Edit a single key in hash - ruby

I have a ruby hash
obj =
{
"context":{
"conversationIDs": [1, 2, 5], "sessionID":2}
}
I want to edit the conversationIds key with new values and take the uniq of that. What I have tried
merged = (
{
context: obj[:context].merge({
"conversationIDs": [*obj[:context]["conversationIDs"], 2,
1, 1].uniq })
}
)
puts merged
Actual Output:
{:context=>{:conversationIDs=>[2, 1]}}
Expected Output:
{:context=>{:conversationIDs=>[2, 1, 5]}, :sessionID=>2 }
5 is missing in the output, how to sustain it?

I would do:
obj = {:context=>{:conversationIDs=>[1, 2, 5], :sessionID=>2}}
obj[:context][:conversationIDs] = (obj[:context][:conversationIDs] + [2, 1, 1]).uniq
obj
#=> {:context=>{:conversationIDs=>[1, 2, 5], :sessionID=>2}}
Or:
obj = {:context=>{:conversationIDs=>[1, 2, 5], :sessionID=>2}}
obj[:context][:conversationIDs] += [1,2,5]
obj[:context][:conversationIDs].uniq!
obj
#=> {:context=>{:conversationIDs=>[1, 2, 5], :sessionID=>2}}
Or – as engineersmnky suggested:
obj = {:context=>{:conversationIDs=>[1, 2, 5], :sessionID=>2}}
obj[:context][:conversationIDs].concat([2, 1, 1]).uniq!
obj
#=> {:context=>{:conversationIDs=>[1, 2, 5], :sessionID=>2}}

Related

How to shift the values in a hash to another key

I am having trouble figuring the best approach to shift some values to another key. Every value most go to the next key, and the last key's values need to be completely removed. For example
hash1 = { a: [1, 2, 3], b: [4, 5, 6], c: [7, 8, 9] }
desired_hash = hash1.some_method
desired_hash === { a: [], b: [1, 2, 3], c: [4, 5, 6] }
My thought is to rename the hash keys but was not sure if this was the best approach.
hash1 = { a: [1, 2, 3], b: [4, 5, 6], c: [7, 8, 9] }
keys = hash1.keys
=> [:a, :b, :c]
values = hash1.values
=> [[1, 2, 3], [4, 5, 6], [7, 8, 9]]
keys.zip(values.unshift([])).to_h
=> {:a=>[], :b=>[1, 2, 3], :c=>[4, 5, 6]}
pv = []
hash1.each_with_object({}) do |(k,v),h|
h[k] = pv
pv = v
end
#=> {:a=>[], :b=>[1, 2, 3], :c=>[4, 5, 6]}
Assuming the hash keys are "in correct order" already.
You can try this:
def shift(input)
output = {}
keys = input.keys
0.upto(keys.size - 1) do |index|
current = keys[index]
if index > 0
previous = keys[index - 1]
output[current] = input[previous]
else
output[current] = []
end
end
output
end
input = { a: [1, 2, 3], b: [4, 5, 6], c: [7, 8, 9] }
p shift(input)
it's not very elegant, there is probably a much nicer solution but it's a starting point.

