Iterating through a hash

B

Brian Ross

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

How can I iterate through a hash so that each key is modified and saved into
a new hash?

# Beginning of code

hash = {"name"=>"greg", "job"=>"boring", "hair"=>"plenty"}
p hash

map = hash.each_key do |key|
key.upcase
end

p map

# End of code

I'd imagine that this would return map as a new hash with the keys modified.
Is there anything like collect! for hashes?

Brian
 
P

Patrick Doyle

How about something like:

map = {}
hash.each_pair do |k,v|
map[k.upcase] = v
end

--wpd
 
F

Farrel Lifson

2008/8/20 Brian Ross said:
How can I iterate through a hash so that each key is modified and saved into
a new hash?

# Beginning of code

hash = {"name"=>"greg", "job"=>"boring", "hair"=>"plenty"}
p hash

map = hash.each_key do |key|
key.upcase
end

p map

# End of code

Hash[*hash.map{|key,value| [key.upcase, value]}.flatten]

Farrel
 
F

Florian Gilcher

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

#upcase doesn't modify the receiver.

#upcase! doesn't work because you can't modify String keys, as they
are frozen.

So, you have to rebuild the hash, the others pointed out some
solutions for this.

Regards,
Florian Gilcher

How can I iterate through a hash so that each key is modified and
saved into
a new hash?

# Beginning of code

hash = {"name"=>"greg", "job"=>"boring", "hair"=>"plenty"}
p hash

map = hash.each_key do |key|
key.upcase
end

p map

# End of code

I'd imagine that this would return map as a new hash with the keys
modified.
Is there anything like collect! for hashes?

Brian

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.8 (Darwin)

iEYEARECAAYFAkisX3AACgkQJA/zY0IIRZYC0ACfVfK1H80uupfXn/g0jhp69uzB
O4YAoMIao4OCp1WfSUkpTWKW3LQA3tKP
=WHKL
-----END PGP SIGNATURE-----
 
R

Robert Klemme

How can I iterate through a hash so that each key is modified and saved into
a new hash?

# Beginning of code

hash = {"name"=>"greg", "job"=>"boring", "hair"=>"plenty"}
p hash

map = hash.each_key do |key|
key.upcase
end

p map

# End of code

I'd imagine that this would return map as a new hash with the keys modified.
Is there anything like collect! for hashes?

Folks, thanks for leaving the #inject solution to me. :)

map = hash.inject({}) do |h,(k,v)|
h[k.upcase] = v
h
end

Cheers

robert
 
M

Martin DeMello

Folks, thanks for leaving the #inject solution to me. :)

map = hash.inject({}) do |h,(k,v)|
h[k.upcase] = v
h
end

without the ugly "h" at the end :) :

map = hash.inject({}) {|h, (k,v)| h.update({k.upcase => v})}

martin
 
B

Brian Ross

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

without the ugly "h" at the end :) :

map = hash.inject({}) {|h, (k,v)| h.update({k.upcase => v})}

Thanks everyone! Using Martin's example, how would I then use upcase for an
array of hashes?

I'd imagine it would be

this_array.collect do |hash|
@map = {}
hash.each_pair do |k,v|
@map[k.upcase] = v
end
end

but while the @map gives me the scope to see it outside of the block, it
looks like only the first hash within the array is passed through.

Brian
 
M

Martin DeMello

this_array.collect do |hash|
@map = {}
hash.each_pair do |k,v|
@map[k.upcase] = v
end
end

but while the @map gives me the scope to see it outside of the block, it
looks like only the first hash within the array is passed through.

new_array = this_array.collect do |hash|
map = {}
hash.each_pair do |k,v|
map[k.upcase] = v
end
map
end

collect collects the return value of the block

martin
 
R

Robert Klemme

2008/8/20 Martin DeMello said:
Folks, thanks for leaving the #inject solution to me. :)

map = hash.inject({}) do |h,(k,v)|
h[k.upcase] = v
h
end

without the ugly "h" at the end :) :

map = hash.inject({}) {|h, (k,v)| h.update({k.upcase => v})}

But with ugly and unnecessary brackets. This works as well

map = hash.inject({}) {|h, (k,v)| h.update(k.upcase => v)}

Note though that this unfortunately creates a lot of temporary Hashes
that are thrown away immediately. I personally find that more ugly
than the h at the end. You can reformat to make it less stand out as
in

map = hash.inject {} do |h,(k,v)|
h[k.upcase] = v; h
end

Kind regards

robert
 
P

Peña, Botp

From: Brian Ross [mailto:p[email protected]]=20
# I'd imagine that this would return map as a new hash with the=20
# keys modified. Is there anything like collect! for hashes?

