[#96321] parent of TrueClass, FalseClass — "Ara.T.Howard" <Ara.T.Howard@...>

19 messages 2004/04/01
[#96356] Re: parent of TrueClass, FalseClass — matz@... (Yukihiro Matsumoto) 2004/04/02

Hi,

[#96404] Variable names — David King Landrith <dlandrith@...>

About a year ago, I wrote a simple type enforcement library that adds a

33 messages 2004/04/02
[#96406] Re: Variable names — ts <decoux@...> 2004/04/02

>>>>> "D" == David King Landrith <dlandrith@mac.com> writes:

[#96424] Re: Variable names — David King Landrith <dlandrith@...> 2004/04/02

[#96430] Re: Variable names — Dan Doel <djd15@...> 2004/04/02

On Friday 02 April 2004 2:43 pm, David King Landrith wrote:

[#96432] Re: Variable names — David King Landrith <dlandrith@...> 2004/04/02

On Apr 2, 2004, at 3:51 PM, Dan Doel wrote:

[#96447] Learning Ruby, was a C geek... — Nicholas Paul Johnson <nickjohnson@...>

Hello all,

17 messages 2004/04/02

[#96634] Where does the name Rite come from? — Michael Neumann <mneumann@...>

Hi,

15 messages 2004/04/06
[#96642] Re: Where does the name Rite come from? — matz@... (Yukihiro Matsumoto) 2004/04/06

Hi,

[#96652] Re: Where does the name Rite come from? — Eric Hodel <drbrain@...7.net> 2004/04/06

Yukihiro Matsumoto (matz@ruby-lang.org) wrote:

[#96697] Idea: Simplified GTK — Hal Fulton <hal9000@...>

Here's an idea. I've begun implementing it.

74 messages 2004/04/07
[#96699] Re: Idea: Simplified GTK — Chad Fowler <chad@...> 2004/04/07

[#96876] RedCloth bug and suggestion — Jim Menard <jimm@...>

_why_ and fellow RedCloth users,

15 messages 2004/04/09

[#96877] Instiki 0.3.0: Before the Storm — David Heinemeier Hansson <david@...>

What's new in Instiki 0.3.0?

12 messages 2004/04/09

[#97020] test / unit question: facility to mark some tests as "missing" or "incomplete" — "Its Me" <itsme213@...>

I often find myself with some unit tests that run, and several more test

14 messages 2004/04/12

[#97077] Instiki 0.5.0: Getting Serious — David Heinemeier Hansson <david@...>

What's new in Instiki 0.5.0?

16 messages 2004/04/13

[#97083] New Ruby questions... — Jeff Massung <jma@...>

I've just started Ruby a couple days ago (man this is cool). Coming from

14 messages 2004/04/13

[#97109] New Local Variable Scope rule — "Shashank Date" <sdate@...>

In one of Matz's slides at RubyConf ,

33 messages 2004/04/14

[#97134] BlueCloth: a Markdown implementation for Ruby — Michael Granger <ged@...>

Hi all,

17 messages 2004/04/14

[#97201] File locking, portably? — Hugh Sasse Staff Elec Eng <hgs@...>

Searching the web and books for information on this, I can't seem to

12 messages 2004/04/14

[#97277] Hash, ==, key-value comparison — walter@...

Ok,

20 messages 2004/04/15

[#97308] Instiki 0.6.0: Feeds, Exports, Safety, and Compatibility — David Heinemeier Hansson <david@...>

What's new in Instiki 0.6.0?