comparing 2 arrays in every position

So what Im trying to accomplish is write a (shorter) condition that makes sure each element is different from the other array. This is confusing but I hope this example clears it up.
array = [1, 2, 3]
new_array = array.shuffle
until array[0] != new_array[0] &&
array[1] != new_array[1] &&
array[2] != new_array[2]
new_array = array.shuffle
end
So what Im doing is making sure that every single element/index pair does not match in the other array.
# [1, 2, 3] => [3, 1, 2] yayyyy
# [1, 2, 3] => [3, 2, 1] not what I want because the 2 didnt move
Is there a better way to do what I want to do? Ive looked up the .any? and .none? but I cant seem to figure out how to implement them. Thanks!
I would do this:
array.zip(new_array).all? { |left, right| left != right }
Here are two approaches that do not involve repeated sampling until a valid sample is obtained:
Sample from the population of valid permutations
Construct the population from which you are sampling:
array = [1, 2, 3, 4]
population = array.permutation(array.size).reject do |a|
a.zip(array).any? { |e,f| e==f }
end
#=> [[2, 1, 4, 3], [2, 3, 4, 1], [2, 4, 1, 3], [3, 1, 4, 2], [3, 4, 1, 2],
# [3, 4, 2, 1], [4, 1, 2, 3], [4, 3, 1, 2], [4, 3, 2, 1]]
Then just choose one at random:
10.times { p population.sample }
# [4, 3, 1, 2]
# [3, 4, 1, 2]
# [3, 4, 1, 2]
# [4, 3, 1, 2]
# [2, 1, 4, 3]
# [2, 1, 4, 3]
# [4, 1, 2, 3]
# [2, 1, 4, 3]
# [4, 3, 1, 2]
# [3, 4, 1, 2]
Sequentially sample for each position in the array
def sample_no_match(array)
a = array.each_index.to_a.shuffle
last_ndx = a[-1]
a.dup.map do |i|
if a.size == 2 && a[-1] == last_ndx
select = a[-1]
else
select = (a-[i]).sample
end
a.delete(select)
array[select]
end
end
10.times.each { p sample_no_match(array) }
# [2, 4, 3, 1]
# [4, 3, 1, 2]
# [2, 1, 3, 4]
# [1, 3, 4, 2]
# [1, 3, 2, 4]
# [1, 3, 2, 4]
# [1, 4, 3, 2]
# [3, 4, 2, 1]
# [1, 3, 4, 2]
# [1, 3, 4, 2]
I have been unable to prove or disprove that the second method produces a random sample. We can, however, determine relative frequencies of outcomes:
n = 500_000
h = n.times.with_object(Hash.new(0)) { |_,h| h[sample_no_match(array)] += 1 }
h.keys.each { |k| h[k] = (h[k]/(n.to_f)).round(4) }
h #=> {[1, 2, 3, 4]=>0.0418, [2, 1, 3, 4]=>0.0414, [1, 4, 2, 3]=>0.0418,
# [3, 4, 2, 1]=>0.0417, [4, 3, 2, 1]=>0.0415, [3, 1, 4, 2]=>0.0419,
# [2, 3, 1, 4]=>0.0420, [4, 2, 3, 1]=>0.0417, [3, 2, 1, 4]=>0.0413,
# [4, 2, 1, 3]=>0.0417, [2, 1, 4, 3]=>0.0419, [1, 3, 2, 4]=>0.0415,
# [1, 2, 4, 3]=>0.0418, [1, 3, 4, 2]=>0.0417, [2, 4, 1, 3]=>0.0414,
# [3, 4, 1, 2]=>0.0412, [1, 4, 3, 2]=>0.0423, [4, 1, 3, 2]=>0.0411,
# [3, 2, 4, 1]=>0.0411, [2, 4, 3, 1]=>0.0418, [3, 1, 2, 4]=>0.0419,
# [4, 3, 1, 2]=>0.0412, [4, 1, 2, 3]=>0.0421, [2, 3, 4, 1]=>0.0421}
avg = (h.values.reduce(:+)/h.size.to_f).round(4)
#=> 0.0417
mn, mx = h.values.minmax
#=> [0.0411, 0.0423]
([avg-mn,mx-avg].max/avg).round(6)
#=> 0.014388
which means that the maximum deviation from the average was only 1.4% percent of the average.
This suggests that the second method is a reasonable way of producing pseudo-random samples.
Initially, the first line of this method was:
a = array.each_index.to_a
By looking at the frequency distribution for outcomes, however, it was clear that that method did not produce a pseudo-random sample; hence, the need to shuffle a.
Here's one possibility:
until array.zip(new_array).reject{ |x, y| x == y }.size == array.size
new_array = array.shuffle
end
Note, though, that it will break for arrays like [1] or [1, 1, 1, 2, 3], where the number of instances of 1 exceeds half the size of the array. Recommend Array#uniq or similar, along with checking for arrays of sizes 0 or 1, depending on how trustworthy your input is!

ruby uniq with replacement

Normal uniq:
[1, 2, 3, 1, 1, 4].uniq => [1, 2, 3, 4]
I want to replace the duplicate with a replacement at where it was.
Is there a method or way to achieve something like this?
[1, 2, 3, 1, 1, 4].uniq_with_replacement(-1) => [1, 2, 3, -1, -1, 4]
Thanks in advance!
Here's a one-liner:
a.fill{ |i| a.index(a[i]) == i ? a[i] : -1 }
Something like this?:
class Array
def uniq_with_replacement(v)
map.with_object([]){|value, obj| obj << (obj.include?(value) ? v : value) }
end
end
Now:
[1, 2, 3, 1, 1, 4].uniq_with_replacement(-1)
# => [1, 2, 3, -1, -1, 4]
[1, 2, 3, 1, 1, 2, 4].uniq_with_replacement(-1)
# => [1, 2, 3, -1, -1, -1, 4]
1 more:
arr = [1, 2, 3, 1, 1, 4]
value = -1
a = arr.each_with_index.to_a
#=> [[1, 0], [2, 1], [3, 2], [1, 3], [1, 4], [4, 5]]
b = (a - a.uniq(&:first)).map(&:last)
#=> [3, 4]
arr.map.with_index { |e,i| b.include?(i) ? value : e }
#=> [1, 2, 3, -1, -1, 4]

