[#2529] concerns about Proc,lambda,block — "David A. Black" <dblack@...>
Hi --
>>>>> "D" == David A Black <dblack@wobblini.net> writes:
Hi --
Hi,
On Tue, 2 Mar 2004 08:44:25 +0900, Yukihiro Matsumoto wrote:
Hi,
On Wednesday, 3 March 2004 at 8:00:09 +0900, Yukihiro Matsumoto wrote:
Hi,
Hi,
On Wed, Mar 03, 2004 at 07:51:10AM +0900, Yukihiro Matsumoto wrote:
Hi,
On Thu, 4 Mar 2004, Yukihiro Matsumoto wrote:
Hi,
[#2575] Comment football being played... with lib/test/unit.rb — Nathaniel Talbott <nathaniel@...>
[Resent because I accidentally signed it the first time]
[#2577] problem with Net::HTTP in 1.8.1 — Ian Macdonald <ian@...>
Hello,
Hi,
[#2582] One more proc question — Dave Thomas <dave@...>
Sorry about this... :)
Hi,
On Friday, 5 March 2004 at 12:52:15 +0900, Yukihiro Matsumoto wrote:
Hi,
[#2588] Duck typing chapter — Dave Thomas <dave@...>
I've posted a rough first pass at a chapter about duck typing (and
[#2606] Thought about class definitions — Dave Thomas <dave@...>
If we allowed
[#2628] YAML complaint while generating RDoc — Dave Thomas <dave@...>
With the latest CVS, I get
[#2640] patch to tempfile.rb to handle ENAMETOOLONG — Joel VanderWerf <vjoel@...>
[#2644] RDoc proporsal — "H.Yamamoto" <ocean@...2.ccsnet.ne.jp>
Hi, rubyists.
[#2646] Problems rdoc'ing cvs... — Hugh Sasse Staff Elec Eng <hgs@...>
I have just done
On Friday, March 12, 2004, 4:15:42 AM, Dave wrote:
On Fri, 12 Mar 2004, Dave Thomas wrote:
[#2661] Pathological slowdown in 1.8 — Ryan Davis <ryand@...>
Hi all,
[#2697] lib/ruby/1.9/yaml.rb:193: [BUG] Segmentation fault — Mauricio Fern疣dez <batsman.geo@...>
Mauricio Fern疣dez wrote:
On Sun, Mar 28, 2004 at 09:42:42AM +0900, why the lucky stiff wrote:
[#2703] Proposed patch to add SSL support to net/pop.rb — Daniel Hobe <daniel@...>
This patch adds support to Net::POP for doing POP over SSL. Modeled on how
This is v2 of the patch. Cleaned up a bit and added some more docs.
v3 of the patch:
Hi,
I agree that there are a lot of arguments to #start, but I think it is the
On Tue, 30 Mar 2004 16:24:17 +0900, Daniel Hobe wrote:
On Wed, 31 Mar 2004 13:27:31 +0900, Daniel Hobe wrote:
On Tue, Mar 30, 2004 at 04:05:06PM +0900, Minero Aoki wrote:
[#2709] typos in lib/singleton.rb — Ian Macdonald <ian@...>
Hello,
[#2713] more spelling and grammar fixes — Ian Macdonald <ian@...>
Hello,
> Hello,
Hi,
Re: Duck typing chapter
On 11/03/2004, at 3:10 AM, Elliott Hughes wrote:
I tend to take the view of horses for courses,
high level <----------------------------------------> low level
dynamic typing
static typing
generalisation more important efficiency more
important
"scripting" languages compile
to machine code
Ruby approaches this from the left, where as languages such as C/C++
and java approach this from the right.
I hadn't thought of the idea of directors vs enablers, but that's an
interesting thought on "do you trust your fellow programmers to do the
right thing." Or, is that people confusing NASA style mission critical
with business style mission critical?
To get back to the main thought behind this post. If your trying to
write generalised or high level abstract code then dynamic typing is
good as it helps, me at least, avoid worrying about what type I'm
dealing with and concentrate on the problem. However, when I'm writing
library routines I'd like to be able to do signature checking if not to
increase efficiency at least to avoid type checking. Especially when it
comes to writing modules to interface to libraries in other languages.
For example, I'm writing an interface to flow tools, which I'll post
something on at a later date, in which I quiet often check the incoming
VALUE for its type. This not only means that I'm duplicating a lot of
code but it can make the intention of the code less clear.
It would be useful to have at this level of ruby, if not also at a high
level, a version of rb_define_method that has the ability to do the
signature checking for me. For example,
the current open file method is registered, thus,
rb_define_method(vflow, "open", vf_open, 1)
Now I'm not sure how you'd do the next bit may be something like this,
rb_define_method_sig(vflow, "open", vf_open_str, T_STRING)
then if an a method that took an fixnum was defined,
rb_define_method_sig(vflow, "open", vf_open_fixnum, T_FIXNUM)
Overloading the vflow.open() method from within ruby, but making the
code clearer.
I seem to have headed off on a tangent, so I'll stop at this point
before I get any further off subject.
Jeff.