14 messages 2004/04/15

[#97351] "bad file descriptor" in Win32 DLL — "Christian Kaiser" <chk@...>

Using Ruby 1.81, the DLL (msvcrt-ruby18.dll) sometimes raises an exception

14 messages 2004/04/16

[#97363] BlueCloth 0.0.2 (beta) — Michael Granger <ged@...>

Hi,

18 messages 2004/04/16
[#97399] BlueCloth on Instiki (was Re: [ANN] BlueCloth 0.0.2 (beta)) — David Heinemeier Hansson <david@...> 2004/04/16

> Thanks to all of you that have offered your suggestions and code. I

[#97405] RubyConf 2004 — Ian Macdonald <ian@...>

Hello,

46 messages 2004/04/17
[#97409] Re: RubyConf 2004 — Chad Fowler <chad@...> 2004/04/17

[#97460] Re: RubyConf 2004 — Paul Duncan <pabs@...> 2004/04/17

* Chad Fowler (chad@chadfowler.com) wrote:

[#97465] Re: RubyConf 2004 — Hal Fulton <hal9000@...> 2004/04/17

Paul Duncan wrote:

[#97466] Re: RubyConf 2004 — Chad Fowler <chad@...> 2004/04/17

[#97486] Ruby Installer for Windows 1.8.1-12 — "Curt Hibbs" <curt@...> 2004/04/18

The Ruby Installer 1.8.1-12 for Windows has been released and is now

[#97418] Objects in perl6 (rubyish :) — gabriele renzi <surrender_it@...1.vip.ukl.yahoo.com>

Hi gurus and nubys,

20 messages 2004/04/17

[#97426] $0 is messed up — Simon Strandgaard <neoneye@...>

I have just upgraded to 1.9 16-apr-2004 from 1.9 7-apr-2004.

15 messages 2004/04/17

[#97473] convert yield to proc — Simon Strandgaard <neoneye@...>

How do you guys convert yield to block ?

21 messages 2004/04/18

[#97565] Gateway — "Robert Klemme" <bob.news@...>

34 messages 2004/04/19

[#97628] Instiki 0.7.0: Flavors of Expression — David Heinemeier Hansson <david@...>

What's new in Instiki 0.7.0?

15 messages 2004/04/19

[#97631] proposal: call_up() for use in redefined methods — Mark Hubbart <discord@...>

Hi all,

18 messages 2004/04/19

[#97640] Fox --> GTK ? — Hal Fulton <hal9000@...>

Who has experience converting Fox to GTK and might like to

29 messages 2004/04/19

[#97705] The quest for opensource database... — "Ruby Tuesdays" <NoSpamPlease_rubytuzdaiz@...>

Perhaps you database guru able to suggest what would be a good choice for

35 messages 2004/04/20

[#97743] Setting up a wiki when you don't have root — ptkwt@... (Phil Tomson)

12 messages 2004/04/20

[#97785] Creating bang methods — "Jon Hurst" <jon@...>

(newbie) I can't for my life figure out how to create bang methods. Please

20 messages 2004/04/21

[#97797] rexml: how to get element type? — "Its Me" <itsme213@...>

doc = REXML::Document.new <<EOF

14 messages 2004/04/21

[#97808] binding - how to get current script? — Szymon Drejewicz <drejewic@...>

18 messages 2004/04/21

[#97866] Is Ruby is better than PHP ... — "Useko Netsumi" <usenets_remote_this@...>

or perhaps Java for developing web application?

46 messages 2004/04/21

[#97873] Re: how to get ri/rdoc working for 1.8.1 on Windows? — "John W. Kennedy" <jwkenne@...>

Part 1 is to chdir \Ruby\bin and delete the five *.bat files

41 messages 2004/04/21
[#97910] Re: how to get ri/rdoc working for 1.8.1 on Windows? — "Its Me" <itsme213@...> 2004/04/21

[#97964] Re: how to get ri/rdoc working for [user-installed libraries] — Gavin Sinclair <gsinclair@...> 2004/04/21

On Thursday, April 22, 2004, 4:44:09 AM, Its wrote:

[#97984] Re: how to get ri/rdoc working for [user-installed libraries] — Dave Thomas <dave@...> 2004/04/22

[#98000] Re: how to get ri/rdoc working for [user-installed libraries] — "Gavin Sinclair" <gsinclair@...> 2004/04/22

