|
|
|
|
||||||
![]() |
|
|
LinkBack | Outils de la discussion |
|
|
#1 |
|
Messages: n/a
Hébergeur: |
I am reading Hal Fulton's "The Ruby Way" On page 57 he makes the statement: "Class instance variables cannot be referenced from within instance methods and, in general are not very useful" huh?.. This , in my feeble mind, contradicts everything else I have ever read, except that the example he gives is similar to the @y = 7 example below, and in that case, @y is not available to the accessor or any other method that I have played with. In the following code, the original assignment of @x = 7 and @y = 5 don't seem to do anything, even with the attr_accessor. I know that both are read by the compiler as I can assign @y = 5/0 and get a division by zero error. So, could someone please explain why the accessor for @y does not work here. If I hand write a reader and writer for @y, it works just fine. There are different scopes here, but I had the impression that the accessor would break down that barrier and make @y available throughout the class just as the initialize method is able to access the class variable @x and assign the passed in value. That value is then available to the @x accessor. I thought for a while that the @y accessor might work in the singleton class of Myclass, but it doesn't. class Myclass attr_accessor :y, :x @x = 7 @y = 5 def initialize(new_val= l) @x = new_val ? new_val : 0 end end mc = Myclass.new(3) puts mc.y #=> Nil puts mc.x #-> 3 I know I can make the @x and @y class variables available to instance methods by referencing them inside defined methods and the scope seem to be class wide, but it just seems like the first assignments above should work as written. Thanks in advance |
|
|
|
#2 |
|
Messages: n/a
Hébergeur: |
On Mar 29, 11:40 am, Ruby Freak <twscann...@gmail.com> wrote: > I am reading Hal Fulton's "The Ruby Way" On page 57 he makes the > statement: > "Class instance variables cannot be referenced from within instance > methods and, in general are not very useful" I think that just an out-dated way of looking at things. I expect Hal would agree that needs to be updated to current trends. I remember once, Hal and I got in a big argument about floating-point arithmetic. A tad spiteful I pointed out a flaw in his book. Despite the heated argument, he took it quite gracefully, saying he was always happy to receive comments/corrections. That was pretty damn respectable, IMHO. If he doesn't see this thread, write him directly. T. |
|
|
|
#3 |
|
Messages: n/a
Hébergeur: |
[Note: parts of this message were removed to make it a legal post.]
unsuscribe 2008/3/29, Trans <transfire@gmail.com>: > > > > On Mar 29, 11:40 am, Ruby Freak <twscann...@gmail.com> wrote: > > I am reading Hal Fulton's "The Ruby Way" On page 57 he makes the > > statement: > > "Class instance variables cannot be referenced from within instance > > methods and, in general are not very useful" > > > I think that just an out-dated way of looking at things. I expect Hal > would agree that needs to be updated to current trends. > > I remember once, Hal and I got in a big argument about floating-point > arithmetic. A tad spiteful I pointed out a flaw in his book. Despite > the heated argument, he took it quite gracefully, saying he was always > happy to receive comments/corrections. That was pretty damn > respectable, IMHO. If he doesn't see this thread, write him directly. > > > T. > > -- Harshad Joshi |
|
|
|
#4 |
|
Messages: n/a
Hébergeur: |
[Note: parts of this message were removed to make it a legal post.]
