Activity
From 02/28/2017 to 03/06/2017
03/06/2017
-
10:32 PM Bug #13287 (Closed): Stack consistency error (sp: 97, bp: 96)
- This works:
~~~ ruby
/ (?<foo>.*)/ =~ "bar" && true
~~~
This produces a stack consistency error with `2.4.0` but not with `2.3.3`:
~~~ ruby
/ (?<foo>.*)/ =~ "bar" &&
true
~~~
```
$ irb
2.4.0 :001 > / (?<foo>.*)/ =~ "ba... -
09:44 PM Feature #12573: Introduce a straightforward way to discover whether a process is running
- I'm unable to reopen this ticket, so I'd appreciate it if someone would do so for me. My apologies for not responding sooner, I must not have seen the previous message somehow. I came back to see what the status was after reimplementing ...
-
07:44 PM Feature #13263: Add companion integer nth-root method to recent Integer#isqrt
- A further simplification can be done for ``numb = root**n``
root = bit_mask = 1 << (b = (num.bit_length - 1)/n)
numb = root**n
numb = root << b*(n-1)
numb = (1 << b) << b*(n-1)
numb = 1 << (b + b*(n-1))
... -
- More efficient.
```
root = bit_mask = (b = 1 << (num.bit_length - 1)/n)
numb = root << b*(n-1) # fast cpu level root**n
until ((bit_mask >>= 1) == 0) || numb == num
root |= bit_mask
... - 07:42 PM Revision 1ec51631 (git): * 2017-03-07
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57791 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
07:42 PM Revision 8188fb9a (git): io.c: [DOC] open mode can be an integer; fix rdoc syntax
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57790 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
03:54 PM Bug #13270: IRB hangs when printing "\e]"
- I can reproduce `print "\e]"` and Ctrl+C, nothing happens. But I can't reproduce using `puts "\e["`. And I can't reproduce Ctrl+D `prints "30m"`. Ctrl+D causes simply exited.
I think iTerm eats output from "\e]" (OSC) to "\a" (BEL) (o... -
03:43 PM Misc #13283: Disable `&' interpreted as argument prefix warning when passing symbol to Enumerable#map
- Dan Allen wrote:
> I omit using a slightly different style in a chain.
> ...
This looks like Lisp to me more than Ruby.
But is the warning any useful in practice?
I agree the first example is unambiguous. -
03:54 AM Misc #13283: Disable `&' interpreted as argument prefix warning when passing symbol to Enumerable#map
- I omit using a slightly different style in a chain.
~~~
(q.map &:w).e.r.t.y
~~~
Effectively, I group the standalone statement inside of (outer) parens.
For me, this provides the visual consistency of omitted parens for when it... -
03:13 AM Misc #13283: Disable `&' interpreted as argument prefix warning when passing symbol to Enumerable#map
- I'm also a member of the church of dropping parens if possible. I'm not against introducing some new syntax to pass symbol-as-a-proc without parens. But wonder if that works for this specific map case. The main concern is that map ret...
-
02:11 PM Bug #13277: Ruby 2.4.0 gem server 2.6.8
- The related fix in the rubygems-2.6.9 changelog is https://github.com/rubygems/rubygems/blob/a2920e27661a18f60ad76b17b13961dd07f034d3/History.txt#L19
Please backport rubygems-2.6.10 to the next stable ruby-2.4 release. -
11:30 AM Bug #13286 (Closed): Segfault when using rb_debug_inspector_open / rb_debug_inspector_frame_binding_get with Fiddle, but not when directly from C extension
- ## System
```shell
$ uname -a
Linux clients 4.1.5-x86_64-linode61 #7 SMP Mon Aug 24 13:46:31 EDT 2015 x86_64 x86_64 x86_64 GNU/Linux
$ ruby -v
ruby 2.4.0p0 (2016-12-24 revision 57164) [x86_64-linux]
```
## Program
### W... -
- Looking at the source code, FIXABLE tends to be just before LOING2FIX
to check applicability of that operation. Why not try computing first
then check for overflow, which should be optimial.
I also tried the same thing for unsigned typ... -
09:10 AM Revision cf2a6866 (git): Avoid to warning of clang with -Wincompatible-pointer-types.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57788 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
08:10 AM Revision f96c171e (git): readline.c: fix waiting fucntion
- * ext/readline/readline.c (getc_body): use
rb_w32_wait_events_blocking in block region.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57787 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- I was not aware of cases when LONG_LONG_MAX is undefined. Sorry.
Signed-off-by: Urabe, Shyouhei <shyouhei@ruby-lang.org>
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57786 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * vm_insnhelper.c (opt_eq_func): method to dispatch is resolved by
only the receiver's class, not including the argument class.
even if basic operation is redefined, other class conditions
never meet. optimize Float and non-Float ... -
- We already check for __builtin_mul_overflow in configure but never
actually referred it before. Why not call it if available, because
that should render supposedly-optimial assembly outputs.