Swapping two numbers with while loop in Ruby

I'd like to get [[2, 1, 3], [1, 3, 2]] from [1, 2, 3] in Ruby.
For [1, 2, 3, 4], I'd like to get [[2, 1, 3, 4], [1, 3, 2, 4], [1, 2, 4, 3]]
Rule: Within two numbers, if left one is smaller then it swap the position.
I have the following codes so far but it returns [[2, 3, 1], [2, 3, 1]]
What am I doing wrong here? I appreciate any inputs.
In amidakuji.rb
class Amidakuji
def initialize(column, rung)
#column = column
#rung = rung
#myarr = []
#per_arr = []
#build_arr = []
end
def build_initial
#arr = (1..#column).to_a
end
def swap_element
i = 0
arr = build_initial
while i < #column - 1 do
#build_arr << swap(arr, i)
i += 1
end
#build_arr
end
def swap(arr, a)
if arr[a] < arr[a + 1]
arr[a], arr[a + 1] = arr[a + 1], arr[a]
end
arr
end
end
In amidakuji_spec.rb
it 'should create an array with swapped elements' do
expect(#kuji1.swap_element).to eq ([[2, 1, 3], [1, 3, 2]])
end
Results
Failures:
expected: [[2, 1, 3], [1, 3, 2]]
got: [[2, 3, 1], [2, 3, 1]]
You can do this quite compactly by using the methods Enumerable#each_cons and Enumerable#map.
Code
def doit(arr)
(0...arr.size).each_cons(2).map do |i,j|
a = arr.dup
a[i], a[j] = a[j], a[i]
a
end
end
Examples
doit([1,2,3]) #=> [[2, 1, 3], [1, 3, 2]]
doit([1,2,3,4]) #=> [[2, 1, 3, 4], [1, 3, 2, 4], [1, 2, 4, 3]]
doit([1,2,3,4,5]) #=> [[2, 1, 3, 4, 5], [1, 3, 2, 4, 5],
#=> [1, 2, 4, 3, 5], [1, 2, 3, 5, 4]]
Explanation
arr = [1,2,3,4]
b = (0...arr.size).each_cons(2)
#=> #<Enumerator: 0...4:each_cons(2)>
To view the contents of this enumerator:
b.to_a
#=> [[0, 1], [1, 2], [2, 3]]
Lastly
b.map do |i,j|
a = arr.dup
a[i], a[j] = a[j], a[i]
a
end
#=> [[2, 1, 3, 4], [1, 3, 2, 4], [1, 2, 4, 3]]
In the last step, consider the first element of b that is passed to map, which assigns the following values to the block variables:
i => 0
j => 1
We then make a copy of arr, swap the elements offsets 0 and 1, making
a => [2, 1, 3, 4]
and then enter a at the end of the block, causing map to replace [0, 1] with that array.
Given what you're trying to accomplish and the output you're getting, it looks like you're reusing the same array when you want distinct arrays instead. Specifically this line:
#build_arr << swap(arr, i)
is always passing the same 'arr' to swap.
So first time, it swaps the 1 and the 2 to give you [2, 1, 3]
Second time, it swaps the 1 and the 3 give you [2, 3, 1]
You push the same array onto #build_arr twice, which is why it repeats.

How to make an array containing a duplicate of an array

I couldn't find a way to build an array such as
[ [1,2,3] , [1,2,3] , [1,2,3] , [1,2,3] , [1,2,3] ]
given [1,2,3] and the number 5. I guess there are some kind of operators on arrays, such as product of mult, but none in the doc does it. Please tell me. I missed something very simple.
Array.new(5, [1, 2, 3]) or Array.new(5) { [1, 2, 3] }
Array.new(size, default_object) creates an array with an initial size, filled with the default object you specify. Keep in mind that if you mutate any of the nested arrays, you'll mutate all of them, since each element is a reference to the same object.
array = Array.new(5, [1, 2, 3])
array.first << 4
array # => [[1, 2, 3, 4], [1, 2, 3, 4], [1, 2, 3, 4], [1, 2, 3, 4], [1, 2, 3, 4]]
Array.new(size) { default_object } lets you create an array with separate objects.
array = Array.new(5) { [1, 2, 3] }
array.first << 4
array #=> [[1, 2, 3, 4], [1, 2, 3], [1, 2, 3], [1, 2, 3], [1, 2, 3]]
Look up at the very top of the page you linked to, under the section entitled "Creating Arrays" for some more ways to create arrays.
Why not just use:
[[1, 2, 3]] * 5
# => [[1, 2, 3], [1, 2, 3], [1, 2, 3], [1, 2, 3], [1, 2, 3]]
The documentation for Array.* says:
...returns a new array built by concatenating the int copies of self.
Typically we'd want to create a repeating single-level array:
[1, 2, 3] * 2
# => [1, 2, 3, 1, 2, 3]
but there's nothing to say we can't use a sub-array like I did above.
It looks like mutating one of the subarrays mutates all of them, but that may be what someone wants.
It's like Array.new(5, [1,2,3]):
foo = [[1, 2, 3]] * 2
foo[0][0] = 4
foo # => [[4, 2, 3], [4, 2, 3]]
foo = Array.new(2, [1,2,3])
foo[0][0] = 4
foo # => [[4, 2, 3], [4, 2, 3]]
A work-around, if that's not the behavior wanted is:
foo = ([[1, 2, 3]] * 2).map { |a| [*a] }
foo[0][0] = 4
foo # => [[4, 2, 3], [1, 2, 3]]
But, at that point it's not as convenient, so I'd use the default Array.new(n) {…} behavior.

Resources