102256-103399

Extending Objects Without Inheritance
102256 [alexg@eb .a ] I'm no great OO expert (or Ruby expert) so maybe the answer to this is
+ 102264 [news@st d. t] # 3. Create a ExtraAtomFunctionality module that you
| + 102266 [bob.news@gm ] "Kent Dahl" <kentda+news@stud.ntnu.no> schrieb im Newsbeitrag
| + 102283 [alexg@eb .a ] Yes, thanks for your reply, I actually stumbled on this solution
+ 102325 [vjoel@PA H. ] why not define the extra accessors you might need and assign to them
  102356 [alexg@eb .a ] Yes, this works as well, I obviously hadn't realised quite how dynamic
  + 102358 [botp@de mo t] I posted this on gem site but I want to get answer quick so....
  + 102423 [vjoel@PA H. ] Yep, you can reopen classes as often as you like, and add methods to

looking for algorithms: e, sqrt,...
102268 [r_mueller@im] Moin,
+ 102271 [Stephan.Kaem] Moinmoin,
| 102274 [r_mueller@im] Now I'm in Berlin, but lived in Kiel for 14 years.
| 102278 [michael.camp] Better check the licensing before you use/distribute anything with
+ 102285 [ahoward@no a] ...

[ANN] Active Record 0.8.1: Object-level transactions, generic connector
102269 [david@lo dt ] What's new in Active Record 0.8.1?
+ 102302 [carl@yo ng l] I am so excited to try this out!  Thanks for a great tool David and
+ 102340 [kbullock@ri ] Excellent work so far, David. I've been playing around with

linefilter - looking for suggestions
102272 [martindemell] I'm writing a small wrapper around ruby, meant to be used as part of a
102276 [bob.news@gm ] "Martin DeMello" <martindemello@yahoo.com> schrieb im Newsbeitrag
102280 [martindemell] Hm - I was trying to avoid an extraneous 'split', since cols always
102282 [bob.news@gm ] "Martin DeMello" <martindemello@yahoo.com> schrieb im Newsbeitrag
102287 [martindemell] Good point - and it's the nicely polymorphic way to do it too. I'll have

[ANN] Ruby/GtkTrayIcon 0.1.0
102295 [lrz@gn me or] Dear all,

online archive of the missing ruby-lang => clr posts
102301 [dblack@wo bl] David

A question about GC
102310 [ssk@ch l. os] While I was making my own utility functions, I met an interesting situation.
+ 102313 [WBrown@is ft] Sam,
+ 102314 [Ruben.Vandeg] Interesting problem :) You have to 'disconnect' the parameter in the
+ 102316 [drbrain@se m] on.
+ 102364 [bob.news@gm ] "Sam Sungshik Kong" <ssk@chol.nospam.net> schrieb im Newsbeitrag

[ANN]RubyGems/RubyForge integration
102338 [chad@ch df w] problems),
+ 102367 [eule@sp ce c] Good work !
+ 102380 [rich@in oe h] Important additional note...you need to name the file with a .gem
+ 102400 [surrender_it] wonderful, thanks!

[RCR]: Replacing IO.select with predicates
102339 [dejaspam@ba ] When you have to use IO.select, code can get kind of ugly.  Consider

TkButton#command problem with threads
102341 [henst374@st ] I upgraded to FreeBSD's latest version of ruby.
102348 [nagai@ai ky ] Thank you for your report. Please test the following patch.
102363 [henst374@st ] Thank you very much for your patch.

Notice on ruby-lang.org mailing list service restart
102351 [shugo@ru y- ] = Notice on ruby-lang.org mailing list service restart

playing with popen again
102352 [hal9000@hy e] Why doesn't this work?
+ 102355 [usenet-20040] This would be a side-effect of buffering.  prog2.rb won't be sending
+ 102372 [ahoward@no a] stdout is only line buffered when it's a tty.  in the case of an external

Gem 0.4 segfaults on winxp upon install
102359 [botp@de mo t] ooops.. I'm very sorry... forgot to change subject.. pls ignore previous
102361 [chadfowler@g] Strange one.   I'm not seeing this on my computer at work.  I'm
102365 [botp@de mo t] Yes, it does segfaults.
102368 [chadfowler@g] Can anyone else confirm that zlib segfaults on this release of ruby on Windows?
+ 102373 [ jimm@io co ] This does *not* segfault on my XP Pro machine, in either a DOS or Cygwin
+ 102379 [rich@in oe h] This DOES NOT segfault on WinXP Home w/latest Ruby win32 distro...works
  102387 [chad@ch df w] # >>>