Optionally if __builtin_mul_overflow_p is av... -
- Vipala Patil wrote:
> ```
> ...
You are mixing ruby 2.3.1 (rvm) and 2.0.0 (system). They won't work together. -
05:35 AM Bug #13285 (Rejected): You may have encountered a bug in the Ruby interpreter or extension libraries.
- /Users/gargee/.rvm/gems/ruby-2.3.1@poaster/gems/eventmachine-1.2.3/lib/rubyeventmachine.bundle: [BUG] Segmentation fault
ruby 2.0.0p648 (2015-12-16 revision 53162) [universal.x86_64-darwin15]
-- Crash Report log information ---------...
03/05/2017
-
11:56 PM Bug #13284 (Closed): IA64 ruby 2.4 miniruby segfault
- Hello, I faced with this degradation (2.0.0 work perfect)
[ruby-2.4.0]$ gdb ./miniruby
GNU gdb Red Hat Linux (6.3.0.0-1.143.el4rh)
Copyright 2004 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Publi... -
- If the warning can't be removed, can we find some other way to write this statement without the need for parentheses?
-
- A common idiom in Ruby is to pass a symbol reference to `Enumerable#map`, which in turn invokes the corresponding method on each entry.
Case in point:
~~~
%(a b c).map &:upcase
~~~
Yet, when warnings are enabled, this line pro... -
- * compile.c (iseq_compile_each): fix variable type, use VALUE for
Symbol.
http://svn.ruby-lang.org/cgi-bin/viewvc.cgi?revision=57772&view=revision
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57783 b2dd03c8-39d4-4d8f-98ff-823fe... -
- An optimization for the initiall ``root**n`` can be as follows:
Given any number **num** with only one bit set, and thus: ``bits = num.bit_length``
then it's exponentiation to any **n** is just: ``num**n => num << (num.bit_length... - 07:43 PM Revision e926245b (git): * 2017-03-06
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57782 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * io.c: [DOC] improve docs for IO
* IO.{write,read}: fix errors (:open_args is not an array of
strings, it might include a perm or options hash argument;
IO.write has no length argument, drop corresponding statement),
impro... -
- Dario Daic wrote:
> * Add "and is a string" where usage of $' and $\ is explained.
> ...
Applied without this restraint, since it doesn't make sense: these variables must reference a string (or nil), since the type is validated during ... -
- Applied in changeset r57780.
----------
docs for IO#print
* io.c: [DOC] split documentation for IO#print into smaller paragraphs,
delete duplicate sentence, fix call-seq.
Based on a patch by Dario Daic. [ruby-core:78291] [Bug #12975] -
- * io.c: [DOC] split documentation for IO#print into smaller paragraphs,
delete duplicate sentence, fix call-seq.
Based on a patch by Dario Daic. [ruby-core:78291] [Bug #12975]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57780 ... -
12:29 PM Bug #13277: Ruby 2.4.0 gem server 2.6.8
- I guess when a new ruby main release will happen, the distributed gem in the bundle will also be
updated. The last release happened at around xmas 2016 so I guess we may get a new ruby release
in a not too terribly long time (currently... -
08:12 AM Bug #13282: opt_str_freeze does not always dedupe
- Eric Wong <normalperson@yhbt.net> wrote:
> Now, I wonder what's causing this to interact badly with
> test-unit...
power_assert/enable_tracepoint_events - this disables all
sorts of MRI optimizations for testing.
For background... -
- Eric Wong <normalperson@yhbt.net> wrote:
> Now, I wonder what's causing this to interact badly with
> test-unit...
Same test with minitest (5.10.1) + ruby 2.5.0dev r57744 works fine:
require 'minitest/autorun'
class TestString ... -
- normalperson@yhbt.net wrote:
> Testing with some old installs I had around, 2.1.9p490 seems fine,
Too Bad I'm Wrong.
It's seems to be only tickled by the gemification of test-unit
in 2.2. Using the test-unit gem with 2.1 still ... -
- Attached is the same test extracted from test/ruby/test_optimization.rb
but it fails in a standalone context. Not sure why...
It's not a new breakage in trunk, either.
Testing with some old installs I had around, 2.1.9p490 seems ... -
- Applied in changeset r57779.
----------
math.c: more accuracy
* math.c (math_log10): calculate log10(2) for more accuracy.
[ruby-core:79907] [Bug #13279] -
- * math.c (math_log10): calculate log10(2) for more accuracy.
