Re: [GRASS-user] map algebra logical operators on DCELL rasters not admitted? maybe docs should outline it

[Markus Neteler]

On Fri, Oct 23, 2015 at 12:20 AM, G. Allegri  wrote:
> Thanks for the clear explanation Glynn. Maybe this behaviour could be
> esplicited inside the docs...

Yes.
I'm happy to add a summary of this discussion to the manual if someone
sends it to me.

thanks
Markus
___
grass-user mailing list
grass-user@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-user

Re: [GRASS-user] map algebra logical operators on DCELL rasters not admitted? maybe docs should outline it

[G. Allegri]

Ah ok Glynn, I always meant logical operators. Probably I copied the error
from a different test.

If logical operator is applied to, sat, Byte data it gives true/false
results, e.g. if(A && B,1, 0) behaves correctly even with cell values
values greater then 1.
That's why I expected it to work with floats operands too.
>From a user perspective if it works for a cell with value 2 it should also
work for a cell with value 2.0.
That'all.

giovanni
Il 22/ott/2015 02:44, "Glynn Clements"  ha
scritto:

>
> G. Allegri wrote:
>
> > > This isn't about the if() function. The bitand() function (to which
> > > the error message refers) corresponds to the bitwise-and operator "&".
> > >
> > > The if() function accepts either integer or floating-point values for
> > > its arguments.
> >
> > That's what disappointed my client (and me) Glynn: why integer yes and
> not
> > floats? Is it a technical limitation or there's a ratio behind it?
> > Anyway did I miss something from the docs? I couldn't find it.
>
> What would you expect the value of (x & y) to be if x and/or y were
> floating-point values?
>
> > Why do you that I'm confusing the operators? I've always talked about
> > logical operators...
>
> Because you quote an error message which specifically refers to a
> *bitwise* operator:
>
> > "Incorrect argument types to function bitand()".
>
> in a message which otherwise only mentions logical operators.
>
> "bitand()" is the name of the bitwise-and function, which corresponds to
> the infix "&" operator. I.e. (x & y) is exactly equivalent to bitand(x, y).
>
> This is the same as the "&" operator in C or C++: it evaluates to an
> integer where each bit in the result is set (one) if the corresponding
> bit is set in both operands, and unset (zero) otherwise. E.g. (23 & 34)
> is
>   00010111 (23)
> & 00100010 (34)
>   
> = 0010 ( 2)
>
> Perhaps it would help if you posted the exact expression which is
> causing you problems.
>
> --
> Glynn Clements 
>
___
grass-user mailing list
grass-user@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-user

Re: [GRASS-user] map algebra logical operators on DCELL rasters not admitted? maybe docs should outline it

[Glynn Clements]

G. Allegri wrote:

> If logical operator is applied to, sat, Byte data it gives true/false
> results, e.g. if(A && B,1, 0) behaves correctly even with cell values
> values greater then 1.
> That's why I expected it to work with floats operands too.
> >From a user perspective if it works for a cell with value 2 it should also
> work for a cell with value 2.0.
> That'all.

r.mapcalc's interpretation of booleans is borrowed from C. Booleans
are just integers where zero is false and non-zero is true.

But that doesn't necessarily make much sense for floats; the
conventional wisdom is that comparing a float for equality with zero
is usually a mistake, as results which should mathematically be equal
to zero often aren't actually equal to zero due to intermediate
rounding errors. In some cases, simply subtracting a number from
itself won't yield zero (e.g. due to the x87 FPU using 80 bits for
floating-point registers while values stored in memory use 32 or 64
bits).

You can obtain C-like behaviour by converting to an integer with the
int() function then using that; the result is that any value between
-1.0 and 1.0 (excluding both endpoints) will be considered false while
anything else is true. Or you can convert to an integer with the
round() function and use that, in which case any value between -0.5
and 0.5 (including both endpoints) will be false while anything else
is true. Or you can compare for equality with 0.0, in which case -0.0
and 0.0 will be false while anything else is true. Or you can compare
the magnitude (obtained using abs()) to an "epsilon" value so that
anything smaller than the expsilon is false while anything larger is
true.

