# How to create hash of hash? **URL:** \ **Category:** Rails\ **Created:** [September 18, 2008, 2:28am UTC](https://www.ruby-forum.com/t/how-to-create-hash-of-hash/148891 "2008-09-18T02:28:45Z")\ **Posts on this page:** 5\ **Page:** 1
**Author:** ![jdunneuk](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/j/35a633/32.png) [@jdunneuk](https://www.ruby-forum.com/u/jdunneuk)\ **Post date:** [September 18, 2008, 2:28am UTC](https://www.ruby-forum.com/t/how-to-create-hash-of-hash/148891/1 "2008-09-18T02:28:45Z")
hashx = { key1 =\> {key1 =\> key1val1, key2 =\> key1val2}, key2 =\> {key1 =\> key2val1, key2 =\> key2val2} } If I want to insert a hash in the 1st level, how to do? key3\_hash = {key1 =\> key3val1, key2 =\> key3val2} hashx[key3] = key3\_hash # Is there a better way to write this? I want to assign the programmatically How about adding to the 2nd level hash? Say anon\_hash = hashx[key1] anon\_hash[key3] = {key3 =\> key1val3} hashx[key1] = anon\_hash # is there a better way to do this? ---
**Author:** ![jdunneuk](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/j/35a633/32.png) [@jdunneuk](https://www.ruby-forum.com/u/jdunneuk)\ **Post date:** [September 18, 2008, 9:09am UTC](https://www.ruby-forum.com/t/how-to-create-hash-of-hash/148891/2 "2008-09-18T09:09:01Z")
On Sep 18, 1:28 am, John Do [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) wrote: > hashx = { key1 =\> {key1 =\> key1val1, key2 =\> key1val2}, key2 =\> {key1 =\> > key2val1, key2 =\> key2val2} } > > If I want to insert a hash in the 1st level, how to do? > > key3\_hash = {key1 =\> key3val1, key2 =\> key3val2} > hashx[key3] = key3\_hash > > # Is there a better way to write this? I want to assign the > > programmatically Not quite sure what you’re getting at, but what’s wrong with hashx[key3] = {key1 =\> key3val1, key2 =\> key3val2} ? > How about adding to the 2nd level hash? > > Say > anon\_hash = hashx[key1] > anon\_hash[key3] = {key3 =\> key1val3} > hashx[key1] = anon\_hash > > # is there a better way to do this? why not just hashx[key1][key3]=key1val3 ? You don’t need all those intermediate steps. Fred ---
**Author:** ![jdunneuk](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/j/35a633/32.png) [@jdunneuk](https://www.ruby-forum.com/u/jdunneuk)\ **Post date:** [September 18, 2008, 10:30am UTC](https://www.ruby-forum.com/t/how-to-create-hash-of-hash/148891/3 "2008-09-18T10:30:31Z")
On Sep 18, 3:08 pm, Frederick C. [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) wrote: > why not just hashx[key1][key3]=key1val3 ? You don’t need all those > intermediate steps. > > Fred Maybe he’s dealing with really large hashes - or just many operations on them - that having to reference the first sub-hash via the parent hash every time would give a significant degradation in performance. I still prefer the approach you mentioned, though. ---
**Author:** ![jdunneuk](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/j/35a633/32.png) [@jdunneuk](https://www.ruby-forum.com/u/jdunneuk)\ **Post date:** [September 18, 2008, 8:56pm UTC](https://www.ruby-forum.com/t/how-to-create-hash-of-hash/148891/4 "2008-09-18T20:56:45Z")
On Sep 18, 6:08 pm, John Do [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) wrote: > Erol F. wrote: > > > On Sep 18, 3:08 pm, Frederick C. [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) > > wrote: > > hashx[key1][key3] = key1val3 will give an error if hashx[key1] is null. > This is how I did it in perl, but ruby isn’t as forgiving. hash = Hash.new {|h,k| h[k]=Hash.new &h.default\_proc} Mwahahahahaha Fred ---
**Author:** ![jdunneuk](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/j/35a633/32.png) [@jdunneuk](https://www.ruby-forum.com/u/jdunneuk)\ **Post date:** [September 18, 2008, 7:08pm UTC](https://www.ruby-forum.com/t/how-to-create-hash-of-hash/148891/5 "2008-09-18T19:08:35Z")
Erol F. wrote: > On Sep 18, 3:08�pm, Frederick C. [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) > wrote: > > > why not just hashx[key1][key3]=key1val3 ? You don’t need all those > > intermediate steps. > > > > Fred > > Maybe he’s dealing with really large hashes - or just many operations > on them - that having to reference the first sub-hash via the parent > hash every time would give a significant degradation in performance. I > still prefer the approach you mentioned, though. hashx[key1][key3] = key1val3 will give an error if hashx[key1] is null. This is how I did it in perl, but ruby isn’t as forgiving. Also, I want to be able to build an arbitrarily deep hash. Say hashx[key1][key11][key111]…[key11111]. The end goal is to build a hash that I can build an YUI treeview on. So I need to build the tree, then loop through the keys to create the TreeNode.