[ruby-core:79907] [Bug #13279]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57779 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
01:07 AM Bug #13270: IRB hangs when printing "\e]"
- Nobuyoshi Nakada wrote:
> What terminal emulator are you using, the standard `Terminal.app`?
I'm using iTerm 3.0.14.
And on Terminal.app (v2.7.1 (387)) I get
~~~
> ...
il
>> puts "\e["
nil
~~~
(notice the first "n" mis... -
- Please request to the upstream.
03/04/2017
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57778 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * lib/unicode_normalize.rb: [DOC] prevent a comment from
showing up in the class documentation for String.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57777 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Applied in changeset r57775.
----------
nodoc OptParse
* lib/optparse.rb: [DOC] nodoc OptParse, introduced with r46126,
to avoid leaking of its documentation (OptionParser's docs) into
the class documentation of Object. [ruby-core... -
- Since r46126 [Feature #9864], where OptParse was added as an alias for OptionParser, the complete class documentation for OptionParser leaks into the class documentation for Object.
This affects docs since 2.2:
* https://docs.ruby-... - 07:52 PM Revision a8fca126 (git): * 2017-03-05
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57776 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * lib/optparse.rb: [DOC] nodoc OptParse, introduced with r46126,
to avoid leaking of its documentation (OptionParser's docs) into
the class documentation of Object. [ruby-core:79909] [Bug #13281]
git-svn-id: svn+ssh://ci.ruby-lang.... -
- It would be really helpful if people produce empirical results of actual coded
examples of techniques, to establish their efficacies.
**Math may suggest theoretical possibilities, but engineering determines their feasibility.**
I'... -
06:15 PM Bug #13280 (Rejected): net/ftp: Putbinaryfile (on Windows) requires blocksize equal to file size
- I searched this issue list (for Descriptions including `ftp`), but didn't find this problem discussed.
Recently, in the process of developing a new program, I invoked `putbinaryfile` (in passive mode) on a large file (about 1.3 megaby... -
05:18 PM Bug #13279 (Closed): Math.log10 accuracy regression
- I've notice a degradation in the precision of Math.log10 in Ruby 2.4 with respect to 2.1:
~~~
2.1.2 :136 > Math.log10 10**1000
=> 1000.0
2.4.0 :009 > Math.log10 10**1000
=> 999.9999999999999
~~~
The problem has been introd... -
04:30 PM Feature #13278 (Third Party's Issue): Rubygems Command Reference added to repo
- After working with Ruby documentation for while, there's only one file I'd really like to see added to the repo --
[RubyGems 2.6.10 Command Reference](https://msp-greg.github.io/ruby_trunk/rubygems/file.command-reference.html)
Coul... -
- * compile.c (ibf_load_code): raise an exception on invalid operand
size. should not use assert() to validate incoming data.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57774 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * compile.c (iseq_set_sequence): fix potential memory leaks on an
invalid instruction sequence.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57773 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * compile.c (validate_label): fix invalid label message.
registered labels are Symbols, not IDs.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57772 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Fixed with r51859 (Array) and r57770 (Enumerable).
-
- * array.c: [DOC] fix grammar in Array#sort, #sort!, #sort_by!,
move references below the code example, add a missing reference.
* enum.c: [DOC] fix grammar in Enumerable#sort, #sort_by.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trun... -
- * enum.c: [DOC] improve structure of docs for Enumerable#sort_by,
adopt explanation of the comparison block from Array#sort_by,
drop mention of 1.8, fix typos.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57770 b2dd03c8-39d4-4d... -
- * ext/readline/readline.c (getc_body): fix variable type, and
extract _get_osfhandle.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57769 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- People involved with [larskanis/rubyinstaller2](https://github.com/larskanis/rubyinstaller2), a new Windows build system for Ruby based on msys2 packages and mingw, found that `gem server` (v2.6.8) is not working properly in 2.4.0.
On...
03/03/2017
-
- The points Kevin makes are exactly some of the reason I think, **from a users perspective**,
its clearer to provide a separate name and API for this resource.
A normal user will have no knowledge of how the things works under the hoo... -
01:32 AM Feature #13166: Feature Request: Byte Arrays for Ruby 3
- I'm in favor of a separate byte type as well. I think it conveys intent much more clearly, is easier to reason about, is easier to optimize, and is less error-prone.
While an ASCII-8BIT string can do the work, it leads to two use case... -
- What terminal emulator are you using, the standard `Terminal.app`?
-
04:56 PM Bug #13270: IRB hangs when printing "\e]"
- Even more information, this is reproducible on ruby -v ruby 2.4.0p0 (2016-12-24 revision 57164) [x86_64-linux], but instead of printing out "30m" after CMD+D it prints out nothing. This seems to be a bug within IRB. Interesting, this beh...
