# Cleaner syntax for .map (is there already a way, or ruby2 idea?) **URL:** \ **Category:** ruby-talk\ **Created:** [27 October 2005 06:05 UTC](https://rubytalk.org/t/cleaner-syntax-for-map-is-there-already-a-way-or-ruby2-idea/21938 "2005-10-27T06:05:41Z")\ **Posts on this page:** 20\ **Page:** 1
**Author:** ![Ron\_M](https://avatars.discourse-cdn.com/v4/letter/r/ce7236/32.png) [@Ron\_M](https://rubytalk.org/u/Ron_M)\ **Post date:** [27 October 2005 06:05 UTC](https://rubytalk.org/t/cleaner-syntax-for-map-is-there-already-a-way-or-ruby2-idea/21938/1 "2005-10-27T06:05:41Z")
I find that a bunch of my code looks like    people.map{|person| person.email\_addr} where I use map to apply a single method to elements inside arrays. That sure beats C, but wouldn't it be nice if there were a shorthand for doing so. I think something like this would be very nice:    people[\*].email\_addr Even better, I often have strings of such stuff that looks like this, creating nested arrays    departments.map{|dept| dept.people.map{|person| person.email\_addr}} where a much cleaner alternative would be    departments[\*].people[\*].email\_addr so the "[\*]" operator would descend into nested arrays. In total ignorance of whether the parser would allow it, I'd like to say that I think that'd be a nice addition to ruby2 if there's no other clean shorthand out there already. ---
**Author:** ![Joel\_VanderWerf1](https://avatars.discourse-cdn.com/v4/letter/j/94ad74/32.png) [@Joel\_VanderWerf1](https://rubytalk.org/u/Joel_VanderWerf1)\ **Post date:** [27 October 2005 06:25 UTC](https://rubytalk.org/t/cleaner-syntax-for-map-is-there-already-a-way-or-ruby2-idea/21938/2 "2005-10-27T06:25:12Z")
Ron M wrote: > I find that a bunch of my code looks like > >   people.map{|person| person.email\_addr} > > where I use map to apply a single method to elements inside arrays. > > That sure beats C, but wouldn't it be nice if there > were a shorthand for doing so. I think something > like this would be very nice: > >   people[\*].email\_addr google for "symbol to\_proc". For example: [http://extensions.rubyforge.org/rdoc/classes/Symbol.html](http://extensions.rubyforge.org/rdoc/classes/Symbol.html) > **···** > > -- >       vjoel : Joel VanderWerf : path berkeley edu : 510 665 3407 ---
**Author:** ![Dave\_Burt2](https://yyz1.discourse-cdn.com/flex029/user_avatar/rubytalk.org/dave_burt2/32/2036_2.png) [@Dave\_Burt2](https://rubytalk.org/u/Dave_Burt2)\ **Post date:** [27 October 2005 06:37 UTC](https://rubytalk.org/t/cleaner-syntax-for-map-is-there-already-a-way-or-ruby2-idea/21938/3 "2005-10-27T06:37:05Z")
Ron M: > I find that a bunch of my code looks like > >   people.map{|person| person.email\_addr} > > where I use map to apply a single method to elements inside arrays. I've done something like this before: module Enumerable   def map\_send(\*methods)     map {|obj| methods.inject(obj) {|obj, meth| obj.send(meth) }}   end end a = %w{ foo bar baz} a.map\_send(:capitalize, :reverse) #=\> ["ooF", "raB", "zaB"] > .... I think something > like this would be very nice: > >   people[\*].email\_addr That's an interesting-looking operator there. I think it's very unclear what it does. I'd prefer the mouse-poo version   people.map { @.email\_addr } or   people.map { it.email\_addr } I don't think Enumerable wants a special operator like this; method names convey meaning. Cheers, Dave ---
