From: Mike Wilson <wmwilson01@...>
Date: 2001-03-06T02:30:25+09:00
Subject: [ruby-talk:12045] RCR Summary 03/05/01

RCR's which need to be or are being discussed


#U001 cut operator for short-circuiting method chains

       The idea of this is that methods such as sort!() would be ideal
       to use in a chain, except for its sometimes returning nil.
       I proposed that an operator (which I called cut, vaguely after
       the Prolog cut) would break the chain in the case where it was
       used with a nil receiver. Otherwise it would just pass on the
       receiver.

       At present I cannot access the archives for some unknown reason,
       but I hope to come back and amend this entry when I can find pointers
       to the discussions on this.
       <HughSasse>

       previously discussed - [ruby-talk:3893] [ruby-talk:3932] 
[ruby-talk:6131] [ruby-talk:6178]
       posted to ruby-talk 02/13/01 [ruby-talk:10804] by Mike Wilson (for 
Hugh)


#U002 new proper name for Hash#indexes, Array#indexes

       I have to admint `indexes' is not good name, since its behavior
       (returning values corresponded to keys/indexes) not consistent with
       one of `index' (returning key/index of value).
       Let's face it.

       I decided to declare them obsolete and give proper new names in the
       future (1.7 and forth coming 1.8/2.0).  Do you have any idea of good
       name for them?

       I'm not going to remove them, just give warning if you use them.
       matz.

       previously discussed - [ruby-talk:8916] [ruby-talk:8920] 
[ruby-talk:8921] [ruby-talk:8973]
       posted to ruby-talk 02/14/01 [ruby-talk:10830] by matz


#U003 infix 'function composition' operator

       I'd sure like an infix 'function composition' operator for
       "chaining" proc's but I'm not sure this is important or generally 
useful
       to include in the language as opposed to requiring a file with the 
code:

       class Proc
        def *(other)
          raise unless arity == other.arity
          proc {|*a| self.call(other.call(*a))}
        end
       end

      (and then, as a side-note being able to do:

        f = lambda{|x| 2*x}
        g = lambda{|x| x+1}
        h = f * g
        h.call 1         # => 4

        # and why not throw in some syntactic sugar for the "lambda{|x| 
...}"
        # stuff � la Haskells "\x -> 2*x" while we're at it!
      )

      previously discussed - no archives found
      posted to ruby-talk 02/14/01 [ruby-talk:10837] by Robert Feldt


#U004 More list operators

      I'd like more list operators in Ruby, like those found in languages 
such
      as Haskell, Lisp, Prolog, Icon, Erlang,.... Ideally I'd like them as
      methods of Array, implemented in C for optimal performance.

      previously discussed - [ruby-talk:9479]
      posted to RubyWishList(the wish list is obsolete) by Hugh Sasse


#U005 Discuss possibility of icon-like suspend or "inside-out yield"

      Icon has the suspend command (see
      http://www.cs.arizona.edu/icon/refernce/infix3.htm#suspend for 
example)
      which allows a function to produce a result in the way that return 
does,
      but it suspends execution so that if the function is called again
      in the same context execution can continue from where the suspend left
      off. It could be considered to be an inside-out yield. This is used to
      create generators rather than iterators. It is achieved by leaving the
      state of the function on the stack.

      I suspect this may give a more functional style of programming than 
Ruby's
      iterators, but messing around with the stack in this way may be 
considered
      messy. It might be possible for the function to carry its state with 
it in
      some other way like closures do.

      previously discussed - [ruby-talk:6641] ?
      posted to RubyWishList(the wish list is obsolete) by Hugh Sasse


#U006 Alternate variable declaration

      Another way to declare that something is a variable.  Like "let" in 
Lisp or
      "my" in Perl.  By itself this is useful as a way of telling Ruby that 
I don't
      care if there is already something in scope called foo, this foo is 
private.

