# 1.9's New Methods **URL:** \ **Category:** Ruby\ **Created:** [April 15, 2010, 1:10pm UTC](https://www.ruby-forum.com/t/1-9s-new-methods/186997 "2010-04-15T13:10:37Z")\ **Posts on this page:** 8\ **Page:** 1
**Author:** ![tcsnide](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/t/eb8c5e/32.png) [@tcsnide](https://www.ruby-forum.com/u/tcsnide)\ **Post date:** [April 15, 2010, 1:10pm UTC](https://www.ruby-forum.com/t/1-9s-new-methods/186997/1 "2010-04-15T13:10:37Z")
First a big shout-out to Marc-Andre L. and his backports project. Nice work. Now I want to ask if others have noticed all the new methods being add to 1.9+? I’m quite happy about vast majority of it, but there was at least one method I thought pretty peculiar. This Enumerable method: def flat\_map(&block) return to\_enum(:flat\_map) unless block\_given? map(&block).flatten(1) end unless method\_defined? :flat\_map Backports.alias\_method self, :collect\_concat, :flat\_map I am very curious to know how it was decided that a normal map followed by a 1-deep flatten is common enough to warrant its own method? Two in fact! ---
**Author:** ![tcsnide](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/t/eb8c5e/32.png) [@tcsnide](https://www.ruby-forum.com/u/tcsnide)\ **Post date:** [April 15, 2010, 1:31pm UTC](https://www.ruby-forum.com/t/1-9s-new-methods/186997/2 "2010-04-15T13:31:07Z")
On Apr 15, 2010, at 4:06 AM, Intransition [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) wrote: > end unless method\_defined? :flat\_map > Backports.alias\_method self, :collect\_concat, :flat\_map > > I am very curious to know how it was decided that a normal map > followed by a 1-deep flatten is common enough to warrant its own > method? Two in fact! i use Hash[\*collection.map {…}.flatten] all the time. ---
**Author:** ![tcsnide](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/t/eb8c5e/32.png) [@tcsnide](https://www.ruby-forum.com/u/tcsnide)\ **Post date:** [April 15, 2010, 1:49pm UTC](https://www.ruby-forum.com/t/1-9s-new-methods/186997/3 "2010-04-15T13:49:16Z")
On Apr 15, 7:30 am, Ryan D. [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) wrote: > i use Hash[\*collection.map {…}.flatten] all the time. I see. But really? So now you will use: ``` Hash[*collection.flat_map {...}] ``` But that explains why this would never have occurred to me. I use: ``` collection.map {...}.to_h ``` Thanks. ---
**Author:** ![tcsnide](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/t/eb8c5e/32.png) [@tcsnide](https://www.ruby-forum.com/u/tcsnide)\ **Post date:** [April 15, 2010, 1:55pm UTC](https://www.ruby-forum.com/t/1-9s-new-methods/186997/4 "2010-04-15T13:55:43Z")
On Thu, Apr 15, 2010 at 8:30 PM, Ryan D. [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) wrote: > > Â def flat\_map(&block) Hash[[1,2,3].map{|i| [i,i\*2]}] # {1=\>2, 2=\>4, 3=\>6} RUBY\_DESCRIPTION # “ruby 1.9.1p378 (2010-01-10 revision 26273) [x86\_64-linux]” ---
**Author:** ![tcsnide](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/t/eb8c5e/32.png) [@tcsnide](https://www.ruby-forum.com/u/tcsnide)\ **Post date:** [April 15, 2010, 3:56pm UTC](https://www.ruby-forum.com/t/1-9s-new-methods/186997/5 "2010-04-15T15:56:20Z")
On Apr 15, 2010, at 1:06 PM, Intransition wrote: > end unless method\_defined? :flat\_map > Backports.alias\_method self, :collect\_concat, :flat\_map > > I am very curious to know how it was decided that a normal map > followed by a 1-deep flatten is common enough to warrant its own > method? Two in fact! One advantage of :flat\_map over map(…).flatten(…) is that it can be chained in an Enumerator. Sometimes, this can come in handy. Regards, Florian G. ---
**Author:** ![tcsnide](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/t/eb8c5e/32.png) [@tcsnide](https://www.ruby-forum.com/u/tcsnide)\ **Post date:** [April 16, 2010, 3:24am UTC](https://www.ruby-forum.com/t/1-9s-new-methods/186997/6 "2010-04-16T03:24:56Z")
On Apr 15, 2010, at 06:34 , James Edward G. II wrote: > On Apr 15, 2010, at 6:30 AM, Ryan D. wrote: > > > i use Hash[\*collection.map {…}.flatten] all the time. > > Me too, but it’s needed less in 1.9 I think. Hash[…] now accepts an Array of Arrays. nice. too bad I don’t use 1.9 at all. ---
**Author:** ![tcsnide](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/t/eb8c5e/32.png) [@tcsnide](https://www.ruby-forum.com/u/tcsnide)\ **Post date:** [April 15, 2010, 3:34pm UTC](https://www.ruby-forum.com/t/1-9s-new-methods/186997/7 "2010-04-15T15:34:40Z")
On Apr 15, 2010, at 6:30 AM, Ryan D. wrote: > i use Hash[\*collection.map {…}.flatten] all the time. Me too, but it’s needed less in 1.9 I think. Hash[…] now accepts an Array of Arrays. James Edward G. II ---
**Author:** ![tcsnide](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/t/eb8c5e/32.png) [@tcsnide](https://www.ruby-forum.com/u/tcsnide)\ **Post date:** [April 18, 2010, 12:55am UTC](https://www.ruby-forum.com/t/1-9s-new-methods/186997/8 "2010-04-18T00:55:40Z")
Intransition wrote: > end unless method\_defined? :flat\_map > Backports.alias\_method self, :collect\_concat, :flat\_map > > I am very curious to know how it was decided that a normal map > followed by a 1-deep flatten is common enough to warrant its own > method? Two in fact! #flat\_map is monadic bind (also known as flatMap in Scala, SelectMany in .NET and (\>\>=) in Haskell). The other monadic operator that is needed to build a monad, is unit (aka return in Haskell), but in an OO language, that’s just a factory method, IOW it’s just .new in Ruby. So, #flat\_map can potentially turn any class that mixes in Enumerable into a monad, provided that it also obeys the monad laws, of course. Whether or not that warrants its own method … well, I have no idea. jwm