-
- I can reproduce it on Darwin, so I'm switching it back to open.
My `ruby -v` is `ruby 2.4.0p0 (2016-12-24 revision 57164) [x86_64-darwin14]`
Also it occurs with `irb -f` but not with `ruby -e` -
- I can't reproduce it on darwin15.
Does it happen without irb, just `ruby -e print "\e]"`?
If only with irb, does it with `irb -f`? -
10:09 PM Feature #13263: Add companion integer nth-root method to recent Integer#isqrt
- If you get the wrong answer from Newton's, then you are doing it wrong. It may fail to converge, (which seems MOST unlikely in this case) but that is a different matter. But ultimately, there is 0 credibility to benchmarks that show BB...
-
-
- `puts` (rb_io_puts) uses `rb_default_rs` ("\n"), which isn't supposed to be customizable from within Ruby code, right? (Different from `$\` = `rb_output_rs`.)
So IMO this is a good improvement and I'll apply the patch. - 07:39 PM Revision 2e97d5bc (git): * 2017-03-04
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57768 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * io.c: [DOC] clarify that the 'record separator' between
arguments passed to 'puts' is always a newline.
Based on a patch by Mark Amery. [ruby-core:65801] [Misc #10403]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57767 b2dd03... -
06:32 PM Bug #13276 (Closed): Dir.glob returns empty array when OS has no more file handles (expected exception)
- The following terminal session demonstrates how `Dir.glob` returns an empty array when the OS has run out of file handles; whereas `File.new` raises a `Errno::EMFILE` exception.
I would expect `Dir.glob` to fail fast in a similar way ... -
- I agree it is easier to read.
In the other thread, Shyouhei Urabe said that nobody at the developer
meeting was against it (I assume this also includes matz).
So I suppose that this may be implemented eventually.
-
04:10 AM Feature #13272 (Closed): Keyword argument to instantiate a subclass of Struct
- When we define a subclass of Struct, it take arguments of values of fields when instantiated:
```ruby
ServiceAccount = Struct(:account_name, :full_name, :nickname, :name_invited_by, :twitter_account_name)
a = ServiceAccount.new("tag... -
- RDoc doesn't handle Document-method directives for `#[]` correctly. When multiple directives are given, "Document-method: []" shows up as a code block, at the beginning of the docs for the method of the directive before:
```
$ cat do... -
-
- 2.2.2 is outdate, does it happen with recent versions?
-
10:17 AM Bug #13274 (Closed): Ruby interpreter when using Puma.
- ```
=== puma startup: 2017-03-02 09:38:34 +0000 ===
/var/www/redacted_app_name/vendor/gems/ruby/2.2.0/gems/activesupport-5.0.0/lib/active_support/dependencies.rb:458: warning: already initialized constant Stats
/var/www/redacted_app_n... -
- Does it happen with recent versions?
-
- * ext/zlib/extconf.rb: fix building zlib for mingw, and for
cross-compiling.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57766 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Applied in changeset r57765.
----------
proc.c: documentation for Proc#{call,yield,[]}
* proc.c: [DOC] fix and improve docs for Proc#{call,yield,[]}:
* change order of Document-method directives as workaround for an
RDoc renderi... -
- The documentation for Proc#call displays one of its Document-method directives as a code block:
```
$ ri Proc.call
------------------------------------------------------------------------------
prc.call(params,...) -> obj
--... -
- * proc.c: [DOC] fix and improve docs for Proc#{call,yield,[]}:
* change order of Document-method directives as workaround for an
RDoc rendering problem where the documentation for Proc#call displays
a "Document-method: []" cod... -
- Benoit Daloze wrote:
> Shyouhei Urabe wrote:
> ...
True. There are two ways to tackle this problem:
1. add a new construct that does not capture variables outside of a given block.
2. just prohibit obtaining bindings; make lambdas... -
- * template/configure-ext.mk.tmpl (EXTMK_ARGS): add MINIRUBY for
cross-compling, which is used in extmk.rb.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57764 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * util.c: define MINGW_HAS_SECURE_API for qsort_s on recent
mingw64.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57763 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Applied in changeset r57762.
----------
broken mingw
* configure.in: check whether frexp and modf are broken.
* include/ruby/win32.h (frexp, modf): ignore bad declarations when
compiling as C++. [ruby-core:79859] [Bug #13267] -
- * configure.in: check whether frexp and modf are broken.
* include/ruby/win32.h (frexp, modf): ignore bad declarations when
compiling as C++. [ruby-core:79859] [Bug #13267]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57762 b2d...
03/02/2017
-
11:36 PM Revision 2f3ff53f (git): [DOC] update doc.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57761 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 11:33 PM Revision b7336a24 (git): * 2017-03-03
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57760 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- The "n" option for regexp, /.../n, is historical.