[ANN] Active Record 0.8.2: Inheritable callback ques
102366 [david@lo dt ] What's new in Active Record 0.8.2?
+ 102375 [raphael.baud] I just tested it and it seems really easy to link Classes to tables. However, I encountered
| + 102388 [carl@yo ng l] I usually prefer to name my primary ID columns <tablename>id just to
| | + 102396 [khaines@en g] I still haven't delved into ActiveRecord's code as much as I want to, but in
| | + 102407 [aredridel@nb] ...
| |   102426 [david@lo dt ] Great patch, Ari! Experimentors beware, though. This patch has a few
| |   102427 [aredridel@nb] Yeah, sorry for the low quality of the patch -- it's from my development
| + 102474 [usenet2@no p] Couldn't the table meta data be used to find out which field(s) is / are
|   102478 [rb@ra hi ou ] That's what I meant by 'having the name of the "id column" extracted
|   102480 [khaines@en g] Most of them support it.  That's how Kansas works.
+ 102385 [carl@yo ng l] This all does look pretty cool, because it has the potential to make
+ 102391 [carl@yo ng l] One other question: Are these operations atomic?  Does a transaction get
  102425 [david@lo dt ] They will be in 0.8.3 ;). It's in the CVS right now. I'll probably

[ANN] Ruby/GraphViz 0.2.0
102369 [greg@we ti e] Ruby/GraphViz 0.2.0 was released yesterday

Reminder: RubyConf pre-registration still open!
102378 [dblack@wo bl] We're getting decent but not overwhelming pre-registration traffic.
102386 [dblack@wo bl] David

Transfering files using drb (distributed Ruby)
102384 [fjoanis@vi e] I'd like to know if it is possible to use drb to transfer files from the
102399 [khaines@en g] It's definitely possible.  As to have the infrastructure behaves, you will
102419 [vjoel@PA H. ] And if the file just gets slurped into a string, the marshalling cost
102485 [fjoanis@vi e] Thanks guys, I'm going to try it out and give you some news.
102526 [m_seki@mv .b] # setup server

Taming Test::Unit
102392 [bg-rubytalk@] Recently I've been trying to use Ruby's Test::Unit framework to test a
+ 102422 [vjoel@PA H. ] IIRC, a new instance is created for each test_XXX method, and setup and
+ 102439 [nathaniel@ta] Hopefully I can help out a bit...
  103054 [bret@pe ti h] What about putting the teardown in an END block?

ANN: REXML 3.1.0
102395 [ser@ge ma e-] This apparently didn't post the first time around, so I'm trying
102411 [clr1.10.rand] Wow, this sounds great! I just need such a feature and thought about

utf8 strings and inspect
102398 [linux@ma cr ] if I have an utf8-string
102404 [angus@qu va ] [Benny <linux@marcrenearns.de>, 2004-06-04 18.53 CEST]
102405 [linux@ma cr ] how can I do that?
102408 [angus@qu va ] [Benny <linux@marcrenearns.de>, 2004-06-04 19.28 CEST]

Problem Installing Ruby Gems 0.4.0 on XP not segfault
102401 [erne@po er a] I get the following error I'm running Ruby 1.8 windows version installed
+ 102403 [chad@ch df w] # I get the following error I'm running Ruby 1.8 windows version installed
| 102486 [erne@po er a] undefined
| 102487 [chadfowler@g] Ernie, would you mind trying with the latest ruby installer version
| 102489 [erne@po er a] I just upgraded to 1.8.1 -12
| 102490 [chadfowler@g] Thanks, Ernie.
+ 102406 [surrender_it] ruby 1.8.1 (2003-12-25) [i386-mswin32]
  102418 [chad@ch df w] # ruby 1.8.1 (2003-12-25) [i386-mswin32]
  102420 [chad@ch df w] # # Bu, one thing: now that gems are growing in number it would be nice to
  102462 [surrender_it] tried, works like a charm, thanks!