the required pairings for hashes makes it fragile to implement a =
map/collect like feature similar to plain arrays.

you can create one if you like. just be careful.

eg, here is my simple-minded implementation,

irb(main):084:0> class Hash
irb(main):085:1> def map2
irb(main):086:2> h=3D{}
irb(main):087:2> self.each do |k,v|
irb(main):088:3* kk,vv=3Dyield(k,v)
irb(main):089:3> key =3D kk || k
irb(main):090:3> val =3D vv || v
irb(main):091:3> h[key] =3D val
irb(main):092:3> end
irb(main):093:2> h
irb(main):094:2> end
irb(main):095:1> end
=3D> nil
irb(main):096:0> hash.map2{|k,_| k.upcase}
=3D> {"NAME"=3D>"greg", "JOB"=3D>"boring", "HAIR"=3D>"plenty"}
irb(main):097:0> hash.map2{|k| k.upcase}
=3D> {"NAME"=3D>"greg", "JOB"=3D>"boring", "HAIR"=3D>"plenty"}
irb(main):098:0> hash.map2{|k,v| k.upcase}
=3D> {"NAME"=3D>"greg", "JOB"=3D>"boring", "HAIR"=3D>"plenty"}
irb(main):099:0> hash.map2{|k,v| [k.upcase,v.capitalize]}
=3D> {"NAME"=3D>"Greg", "JOB"=3D>"Boring", "HAIR"=3D>"Plenty"}


kind regards -botp
 
R

Robert Klemme

2008/8/21 Pe=F1a said:
From: Brian Ross [mailto:p[email protected]]
# I'd imagine that this would return map as a new hash with the
# keys modified. Is there anything like collect! for hashes?

the required pairings for hashes makes it fragile to implement a map/coll=
ect like feature similar to plain arrays.
you can create one if you like. just be careful.

eg, here is my simple-minded implementation,

irb(main):084:0> class Hash
irb(main):085:1> def map2
irb(main):086:2> h=3D{}
irb(main):087:2> self.each do |k,v|
irb(main):088:3* kk,vv=3Dyield(k,v)
irb(main):089:3> key =3D kk || k
irb(main):090:3> val =3D vv || v
irb(main):091:3> h[key] =3D val
irb(main):092:3> end
irb(main):093:2> h
irb(main):094:2> end
irb(main):095:1> end
=3D> nil

I'd like to use that bit for a small discussion if you allow. I
believe the implementation would be better off by not messing with
return values from the block because the changes lead to more complex
code and results may be surprising (i.e. if you relied on returning
nil or false and expected that to show up in the new Hash). By having
the method do less things it becomes more flexible to use.

Another interesting question is which is the proper method for
creating the new Hash? I can think at least of these as reasonable
solutions

h =3D {}
h =3D Hash.new(default)
h =3D Hash.new(&default_proc)
h =3D default_proc ? Hash.new(&default_proc) : Hash.new(default)
h =3D self.class.new
h =3D self.class.new(default)
h =3D self.class.new(&default_proc)
h =3D dup.clear

The solution might be to do

def map2(h =3D {})
each do |k,v|
k, v =3D yield k, v
h[k] =3D v
end
h
end

i.e. allow the caller to decide where he wants the mapped data to go
while defaulting to the most straightforward case, a new Hash. This is
similar to what Enumerable#map does, i.e. it creates a new Array
regardless of the type at hand.

Kind regards

robert

--=20
use.inject do |as, often| as.you_can - without end
 
C

clay

For fun, a solution with zip:

class Hash
def map_keys &block
Hash[ * keys.map(&block).zip(values) ]
end
end

Clay
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Members online

No members online now.

Forum statistics

Threads
473,755
Messages
2,569,537
Members
45,022
Latest member
MaybelleMa

Latest Threads

Top