It doesn't mean the regexp works as binary match since Ruby 1.9.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57759 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
09:38 PM Bug #13271 (Closed): Clarifications on refinement spec
- Consider the following code:
~~~ruby
class Foo
def foo
"Foo#foo"
end
def bar
"Foo#bar"
end
end
class Bar < Foo
def foo
"Bar#foo -> "+super
end
def bar
"Bar#bar -> "+super
end
end
modu... -
- Is this darwin-specific? It appears to work fine on my linux system here.
ruby 2.4.0p0 (2016-12-24 revision 57164) [i686-linux] -
- Steps to reproduce:
- `irb`
- `print "\e]"`
- Or: `puts "\e["`
- try CMD+C, nothing happens
- try CMD+D, prints "30m"
Expected behavior:
- just prints "30m" (that's what `pry` does)
Ruby versions tried:
- ruby 2.3.3p22... -
- When running test-all on a mingw build, [`test_readline.rb`](https://github.com/ruby/ruby/blob/trunk/test/readline/test_readline.rb) generates failures and errors. Several of the errors are due to the `with_temp_stdio` method in `test_r...
-
03:08 PM Feature #13257: Symbol#singleton_class should be undef
- Is that a use case where you need a method for checking without allocating a singleton class?
As I read it diagonally, it looks like you are always accessing and using the singleton class when an object has one. -
- After chat with narse, I found that what I really want is the list of classes and modules on method search path of objects (nearly equal to `obj.singleton_class.ancestors`).
If I can get it without creating/allocating singleton classes,... -
- My motivation is to know whether there is an class/module in ancestors or not.
I needed it at this patch: https://github.com/msgpack/msgpack-ruby/pull/132/files
The feature implemented in this patch serialize/deserialize objects in a... -
- What's the motivation behind this? You cannot tell if an object has no singleton class just because it doesn't define or respond to #singleton_class anyway. BasicObject is an exception, and I could imagine certain kinds of objects (for...
-
- I understood that `#respond_to?` doesn't matter in fact.
What I really want is to undef `Symbol#singleton_class` (and Integer, Float too). -
09:27 AM Feature #13257 (Feedback): Symbol#singleton_class should be undef
- Fundamentally, `#respond_to?` is a method to check method existence. The class Symbol does have `#singleton_class` so that it's natural for it to return `true`, even though `#singleton_class` always fails.
It seems you want to check t... -
02:56 PM Bug #7976: TracePoint call is at call point, not call site
- I recently found this behaviour while attempting to build a gem that outputs the full call graph for our Rails application.
Using TracePoint, we can capture all calls/returns with the object the method is invoked upon, but finding th... -
-
- Closing, fixed with r51859.
-
- Closing, this should have been solved with #12332.
-
- To come back to the main topic of this issue,
I think having some syntax support for a non-capturing lambda
can be a useful feature in Ruby, but should not be for performance reasons.
It should be a design tool.
If a lambda/proc do... -
- Koichi Sasada wrote:
> I haven't read discussion details, but some optimizations eliminate
> ...
No, that's the important part, proper deoptimization *must* be able to restore
local variables. It can do so how it wants, but typically... -
03:08 AM Feature #12901: Anonymous functions without scope lookup overhead
- On 2017/03/01 5:47, eregontp@gmail.com wrote:
> It is possible with deoptimization though.
> An optimizer might assume that a block does not call binding and in case it does, deoptimize.
I haven't read discussion details, but some ... -
- Benoit Daloze wrote:
> > now, should what be visible from the obtained binding? Should we re-calculate the optimized-out variables?
> ...
This (recomputation) sounds practically very difficult to me, if not impossible. Much easier to... -
- Closing, because:
* in 2.1, 2.2, 2.3, the example code shows uses of both `#<<` and `#concat`,
* since 2.4, they are not aliases any more (`#concat` now accepts multiple arguments). -
- Applied in changeset r57758.
----------
string.c: restore documentation for String#<<
* string.c: [DOC] restore documentation for String#<<
which became undocumented with r56021; fix a typo.
[ruby-core:79865] [Bug #13268] -
- Since r56021 (Feature #12333), `String#<<` is not documented; the patch fixes this.
A backport should only be needed for 2.4. -
- * string.c: [DOC] restore documentation for String#<<
which became undocumented with r56021; fix a typo.
[ruby-core:79865] [Bug #13268]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57758 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
09:01 AM Bug #13267: ruby 2.4.0 i686-w64-mingw32 + boost compilation error
- When compiling the same program, with the same libraries for linux, this issue does not arise.