Check assertion in Ruby
102412 [ehames@va es] number of rows returned by a query. Something like Eiffel's
+ 102413 [mneumann@nt ] def check
| + 102414 [ehames@va es] Awesome! I love Ruby!
| + 102421 [jean_hugues_] if DEBUG
| | 102424 [pbrannan@at ] That's the best I know of that you could do in pure ruby.
| | 102447 [lists@za a. ] Is anyone using cpp (C preprocessor) along with Ruby? I've always wanted
| | 102454 [bob.news@gm ] "David Garamond" <lists@zara.6.isreserved.com> schrieb im Newsbeitrag
| | + 102463 [lists@za a. ] I don't think preprocessing is obsolete as a concept. We can still do
| | + 102606 [jean_hugues_] Cool !
| |   102639 [bob.news@gm ] ...
| |   + 102641 [martindemell] I wonder how adaptable camlp4 is.
| |   + 102657 [jean_hugues_] Perfect fro me. Thanks a lot. Ruby is amazing.
| + 102428 [dblack@wo bl] def check(a)
|   + 102453 [bob.news@gm ] "David A. Black" <dblack@wobblini.net> schrieb im Newsbeitrag
|   + 102458 [mneumann@nt ] def check(a=nil, &block)
|     + 102461 [news@st d. t] I don't like unnecessary object construction, such as that Proc-wrapping.
|     + 102464 [surrender_it] it is not possible to use the assert-stuff in test::unit ?
|       102468 [mneumann@nt ] Sure it is...
+ 102415 [walker@le ha] result.num_rows == 1 or raise "num_rows is not 1"
  102429 [dblack@wo bl] I didn't know idioms had to be unreadable :-))