>

[#97926] Re: how to get ri/rdoc working for 1.8.1 on Windows? — Lothar Scholz <mailinglists@...> 2004/04/21

Hello John,

[#97936] Re: how to get ri/rdoc working for 1.8.1 on Windows? — Dave Thomas <dave@...> 2004/04/21

[#97971] Re: how to get ri/rdoc working for 1.8.1 on Windows? — Jos Backus <jos@...> 2004/04/21

On Thu, Apr 22, 2004 at 05:50:40AM +0900, Dave Thomas wrote:

[#97985] Re: how to get ri/rdoc working for 1.8.1 on Windows? — Dave Thomas <dave@...> 2004/04/22

[#97997] RedCloth 2.0.7 -- A Textile Humane Web Text Generator — why the lucky stiff <ruby-talk@...>

more and more, you've seen it all before (i swear it's slowin down):

12 messages 2004/04/22
[#98008] Re: [ANN] RedCloth 2.0.7 -- A Textile Humane Web Text Generator — David Heinemeier Hansson <david@...> 2004/04/22

> I've wanted this feature to work right for awhile. It took a rewrite

[#98101] Extract all occurences from a text — Michael Weller <michael@...>

Hi!

11 messages 2004/04/23

[#98135] Problem assigning an Array object to an Array-subclass object — "Richard Lionheart" <NoOne@...>

[ I apologize if this is a second post. My earlier one seems to have gotten

29 messages 2004/04/23

[#98177] Psyco — Jim Moy <web@...>

Interesting stuff for Python, is any work like this being done in Ruby?

14 messages 2004/04/23

[#98181] Playing with sockets... — Hal Fulton <hal9000@...>

I'm writing a little expect-like piece of code and trying to test it

16 messages 2004/04/23

[#98281] String#unpack and null-terminated strings — Michael Neumann <mneumann@...>

Hi,

15 messages 2004/04/24

[#98362] How's ruby compare to it older brother python — "Hunn E. Balsiche" <hunnebal@...>

in term of its OO features, syntax consistencies, ease of use, and their

51 messages 2004/04/26
[#98597] Re: How's ruby compare to it older brother python — "Ken Hilton" <kenosis@...> 2004/04/28

Amen, brother.

[#98778] Re: How's ruby compare to it older brother python — "trevor andrade" <trevor.andrade@...> 2004/04/30

I agree that flaming the question is not appropriate and its also bad for

[#98789] Re: How's ruby compare to it older brother python — Dan Doel <djd15@...> 2004/04/30

I'm not saying the topic isn't appropriate. I don't mind the topic, and I find

[#98365] Introducing myself - Sascha Ebach — Sascha Ebach <se@...>

Hello dear Rubyists,

51 messages 2004/04/26
[#98568] Re: Introducing myself - Sascha Ebach — "Josef 'Jupp' Schugt" <jupp@...> 2004/04/27

Hello from Beethoven's home town,

[#98569] Re: Introducing myself - Sascha Ebach — Sascha Ebach <se@...> 2004/04/27

Hello Jupp

[#98570] Re: Introducing myself - Sascha Ebach — Simon Strandgaard <neoneye@...> 2004/04/27

Sascha Ebach <se@hexatex.de> wrote:

[#98753] Re: Introducing myself - Sascha Ebach — "Josef 'Jupp' Schugt" <jupp@...> 2004/04/29

Simon Strandgaard wrote:

[#98762] Re: Introducing myself - Sascha Ebach — Simon Strandgaard <neoneye@...> 2004/04/29

Josef 'Jupp' Schugt <jupp@gmx.de> wrote:

[#98817] Opportunities and pitfalls; was "Introducing myself - Sascha Ebach" — Mike Calder <ceo@...> 2004/04/30

A word of warning from a potential friend. Please don't take this negatively,

[#98847] Re: Opportunities and pitfalls; was "Introducing myself - Sascha Ebach" — Mark Hubbart <discord@...> 2004/04/30

