Creating hashes with duplicate keys
 

I want to simplify this hash creation:
g = { 1 => "test", 2 => "test", }

I have a particularly complicated hash that I'm generating. It can
have very large values.

Right now, I duplicate values with something like this:

c = { 1 => "one", }
d = { 2 => c[1], }
c.merge!(d)

however, I don't like the way my code looks.. I want to be able to
specify the duplication within the hash generation.

Is this possible?

Re: Creating hashes with duplicate keys
 

On 28/07/06, Sy Ali <sy1234@gmail.com> wrote:
> I want to simplify this hash creation:
> g = { 1 => "test", 2 => "test", }
>
> I have a particularly complicated hash that I'm generating. It can
> have very large values.
>
> Right now, I duplicate values with something like this:
>
> c = { 1 => "one", }
> d = { 2 => c[1], }
> c.merge!(d)
>
> however, I don't like the way my code looks.. I want to be able to
> specify the duplication within the hash generation.
>
> Is this possible?

Something like this?
irb(main):001:0> c={}
=> {}
irb(main):002:0> c[1] = "one"
=> "one"
irb(main):003:0> c[2] = c[1]
=> "one"
irb(main):004:0> c[1].object_id == c[2].object_id
=> true

Farrel

Re: Creating hashes with duplicate keys
 

Sy Ali wrote:
> I want to simplify this hash creation:
> g = { 1 => "test", 2 => "test", }
>
> I have a particularly complicated hash that I'm generating. It can
> have very large values.
>
> Right now, I duplicate values with something like this:
>
> c = { 1 => "one", }
> d = { 2 => c[1], }
> c.merge!(d)
>
> however, I don't like the way my code looks.. I want to be able to
> specify the duplication within the hash generation.
>
> Is this possible?

This will work:

hsh = Hash.new{|h, k| h[k] = (k == 1 ? "one" : h[k - 1])}

Note that all the values will be the same -- the same object! If you
want duplicates, do this instead:

hsh = Hash.new{|h, k| h[k] = (k == 1 ? "one" : h[k - 1].dup)}

Just ask if there are parts of the snippet you don't understand.


Cheers,
Daniel

Re: Creating hashes with duplicate keys
 

fr Sy:
# I want to simplify this hash creation:
# g =3D { 1 =3D> "test", 2 =3D> "test", }
# ...
# however, I don't like the way my code looks.. I want to be able to
# specify the duplication within the hash generation.

not sure if this simple idea will help, but,

irb(main):055:0> class Hash
irb(main):056:1> def init2 range
irb(main):057:2> range.each{|c| self[c]=3Dself.default}
irb(main):058:2> end
irb(main):059:1> end
=3D> nil
irb(main):060:0> h=3DHash.new
=3D> {}
irb(main):065:0> h.default =3D "test"
=3D> "test"
irb(main):066:0> h.init2 1..8
=3D> 1..8
irb(main):067:0> h
=3D> {5=3D>"test", 6=3D>"test", 1=3D>"test", 7=3D>"test", 2=3D>"test", =
8=3D>"test", 3=3D>"test", 4=3D>"test"}
irb(main):068:0> h.default =3D "testing again"
=3D> "testing again"
irb(main):069:0> h.init2 6..10
=3D> 6..10
irb(main):070:0> h
=3D> {5=3D>"test", 6=3D>"testing again", 1=3D>"test", 7=3D>"testing =
again", 2=3D>"test", 8=3D>"testing again", 3=3D>"test", 9=3D>"testing =
again", 4=3D>"test", 10=3D>"testing again"}
irb(main):071:0>

kind regards -botp

Re: Creating hashes with duplicate keys
 

On Fri, 28 Jul 2006, Sy Ali wrote:

> I want to simplify this hash creation:
> g = { 1 => "test", 2 => "test", }
>
> I have a particularly complicated hash that I'm generating. It can
> have very large values.
>
> Right now, I duplicate values with something like this:
>
> c = { 1 => "one", }
> d = { 2 => c[1], }
> c.merge!(d)
>
> however, I don't like the way my code looks.. I want to be able to
> specify the duplication within the hash generation.
>
> Is this possible?

use rbtree (from the raa)

harp:~ > cat a.rb
require 'rbtree'

rb = RBTree.new

rb[1] = 'test'
rb[2] = 'test'

rb.each{|k,v| p [k,v]}


harp:~ > ruby a.rb
[1, "test"]
[2, "test"]

regards.

-a
--
we can never obtain peace in the outer world until we make peace with
ourselves.
- h.h. the 14th dali lama

Re: Creating hashes with duplicate keys
 

On Fri, 28 Jul 2006 Ara.T.Howard@noaa.gov wrote:

> use rbtree (from the raa)
>
> harp:~ > cat a.rb
> require 'rbtree'
>
> rb = RBTree.new
>
> rb[1] = 'test'
> rb[2] = 'test'
>
> rb.each{|k,v| p [k,v]}
>
>
> harp:~ > ruby a.rb
> [1, "test"]
> [2, "test"]

yikes - hit send too soon, here ya go:

harp:~ > cat a.rb
require 'rbtree'

rb = MultiRBTree.new

rb[1] = 'test'
rb[1] = 'test'

rb.each{|k,v| p [k,v]}

p rb.to_a
p rb[1]


harp:~ > ruby a.rb
[1, "test"]
[1, "test"]
[[1, "test"], [1, "test"]]
"test"


-a
--
we can never obtain peace in the outer world until we make peace with
ourselves.
- h.h. the 14th dali lama