Array question - How dynamic is dynamic?

[Ruby Student]

[Note: parts of this message were removed to make it a legal post.]

Hello all you happy people!

I have been thinking about asking this question but did not know how to
start.
So, if my question is ambiguous please let me know and I will try to make it
clearer.

Given an NxN array or even an NxM array mA from which at times I break it
into sub arrays, I would like to be able to manipulate the sub arrays and
have the changes dynamically propagated to the main array mA.
For example, assume I have main array mA as listed below.
From mA I strip say row[2], which I named sAr2 and which has values: sAr2 =
[D, C, B, A]
I also created sub arrays of columns, as listed below and furthermore, I
created quadrants sub arrays.

I would like to make a change on any of the sub arrays and have that change
propagated simultaneously to the main array.
For instance, say I set sAr2[1] = Y, I would like element mA[2,1] to change
to Y from C.
The same should happen if and when I change any of the sub arrays.
The same should happen if and when I change the main array.

Is this possible say in 1.9.1? or do I need some kind of GEM?


Main Array:

mA = [
[A, B, C, D],
[B, A, C, D],
[D, C, B, A],
[C, A, D, B]
]

Sub Arrays:
sAr0 (sub array row 0 to row n, where n is 3 for this case)
sAr0 = [A, B, C, D]
sAr1 = [B, A, C, D]
sAr2 = [D, C, B, A]
sAr3 = [C, A, D, B]

sAc0 (sub array column 0 to column n, where n is 3 for this case)
sAc0 = [A, B, D, C]
sAc1 = [B, A, C, A]
sAc2 = [C, C, B, D]
sAc3 = [D, D, A, B]
Four "quadrants?" arrays as follows:
qA1:
[[A, B
B, A]]

qA3:
[[B, D
A, B]]

Etc.

Thank you

 

[Jesús Gabriel y Galán]

Hello all you happy people!

I have been thinking about asking this question but did not know how to
start.
So, if my question is ambiguous please let me know and I will try to make it
clearer.

Given an NxN array or even an NxM array mA from which at times I break it
into sub arrays, I would like to be able to manipulate the sub arrays and
have the changes dynamically propagated to the main array mA.
For example, assume I have main array mA as listed below.
From mA I strip say row[2], which I named sAr2 and which has values: sAr2 =
[D, C, B, A]
I also created sub arrays of columns, as listed below and furthermore, I
created quadrants sub arrays.

I would like to make a change on any of the sub arrays and have that change
propagated simultaneously to the main array.
For instance, say I set sAr2[1] = Y, I would like element mA[2,1] to change
to Y from C.

First thing that came to mind is to add a level of indirection, so
that the array doesn't
contain the actual values, but an object that contains the values, like so:

irb(main):001:0> O = Struct.new :value
=> O
irb(main):008:0> A = "A"
=> "A"
irb(main):009:0> B = "B"
=> "B"
irb(main):010:0> C = "C"
=> "C"
irb(main):013:0> myAr = [[O.new(A), O.new(B), O.new(C)],[O.new(B),
O.new(A), O.new(C)],[O.new(C), O.new(B), O.new(A)]]
=> [[#<struct O value="A">, #<struct O value="B">, #<struct O
value="C">], [#<struct O value="B">, #<struct O value="A">, #<struct O
value="C">], [#<struct O value="C">, #<struct O value="B">, #<struct O
value="A">]]
irb(main):017:0> first_row = myAr[0]
=> [#<struct O value="A">, #<struct O value="B">, #<struct O value="C">]
irb(main):018:0> Y = "Y"
=> "Y"
irb(main):019:0> first_row[2].value = Y
=> "Y"
irb(main):020:0> myAr
=> [[#<struct O value="A">, #<struct O value="B">, #<struct O
value="Y">], [#<struct O value="B">, #<struct O value="A">, #<struct O
value="C">], [#<struct O value="C">, #<struct O value="B">, #<struct O
value="A">]]

So by changing the value of the object pointed to by first_row[0], you
are really changing the same object that is referenced in the original
array. Don't know if this is convenient for you or not. Probably
someone will come up with a better idea.

Hope this helps,

Jesus.

 

[Mark Thomas]

Main Array:

mA = [
[A, B, C, D],
[B, A, C, D],
[D, C, B, A],
[C, A, D, B]
]

Sub Arrays:
sAr0 (sub array row 0 to row n, where n is 3 for this case)
sAr0 = [A, B, C, D]
sAr1 = [B, A, C, D]
sAr2 = [D, C, B, A]
sAr3 = [C, A, D, B]

I suggest you use the Matrix class (in the standard library).

Though for the row case, you don't need anything special:
sAr0 = mA[0] = [A, B, C, D]
sAr1 = mA[1] = [B, A, C, D]
sAr2 = mA[2] = [D, C, B, A]
sAr3 = mA[3] = [C, A, D, B]
If you change sAr0, it will be reflected in mA[0].

-- Mark.

 

[Robert Klemme]

Main Array:

mA = [
[A, B, C, D],
[B, A, C, D],
[D, C, B, A],
[C, A, D, B]
]

Sub Arrays:
sAr0 (sub array row 0 to row n, where n is 3 for this case)
sAr0 = [A, B, C, D]
sAr1 = [B, A, C, D]
sAr2 = [D, C, B, A]
sAr3 = [C, A, D, B]

I suggest you use the Matrix class (in the standard library).

Are you sure this works? It seems, when I create a row Vector of the
Matrix it is read only. At least []= is defined private.

But I agree to the general advice to use a special class for this and
not try to work with nested Arrays. That's what OO is for. Defining a
two dimensional matrix with Vector proxies isn't too hard.

Kind regards

robert

 

[Ruby Student]

[Note: parts of this message were removed to make it a legal post.]

Main Array:

mA = [
[A, B, C, D],
[B, A, C, D],
[D, C, B, A],
[C, A, D, B]
]

Sub Arrays:
sAr0 (sub array row 0 to row n, where n is 3 for this case)
sAr0 = [A, B, C, D]
sAr1 = [B, A, C, D]
sAr2 = [D, C, B, A]
sAr3 = [C, A, D, B]

I suggest you use the Matrix class (in the standard library).

Are you sure this works? It seems, when I create a row Vector of the
Matrix it is read only. At least []= is defined private.

But I agree to the general advice to use a special class for this and not
try to work with nested Arrays. That's what OO is for. Defining a two
dimensional matrix with Vector proxies isn't too hard.

Kind regards

robert

Thank you all for your input and suggestions!