[#98854] Re: Opportunities and pitfalls; was "Introducing myself - Sascha Ebach" — "Curt Hibbs" <curt@...> 2004/04/30

Mark Hubbart wrote:

[#98858] Re: Opportunities and pitfalls; was "Introducing myself - Sascha Ebach" — Mark Hubbart <discord@...> 2004/04/30

[#98366] How do I scale large Ruby web applications? — Sascha Ebach <se@...>

Hi there,

19 messages 2004/04/26

[#98409] Semantics of Multiple Values — Kristof Bastiaensen <kristof@...>

Hi liszt,

25 messages 2004/04/26

[#98435] Approaches to localization? — ptkwt@... (Phil Tomson)

I'm developing a GUI app using Ruby and FLTK. One of the requirements

22 messages 2004/04/26

[#98714] Ruby under Suse Linux — Mike Calder <ceo@...>

Hi.

14 messages 2004/04/29

[#98750] coerce(), what protocol to implement it — Jean-Hugues ROBERT <jean_hugues_robert@...>

Hi,

14 messages 2004/04/29

[#98758] File.expand_path(__FILE__) — John Platte <john.platte@...>

I'm having a problem with File.expand_path(__FILE__) after a chdir.

30 messages 2004/04/29

[#98796] SciTE Ruby Lexer — Kaspar Schiess <eule@...>

-----BEGIN PGP SIGNED MESSAGE-----

13 messages 2004/04/30

[#98820] ruby CVS can't use shared libs on NetBSD — Dick Davies <rasputnik@...>

11 messages 2004/04/30

[#98832] def [](v) xx; return yy; end # returned value is ignored !? — Jean-Hugues ROBERT <jean_hugues_robert@...>

Hi,

27 messages 2004/04/30

[#98851] Lazy evaluation — Michael Neumann <mneumann@...>

Hi,

51 messages 2004/04/30
[#98871] Re: Lazy evaluation — Jean-Hugues ROBERT <jean_hugues_robert@...> 2004/04/30

At 03:45 01/05/2004 +0900, you wrote:

[#98875] Re: Lazy evaluation — Florian Gross <flgr@...> 2004/04/30

Jean-Hugues ROBERT wrote:

[#98896] Re: Lazy evaluation (evil) — ptkwt@... (Phil Tomson) 2004/05/01

In article <c6uh31$gaq2i$1@ID-7468.news.uni-berlin.de>,

[#98913] Re: Lazy evaluation (evil) — Florian Gross <flgr@...> 2004/05/01

Phil Tomson wrote:

[#98917] Re: Lazy evaluation (evil) — ts <decoux@...> 2004/05/01

>>>>> "F" == Florian Gross <flgr@ccan.de> writes:

[#98919] Re: Lazy evaluation (evil) — Florian Gross <flgr@...> 2004/05/01

ts wrote:

Re: Semantics of Multiple Values (updated RCR)

From: "Its Me" <itsme213@...>
Date: 2004-04-27 18:29:05 UTC
List: ruby-talk #98554
"Yukihiro Matsumoto" <matz@ruby-lang.org> wrote
> |Semantics of Multiple Values
>
> Interesting, but I'm afraid this one is too big (and too complex)
> change.  I'd like to see a language with this semantics though.

Unification as 2-way pattern matching seems belong outside this proposal.

But perhaps Multiple Values could fit consistently into a more general
framework of one-way pattern matching for variable binding/assignment. It
seems duck typing also fits into such a pattern matching framework.