That's four different possibilities, none of which is any more
"correct" than any of the others. Choosing one of them would just mean
that r.mapcalc was probably using the "wrong" one, and doing so
silently. The current behaviour forces the user to be explicit.

-- 
Glynn Clements 
___
grass-user mailing list
grass-user@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-user

Re: [GRASS-user] map algebra logical operators on DCELL rasters not admitted? maybe docs should outline it

[G. Allegri]

Thanks for the clear explanation Glynn. Maybe this behaviour could be
esplicited inside the docs...

giovanni
Il 22/ott/2015 23:35, "Glynn Clements"  ha
scritto:

>
> G. Allegri wrote:
>
> > If logical operator is applied to, sat, Byte data it gives true/false
> > results, e.g. if(A && B,1, 0) behaves correctly even with cell values
> > values greater then 1.
> > That's why I expected it to work with floats operands too.
> > >From a user perspective if it works for a cell with value 2 it should
> also
> > work for a cell with value 2.0.
> > That'all.
>
> r.mapcalc's interpretation of booleans is borrowed from C. Booleans
> are just integers where zero is false and non-zero is true.
>
> But that doesn't necessarily make much sense for floats; the
> conventional wisdom is that comparing a float for equality with zero
> is usually a mistake, as results which should mathematically be equal
> to zero often aren't actually equal to zero due to intermediate
> rounding errors. In some cases, simply subtracting a number from
> itself won't yield zero (e.g. due to the x87 FPU using 80 bits for
> floating-point registers while values stored in memory use 32 or 64
> bits).
>
> You can obtain C-like behaviour by converting to an integer with the
> int() function then using that; the result is that any value between
> -1.0 and 1.0 (excluding both endpoints) will be considered false while
> anything else is true. Or you can convert to an integer with the
> round() function and use that, in which case any value between -0.5
> and 0.5 (including both endpoints) will be false while anything else
> is true. Or you can compare for equality with 0.0, in which case -0.0
> and 0.0 will be false while anything else is true. Or you can compare
> the magnitude (obtained using abs()) to an "epsilon" value so that
> anything smaller than the expsilon is false while anything larger is
> true.
>
> That's four different possibilities, none of which is any more
> "correct" than any of the others. Choosing one of them would just mean
> that r.mapcalc was probably using the "wrong" one, and doing so
> silently. The current behaviour forces the user to be explicit.
>
> --
> Glynn Clements 
>
___
grass-user mailing list
grass-user@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-user

Re: [GRASS-user] map algebra logical operators on DCELL rasters not admitted? maybe docs should outline it

[Glynn Clements]

G. Allegri wrote:

> > This isn't about the if() function. The bitand() function (to which
> > the error message refers) corresponds to the bitwise-and operator "&".
> >
> > The if() function accepts either integer or floating-point values for
> > its arguments.
> 
> That's what disappointed my client (and me) Glynn: why integer yes and not
> floats? Is it a technical limitation or there's a ratio behind it?
> Anyway did I miss something from the docs? I couldn't find it.

What would you expect the value of (x & y) to be if x and/or y were
floating-point values?

> Why do you that I'm confusing the operators? I've always talked about
> logical operators...

Because you quote an error message which specifically refers to a
*bitwise* operator:

> "Incorrect argument types to function bitand()".

in a message which otherwise only mentions logical operators.

"bitand()" is the name of the bitwise-and function, which corresponds to
the infix "&" operator. I.e. (x & y) is exactly equivalent to bitand(x, y).

This is the same as the "&" operator in C or C++: it evaluates to an
integer where each bit in the result is set (one) if the corresponding
bit is set in both operands, and unset (zero) otherwise. E.g. (23 & 34)
is
  00010111 (23)
& 00100010 (34)
  
= 0010 ( 2)

Perhaps it would help if you posted the exact expression which is
causing you problems.