**Author:** ![Nakada\_Nobuyoshi](https://avatars.discourse-cdn.com/v4/letter/n/848f3c/32.png) [@Nakada\_Nobuyoshi](https://rubytalk.org/u/Nakada_Nobuyoshi)\ **Post date:** [27 October 2005 08:46 UTC](https://rubytalk.org/t/cleaner-syntax-for-map-is-there-already-a-way-or-ruby2-idea/21938/4 "2005-10-27T08:46:17Z")
Hi, At Thu, 27 Oct 2005 15:05:41 +0900, Ron M wrote in [ruby-talk:162876]: > That sure beats C, but wouldn't it be nice if there > were a shorthand for doing so. I think something > like this would be very nice: > >    people[\*].email\_addr What about:   module Mappable     class Mapper       def initialize(obj)         @obj = obj       end       def method\_missing(meth, \*args, &block)         @obj.map {|i| i.send(meth, \*args, &block)}       end     end     def self.included(klass)       super       aref = klass.instance\_method(:)       klass.module\_eval do         define\_method(:) do |\*idx|           if idx.empty?             Mapper.new(self)           else             aref[\*idx]           end         end       end     end   end   class Array     include Mappable   end   a = %w[a b c]   p a.upcase > **···** > > -- > Nobu Nakada ---
**Author:** ![7rans](https://avatars.discourse-cdn.com/v4/letter/7/a8b319/32.png) [@7rans](https://rubytalk.org/u/7rans)\ **Post date:** [27 October 2005 12:07 UTC](https://rubytalk.org/t/cleaner-syntax-for-map-is-there-already-a-way-or-ruby2-idea/21938/5 "2005-10-27T12:07:03Z")
require 'facets/enumerable/op\_mod'   people.%.email\_addr A shorthand for:   people.every.email\_addr But I like your [\*]. Hmm... I maybe able to adjust above to use [:\*] --I'll see. T. ---
**Author:** ![Sylvain\_Joyeux](https://avatars.discourse-cdn.com/v4/letter/s/96bed5/32.png) [@Sylvain\_Joyeux](https://rubytalk.org/u/Sylvain_Joyeux)\ **Post date:** [27 October 2005 06:47 UTC](https://rubytalk.org/t/cleaner-syntax-for-map-is-there-already-a-way-or-ruby2-idea/21938/6 "2005-10-27T06:47:11Z")
> google for "symbol to\_proc". For example: > [http://extensions.rubyforge.org/rdoc/classes/Symbol.html](http://extensions.rubyforge.org/rdoc/classes/Symbol.html) Very nice this one ! > **···** > > -- > Sylvain Joyeux ---
**Author:** ![Robert](https://avatars.discourse-cdn.com/v4/letter/r/e95f7d/32.png) [@Robert](https://rubytalk.org/u/Robert)\ **Post date:** [27 October 2005 09:07 UTC](https://rubytalk.org/t/cleaner-syntax-for-map-is-there-already-a-way-or-ruby2-idea/21938/7 "2005-10-27T09:07:14Z")
nobuyoshi nakada wrote: > Hi, > > At Thu, 27 Oct 2005 15:05:41 +0900, > Ron M wrote in [ruby-talk:162876]: > > > That sure beats C, but wouldn't it be nice if there > > were a shorthand for doing so. I think something > > like this would be very nice: > > > >    people[\*].email\_addr > > What about: > >   module Mappable >     class Mapper >       def initialize(obj) >         @obj = obj >       end >       def method\_missing(meth, \*args, &block) >         @obj.map {|i| i.send(meth, \*args, &block)} >       end >     end >     def self.included(klass) >       super >       aref = klass.instance\_method(:) >       klass.module\_eval do >         define\_method(:) do |\*idx| >           if idx.empty? >             Mapper.new(self) >           else >             aref[\*idx] >           end >         end >       end >     end >   end > >   class Array >     include Mappable >   end > >   a = %w[a b c] >   p a.upcase I don't like it because now does two completely different things: access and mapping. I'd prefer something like this: module Enumerable   # replacement for map   def mapx(\*a,&b)     raise "Can only have one" if !a.empty? && b     if a.empty?       map(&b)     else       map {|x| a.inject(x) {|v,m| v.send(m)}}     end   end end > > %w{a b c}.mapx :upcase =\> ["A", "B", "C"] > > %w{a b c}.mapx {|a| a + "x"} =\> ["ax", "bx", "cx"] > > %w{abc bcd cde}.mapx :upcase, :reverse =\> ["CBA", "DCB", "EDC"] I just chose mapx to get it working fast, ideally the original map method of Enumerable classes would be changed. Kind regards     robert ---