Here are a couple of usages examples of pattern match:

## call with multiple values
    def f (x, y, z=nil); end
    def g; return 1, 2; end
    x,y = g() #x=1, y=2
    f(g()) # same as f(1,2,nil)
    f(g(),3) # same as f(1,2,3)

## conditional multiple assignment
    {p: x, q: y} ||= h
    # match x if x is not nil, then x = h[:p]
    # match y if y is not nil, then y = h[:q]
    # overall match never fails


## method call with pattern-matched arg list
  class C
    def f ( a1, [a2,a3], {p: a4}, a5.(q: a6, r: a7, s:) ) ; end
  end
  C.new.f (w, x, y, z)
  # object match for method 'f'
    # multiple values match formal args with w,x,y,z
        # assuming w,x,y,z are not 'multiple values'
        # variable match: a1 = w
        # array match: a2, a3 = x[0],x[1]
        # hash match: a4 = y[:p]
        # object match: respond_to? :q,:r,:s
        #   a5 = a; a6 = z.q;  a7 = z.r

Why?
  - Makes intent and usage clear
  - Duck typing: patterns are duck types .(q:, r:, s:)

So here is my attempt at defining Pattern:

Pattern
-------
A Pattern is a syntactic form containing 0 or more variables. It can be used
in place of a variable, anywhere that variable could be initialized or
assigned (lvalue?). The Pattern is matched at run-time (to the extent
necessary), resulting in either
- successful match + variable bindings + success action
- failed pattern match + failure action

A pattern, to match and bind, walks the target structure:
- by position (array, positional args, multiple values, regex)
- by name (hash, keyword args, object accessors/methods)
- by magic (&block args, others?)

Kinds of patterns + pattern match:

- variable + variable assigment
    x                x = a
    # always matches

- multiple values + multiple assigment
    x,y,z
                    x,y,z = a, b, c
                    x,y,z = a, b # a, b, nil
                    x,y,z = a # a, nil, nil
    def f(p); return 1, 2; end
                    x,y,z = f(a) #1,2,nil
                    x,y,z = f(a), b #1,2,b
    *v = 1,2
                    x,y,z = *v, 3 #1, 2, 3
    # multiple values is a separate type
    # "," extends multiple values, like Array#concat
    # rhs treated as multiple values (extended by nil's)
    # always matches on assigment
    # can fail on method invocation (below)

- *arg vary-ary variable
    *x                *x = 1, 2, 3
    # always matches, assigns an array
    # *x also converts back to multiple values

- method definition + method invocation
    X#f(p,q)           x.f(a,b)
    X#f(p: 10)        x.f(p: a)
    X#f(p, *q)       x.f(a, b, c)
    # fail with exception if parameter mismatch
    # else invoke method

- hash pattern + pattern assigment
    {p: x}          {p: x} = a
    # fail with exception if !a.is_a? Hash
    # else x = a[:p]

- array pattern + pattern assigment
    [x, y, z]       [x, y, z] = [a, b, c]
    # fail if !rhs.is_a? Array
    # else assign rhs[0], rhs[1], ...
    # like hash pattern: [1: x, 2: y, 3: z]

- object pattern + pattern assigment
    .(p: x, q: y)
    .(p: x, q: y) = a
    # match if a.respond_to? :p, :q
    # assign x = a.p;  y = a.q
    o.(p:x, q:y) = a
    # assign o = a; x=a.p; y=a.q
    o.(p,q)
    # match if a.respond_to? :p, :q
    # perhaps extend to match specified classes?

- conditional assigment
    x ||= value
    # match if x was nil; no-op if fail
    # Think of pattern: x||

- regex pattern + regex match
    /abc*/            /abc*/ =~ string
    # perhaps '=' would work for '=~'

Any checking required to bind variables must take place at (by) run-time.
Some checks that define pattern match, but do not bind variables, may be
optionally turned off.

The behavior of success and failure depend on the context in which the
pattern appears. e.g. method invocation x.f(a) will pattern match x's
pattern of methods x.<m> with "x.f(a)". Success means the method is invoked.
Failure means either undefined method, or wrong number of arguments, or
missing 2.0 keyword argument, or missing block, etc. depending on what
failed.

Thoughts? Full of holes?

(btw: Beta pushed this whole pattern/match about as far as it can go).



In This Thread