-- 
Glynn Clements 
___
grass-user mailing list
grass-user@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-user

Re: [GRASS-user] map algebra logical operators on DCELL rasters not admitted? maybe docs should outline it

[Glynn Clements]

Moritz Lennert wrote:

> >> Doing a logical operation with a DCELL raster within an if(x,a) statement
> >> produces an error: "Incorrect argument types to function bitand()".
> >
> > First, you're confusing bitwise operators and logical operators. & and
> > | are bitwise operators, while && and || (and &&& and |||) are logical
> > operators.
> >
> > Second, both bitwise and logical operators only work on integer
> > values. If you want to use floating-point values as boolean values,
> > you need to explicitly convert them to integers (either via int(x) or
> > round(x), or x==0.0, or abs(x)<1e-30, or whatever).
> 
> ???
> 
> On my grass71 and grass64release, these work (using NC demo data set):
> 
> r.mapcalc "test = if(elevation == 100.0, 1)"
> r.mapcalc "test = if(elevation > 100.0, 1)"
> 
> where elevation is FCELL.

I would expect those to work.

What won't work is using bitwise operators (& and |) or logical
operators (&& and ||, &&& and |||) on floats.

-- 
Glynn Clements 
___
grass-user mailing list
grass-user@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-user

Re: [GRASS-user] map algebra logical operators on DCELL rasters not admitted? maybe docs should outline it

[Moritz Lennert]

On 20/10/15 09:44, Moritz Lennert wrote:

On 19/10/15 21:16, Glynn Clements wrote:


G. Allegri wrote:


Doing a logical operation with a DCELL raster within an if(x,a) statement
produces an error: "Incorrect argument types to function bitand()".


First, you're confusing bitwise operators and logical operators. & and
| are bitwise operators, while && and || (and &&& and |||) are logical
operators.

Second, both bitwise and logical operators only work on integer
values. If you want to use floating-point values as boolean values,
you need to explicitly convert them to integers (either via int(x) or
round(x), or x==0.0, or abs(x)<1e-30, or whatever).


???

On my grass71 and grass64release, these work (using NC demo data set):

r.mapcalc "test = if(elevation == 100.0, 1)"
r.mapcalc "test = if(elevation > 100.0, 1)"

where elevation is FCELL.


Sorry, forget that. I missed the "logical" part...

Moritz

___
grass-user mailing list
grass-user@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-user

Re: [GRASS-user] map algebra logical operators on DCELL rasters not admitted? maybe docs should outline it

[Moritz Lennert]

On 19/10/15 21:16, Glynn Clements wrote:


G. Allegri wrote:


Doing a logical operation with a DCELL raster within an if(x,a) statement
produces an error: "Incorrect argument types to function bitand()".


First, you're confusing bitwise operators and logical operators. & and
| are bitwise operators, while && and || (and &&& and |||) are logical
operators.

Second, both bitwise and logical operators only work on integer
values. If you want to use floating-point values as boolean values,
you need to explicitly convert them to integers (either via int(x) or
round(x), or x==0.0, or abs(x)<1e-30, or whatever).


???

On my grass71 and grass64release, these work (using NC demo data set):

r.mapcalc "test = if(elevation == 100.0, 1)"
r.mapcalc "test = if(elevation > 100.0, 1)"

where elevation is FCELL.

Moritz

___
grass-user mailing list
grass-user@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-user

Re: [GRASS-user] map algebra logical operators on DCELL rasters not admitted? maybe docs should outline it

[Paulo van Breugel]

On Mon, Oct 19, 2015 at 5:18 PM, G. Allegri  wrote:

> Doing a logical operation with a DCELL raster within an if(x,a) statement
> produces an error: "Incorrect argument types to function bitand()".
> This doesn't seem to be described inside the docs, is it?
>

That sounds like a bug to me. I just tried it out and it runs fine (using
GRASS 7.1). What GRASS version are you using?


>
> When the if() function is described it only states its behaviour in case
> of NULL, 0 or "otherwise" values.
> It works correctly if the maps are CELL (i.e. e value not eqaul 0 is
> treated correctly whatever it is) and it's quite confusing for users having
> a different behaviour, and sepcifically an error, in case of floating point
> rasters, don't you think?
>
> Regards,
> giovanni
>
> --
> Giovanni Allegri
> http://about.me/giovanniallegri
> Gis3W - http://gis3w.it
> Ikare - http://ikare.it
> Twitter: https://twitter.com/_giohappy_
> blog: http://blog.spaziogis.it
> GEO+ geomatica in Italia http://bit.ly/GEOplus
>
> ___
> grass-user mailing list
> grass-user@lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/grass-user
>
___
grass-user mailing list
grass-user@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-user