**Author:** ![Sean\_O\_Halpin](https://avatars.discourse-cdn.com/v4/letter/s/ac91a4/32.png) [@Sean\_O\_Halpin](https://rubytalk.org/u/Sean_O_Halpin)\ **Post date:** [27 October 2005 10:32 UTC](https://rubytalk.org/t/cleaner-syntax-for-map-is-there-already-a-way-or-ruby2-idea/21938/8 "2005-10-27T10:32:24Z")
As always, there's something magical about your code! Unfortunately, it makes pp blow up for one 😉 For a similar purpose, I use this - not as slick but simple and understandable. Doesn't handle the nesting though. module Enumerable   def where(&block)     self.select{|x| x.instance\_eval(&block) }   end   def project(&block)     self.map{|x| x.instance\_eval(&block) }   end end as in   depts.where{ name == "Apps" }.project{ people.project{ name } } Regards, Sean > **···** > > On 10/27/05, nobuyoshi nakada \ wrote: > > > What about: > > > > module Mappable > >    class Mapper > >      def initialize(obj) > >        @obj = obj > >      end > >      def method\_missing(meth, \*args, &block) > >        @obj.map {|i| i.send(meth, \*args, &block)} > >      end > >    end > >    def self.included(klass) > >      super > >      aref = klass.instance\_method(:) > >      klass.module\_eval do > >        define\_method(:) do |\*idx| > >          if idx.empty? > >            Mapper.new(self) > >          else > >            aref[\*idx] > >          end > >        end > >      end > >    end > > end > > > > class Array > >    include Mappable > > end > > > > a = %w[a b c] > > p a.upcase > > > > -- > > Nobu Nakada ---
**Author:** ![Nakada\_Nobuyoshi](https://avatars.discourse-cdn.com/v4/letter/n/848f3c/32.png) [@Nakada\_Nobuyoshi](https://rubytalk.org/u/Nakada_Nobuyoshi)\ **Post date:** [28 October 2005 02:45 UTC](https://rubytalk.org/t/cleaner-syntax-for-map-is-there-already-a-way-or-ruby2-idea/21938/9 "2005-10-28T02:45:11Z")
Hi, At Thu, 27 Oct 2005 21:07:03 +0900, Trans wrote in [ruby-talk:162907]: > A shorthand for: > >   people.every.email\_addr > > But I like your [\*]. Hmm... I maybe able to adjust above to use [:\*] I feel your "every" much cleaner than [\*]. > **···** > > -- > Nobu Nakada ---
**Author:** ![Martin\_DeMello1](https://avatars.discourse-cdn.com/v4/letter/m/3be4f8/32.png) [@Martin\_DeMello1](https://rubytalk.org/u/Martin_DeMello1)\ **Post date:** [27 October 2005 10:52 UTC](https://rubytalk.org/t/cleaner-syntax-for-map-is-there-already-a-way-or-ruby2-idea/21938/10 "2005-10-27T10:52:03Z")
That is neat - I never thought of unifying map and mapx martin > **···** > > Robert Klemme \ wrote: > > > \>\> %w{a b c}.mapx :upcase > > =\> ["A", "B", "C"] > > \>\> %w{a b c}.mapx {|a| a + "x"} > > =\> ["ax", "bx", "cx"] > > \>\> %w{abc bcd cde}.mapx :upcase, :reverse > > =\> ["CBA", "DCB", "EDC"] ---
**Author:** ![David\_A\_Black3](https://avatars.discourse-cdn.com/v4/letter/d/6a8cbe/32.png) [@David\_A\_Black3](https://rubytalk.org/u/David_A_Black3)\ **Post date:** [28 October 2005 05:11 UTC](https://rubytalk.org/t/cleaner-syntax-for-map-is-there-already-a-way-or-ruby2-idea/21938/11 "2005-10-28T05:11:06Z")
Hi -- > **···** > > On Fri, 28 Oct 2005, nobuyoshi nakada wrote: > > > Hi, > > > > At Thu, 27 Oct 2005 21:07:03 +0900, > > Trans wrote in [ruby-talk:162907]: > > > > > A shorthand for: > > > > > >   people.every.email\_addr > > > > > > But I like your [\*]. Hmm... I maybe able to adjust above to use [:\*] > > > > I feel your "every" much cleaner than [\*]. > > I agree, visually, but I find both of the semantically opaque compared > to people.each {|person| ... } I know that people.every could return > some kind of generator or enumerator, which could then be fed > "email\_addr" symbolically... but it seems to conceal rather than > reveal what's going on. > > David > > -- > David A. Black > dblack@wobblini.net ---
