[#35631] [Ruby 1.9 - Bug #4558][Open] TestSocket#test_closed_read fails after r31230 — Tomoyuki Chikanaga <redmine@...>

23 messages 2011/04/06

[#35632] [Ruby 1.9 - Bug #4559][Open] Proc#== does not match the documented behaviour — Adam Prescott <redmine@...>

13 messages 2011/04/06

[#35637] [Ruby 1.9 - Bug #4561][Open] 1.9.2 requires parentheses around argument of method call in an array, where 1.8.7 did not — Dave Schweisguth <redmine@...>

9 messages 2011/04/07

[#35666] caching of the ancestor chain — Xavier Noria <fxn@...>

Why does Ruby cache the ancestors chain? I mean, not why the implementation implies that, but why it works that way conceptually.

9 messages 2011/04/09

[#35734] [Ruby 1.9 - Feature #4574][Open] Numeric#within — redmine@...

16 messages 2011/04/13

[#35753] [Ruby 1.9 - Bug #4576][Open] Range#step miss the last value, if end-exclusive and has float number — redmine@...

61 messages 2011/04/14
[#39566] [Ruby 1.9 - Bug #4576] Range#step miss the last value, if end-exclusive and has float number — Marc-Andre Lafortune <ruby-core@...> 2011/09/15

[#39590] [Ruby 1.9 - Bug #4576] Range#step miss the last value, if end-exclusive and has float number — Marc-Andre Lafortune <ruby-core@...> 2011/09/16

[#39593] Re: [Ruby 1.9 - Bug #4576] Range#step miss the last value, if end-exclusive and has float number — Tanaka Akira <akr@...> 2011/09/16

2011/9/17 Marc-Andre Lafortune <ruby-core@marc-andre.ca>:

[#39608] Re: [Ruby 1.9 - Bug #4576] Range#step miss the last value, if end-exclusive and has float number — Masahiro TANAKA <masa16.tanaka@...> 2011/09/17

I have not been watching ruby-core, but let me give a comment for this issue.

[#35765] [Ruby 1.9 - Bug #4579][Open] SecureRandom + OpenSSL may repeat with fork — redmine@...

27 messages 2011/04/15

[#35866] [Ruby 1.9 - Bug #4603][Open] lib/csv.rb: when the :encoding parameter is not provided, the encoding of CSV data is treated as ASCII-8BIT — yu nobuoka <nobuoka@...>

13 messages 2011/04/24

[#35879] [Ruby 1.9 - Bug #4610][Open] Proc#curry behavior is inconsistent with lambdas containing default argument values — Joshua Ballanco <jballanc@...>

11 messages 2011/04/25

[#35883] [Ruby 1.9 - Bug #4611][Open] [BUG] Segementation fault reported — Deryl Doucette <me@...>

15 messages 2011/04/25

[#35895] [Ruby 1.9 - Feature #4614][Open] [RFC/PATCH] thread_pthread.c: lower RUBY_STACK_MIN_LIMIT to 64K — Eric Wong <normalperson@...>

10 messages 2011/04/25

[ruby-core:35651] Re: [Ruby 1.9 - Feature #4553] Add Set#pick and Set#pop

From: Michael Edgar <adgar@...>
Date: 2011-04-07 21:34:17 UTC
List: ruby-core #35651
The existence of `Set#first` is an artifact of Set including Enumerable and the very idea of "first" on a Set is an odd notion. A set by itself does not have ordering. I have implemented it to pick the first element because it is simplest, but someone calling "pick" on a set does not care which element it is. If one would prefer implementing it as an alias, I'd be fine with that.

On a personal note, one reason I enjoy using Ruby because often my code reads just as I would describe it to a colleague - reading "element = set.first" is not something many would agree is a sensible way to describe the act of picking an arbitrary element from a mathematical set.

And yes, one could implement #pop with #delete and #first, which again relies on a notion of ordering in a set. One can implement #proper_subset with #subset and a comparison, as can one implement #proper_superset with #supserset and a comparison. Yet they are in the Set class because they are fundamental operations. #add? and #delete? can also be implemented with #add and #include?/#delete and #include? respectively, yet they are included in the Set class because they are useful and common operations on a set.

If `pop` is too close the Array method of the same name, I understand the confusion, but I don't get why picking first/last makes a difference. Anyone asking to remove an arbitrary element from a Set surely doesn't care where it lies in an underlying hash table. I am surprised one would argue that the Set class rarely needs the notion of selecting an element from it. Rarely am I presented with an algorithm using sets that does not require simply getting an element or picking one out of a set.

Michael Edgar
adgar@carboni.ca
http://carboni.ca/

On Apr 7, 2011, at 2:40 PM, Marc-Andre Lafortune wrote:

> Hi.
> 
> There is no difference between your `Set#pick` and the existing `Set#first`. As such, there is no need for `pick`.
> 
> One can get the effect of `Set#pop` by calling the existing `Set#delete` with the result of `Set#first` (with the same O(1) efficiency).
> 
> I find problematic that a method called `pop` would take the first element; it could be renamed to `shift`, say, or take the last element instead. More importantly, though, I am not convinced that this functionality is that frequently used.


In This Thread

Prev Next