Re: [GRASS-user] map algebra logical operators on DCELL rasters not admitted? maybe docs should outline it

[G. Allegri]

GRASS 6.4.4 on Windows.

giovanni
Il 19/ott/2015 17:52, "Paulo van Breugel"  ha
scritto:

>
>
> On Mon, Oct 19, 2015 at 5:18 PM, G. Allegri  wrote:
>
>> Doing a logical operation with a DCELL raster within an if(x,a) statement
>> produces an error: "Incorrect argument types to function bitand()".
>> This doesn't seem to be described inside the docs, is it?
>>
>
> That sounds like a bug to me. I just tried it out and it runs fine (using
> GRASS 7.1). What GRASS version are you using?
>
>
>>
>> When the if() function is described it only states its behaviour in case
>> of NULL, 0 or "otherwise" values.
>> It works correctly if the maps are CELL (i.e. e value not eqaul 0 is
>> treated correctly whatever it is) and it's quite confusing for users having
>> a different behaviour, and sepcifically an error, in case of floating point
>> rasters, don't you think?
>>
>> Regards,
>> giovanni
>>
>> --
>> Giovanni Allegri
>> http://about.me/giovanniallegri
>> Gis3W - http://gis3w.it
>> Ikare - http://ikare.it
>> Twitter: https://twitter.com/_giohappy_
>> blog: http://blog.spaziogis.it
>> GEO+ geomatica in Italia http://bit.ly/GEOplus
>>
>> ___
>> grass-user mailing list
>> grass-user@lists.osgeo.org
>> http://lists.osgeo.org/mailman/listinfo/grass-user
>>
>
>
___
grass-user mailing list
grass-user@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-user

Re: [GRASS-user] map algebra logical operators on DCELL rasters not admitted? maybe docs should outline it

[G. Allegri]

That's what disappointed my client (and me) Glynn: why integer yes and not
floats? Is it a technical limitation or there's a ratio behind it?
Anyway did I miss something from the docs? I couldn't find it.

Why do you that I'm confusing the operators? I've always talked about
logical operators...

giovanni

Il 19/ott/2015 21:19, "Glynn Clements"  ha
scritto:
>
>
> G. Allegri wrote:
>
> > Doing a logical operation with a DCELL raster within an if(x,a)
statement
> > produces an error: "Incorrect argument types to function bitand()".
>
> First, you're confusing bitwise operators and logical operators. & and
> | are bitwise operators, while && and || (and &&& and |||) are logical
> operators.
>
> Second, both bitwise and logical operators only work on integer
> values. If you want to use floating-point values as boolean values,
> you need to explicitly convert them to integers (either via int(x) or
> round(x), or x==0.0, or abs(x)<1e-30, or whatever).
>
> > This doesn't seem to be described inside the docs, is it?
> >
> > When the if() function is described it only states its behaviour in
case of
> > NULL, 0 or "otherwise" values.
>
> This isn't about the if() function. The bitand() function (to which
> the error message refers) corresponds to the bitwise-and operator "&".
>
> The if() function accepts either integer or floating-point values for
> its arguments.
>
> --
> Glynn Clements 
___
grass-user mailing list
grass-user@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-user

Re: [GRASS-user] map algebra logical operators on DCELL rasters not admitted? maybe docs should outline it

[Glynn Clements]

G. Allegri wrote:

> Doing a logical operation with a DCELL raster within an if(x,a) statement
> produces an error: "Incorrect argument types to function bitand()".

First, you're confusing bitwise operators and logical operators. & and
| are bitwise operators, while && and || (and &&& and |||) are logical
operators.

Second, both bitwise and logical operators only work on integer
values. If you want to use floating-point values as boolean values,
you need to explicitly convert them to integers (either via int(x) or
round(x), or x==0.0, or abs(x)<1e-30, or whatever).

> This doesn't seem to be described inside the docs, is it?
> 
> When the if() function is described it only states its behaviour in case of
> NULL, 0 or "otherwise" values.

This isn't about the if() function. The bitand() function (to which
the error message refers) corresponds to the bitwise-and operator "&".

The if() function accepts either integer or floating-point values for
its arguments.

-- 
Glynn Clements 
___
grass-user mailing list
grass-user@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-user