**Author:** ![Zach\_Dennis1](https://avatars.discourse-cdn.com/v4/letter/z/ea5d25/32.png) [@Zach\_Dennis1](https://rubytalk.org/u/Zach_Dennis1)\ **Post date:** [28 October 2005 05:45 UTC](https://rubytalk.org/t/cleaner-syntax-for-map-is-there-already-a-way-or-ruby2-idea/21938/12 "2005-10-28T05:45:46Z")
nobuyoshi nakada wrote: > Hi, > > At Thu, 27 Oct 2005 21:07:03 +0900, > Trans wrote in [ruby-talk:162907]: > > > A shorthand for: > > > > people.every.email\_addr > > > > But I like your [\*]. Hmm... I maybe able to adjust above to use [:\*] > > I feel your "every" much cleaner than [\*]. So...    people = [person1, person2, person3]    emails = people.every.email\_addr      =\> ['test@me.com', 'test2@you.com', 'test3@us.com']    # people still is [person1, person2, person3]    people.every!.nickname    # people still is now ['Joey', 'Zeke', 'Flava Flav'] ? Zach ---
**Author:** ![Daniel\_Schierbeck](https://avatars.discourse-cdn.com/v4/letter/d/dfb087/32.png) [@Daniel\_Schierbeck](https://rubytalk.org/u/Daniel_Schierbeck)\ **Post date:** [28 October 2005 12:12 UTC](https://rubytalk.org/t/cleaner-syntax-for-map-is-there-already-a-way-or-ruby2-idea/21938/13 "2005-10-28T12:12:04Z")
David A. Black wrote: > I know that people.every could return > some kind of generator or enumerator, which could then be fed > "email\_addr" symbolically... module Enumerable    def every      enum, obj = self, Object.new      obj.define\_method :method\_missing do |name, \*args|        enum.map { |element| element.send(name, \*args) }      end      return obj    end end ary = ["john", "sylvia", "sarah"] ary.every.capitalize! puts ary.join(", ") -\> John, Sylvia, Sarah Cheers, Daniel ---
**Author:** ![Simon\_Strandgaard2](https://avatars.discourse-cdn.com/v4/letter/s/ac91a4/32.png) [@Simon\_Strandgaard2](https://rubytalk.org/u/Simon_Strandgaard2)\ **Post date:** [28 October 2005 17:17 UTC](https://rubytalk.org/t/cleaner-syntax-for-map-is-there-already-a-way-or-ruby2-idea/21938/14 "2005-10-28T17:17:54Z")
Maybe this can be useful? > **···** > > -- > Simon Strandgaard > > class Array >   def xmap(\*symbols) >     symbols.each do |symbol| >       s = symbol.to\_s >       eval "def #{s}(\*args);map{|i|i.send(:#{s}, \*args)};end" >     end >   end > end > > ary = %w(a b c d e) > > ary.xmap(:upcase, :gsub) > > p ary[0].upcase # "A" > p ary.upcase # ["A", "B", "C", "D", "E"] > p ary.gsub(/[bd]/, 'X') # ["a", "X", "c", "X", "e"] ---
**Author:** ![Daniel\_Schierbeck](https://avatars.discourse-cdn.com/v4/letter/d/dfb087/32.png) [@Daniel\_Schierbeck](https://rubytalk.org/u/Daniel_Schierbeck)\ **Post date:** [28 October 2005 12:47 UTC](https://rubytalk.org/t/cleaner-syntax-for-map-is-there-already-a-way-or-ruby2-idea/21938/15 "2005-10-28T12:47:04Z")
Daniel Schierbeck wrote: > David A. Black wrote: > > > I know that people.every could return > > some kind of generator or enumerator, which could then be fed > > "email\_addr" symbolically... > > module Enumerable >   def every >     enum, obj = self, Object.new >     obj.define\_method :method\_missing do |name, \*args| >       enum.map { |element| element.send(name, \*args) } >     end >     return obj >   end > end > > ary = ["john", "sylvia", "sarah"] > ary.every.capitalize! > > puts ary.join(", ") -\> John, Sylvia, Sarah > > Cheers, > Daniel By the way, you need the Object#define\_method if you want it to work: class Object    def define\_method(\*args, &block)      singleton\_class = class \<\< self; self; end      singleton\_class.module\_eval do        define\_method(\*args, &block)      end    end end Cheers, Daniel ---
**Author:** ![Martin\_DeMello1](https://avatars.discourse-cdn.com/v4/letter/m/3be4f8/32.png) [@Martin\_DeMello1](https://rubytalk.org/u/Martin_DeMello1)\ **Post date:** [30 October 2005 10:42 UTC](https://rubytalk.org/t/cleaner-syntax-for-map-is-there-already-a-way-or-ruby2-idea/21938/16 "2005-10-30T10:42:06Z")
I don't like that one, because it provides no visual distinction between methods that act on the array as a whole and methods that map over it. Would be a nice place to introduce the -\> operator, though 🙂 martin > **···** > > Simon Strandgaard \ wrote: > > > ary = %w(a b c d e) > > > > ary.xmap(:upcase, :gsub) > > > > p ary[0].upcase # "A" > > p ary.upcase # ["A", "B", "C", "D", "E"] > > p ary.gsub(/[bd]/, 'X') # ["a", "X", "c", "X", "e"] ---
