  Severity:  normal       |   Resolution:                               
> >  Keywords:  patch, piv   |  
> > -------------------------+--------------------------------------------------
> > Comment (by harningt):
> > 
> >  Attached is a new update to the patch, it accounts for the location of
> >  'malloc' on OSX (stdlib.h, not malloc.h) and cleans up one of the 'hacks'
> >  that I had in place.
> 
> according to c99 malloc(3) should be in stdlib.h so this isn't really
> a osx issue. Btw: why do include "memory.h" in compression.c ?
> 
Oops.. 'wrong' header for memset, string.h appears to be the correct
header (my Linux implementation #includes 'string.h' for 'memory.h's
content).

> some first comments: your patch exports a new header file compression
> (btw: why export it ? it looks more like an internal header). But the
> names used for constanst and functions are very generic. *IF* we want
> to export this header file the functions/constants within the file
> should have a opensc prefix (i.e. normally "sc_" for functions and "SC_"
> for constants). Furthermore it would be nice if the compilation of
> compression.c|h is dependent on the presence of zlib support.
> 
It exports it because I figure it would be useful in developing external
card plugins w/o duplicating the compression.h header/sources.  I also
suppose for that case I should have exported p15card-helper.h.
Not exporting it could work just as well, in which case any external
plugins would just have duplicate code.

I do agree w/ the fact that if compression.h is to be exported, then
functions/constants should probably be changed.
-- 
Thomas Harning Jr.
Authentication Engineer @ Identity Alliance
http://www.trustbearer.com/

_______________________________________________
opensc-devel mailing list
[email protected]
http://www.opensc-project.org/mailman/listinfo/opensc-devel

Reply via email to