Binary numbers starting with 0b

Binary numbers starting with 0b

[Marc Khouzam]

Hi,

A couple of months ago it was discussed to add a 0b in front of binary numbers.

http://sourceware.org/ml/gdb/2008-01/msg00327.html

I'm not sure what happened to this.  Was it decided against?
Or maybe there was a backwards compatibility concern?  I actually am a bit worried
about adding the 0b as not being backwards compatible, although I do think it would
be a nice thing for GDB to accept.

Thanks

marc

-- 
Marc Khouzam
Software Designer, Methods and Tools
Ericsson Canada Inc

Re: Binary numbers starting with 0b

[Daniel Jacobowitz]

On Tue, Mar 25, 2008 at 04:17:44PM -0400, Marc Khouzam wrote:
> Hi,
> 
> A couple of months ago it was discussed to add a 0b in front of binary numbers.
> 
> http://sourceware.org/ml/gdb/2008-01/msg00327.html
> 
> I'm not sure what happened to this.  Was it decided against?
> Or maybe there was a backwards compatibility concern?  I actually am a bit worried
> about adding the 0b as not being backwards compatible, although I do think it would
> be a nice thing for GDB to accept.

No, there was no decision against it.  I still think it's a good idea,
I just didn't have time to update the parsers.

I do not see any compatibility problems.

-- 
Daniel Jacobowitz
CodeSourcery

Re: Binary numbers starting with 0b

[Nick Roberts]

 > A couple of months ago it was discussed to add a 0b in front of binary
 > numbers.
 > 
 > http://sourceware.org/ml/gdb/2008-01/msg00327.html
 > 
 > I'm not sure what happened to this.  Was it decided against?

If approved, I could commit my change to printcmd.c which would fix your
problem with variable objects.  However, I'm not planning on going through
all the yacc parser files anytime soon.

-- 
Nick                                           http://www.inet.net.nz/~nickrob

RE: Binary numbers starting with 0b

[Marc Khouzam]

> On Behalf Of Nick Roberts
>  > A couple of months ago it was discussed to add a 0b in front of binary
>  > numbers.
>  > 
>  > http://sourceware.org/ml/gdb/2008-01/msg00327.html
>  > 
>  > I'm not sure what happened to this.  Was it decided against?
> 
> If approved, I could commit my change to printcmd.c which would fix your
> problem with variable objects.  However, I'm not planning on going through
> all the yacc parser files anytime soon.

Just to clarify: my original problem with variable objects was already fixed by
http://sourceware.org/ml/gdb/2008-02/msg00006.html

But it seems we all agree that the 0b prefix has value in itself.
But there is not urgency.

Thanks

Marc

RE: Binary numbers starting with 0b

[Marc Khouzam]

> From: Daniel Jacobowitz
> > Or maybe there was a backwards compatibility concern?  I actually am a bit worried
> > about adding the 0b as not being backwards compatible, although I do think it would
> > be a nice thing for GDB to accept.
> 
> No, there was no decision against it.  I still think it's a good idea,
> I just didn't have time to update the parsers.
> 
> I do not see any compatibility problems.

I'm not sure what is considered bacwards-compatible but I figured
that if this patch broke the test suite, then there was some compatibility issue :-)
One example is if a frontend prepends binary numbers with 0b to display to the user;
they may now end up with 0b0b1010 if they didn't first check for the presence of 0b.
But maybe this is being overly careful.  

I still vote for the 0b, though.

Thanks

Marc

RE: Binary numbers starting with 0b

[Nick Roberts]

 > > If approved, I could commit my change to printcmd.c which would fix your
 > > problem with variable objects.  However, I'm not planning on going through
 > > all the yacc parser files anytime soon.
 > 
 > Just to clarify: my original problem with variable objects was already fixed
 > by http://sourceware.org/ml/gdb/2008-02/msg00006.html

I just mean that the change to printcmd.c ensures that variable objects
get displayed with 0b when using the binary format specifier.

-- 
Nick                                           http://www.inet.net.nz/~nickrob