**Author:** ![David\_A\_Black3](https://avatars.discourse-cdn.com/v4/letter/d/6a8cbe/32.png) [@David\_A\_Black3](https://rubytalk.org/u/David_A_Black3)\ **Post date:** [28 October 2005 13:45 UTC](https://rubytalk.org/t/cleaner-syntax-for-map-is-there-already-a-way-or-ruby2-idea/21938/17 "2005-10-28T13:45:40Z")
Hi -- > Daniel Schierbeck wrote: > > > David A. Black wrote: > > > > > I know that people.every could return > > > some kind of generator or enumerator, which could then be fed > > > "email\_addr" symbolically... > > > > module Enumerable > >   def every > >     enum, obj = self, Object.new > >     obj.define\_method :method\_missing do |name, \*args| > >       enum.map { |element| element.send(name, \*args) } If you must do this, you'd probably want each rather than map there. > >     end > >     return obj > >   end > > end > > > > ary = ["john", "sylvia", "sarah"] > > ary.every.capitalize! > > > > puts ary.join(", ") -\> John, Sylvia, Sarah > > > > Cheers, > > Daniel > > By the way, you need the Object#define\_method if you want it to work: I definitely \*don't\* want it to work 🙂 I dislike using dot syntax for non-dot semantics. I've never liked things like:    hash.where.the.key.equals(10) even though they can usually be made to work. I continue to struggle to understand what people find so horrible about ary.each {|item| .... }, ary.map {|item| ... }, and so forth. I'll have some more coffee and maybe I'll start to see the light.... David > **···** > > On Fri, 28 Oct 2005, Daniel Schierbeck wrote: > > -- > David A. Black > dblack@wobblini.net ---
**Author:** ![Daniel\_Schierbeck](https://avatars.discourse-cdn.com/v4/letter/d/dfb087/32.png) [@Daniel\_Schierbeck](https://rubytalk.org/u/Daniel_Schierbeck)\ **Post date:** [28 October 2005 14:07 UTC](https://rubytalk.org/t/cleaner-syntax-for-map-is-there-already-a-way-or-ruby2-idea/21938/18 "2005-10-28T14:07:05Z")
David A. Black wrote: > Hi -- > > > Daniel Schierbeck wrote: > > > > > David A. Black wrote: > > > > > > > I know that people.every could return > > > > some kind of generator or enumerator, which could then be fed > > > > "email\_addr" symbolically... > > > > > > module Enumerable > > >   def every > > >     enum, obj = self, Object.new > > >     obj.define\_method :method\_missing do |name, \*args| > > >       enum.map { |element| element.send(name, \*args) } > > If you must do this, you'd probably want each rather than map there. Nope. I wan't it to return an array of the values returned.    ["john", "sylvia", "sarah"].every.upcase -\> ["JOHN", "SYLVIA", "SARAH"] That's of course not a very good example. This would probably be better:    addresses = contacts.every.email\_addr As opposed to    addresses = contacts.collect { |contact| contact.email\_addr } But I agree that the dot syntax is bad, I was just proving that it could easily be done. This would be better:    addresses = contacts.every(:email\_addr) And maybe even have a `with\_every' method:    contacts.with\_every(:email\_addr) do |email\_addr|      puts " - " + email\_addr    end Which could be implemented this way    module Enumerable      def with\_every(\*args)        each do |element|          yield element.send(\*args)        end      end    end Cheers, Daniel > **···** > > > On Fri, 28 Oct 2005, Daniel Schierbeck wrote: ---