How to ducktype a Hash?
102431 [sean@ce so t] I need to detect when an object is a hash-like container for other objects,
+ 102433 [flgr@cc n. e] obj.respond_to?(:keys) # ? :)
| 102434 [nobu.nokada@] I'd prefer :each_pair.
| 102436 [sean@ce so t] In both cases, does that mean that the object has hash-like functionality, or
| + 102440 [DocBoobenste] Well, what exactly do all your hash-like objects have in common that
| | 102492 [sean@ce so t] The problem is, there's no definition of what a hash-like object really is.
| + 102441 [burtdav@ho m] It seems to me that listing (#keys) and iterating through (#each_pair) its
| | 102456 [bob.news@gm ] Maybe I got something wrong here, but I thought the idea of duck typing was
| | 102495 [sean@ce so t] The problem isn't that the object doesn't respond to the methods I need, it's
| | + 102496 [dave@pr gp o] Perhaps you could design the interface to your classes so that the
| | | 102499 [sean@ce so t] It's not my classes that are the problem, it's all of the other objects that
| | | + 102505 [quixoticsyco] Could you use explicit names such as #lookup and #store ?  I realize it's
| | | + 102507 [dave@pr gp o] Then it's not your problem! Code on, young turk.
| | |   102514 [sean@ce so t] No, it IS my problem.  My code quietly does some really undesired things when
| | |   + 102524 [flgr@cc n. e] Duck typing isn't really "quacks like a duck", but rather "quacks" in
| | |   | 102533 [sean@ce so t] It would be simpler to just have some typing.
| | |   + 102528 [dave@pr gp o] Then you need to test to make sure your parameter is a Hash. Duck
| | |   | 102534 [sean@ce so t] The objects are not always a Hash per se, just objects with hash-like
| | |   | 102537 [dave@pr gp o] What gives them this hash-like functionality? Can you test for it? And
| | |   | 102543 [sean@ce so t] It's tough to explain, the code affecting this is spread out in different
| | |   | + 102562 [lists@za a. ] You probably meant 'type checking' instead of 'typing'.
| | |   | + 102568 [kristof@vl e] Hm, just an idea, but perhaps is tagging not such an ugly solution?
| | |   | | 102576 [sean@ce so t] It's a decent solution, I think.  My only complaint is that I have to hack it
| | |   | + 102593 [ng@jo nw on ] I have only a small amount of experience with using Ruby on large
| | |   |   102599 [sean@ce so t] No.  As the project grows, errors crop up in new places.  Unit tests only help
| | |   |   + 102602 [jamesUNDERBA] I believe the purpose of unit testing is to ensure correct program
| | |   |   + 102617 [billk@ct .c ] There seems to be a contradiction here.  You're reporting a
| | |   |     102629 [sean@ce so t] I tend to have a low defect rate these days.  Not that they don't happen, it's
| | |   |     102634 [billk@ct .c ] Well, Gramps - that's 1 year longer than me.  :)
| | |   + 102530 [ng@jo nw on ] The problem Sean is that even if you know the method has the right
| | |     102535 [sean@ce so t] That's just a confidence issue.  I don't care if the object coming in is
| | |     + 102538 [dave@pr gp o] I think to be fair you're asking too much of duck typing. It really is
| | |     | 102541 [sean@ce so t] That would be sad to deprecate Ruby to just little scripts.  It already does
| | |     + 102545 [g_ogata@op u] So if throughout the standard library, all "Hash-like" objects would
| | |     | 102577 [sean@ce so t] Whatever the "interface" description is, that's the rule.  If an object said
| | |     | + 102604 [kristof@vl e] Like I said in another post, this is very easy :-)
| | |     | + 102627 [g_ogata@op u] Well, as Kristof replied, that's easily done.  Doing it on a per-Class
| | |     |   102630 [sean@ce so t] My complaint is that it's not already in the stdlib.  I don't expect
| | |     |   + 102644 [dblack@wo bl] Matz is definitely giving thought to this area; see
| | |     |   + 102707 [kristof@vl e] Kristof Bastiaensen
| | |     |     102710 [Austin.Ziegl] Well, I still think that this is probably a good compromise, as (1) it
| | |     |     + 102711 [dblack@wo bl] I agree that that can work in principle, but I think it would backfire
| | |     |     + 102712 [msparshatt@y] I think you're reffering to needing a $ in front of global variables.
| | |     |     | 102719 [jim@we ri hh] I always thought that doing finalizers that was made cumbersome to
| | |     |     + 102720 [jim@we ri hh] I missed what this syntax (and the other proposals) is supposed to
| | |     |     | + 102722 [dblack@wo bl] Go to http://www.ruby-talk.org/102672 and follow what's nested under
| | |     |     | | 102767 [jim@we ri hh] Thanks.  Now I wish I hadn't asked.  :-)
| | |     |     | | + 102769 [gsinclair@so] I like the idea.  Can you propose some details?  And can it be called Map
| | |     |     | | + 102772 [dblack@wo bl] I tend to agree with the cannon point.  My first choice, by a wide
| | |     |     | |   102774 [david.naseby] If you head down this path a bit further (the path of decomposing existing
| | |     |     | |   102778 [surrender_it] did I already asked you 'please join the ruby buzz at artima.com' ? :)
| | |     |     | + 102725 [rich@in oe h] The key is expressing that a method you are defining is semantically