-
- When building https://github.com/Ancurio/mkxp for windows (via i686-w64-mingw32), the following build error occurs:
```
In file included from /usr/i686-w64-mingw32/include/ruby-2.4.0/ruby/defines.h:243:0,
from /usr/... -
- Jabari Zakiya wrote:
> An **Array8** is an array and a **String** a string.
I think I understand what you mean but in Ruby, classes tend not be split when they share same backends. For instance Array class can also be used as stacks... -
- > Additionally, an empty array is returned when the starting index for
an element range is at the end of the array. -
04:35 AM Bug #13266 (Rejected): Slicing in array producing wrong results
- ```
even_nums = [2,4,6,8,10]
2.3.1 :007 > even_nums[5,0]
=> []
# it must return nil as index is out of bound when slicing
# it may be due to the code written in ruby.c
=begin
if (argc == 2) {
if (SYMBOL_P(argv[0])) {
... -
- This is possible in perl:
```
use strict;
use warnings;
use Data::Dumper qw(Dumper);
my %hash = qw();
$hash{q}{w}{e}{r}{t} = 'y';
warn(Dumper(\%hash));
```
From perl's POV it seems ruby lacks this feature. -
- * test/lib/envutil.rb: rbconfig/sizeof may not be available when
cross-compiling. use RbConfig::Limits instead of calculating
limits of Fixnum with hard-coded CHAR_BIT.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57757 b2dd03... -
- * common.mk (test-rubyspec): add dependency for cross-compiling.
* defs/gmake.mk: ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57756 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- How about we allow some inter-thread TLS operations, but not others?
Maybe:
Allowed:
+ reading values
+ updating value of existing keys (maybe)
Disallowed:
- setting value to a new key
In other words, the Allowed stuf... -
01:14 AM Bug #12834: `prepend` getting prepended even if it already exists in the ancestors chain
- I feel like this is a bug, and should be fixed. Prepending the same thing multiple times should be a warning IMHO. I'm not even sure what situation this is useful.
```
2.4.0 :001 > module M
2.4.0 :002?> end
=> nil
2.4.0 :003 >...
03/01/2017
-
- Actually, this **bbm** version is generally a smidgen faster than the original, especially for perfect roots.
```
class Integer
def irootn3(n) # binary bit method (bbm) for nth root
return nil if self < 0 && n.even?
ra... -
- Further testing shows Newton's method is sensitive to its implementation as you take larger roots.
Shown below are test results that show the **irootn1** Newton's implementation starts to give incorrect (smaller)
results past a certa... -
- Jabari Zakiya wrote:
> This will enhance Ruby's use even more in fields like number theory, advanced math, cryptography,
> ...
I'm not immediately against this but can I ask you when is this method useful? Because being amateur, I don... -
- Julien BONNET wrote:
> Between version 2.0.0 and 2.3.1, documentation has shrunk a lot.
Could you please be more specific about which parts are missing?
I can not see big differences:
* http://ruby-doc.org/stdlib-2.0.0/libdoc/soc... -
- * struct.c: [DOC] improve examples for ::new, improve #dig example,
simplify #select example, use consistent style for return values,
fix typos and example code style, remove duplicate cross reference.
git-svn-id: svn+ssh://ci.ruby-... - 07:57 PM Revision 9386e511 (git): * 2017-03-02
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57754 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57753 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
02:41 PM Feature #13265 (Rejected): TracePoint for basic operation redefinition
-
Add TracePoint for basic operation redefinition.
This infrequent event can have an outsized performance impact by eliminating a lot of the optimization that exists in the Ruby interpreter.
In order to know when these sor... -
02:00 PM Bug #13264 (Closed): Binding#irb does not work in context of frozen object
- This works as expected (starts IRB):
```
$ ruby -r irb -e '"".instance_eval{ binding.irb }'
```
This throws an error:
```
$ ruby -r irb -e '"".freeze.instance_eval{ binding.irb }'
```
```
/home/jan/.rvm/rubies/ruby-2.4.0... -
11:39 AM Bug #12774: TkDND cannot handle text/plain, text/uri-list type well
- There is no Issue tracker in https://github.com/ruby/tk
Where should we report bugs related with tk? -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57752 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- Shyouhei Urabe wrote:
> Let me ask a (possibly stupid) question. Consider this scenario:
> ...
I assume you mean a binding via Kernel#binding.
I think that's the general contract of deoptimization: it must not be observable from the ... -
- Math Ieu wrote:
> Shyouhei Urabe wrote:
> ...
I (broadly) agree with you. Extending or modifying the language to make optimization possible is a great idea. Matz said in comment #9 that Koichi has one of such ideas. Let us see what ... -
- Benoit Daloze wrote:
> Shyouhei Urabe wrote:
> ...
Let me ask a (possibly stupid) question. Consider this scenario:
1. create a proc.