**Author:** ![Dave\_Burt2](https://yyz1.discourse-cdn.com/flex029/user_avatar/rubytalk.org/dave_burt2/32/2036_2.png) [@Dave\_Burt2](https://rubytalk.org/u/Dave_Burt2)\ **Post date:** [28 October 2005 14:47 UTC](https://rubytalk.org/t/cleaner-syntax-for-map-is-there-already-a-way-or-ruby2-idea/21938/19 "2005-10-28T14:47:04Z")
David A. Black wrote: > I continue to struggle to understand what people find so horrible > about ary.each {|item| .... }, ary.map {|item| ... }, and so forth. > I'll have some more coffee and maybe I'll start to see the light.... There's nothing really wrong with them, but they are verbose in simple (and common) cases: result = array.map {|element| element.transform } The only issue is that you're saying "element" twice when you would only say it once if you were describing the operation to a person. Groovy's (optional) implicit parameter is the counterpoint: // Groovy: result = array.map { it.transform() } Tangentially, I wonder if the proposed Ruby 2 block syntax makes this slightly more possible. # Pseudo-Ruby 2.0: result = array.map -\> { it.transform } Actually, probably less possible - IIRC the new block style is meant to be more like the method semantics, and so stricter on parameters. So, anyway, I don't mind the idea of an implicit parameter, but I realise it's not going to happen before Ruby 3. Cheers, Dave ---
**Author:** ![7rans](https://avatars.discourse-cdn.com/v4/letter/7/a8b319/32.png) [@7rans](https://rubytalk.org/u/7rans)\ **Post date:** [28 October 2005 15:52 UTC](https://rubytalk.org/t/cleaner-syntax-for-map-is-there-already-a-way-or-ruby2-idea/21938/20 "2005-10-28T15:52:09Z")
David A. Black wrote: > I definitely \*don't\* want it to work 🙂 I dislike using dot syntax > for non-dot semantics. I've never liked things like: > >    hash.where.the.key.equals(10) > > even though they can usually be made to work. I sort-of agree. It's definitely a semantic we're not used to --using a method call to give us a "reoriented" version of the same thing --kind of like having Roles. I suspect this will become a more common paradigm over time and me may grow acustomed to it, but I'm with you in that I'd rather have a distinguishing indication of when that's occuring. Perhaps:   people:every.email\_addr And maybe that's what Matz already has in mind. And besides, it would be nice to have it as a language feature becasue current implementations are not very efficient:   def every     Functor.new do |op,\*args|       self.collect{ |a| a.send(op,\*args) }     end   end or at best   def every     @\_\_every\_functor\_\_ ||= Functor.new do |op,\*args|       self.collect{ |a| a.send(op,\*args) }     end   end > I continue to struggle to understand what people find so horrible > about ary.each {|item| .... }, ary.map {|item| ... }, and so forth. > I'll have some more coffee and maybe I'll start to see the light.... Not horrible, but looking at a language like R with it's elemewise operations, one kind of wishes we had shorthands techinques as nice. BTW, I have another version of #every which is even more R-like which DeMello helped write. But I'm searching for a another name for it. Here it is: module Enumerable   # Returns an elementwise Functor designed to make R-like   # elementwise operations possible. > **···** > > # >   # [1,2].ew + 3 #=\> [4,5] >   # [1,2].ew + [4,5] #=\> [5,7] >   # [1,2].ew + [[4,5],3] #=\> [[5,7],[4,5]] >   # >   #-- >   # Special thanks to Martin DeMello for helping to develop this. >   #++ >   def elementwise >     Functor.new do |op,\*args| >       a = args.collect do |arg| >         if arg.kind\_of?(Enumerable) >           ln = ( arg.length \> self.length ? self.length : arg.length ) >           self[0...ln].zip(arg[0...ln]).collect{ |a,b| a.send(op,b) } >           #self[0...ln].zip(arg[0...1n]).collect{ |a,b| b ? > a.send(op,b) : nil } >         else >           self.collect{ |a| a.send(op,arg) } >         end >       end >       a.flatten! if args.length == 1 >       a >     end >   end > >   alias\_method :ew, :elementwise > > end > > Any suggestions? > > T. [Next page](https://rubytalk.org/t/cleaner-syntax-for-map-is-there-already-a-way-or-ruby2-idea/21938.md?page=2)