unsubscribe 2008/3/29, surreal <firewalrus@gmail.com>: > > unsuscribe > > 2008/3/29, Trans <transfire@gmail.com>: > > > > > > > > On Mar 29, 11:40 am, Ruby Freak <twscann...@gmail.com> wrote: > > > I am reading Hal Fulton's "The Ruby Way" On page 57 he makes the > > > statement: > > > "Class instance variables cannot be referenced from within instance > > > methods and, in general are not very useful" > > > > > > I think that just an out-dated way of looking at things. I expect Hal > > would agree that needs to be updated to current trends. > > > > I remember once, Hal and I got in a big argument about floating-point > > arithmetic. A tad spiteful I pointed out a flaw in his book. Despite > > the heated argument, he took it quite gracefully, saying he was always > > happy to receive comments/corrections. That was pretty damn > > respectable, IMHO. If he doesn't see this thread, write him directly. > > > > > > T. > > > > > > > -- > Harshad Joshi -- Harshad Joshi |
|
|
|
#5 |
|
Messages: n/a
Hébergeur: |
On Mar 29, 9:39 am, surreal <firewal...@gmail.com> wrote:
> [Note: parts of this message were removed to make it a legal post.] > > unsubscribe > > 2008/3/29, surreal <firewal...@gmail.com>: > > > > > > > unsuscribe > > > 2008/3/29, Trans <transf...@gmail.com>: > > > > On Mar 29, 11:40 am, Ruby Freak <twscann...@gmail.com> wrote: > > > > I am reading Hal Fulton's "The Ruby Way" On page 57 he makes the > > > > statement: > > > > "Class instance variables cannot be referenced from within instance > > > > methods and, in general are not very useful" > > > > I think that just an out-dated way of looking at things. I expect Hal > > > would agree that needs to be updated to current trends. > > > > I remember once, Hal and I got in a big argument about floating-point > > > arithmetic. A tad spiteful I pointed out a flaw in his book. Despite > > > the heated argument, he took it quite gracefully, saying he was always > > > happy to receive comments/corrections. That was pretty damn > > > respectable, IMHO. If he doesn't see this thread, write him directly. > > > > T. > > > -- > > Harshad Joshi > > -- > Harshad Joshi Thanks Trans. I guess Harshad didn't like my post! :-) Anyway, I would still appreciate an explanation for the difference in scope, or the reason the first assignment doesn't work. Thanks |
|
|
|
#6 |
|
Messages: n/a
Hébergeur: |
On Sat, Mar 29, 2008 at 4:40 PM, Ruby Freak <twscannell@gmail.com> wrote:
> > I am reading Hal Fulton's "The Ruby Way" On page 57 he makes the > statement: > "Class instance variables cannot be referenced from within instance > methods and, in general are not very useful" > > huh?.. > This , in my feeble mind, contradicts everything else I have ever > read, except that the example he gives is similar to the @y = 7 > example below, and in that case, @y is not available to the accessor > or any other method that I have played with. > > In the following code, the original assignment of @x = 7 and @y = 5 > don't seem to do anything, even with the attr_accessor. I know that > both are read by the compiler as I can assign @y = 5/0 and get a > division by zero error. > > So, could someone please explain why the accessor for @y does not work > here. If I hand write a reader and writer for @y, it works just fine. > There are different scopes here, but I had the impression that the > accessor would break down that barrier and make @y available > throughout the class just as the initialize method is able to access > the class variable @x and assign the passed in value. That value is > then available to the @x accessor. > > I thought for a while that the @y accessor might work in the singleton > class of Myclass, but it doesn't. > > class Myclass ### make this class << self > attr_accessor :y, :x end Now this is an accessor to the singleton class of the class where the class instance variables are stored (that is not completely correct, maybe I shall say from where one has access to them). > > @x = 7 > @y = 5 > > def initialize(new_val= l) > @x = new_val ? new_val : 0 # and this shall read self.class.x = new_val || 0 > end > end > > > puts mc.y #=> Nil Myclass.x --> 7 mc = Myclass::new( 42 ) Myclass.x --> 42 mc = Myclass::new 101010 Myclass.x --> 101010 there is of course no mc.x but please see below. I do however fear that you confuse instance variables with class instance variables. The former exist on an per object base and the later on a per class base ( a class being an object of course ). For completeness I'll show you how to use the former class A attr_accessor :a def initialize; @a = 42 end end a= A.new a.a --> 42 b = A.new a.a = 101010 a.a -> 101010 b.a -> 42 HTH Robert -- http://ruby-smalltalk.blogspot.com/ --- Whereof one cannot speak, thereof one must be silent. Ludwig Wittgenstein |
|
![]() |
| Outils de la discussion | |
|
|