2. it gets optimized.
3. lots of calculations goes on without variable captures.
4. obtain a b... -
- * lib/fileutils.rb: [DOC] fix invalid example code to make it
syntax highlighted, fix rdoc for lists, nodoc internal methods,
avoid a dangerous example.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57751 b2dd03c8-39d4-4d8f-98ff... -
10:15 AM Revision 3acda398 (git): openssl: avoid segfault during running tests on Ubuntu trusty
- Import the commit 6693a549d673 ("test/test_pkey_ec: do not use dummy 0
order", 2017-02-03) from upstream. Hopefully this will fix the segfault
on RubyCI icc16-x64:
http://rubyci.org/logs/rubyci.s3.amazonaws.com/icc-x64/ruby-trunk/log/... -
-
- I used `/dev/urandom` in the example because `/dev/random` might block.
-
- * lib/fileutils.rb (compare_file): [DOC] clarify documentation
by avoiding confusing return value of "maybe false".
[ruby-core:75431] [Misc #12365] reported by Robert A. Heiler.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57749... -
- shyouhei@ruby-lang.org wrote:
> Nobuyoshi Nakada wrote:
> > It is not guaranteed that `ruby_xfree` can free a pointer allocated by other than `ruby_xmalloc` and so on.
> > You can't mix bare `malloc` and `ruby_xfree`.
>
> Agreed. ... -
- Nobuyoshi Nakada wrote:
> It is not guaranteed that `ruby_xfree` can free a pointer allocated by other than `ruby_xmalloc` and so on.
> ...
Agreed. I heard that DLLs can have their own malloc() implementation on Windows. Even on POSI... -
- * common.mk (clean-ext): separate clean-up of makefiles and
timestamps, and clean-up of subdirectories under ext and gems.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57748 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
07:39 AM Feature #13110: Byte-based operations for String
- Yukihiro Matsumoto wrote:
> Are byteindex and byteoffset good enough for your use-case?
I also want bytesplice, but its priority is lower.
> ...
byteoffsets may be better, but we should consider consistency with MatchData#offset.
-
- Are byteindex and byteoffset good enough for your use-case?
Should byteoffset be byteoffsets since it returns both edge?
Matz.
-
07:00 AM Feature #12515: Create "Boolean" superclass of TrueClass / FalseClass
- Shyouhei Urabe wrote:
> Radovan Smitala wrote:
> ...
Just deprecated warnings for Fixnum and Bignum :) And that is OK. Rails is biggest project/framework/tool in Ruby community. It takes some time to be fully compatible, not only for I... -
- Radovan Smitala wrote:
> There should be some circumstances where implemented Boolean class should be problematic. But many of existed gems just reopen class and life goes on. Rails 5 isn't till now compatible with Ruby 2.4 with introdu... -
- * ext/zlib/extconf.rb: clean zlib libraries generated in the
place.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57747 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 01:16 AM Revision c0be1c7b (git): * 2017-03-01
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57746 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 01:16 AM Revision d1a506ec (git): deduplicate strings sooner at load and compile
- We can use rb_fstring_cstr in some places to prevent an
intermediate object from being created before deduplication
via rb_fstring.
* compile.c (iseq_compile_each): use rb_fstring_cstr
(rb_insns_name_array): ditto
* load.c (rb_load_in...
02/28/2017
-
- Newton's method has quadratic convergence. This means that a properly implemented Newton's method will blow away any BBM if more than just a few bits are needed.
"Properly implemented" is a big deal, because, as I have found with som... -
- Following the heels of adding the method ``Integer#isqrt``, to create exact integer
squareroot values for arbitrary sized integers, based on the following threads:
https://bugs.ruby-lang.org/issues/13219
https://bugs.ruby-lang.org/i... -
10:37 PM Feature #12901: Anonymous functions without scope lookup overhead
- Shyouhei Urabe wrote:
> Math Ieu wrote:
> ...
But it could be enabled on-demand. Say with a magic comment, then all closures created from code of this file would be optimized to not capture everything. I guess there might as well be sp... -
- Shyouhei Urabe wrote:
> I think it's impossible to optimize lambdas without breaking backwards compatibility.
It is possible with deoptimization though.
An optimizer might assume that a block does not call binding and in case it doe... -
12:35 PM Feature #12901: Anonymous functions without scope lookup overhead
- I didn't think about (er, *know* about!) those ways of accessing the inner scope. That definitely rules out skipping the binding in those cases! Thanks for the info.
-
10:30 AM Feature #12901: Anonymous functions without scope lookup overhead
- Robert Mosolgo wrote:
> What about checking at the block level? I mean, could we check that:
> ...