| | |     |     |   + 102726 [dblack@wo bl] What happens if the method in the other class or module gets
| | |     |     |   | 102727 [rich@in oe h] What happens if you override the << method on Array to say, just ignore the
| | |     |     |   + 102728 [dblack@wo bl] It's interesting; I'm actually having trouble coming up with much in
| | |     |     |     102730 [sean@ce so t] I'm working on an typing implementation right now that uses only arity to make
| | |     |     |     102805 [batsman.geo@] => nil
| | |     |     + 102754 [discord@ma .] hash<Hash>[:key]
| | |     + 102563 [lists@za a. ] Okay, I'm trying to sum up this thread as well as test whether my
| | |       + 102573 [dblack@wo bl] No; classes are called classes :-)  In fact, Matz has deprecated the
| | |       | + 102578 [lists@za a. ] What other things characterizes an object behaviour other than its
| | |       | | + 102580 [dblack@wo bl] Yes, but that class doesn't show up in things like #kind_of?, #class,
| | |       | | + 102582 [sean@ce so t] In Ruby, classes can change radically; methods can go from taking one
| | |       | |   + 102583 [dave@pr gp o] Perhaps you'd like Objective-C, which has the concept of protocols, and
| | |       | |   | 102591 [lists@za a. ] Interesting. Will Ruby ever have something like this? Will it be a good
| | |       | |   + 102585 [gsinclair@so] What if a class tags itself as a "hash type" but an instance of it
| | |       | |     102590 [sean@ce so t] Derivatives should, by default, inherit the type tag.  If the developer really
| | |       | + 102581 [sean@ce so t] This is brilliantly put.  This is what I mean to say.  I like that classes can
| | |       | + 102584 [jamesUNDERBA] <snip type='Superb illumination on Ruby typing/>
| | |       |   102752 [surrender_it] I wonder if that was my message.
| | |       |   103379 [DocBoobenste] (Disclaimer - I haven't read all of the thread, not am I following it
| | |       |   103397 [surrender_it] why does it? I'm not an AOP expert, but we have some degree of AOP in
| | |       + 102579 [sean@ce so t] Types and classes are not the same thing in Ruby.  In Java/C++, yes, but not
| | + 102502 [g_ogata@op u] I thought the idea was that if you unit test each piece as you write
| |   102515 [sean@ce so t] Unit tests are not the solution.  I don't even do them anymore.  They really,
| |   102520 [mailinglists] Yes they do slow you down. No doubt about this.
| + 102465 [surrender_it] why do you need to say 'it is hash like' ?
| | 102509 [ng@jo nw on ] This is even worse than object.kind_of(Hash). The whole point of "duck
| | 102516 [sean@ce so t] Which always sounded fine to me until I was actually exposed to this kind of
| | + 102519 [ljz@as as .c] I, for one, totally agree with you.
| | | + 102521 [aredridel@nb] I'd say I agree and disagree.
| | | | 102536 [sean@ce so t] Dynamic typing doesn't automatically eliminate the possibility of having some
| | | + 102522 [vjoel@PA H. ] But "same" and "different" are on a continuum. Sometimes Array#[] and
| | |   102523 [ljz@as as .c] Yes, sometimes. Recall the statement in my original post where I said
| | |   102532 [austin@ha os] Um ... not the way that I understand it. Duck typing, per se, isn't
| | |   102539 [sean@ce so t] I'm checking for hash-like behavior, because I have no control over the data
| | + 102529 [ng@jo nw on ] Duck typing may not be the best thing for you. =) Sorry. Didn't mean to
| | + 102531 [gsinclair@so] Well, that's like static typing, which is a kludge to many people.
| | | 102540 [sean@ce so t] No it's not.  Static typing is done at load/compile time.  I'm not asking for
| | + 102547 [quixoticsyco] Why should you expect [] to imply hash-like behavior in the first place?  It's
| |   + 102548 [quixoticsyco] Oops, I meant #store and #fetch.  Hash already implements these.
| |   + 102551 [surrender_it] I believe Sean is correct in saying that what he defines as hash-like
| |     102553 [quixoticsyco] This is the fault of PStore or of the lack of agreed-upon semantics
| |     + 102564 [surrender_it] I may be wrong but what you're asking has been throw out many times.
| |     | 102565 [bob.news@gm ] "gabriele renzi" <surrender_it@remove.yahoo.it> schrieb im Newsbeitrag
| |     | + 102566 [quixoticsyco] It is not necessary to introduce a Map mixin, nor would I even suggest it.
| |     | | 102588 [sean@ce so t] The problem is, it's not a matter of whether or not the methods exist, it's a
| |     | + 102587 [sean@ce so t] Please go into detail on this.  Why would type checking be an indication of
| |     |   102635 [flori@ni e. ] You can usually get rid of those checks by using polymorphism. In Ruby
| |     |   + 102663 [bob.news@gm ] "Florian Frank" <flori@nixe.ping.de> schrieb im Newsbeitrag
| |     |   + 102678 [sean@ce so t] You can't morph everything.  What if you expect an entire tree of data loaded
| |     + 102586 [sean@ce so t] I don't think such casual contracts as saying "this method identifies an
| + 102658 [vladare@ya o] Guess, you will be happy if Ruby can do some like ...
|   + 102659 [Austin.Ziegl] Actually, Ruby has both types and classes. Many developers confuse
