Re: setting nil to zero

Robert K. [mailto:[email protected]] :

#f[w] = (f[w] || 0) + 1

clear rubyish solution, Robert.
thanks again for generous help.

i am still hoping for the plain f[w] += 1 though :slight_smile:

for the record, i played w the ff

ruby>cat test.rb

class NilClass
def method_missing(methId, *args)
str = methId.id2name
if str == “+” and args[1].nil?
#str is “+”
#args[1] is nil
#args[0] is rhs value
return args[0]
end
end
end

f = {}
f[“test”] += 1
f[“test2”] += 2
p f

f = []
f[1] += 1
f[5] += 2
p f

ruby>ruby test.rb
{“test2”=>2, “test”=>1}
[nil, 1, nil, nil, nil, 2]

ruby>

I doubted if it’s robust enough, so my q…

-botp

#Kind regards

robert

Botp wrote:

Robert K. [mailto:[email protected]] :

#f[w] = (f[w] || 0) + 1

clear rubyish solution, Robert.
thanks again for generous help.

i am still hoping for the plain f[w] += 1 though :slight_smile:

As mentioned already you can do that, if it’s a Hash:

f=Hash.new 0
=> {}
f[:foo]+=1
=> 1
f
=> {:foo=>1}

     #args[0]  is rhs value
     return args[0]
  end

end
end

Why so complicated? Why not just

class NilClass
def +(n) n end
end

h={}
=> {}
h[:foo]+=1
=> 1
h
=> {:foo=>1}

Note, I recommend to use the veriant with 0 as default value for Hash
lookups because that avoids tampering with standard classes.

Cheers

robert