www.digitalmars.com         C & C++   DMDScript  

digitalmars.D.learn - terminology: "l-value" & "r-value"

reply spir <denis.spir gmail.com> writes:
Hello,

I'm bluffed by the 2 terms "l-value" & "r-value" used in C-line language co=
mmon terminologies. I think I guess what they mean, but I don't understand =
the need for such absconse idioms. Why not:
	l-value <-> variable
	r-value <-> value (or expression)
?

I guess (*p) is considered an l-value. Indeed, it's a special way of denoti=
ng a variable, matching the special case of a pointer. If correct, this req=
uires slightly extending the notion of variable (and/or of identifier).
On the r-value side, I cannot find anything that makes it a distinct concep=
t from the one of value, or of expression.

Explanations welcome, thank you,
Denis
-- -- -- -- -- -- --
vit esse estrany =E2=98=A3

spir.wikidot.com
Jan 04 2011
next sibling parent Pelle <pelle.mansson gmail.com> writes:
On 01/04/2011 02:55 PM, spir wrote:

 Hello,

 I'm bluffed by the 2 terms "l-value"&  "r-value" used in C-line language
common terminologies. I think I guess what they mean, but I don't understand
the need for such absconse idioms. Why not:
 	l-value<->  variable
 	r-value<->  value (or expression)
 ?

 I guess (*p) is considered an l-value. Indeed, it's a special way of denoting
a variable, matching the special case of a pointer. If correct, this requires
slightly extending the notion of variable (and/or of identifier).
 On the r-value side, I cannot find anything that makes it a distinct concept
from the one of value, or of expression.

 Explanations welcome, thank you,
 Denis
 -- -- -- -- -- -- --
 vit esse estrany ☣

 spir.wikidot.com


rvalue is easier than value-not-bound-to-a-memory-address.

lvalue is easier than value-with-memory-address.

Both lvalues and rvalues are values, both can be expressions, and 
lvalues doesn't have to be variables.

Perhaps a better terminology could have been chosen, but changing them 
doesn't provide real benefits, as far as I can tell.
Jan 04 2011
prev sibling parent reply "Steven Schveighoffer" <schveiguy yahoo.com> writes:
On Tue, 04 Jan 2011 08:55:07 -0500, spir <denis.spir gmail.com> wrote:


 Hello,

 I'm bluffed by the 2 terms "l-value" & "r-value" used in C-line language  
 common terminologies. I think I guess what they mean, but I don't  
 understand the need for such absconse idioms. Why not:
 	l-value <-> variable
 	r-value <-> value (or expression)
 ?

 I guess (*p) is considered an l-value. Indeed, it's a special way of  
 denoting a variable, matching the special case of a pointer. If correct,  
 this requires slightly extending the notion of variable (and/or of  
 identifier).
 On the r-value side, I cannot find anything that makes it a distinct  
 concept from the one of value, or of expression.

 Explanations welcome, thank you,
 Denis


lvalue stands for left value, rvalue stands for right value.  They  
describe which side of the equation they are on:

lvalue = rvalue;

Why these terms instead of something more natural?  Well, I have no idea  
:)  If I were to guess, it would be that no natural term could exactly  
describe the meaning, and also that using natural terms are subjective.   
There's no mistaking what lvalue and rvalue mean becausey they are defined  
by the language itself.

I'm sure Walter probably knows the origin, l and r values are really  
compiler terms, and he's been writing compilers for a long time.

-Steve
Jan 04 2011
parent reply "Manfred_Nowak" <svv1999 hotmail.com> writes:
Steven Schveighoffer wrote:


 They describe which side of the equation they are on


arg, no! please replace "equation" by "assignExpression".

-manfred
Jan 04 2011
next sibling parent reply "Steven Schveighoffer" <schveiguy yahoo.com> writes:
On Tue, 04 Jan 2011 12:56:53 -0500, Manfred_Nowak <svv1999 hotmail.com>  
wrote:


 Steven Schveighoffer wrote:


 They describe which side of the equation they are on


 arg, no! please replace "equation" by "assignExpression".


please replace arg with argh!

;)

But whatever floats your boat.

-Steve
Jan 04 2011
parent "Manfred_Nowak" <svv1999 hotmail.com> writes:
Steven Schveighoffer wrote:


 please replace


done. :-)

-manfred
Jan 04 2011
prev sibling parent spir <denis.spir gmail.com> writes:
On Tue, 4 Jan 2011 17:56:53 +0000 (UTC)
"Manfred_Nowak" <svv1999 hotmail.com> wrote:


 They describe which side of the equation they are on =20


=20
 arg, no! please replace "equation" by "assignExpression".


lol, great! this is one of the reasons why in my dream language, assignment=
 would be denoted by any other sign *but* "=3D".

Denis
-- -- -- -- -- -- --
vit esse estrany =E2=98=A3

spir.wikidot.com
Jan 04 2011