|   | + 102660 [Austin.Ziegl] Then perhaps the way that you're looking at the solution isn't
|   | | + 102661 [Austin.Ziegl] No, you're not being alarmist at all; even in statically typed
|   | | | 102686 [sean@ce so t] That's my point about not tightly constraining the objects to their described
|   | | | 102687 [Austin.Ziegl] Could be. Your description of the problem/solution space suggests
|   | | + 102685 [sean@ce so t] Or perhaps it's very Ruby and you don't realize it.
|   | + 102684 [sean@ce so t] Ruby has no types.  All of Ruby's classes are described, and objects
|   |   102714 [gsinclair@so] So are there types in Java, such that would fit the above description?
|   |   102718 [sean@ce so t] I like interfaces better than classes, but C++ classes serve pretty well for
|   + 102679 [sean@ce so t] Yes, if obj.respond_to?(Hash::[]) can be called for non-Hash objects, and for
|     102689 [jim@we ri hh] I think the problem Sean is dealing with is one of "type discovery" rather
+ 102442 [ahoward@no a] is this totally out of the question?
| + 102443 [Steven@ru y-] You have two requirements: that it respond to [], and that it can
| | + 102467 [surrender_it] you don't need the begin..end, a method block is an "exception
| | + 102494 [sean@ce so t] On Friday 04 June 2004 22:03, Steven@ruby-lang.org, Grady@ruby-lang.org,
| + 102476 [dblack@wo bl] I think that could well work in practice, but wouldn't call it duck
| | 102510 [ahoward@no a] hmmmm... i see where you are coming from, but actualy disagree.  the way i see
| | + 102517 [sean@ce so t] This is the solution I am going with now.  In fact, if Ruby itself had this
| | | + 102555 [batsman.geo@] A protocol imposes constraints to the acceptable behavior of the object, so
| | | | 102589 [sean@ce so t] No constraint is necessary, an interface is just a declaration.  A complete
| | | + 102605 [jean_hugues_] This may be somehow stupid but what about simply inheriting
| | |   + 102609 [dblack@wo bl] I think this didn't come up because Sean was asking more about duck
| | |   + 102616 [sean@ce so t] Not all hash-like objects inherit from Hash, and the one real strength of
| | + 102610 [dblack@wo bl] But if being a Hash instance is really where all this is going, then
| + 102493 [sean@ce so t] Funny, this is similar to an idea I had once about flagging objects as
+ 102525 [elbows@sp mc] How about using the fetch and store methods? These work the same as
| 102567 [news@st d. t] ObjectSpace.each_object(Module){|m| p m if m.instance_methods.include?
+ 102649 [burtdav@ho m] param.respond_to? '[]' and
| + 102650 [dblack@wo bl] That's not duck typing (see Dave's description(s), especially recent
| | + 102656 [rich@in oe h] 1) Having a fairly large 10K Ruby framework Running across over 300 machines
| | | 102668 [paul@NO rg l] Rich,
| | | + 102672 [rich@in oe h] class Foo
| | | | + 102674 [dblack@wo bl] It would have to be something slightly different from that, though,
| | | | | 102690 [rich@in oe h] Actually no...I don't think so.  Ruby's scoping handles whether something is
| | | | | 102693 [dblack@wo bl] My point, though, was that the syntax you're proposing already means
| | | | | 102696 [rich@in oe h] f.{Array}['rich']
| | | | | 102701 [Austin.Ziegl] f<Array>['rich']
| | | | | 102702 [dblack@wo bl] I think it has to be assumed that if Matz adds any type-related stuff,
| | | | + 102682 [sean@ce so t] This is interesting, although I would like if there were a way to quickly test
| | | + 102681 [sean@ce so t] I'm not sure what this means.  I'm not looking for duck typing per se, just a
| | + 102662 [gsinclair@so] Let me just say that Sean, despite his protestations, doesn't really
| |   + 102669 [bob.news@gm ] "Gavin Sinclair" <gsinclair@soyabean.com.au> schrieb im Newsbeitrag
| |   | 102670 [gsinclair@so] Yeah, but that's a crappy kind of map :)  That's more like an array :\
| |   | 102671 [bob.news@gm ] "Gavin Sinclair" <gsinclair@soyabean.com.au> schrieb im Newsbeitrag
| |   | 102673 [gsinclair@so] => [1, 2, 3, 4]
| |   | 102802 [burtdav@ho m] You stole my next answer!
| |   + 102683 [sean@ce so t] Whatever you need to call it, I'm up for it.  =)
| + 102680 [sean@ce so t] Other objects also respond to [] and []= which are not Strings or Arrays.  The
+ 103311 [rvb@rv .d nd] Use the method #to_hash instead. All hash-like objects should be able to
  + 103330 [sean@ce so t] Not necessarily.  A database abstraction might act like a hash, but wouldn't
  | + 103331 [rvb@pr -l nu] It can act like a hash but doesn't have #to_hash. :/
  | | 103332 [sean@ce so t] It's about time someone finally admitted I'm right.  I'm going to go tell my
  | + 103369 [gsinclair@so] Such an abstraction could give a lazy-loading Hash, as in
  |   103375 [sean@ce so t] I don't see how this works.  When you call to_hash, no arguments are provided,
  |   103377 [gsinclair@so] Try 'ri Hash.new'.  All will be explained.
  |   103384 [sean@ce so t] No such command.  It doesn't ship with Ruby?  I installed ri as a gem, but it
  |   103398 [surrender_it] it did not in the past, now it is included.
  + 103394 [jean_hugues_] This makes a lot of sense to me. It seems to be symmetric
    103399 [surrender_it] [ot]