Re: [Flightgear-devel] Chasing warnings...

2009-08-25 Thread Tim Moore
Torsten Dreyer wrote:
 While iterating through the source files, trying to eliminate the warnings, I 
 just found this precious line of code (some more of them in the same file)
 
  *((float*)buf[length]) = sg_bswap_32(*(uint32_t*)val);
 
 The beauty of pointers - poor Java, C# et al developers. You will never see 
 something alike ;-)
 
 Torsten
The worst thing about that line is that it is broken :) The fix is checked
in.  I used to think that a lot of the warnings produced by -Wall were stupid,
but I'm now convinced that we should make -Wall -Werror the default for gcc.

Tim

--
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with 
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
___
Flightgear-devel mailing list
Flightgear-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/flightgear-devel


Re: [Flightgear-devel] Chasing warnings...

2009-08-25 Thread Erik Hofman


Tim Moore wrote:
 The worst thing about that line is that it is broken :) 

I can't find anything about it that makes it 'broken', knowing that 
doubles are 64-bit and floats are 32-bit. It might be a bit better this 
way though.

Erik

--
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with 
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
___
Flightgear-devel mailing list
Flightgear-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/flightgear-devel


Re: [Flightgear-devel] Chasing warnings...

2009-08-25 Thread Anders Gidenstam
On Tue, 25 Aug 2009, Erik Hofman wrote:



 Tim Moore wrote:
 The worst thing about that line is that it is broken :)

 I can't find anything about it that makes it 'broken', knowing that
 doubles are 64-bit and floats are 32-bit. It might be a bit better this
 way though.

The line:
   *((float*)buf[length]) = sg_bswap_32(*(uint32_t*)val);

Unless I'm mistaken the line writes a uint_32 (supposedly containing a 
byte-swapped float) to a float location, triggering a automatic value 
conversion from uint32 to float. Hmm.. and in addition to that val is a 
double so I would worry that *(uint32_t*)val in fact only gives you the 
uint32_t value of the first half of the double. Time to fetch the C LRM.

Thanks to Tim for noticing this.

Cheers,

Anders
-- 
---
Anders Gidenstam
WWW: http://www.gidenstam.org/FlightGear/

--
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with 
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
___
Flightgear-devel mailing list
Flightgear-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/flightgear-devel


Re: [Flightgear-devel] Chasing warnings...

2009-08-25 Thread Erik Hofman


Anders Gidenstam wrote:
 On Tue, 25 Aug 2009, Erik Hofman wrote:
 

 Tim Moore wrote:
 The worst thing about that line is that it is broken :)
 I can't find anything about it that makes it 'broken', knowing that
 doubles are 64-bit and floats are 32-bit. It might be a bit better this
 way though.
 
 The line:
*((float*)buf[length]) = sg_bswap_32(*(uint32_t*)val);
 
 Unless I'm mistaken the line writes a uint_32 (supposedly containing a 
 byte-swapped float) to a float location, triggering a automatic value 
 conversion from uint32 to float. Hmm.. and in addition to that val is a 
 double

Hm, that was only obvious after reading the source. I had another part 
of the code in mind. Good catch indeed.

Erik

--
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with 
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
___
Flightgear-devel mailing list
Flightgear-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/flightgear-devel


[Flightgear-devel] Chasing warnings...

2009-08-21 Thread Torsten Dreyer
While iterating through the source files, trying to eliminate the warnings, I 
just found this precious line of code (some more of them in the same file)

 *((float*)buf[length]) = sg_bswap_32(*(uint32_t*)val);

The beauty of pointers - poor Java, C# et al developers. You will never see 
something alike ;-)

Torsten

--
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with 
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
___
Flightgear-devel mailing list
Flightgear-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/flightgear-devel


Re: [Flightgear-devel] Chasing warnings...

2009-08-21 Thread Curtis Olson
Yeah, but just try byte swapping in visual basic ... it can be done!  Thanks
for the warning clean up pass, it's very much appreciated!

Curt.


On Fri, Aug 21, 2009 at 4:45 PM, Torsten Dreyer wrote:

 While iterating through the source files, trying to eliminate the warnings,
 I
 just found this precious line of code (some more of them in the same file)

  *((float*)buf[length]) = sg_bswap_32(*(uint32_t*)val);

 The beauty of pointers - poor Java, C# et al developers. You will never see
 something alike ;-)

 Torsten


 --
 Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
 trial. Simplify your report design, integration and deployment - and focus
 on
 what you do best, core application coding. Discover what's new with
 Crystal Reports now.  http://p.sf.net/sfu/bobj-july
 ___
 Flightgear-devel mailing list
 Flightgear-devel@lists.sourceforge.net
 https://lists.sourceforge.net/lists/listinfo/flightgear-devel




-- 
Curtis Olson: http://baron.flightgear.org/~curt/
--
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with 
Crystal Reports now.  http://p.sf.net/sfu/bobj-july___
Flightgear-devel mailing list
Flightgear-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/flightgear-devel