In addition to what Shyouhei said, somebody could also redefine addition, like so:
~~~ ruby
alias :old_plus :'+'
def + (other)
b... -
- Robert Mosolgo wrote:
> What about checking at the block level? I mean, could we check that:
> ...
Sorry, no. You can look at outside scope from within the lambda's binding like this:
```
zsh % rbenv exec irb
irb(main):001:0> a =... -
- What about checking at the block level? I mean, could we check that:
- All local variable reads come from the block's local table
- There are no method calls on implicit receiver
So, this could be optimized:
~~~ ruby
lambda... -
- Math Ieu wrote:
> The introspection features like the "binding" method and maybe the way "eval" works might make this hard to optimize though (since they essentially capture everything?).
Yes. This is why we capture everything. And... -
- Fixed with r57539 (#11752).
-
12:09 PM Feature #13259: Kernel#Date
- Marcus Stollsteimer wrote:
> Note also the ongoing discussion about the future of Date (#13072).
Thanks for the notification. I actually do think there should be a similar thing with `Time`, and if `Date` is supposed to become obsole... -
- Note also the ongoing discussion about the future of Date (#13072).
-
- I have nothing against it per se but what happens to the Date namespace?
People may be confused when they see Date like the above that used to be part of the
require 'date' call, but now suddenly is part of Kernel.
On a side note ... -
- I often see a piece of code like this:
```ruby
require "date"
begin
Date.strptime(some_string, some_format)
rescue
nil
end
```
Since we now have (https://bugs.ruby-lang.org/issues/12732) `Kernel#Integer` with a parame... -
- I first stumbled upon it when doing `ri Integer` (before the proper documentation there appear snippets introduced by BigDecimal, OpenSSL (both fixed upstream now), and mathn), and then also noticed in the context of #13219 that `ri Math...
-
- Very interesting find. I would not have thought that this is a possibility.
May I ask how you did discover this? -
- The deprecated (#10169) mathn library leaks documentation into some core classes, or even replaces it (see example further down).
I would like to fix this, but I could need some ideas what would be the best way to do that. The problem... -
- > I think we have lots of occasions to receive a true or false value as a string input, and
> ...
I agree. :)
Not saying that I agree with the API, it feels a bit cumbersome; but I agree with the
comment above.
By the way, I thi... -
- I agree that the proposal was vague. Instead of using `strict`, perhaps we can pass a string or strings expressing the language/mode that we want to allow.
```ruby
Boolean("はい", allow: ["JA", "EN"])
Boolean("0", allow: ["NUMBER"])
... -
- The proposal is still vague, especially when `strict: false`. The criteria are totally culture dependent.
"yes" may be true for English speaking people, but what about "はい" (Japanese) or "Ja" (German)? The list would go on.
Until the... -
- Please note that this proposal is distinct from any proposal to introduce the `Boolean` class into Ruby. In think the latter does not make sense, and I am against it.
-
- I think we have lots of occasions to receive a `true` or `false` value as a string input, and want to convert it to `true` or `false`. Perhaps we can have a method `Kernel#Boolean` in a similar spirit to `Kernel#Integer` and its kins, wh...
-
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57744 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * Makefile.in (clean-ext): clean gem directories too.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57743 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Applied in changeset r57742.
----------
Added initial gemspec for SDBM module.
[Feature #13261]
* doc/*.rdoc: Move SDBM entry to default gem section.
* ext/sdbm/sdbm.gemspec: first gemspec. -
- This issue is tracking for gem release of SDBM module.
-
- [Feature #13261]
* doc/*.rdoc: Move SDBM entry to default gem section.
* ext/sdbm/sdbm.gemspec: first gemspec.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57742 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Applied in changeset r57741.
----------
disable critical-error-handler
* win32/win32.c (rb_w32_sysinit): disable critical-error-handler
message box even on mswin, regardless of runtime DLL version.
[ruby-dev:49988] [Bug #13254] -
- * win32/win32.c (rb_w32_sysinit): disable critical-error-handler
message box even on mswin, regardless of runtime DLL version.
[ruby-dev:49988] [Bug #13254]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57741 b2dd03c8-39d4-4d8f-... -
- * win32/Makefile.sub (config.h): define ULL_TO_DOUBLE for
conversion from unsigned __int64 to double, which is not
implemented in till Visual Studio.NET 2003, aka VC7.1.
* bignum.c (estimate_initial_sqrt): use ULL_TO_DOUBLE if defin... -
- * numeric.c (DEFINE_INT_SQRT): suppress a type-limits warning
where int is always smaller than double.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57739 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
-
- * Makefile.in (update-mspec, update-rubyspec): use CHDIR, which may
use -P, to resolve symbolic links.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57738 b2dd03c8-39d4-4d8f-98ff-823fe69b080e