[#1215] Tk widget demo; English Tk docs?; Java 1.2 Swing — "Conrad Schneiker" <schneiker@...>
Hi,
[#1218] Trivial FAQ bug — Dave Thomas <Dave@...>
[#1229] A vote for old behavior — Dave Thomas <Dave@...>
[#1232] Any FAQ requests, updates, ... — Dave Thomas <Dave@...>
[#1233] Singleton classes — Dave Thomas <Dave@...>
[#1263] Draft of the updated Ruby FAQ — Dave Thomas <Dave@...>
[#1307] Ruby/GTK 0.23 released — Hiroshi IGARASHI <igarashi@...>
Hi all,
From: Hiroshi IGARASHI <igarashi@ueda.info.waseda.ac.jp>
From: "Conrad Schneiker" <schneiker@jump.net>
On Fri, Feb 18, 2000 at 09:37:27PM -0500, Yasushi Shoji wrote:
[#1322] FAQ: Ruby acronyms — "Conrad Schneiker" <schneiker@...>
In the spirit of TABWTDI (there are better ways to do it), I'd like to
[#1341] Vim syntax file — Mirko Nasato <mirko.nasato@...>
Hi,
On Mon, Feb 14, 2000 at 05:44:39PM +0100, Mirko Nasato wrote:
[#1354] Say hi (bis) — Pixel <pixel_@...>
hi all,
[#1355] nice sample for functional stuff — Pixel <pixel_@...>
what about having map in standard (and map_index too)?
[#1373] Ruby Language Reference Manual--Glossary — "Conrad Schneiker" <schneiker@...>
I was going to print the Ruby Language Reference Manual when I noticed that
[#1376] Re: Scripting versus programming — Andrew Hunt <andy@...>
Conrad writes:
[#1379] Re: Yield — Andrew Hunt <andy@...>
>From: "Conrad Schneiker" <schneiker@jump.net>
[#1384] Re: Say Hi — mengx@...
My suggestion was to try to find a more comfortable method name (to me, and
[#1392] Re: Some Questions - Parameterised Types / Invariants — Andrew Hunt <andy@...>
>1. Parameterised Types / Template Classes
[#1398] Bignum aset — Andrew Hunt <Andy@...>
[#1488] Discussion happens on news.groups — Clemens Hintze <c.hintze@...>
Hi,
[#1508] Ruby/GTK and the mainloop — Ian Main <imain@...>
Hello Ian,
On Wed, Feb 23, 2000 at 02:56:10AM -0500, Yasushi Shoji wrote:
[#1516] Ruby: PLEASE use comp.lang.misc for all Ruby programming/technical questions/discussions!!!! — "Conrad Schneiker" <schneiker@...>
((FYI: This was sent to the Ruby mail list.))
From: "Conrad Schneiker" <schneiker@jump.net>
[#1528] ruby <=> python — Quinn Dunkan <quinn@...>
Hello! I'm new to ruby-talk, and mostly new to ruby. I'm making a document
[#1551] Ruby thread scheduling buglet — Ian Main <imain@...>
[#1569] Re: Ruby: constructors, new and initialise — Yukihiro Matsumoto <matz@...>
The following message is a courtesy copy of an article
[#1591] Certain char's not recognized by "." in regex? — Wes Nakamura <wknaka@...>
[#1592] Race condition in Singleton — Dave Thomas <Dave@...>
[ruby-talk:01381] Re: Say hi (bis)
matz@netlab.co.jp (Yukihiro Matsumoto) writes:
> Hi,
>
> In message "[ruby-talk:01354] Say hi (bis)"
> on 00/02/15, Pixel <pixel_@mandrakesoft.com> writes:
> |--------------------------------------------------------------------------------
> |bad points:
> |
> |--- `||' and `or' treats everything except nil as true
> | (so 0 || 4 == 0, "" || 4 == "")
oups, sorry it should have been (0 || 4) == 0, ("" || 4) == ""
>
> It is for performance and unambiguity reason.
i still find it a moot point:
- either 0 || 4 should return true (C behaviour)
- either 0 || 4 should return 4 (python and perl behaviour)
- either 0 || 4 should raise "undefined method `||' for 0''
(by the way can you redefine `||' ?) (ML family behaviour)
tis the first time i see 0 || 4 returning 0, kinda weird.
of course the python/perl behaviour is maybe less orthogonal as special cases
are: [] || 1, "" || 1 and 0 || 1 all gives 1
>
> |-- ?? no explicit references (aka \ in perl)
>
> In Ruby, everything can be considered as reference, so that \ is
> needless.
yeah, i had in mind things like my $p = \{}; my $pp = \$p; which i did use in
perl, but seems like it can be achieved nicely with an object, forget it :)
one other thing for which it was missing is giving a function instead of a
block, eg:
def f(n); [n,n+1]; end;
#l.collect \f
#instead of
l.collect{|n| f(n)}
>
> |-- ?? no slices
>
> a=[1,2,3]
> a[1,2] #=> [2,3]
> a[1..2] #=> [1,2]
i was thinking of hashes, some kind of extension to method fetch:
class Hash
def [](*l)
l = l[0] if l.length == 1 && l[0].type == Range
l.collect{|i| fetch(i) }
end
end
also interesting would []=
>
> |-- ?? can't redefine a function
>
> you can.
i was thinking of:
# !warning! perl here
sub f { 1 }
{
my $f = \&f;
*f = sub { print "f called\n"; goto &f };
}
maybe subclassing should be the solution?
>
> |-- no overloading (can be achieved manually, but hell...)
>
> I don't know what you mean. By argument number and types?
by argument number, only simple sugar needed. But i guess it would interfere
with method redefinition, and wouldn't be easy to add...
by type could be cool, but needs more sugar...
def f(Fixnum i); i + 1; end
def f(String i); i + "\n"; end
[...]
> |- no flattening of lists (eg: trans(x1,y1,x2,y2) can't be called with trans(*f,*g))
> | (only for blocks)
>
> trans(*(f+g)) will work.
cool :)
[...]
> |- you have to initialize before doing +=
>
> Explicit initialization is good thing I think.
>
> You can avoid them by
>
> def <<nil
> def +(other)
> other
> end
> end
i don't quite understand that one :(
i'd have said:
class NilClass
def +(other); other; end
end
but anyway, that's only += that i would like to work, not full `+'
>
> |- no named arguments (see python)
>
> It's planned.
:)
[...]
> |- each_index useless? why not passing both the index and the value (usefull!)
>
> use each_with_index. In general, methods described under Enumerable
> are useful, e.g. collect, select, each_with_index.
ok, still missing collect_with_index :)