[#17566] rubychecker - runs checks on a Ruby interpreter — Igal Koshevoy <igal@...>

I've put together a shell script that runs checks on a Ruby interpreter.

14 messages 2008/07/03

[#17615] [PATCH] ruby-mode.el: Fix here-doc strings with inner quotes — Nathan Weizenbaum <nex342@...>

At the moment, ruby-mode.el uses font-lock-keywords as opposed to

22 messages 2008/07/05
[#17657] Re: [PATCH] ruby-mode.el: Fix here-doc strings with inner quotes — Yukihiro Matsumoto <matz@...> 2008/07/08

[#17678] Re: [PATCH] ruby-mode.el: Fix here-doc strings with inner quotes — Nathan Weizenbaum <nex342@...> 2008/07/09

It was designed to fix the following case:

[#17755] Re: [PATCH] ruby-mode.el: Fix here-doc strings with inner quotes — Nathan Weizenbaum <nex342@...> 2008/07/13

Here's a third patch that fixes a bug in the second and uses a quicker

[#17772] Re: [PATCH] ruby-mode.el: Fix here-doc strings with inner quotes — Nathan Weizenbaum <nex342@...> 2008/07/15

One more patch which fixes a few bugs in the the last one.

[#17773] Re: [PATCH] ruby-mode.el: Fix here-doc strings with inner quotes — Nobuyoshi Nakada <nobu@...> 2008/07/15

Hi,

[#17776] Re: [PATCH] ruby-mode.el: Fix here-doc strings with inner quotes — Nathan Weizenbaum <nex342@...> 2008/07/15

Looks like version 22 doesn't support explicitly numbered regexp groups.

[#17779] Re: [PATCH] ruby-mode.el: Fix here-doc strings with inner quotes — Nobuyoshi Nakada <nobu@...> 2008/07/15

Hi,

[#17783] Re: [PATCH] ruby-mode.el: Fix here-doc strings with inner quotes — Nobuyoshi Nakada <nobu@...> 2008/07/15

Hi,

[#17788] Re: [PATCH] ruby-mode.el: Fix here-doc strings with inner quotes — Nathan Weizenbaum <nex342@...> 2008/07/15

Alright, here's a version that fixes both the highlighting bug and the

[#17793] Re: [PATCH] ruby-mode.el: Fix here-doc strings with inner quotes — Nobuyoshi Nakada <nobu@...> 2008/07/16

Hi,

[#17644] Features to be included in Ruby 1.9.1 — "Yugui (Yuki Sonoda)" <yugui@...>

Hi, all

27 messages 2008/07/08

[#17674] [Ruby 1.8 - Bug #238] (Open) Ruby doesn't respect the Windows read-only flag — Jim Deville <redmine@...>

Issue #238 has been reported by Jim Deville.

10 messages 2008/07/08

[#17708] [Ruby 1.8 - Bug #252] (Open) Array#sort doesn't respect overridden <=> — Ryan Davis <redmine@...>

Issue #252 has been reported by Ryan Davis.

13 messages 2008/07/09

[#17871] duping the NilClass — "Nasir Khan" <rubylearner@...>

While nil is an object, calling dup on it causes TypeError. This doesnt seem

33 messages 2008/07/20
[#17872] Re: duping the NilClass — Urabe Shyouhei <shyouhei@...> 2008/07/20

Nasir Khan wrote:

[#17873] Re: duping the NilClass — "Meinrad Recheis" <meinrad.recheis@...> 2008/07/20

On Sun, Jul 20, 2008 at 7:55 PM, Urabe Shyouhei <shyouhei@ruby-lang.org>

[#17877] Re: duping the NilClass — Urabe Shyouhei <shyouhei@...> 2008/07/20

Meinrad Recheis wrote:

[#17879] Re: duping the NilClass — Kurt Stephens <ks@...> 2008/07/20

Urabe Shyouhei wrote:

[#17880] Re: duping the NilClass — "Nasir Khan" <rubylearner@...> 2008/07/21

I write a lot of hand crafted dup or clone because I want control as well as

[#17881] Re: duping the NilClass — "David A. Black" <dblack@...> 2008/07/21

Hi --

[#17882] Re: duping the NilClass — Urabe Shyouhei <shyouhei@...> 2008/07/21

+1 to David. A convenient way to do Marshal idiom should be a new

[#17885] Re: duping the NilClass — "Robert Dober" <robert.dober@...> 2008/07/21

On Mon, Jul 21, 2008 at 8:21 AM, Urabe Shyouhei <shyouhei@ruby-lang.org> wrote:

[#17887] Re: duping the NilClass — "David A. Black" <dblack@...> 2008/07/21

Hi --

[#17889] Re: duping the NilClass — "Robert Dober" <robert.dober@...> 2008/07/21

On Mon, Jul 21, 2008 at 1:02 PM, David A. Black <dblack@rubypal.com> wrote:

[#17883] [Ruby 1.9 - Bug #340] (Open) 1.9/trunk does not work when compiled with llvm-gcc4 2.3 (gcc 4.2.1) — Ollivier Robert <redmine@...>

Issue #340 has been reported by Ollivier Robert.