      previously discussed - no archives found
      posted to ruby-talk 02/16/01 [ruby-talk:10975] [ruby-talk:10990] by 
Ben Tilly


#U007 Pragma to disallow variable creation with just an assignment

      request that there be a pragma to (lexically) disallow declaring a 
variable
      by just assigning to it.  Combining #U007 with #U006 could protect 
users from
      typos ala perl's use strict "vars"

      previously discussed - no archives found
      posted to ruby-talk 02/16/01 [ruby-talk:10975] [ruby-talk:10990] by 
Ben Tilly


#U008 Keep track of implicitly called methods

      request to keep track of what methods have been implicitly called
      and have a function you can call that checks that they all resolve 
properly.
      If you make a habit of adding this at the end of classes, this will 
catch
      the common error where a method will blow up later due to a typo.  
(There
      is, of course, no good way to test for typos in explicit method calls.
      That wouldn't stop it from being a useful feature though.)
      ala perl's use strict "subs"

      previously discussed - no archives found
      posted to ruby-talk 02/16/01 [ruby-talk:10975] [ruby-talk:10990] by 
Ben Tilly


#U009 Allow "#{foo} bar" to be used as "#foo bar"

      Would like to be able to use #foo in place of #{foo} (Just to stop
      people complaining that interpolation requires 2 extra characters.)
      ala perl's "$foo bar"

      previously discussed - no archives found
      posted to ruby-talk 02/16/01 [ruby-talk:10990] by Ben Tilly


#U010 new method Enumerable#hashify(value)

      Return a hash based on iterating through the receiver (an enumerable),
      either setting every value to the parameter "value" or setting the 
values
      via a block.

      previously discussed - [ruby-talk:6593] [ruby-talk:6611] 
[ruby-talk:6663]
      [ruby-talk:7186] [ruby-talk:7367] [ruby-talk:7368] [ruby-talk:7371] 
[ruby-talk:7427]
      posted to ruby-talk 02/20/01 [ruby-talk:11195] by David Alan Black


#U011 Add a method: Time#set(seconds <,microseconds)

      This method, which is much like Time.at, will alter an existing Time 
object.

      I think making Time mutable is a big change.
      We must discuss carefully before making decision.

      previously discussed - [ruby-talk:11426] [ruby-talk:11381]
      posted to ruby-talk 02/25/01 [ruby-talk:11531] by matz


#U012 Discuss making Ruby XML-ready "out of the box"

      IMHO, all future Ruby distributions (tgz, InstallShield?) ought to be
      "XML-ready" out of the box.

      I think XML is going to be the "sort-of next big thing"--i.e. not 
*that*
      big, but big *enough* to matter. (Where's my marketing hat. Ahh, here 
it
      is. "Why dork around with Perl, Python, or Tcl when Ruby supports XML 
out
      of the box?")

      previously discussed - no archives found
      posted to ruby-talk 02/26/01 [ruby-talk:11619] by Conrad Schneiker


#U013 shortcut for instance variable initialization

      I'm constantly creating small classes, and many of them start
      something like:

      class Msg
        def initialize(name, type, txt)
          @name, @type, @txt = name, type, txt
        end
        # ...
      end

      So, could we change parameter passing slightly so that if an instance
      method has a formal parameter starting with an '@' sign, the incoming
      value is assigned directly to the corresponding instance variable?
      Using this scheme I could write the above as:

      class Msg
        def initialize(@name, @type, @txt)
        end
      end

      previously discussed - no archives found
      posted to ruby-talk 02/26/01 [ruby-talk:11633] by Dave Thomas


#U014 new method, instance_eval + value passing

      matz has proposed that there should be a method with functionality
      similar to instance_eval but with the ability to pass values to the
      evaluating block.  matz suggests that the behavior contradicts the
      name and role of instance_eval, and so a new method with a new name
      should be implemented.

      instance_yield is noted as being one possible name.

      previously discussed - no archives found
      posted to ruby-talk 03/01/01 [ruby-talk:11881] by matz




_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com