14 messages 2008/07/21

[#17943] RUBY_ENGINE? — "Vladimir Sizikov" <vsizikov@...>

Hi,

56 messages 2008/07/24
[#17950] Re: RUBY_ENGINE? — Tanaka Akira <akr@...> 2008/07/25

In article <3454c9680807241200xf7cc766qb987905a3987bb78@mail.gmail.com>,

[#17958] Re: RUBY_ENGINE? — "Vladimir Sizikov" <vsizikov@...> 2008/07/25

Hi,

[#17981] Re: RUBY_ENGINE? — Tanaka Akira <akr@...> 2008/07/26

In article <3454c9680807250054i70db563duf44b42d92ba41bfb@mail.gmail.com>,

[ruby-core:17686] default compiler behaviour about trace insn.

From: SASADA Koichi <ko1@...>
Date: 2008-07-09 03:13:35 UTC
List: ruby-core #17686
Please give me comments.

A compiler of Ruby 1.9 compile ruby script, without "trace
instruction" by default.  This means that some "trace event" used by
set_trace_func() is not active (line, call, leave).  "trace
instruction" invoke some hook procs.

This is a reasone of this issue:
http://redmine.ruby-lang.org/issues/show/161

lib/profile.rb enable to add trace insn.  You need to require
lib/profile.rb *before* target programs are compiled.  require *in*
target program is not affect profiling because target programs are
already compiled.  If you run ruby with profiler specified in
command line, you have correct result:

> ./ruby -v -rprofile ../trunk/test.rb
> ruby 1.9.0 (2008-07-08 revision 0) [i686-linux]
>   %   cumulative   self              self     total
>  time   seconds   seconds    calls  ms/call  ms/call  name
>  67.44     0.29      0.29     2432     0.12     4.84  Object#ackerman
>  20.93     0.38      0.09     3676     0.02     0.02  Fixnum#==
>   6.98     0.41      0.03     1188     0.03     0.03  Fixnum#+
>   4.65     0.43      0.02     2431     0.01     0.01  Fixnum#-
>   0.00     0.43      0.00        1     0.00     0.00  Module#method_added
>   0.00     0.43      0.00        1     0.00     0.00  Kernel.require
>   0.00     0.43      0.00       57     0.00     0.00  Fixnum#>
>   0.00     0.43      0.00        1     0.00     0.00  RubyVM::FrozenCore.core_define_method
>   0.00     0.43      0.00        2     0.00     0.00  IO#set_encoding
>   0.00     0.43      0.00        1     0.00   430.00  #toplevel

The reason why I eliminate "trace instruction" is performance.  In
most cases, we don't need "trace instruction".  If there is no
hooks, "trace instruction" means "nop instruction", waste some
performance.

This is a benchmark result:
http://www.atdot.net/sp/readonly/30xp3k
ruby 1.9.0 (2008-07-09 revision 17965) [x86_64-linux]

"wo-trace" is current 1.9, compile without trace instruction.
another one "ruby 1.9.0 (2008-07-09 revision 17965) [x86_64-linux]"
is modified version which include trace instruction everytime.

"wo-trace" is fast as a few percent.


But there are some drawback:
- spec changing from 1.8
- difficult to describe especially for newbe
 (or 1.8 users except ruby-core readers?)
- dave should re-write his book :)

There are some technical solutions.  One is to insert one-byte trace
instruction ([ruby-core:17653]) dynamically.  But I think we can't
support it in high-quality by 1.9.1 release.


I consier this issue and current my solutions are:

- enable (include) trace insn on 1.9.1 everytime
- speed up with one-byte trace instruction on 1.9.x (x > 1)
- you can disable to compile with trace insn by require some
  library (ruby -r without-trace-insn ...) on 1.9.1
  in concrete term, modify
  RubyVM::InstructionSequence.compile_option.
- you can write about it on performance tips page :)


Any comments?
-- 
// SASADA Koichi at atdot dot net

In This Thread

Prev Next