Activity
From 10/18/2016 to 10/24/2016
10/24/2016
-
09:44 PM Feature #12861: super in a block can be either lexically or dynamically scoped depending on how the block is invoked
- super is in sync with `__method__` because they are designed to be in sync, `__method__` called from a block typically returns the enclosing method of the block but not always, when the block is invoked as a method, `__method__` returns ...
-
09:03 PM Feature #12861: super in a block can be either lexically or dynamically scoped depending on how the block is invoked
- bug hit wrote:
> Jeremy Evans wrote:
> ...
When the block with super is invoked by Class2.call_stored_block, `foo` is the current method at that point, according to `__method__`. This is simple to see by changing Class1.foo to raise an... -
- Jeremy Evans wrote:
>
> ...
Except it's not.
```ruby
class Class1
def self.foo
'Class1::foo'
end
end
class Class2 < Class1
def self.store_block(&block)
@block = block
end
def self.foo
store_bloc... -
- bug hit wrote:
> Jeremy Evans wrote:
> ...
If method resolution is always dynamic, and super is directly related to method resolution, it would certainly be odd for super to always be lexical, right?
One could argue that super is cu... -
- Jeremy Evans wrote:
> method calls in blocks are sometimes lexical and sometimes dynamic, depending on how the block is invoked.
> ...
Method resolution is never lexical, it is always relative to the current, dynamic self. -
- bug hit wrote:
> it would have to fail the same way instance_eval of the block (with super) fails when self is wrong (self has wrong type to call super in this context).
to expand on that, super normally binds to the method (class + ... -
- Shyouhei Urabe wrote:
> bug hit wrote:
> ...
you probably meant to call the defined :foo?
```ruby
class Foo
def self.foo
'foo@foo'
end
end
class Bar < Foo
def self.bar(&block)
define_singleton_method :foo, &b... -
- I don't think this is a bug, the ability to change the scope of a block is just part of ruby, and arguably one of the things that makes ruby flexible and a joy to program in:
~~~ ruby
foo = proc do
bar
end
foo.call # main.bar ... -
03:59 AM Feature #12861: super in a block can be either lexically or dynamically scoped depending on how the block is invoked
- bug hit wrote:
> I think it would be better if super were always lexically scoped
Agreed but... What should then happen for inter-class block passing situation, like this?
```ruby
class Foo
def self.foo
'foo@foo'
end
... -
05:24 PM Feature #10208: Passing block to Enumerable#to_h
- So, we need a different name for this.
Here's some candidates I can think of:
- hash_by (proposed above)
- to_h_by
- hash_map
- map_h
- map_to_h
-
05:24 PM Bug #12832: Calling Object#method hangs for private method defined on module then made public once it's been used to extend class
- I'm not sure whether this is useful information but calling `#instance_method` on `Bar`'s metaclass also causes Ruby to hang. Using James's most recent code snippet, the following code seems to cause the same problem as `Bar.method(:foo)...
-
12:13 PM Feature #12867: Add ability to check validity of a URL
- Shlok Srivastava wrote:
> Shyouhei Urabe wrote:
> ...
`http:invalid-url` is invalid according to the scheme-specific definition in RFC7230, but is valid and parses according to RFC3986. Depending on one's viewpoint, there's either no w... -
11:54 AM Feature #12867: Add ability to check validity of a URL
- Shyouhei Urabe wrote:
> Doesn't URI.parse work for you?
As URI.parse actually check for valid URIs and not URLs
I tried something like this:
~~~ ruby
URI.parse(some_string).scheme == "http"
~~~
But it returns true for inva... -
- Doesn't URI.parse work for you?
-
10:47 AM Feature #12869 (Open): open-uri's open rejects `BOM' in encoding
- Kernel.#openのエンコード指定では、BOMへの対応も指定できます。
```ruby
open('foo.txt', 'r:BOM|UTF-8') ...
```
open-uriが再定義したopenでURLをオープンするときにも同じ動作を期待して実行してみます。
```ruby
require 'open-uri'
open('http://example.com/foo.txt', 'r:BOM|UTF-8') ...
```
... -
08:51 AM Bug #12830: OpenSSL 1.1.0+ support?
- Thx for investigation. I can confirm that r55074 make the Typhoeus (and Ethon) to pass its test suite.
-
- I am going to use this patch [1] in Fedora if it helps.
[1]: http://pkgs.fedoraproject.org/cgit/rpms/ruby.git/tree/ruby-2.4.0-openssl-register-ex_data-index-for-X509_STORE-_CTX-r.patch -
07:29 AM Bug #9839 (Rejected): Segment fault in http
- This should have been fixed at [Bug #9592]. I suspect Paul's is another issue ([Bug #12292]), though, I can't tell without a reproducer.
-
- It seems it does. TestWEBrickHTTPProxy#test_connect in test/webrick/test_httpproxy.rb is testing.
-
- It turned out to be an issue of mixing two different OpenSSL installations.
-
- Fixed by 66c1da57eb03 at new upstream[1] and imported by r56027 ("import Ruby/OpenSSL 2.0.0.beta.1").
[1]: https://github.com/ruby/openssl/commit/66c1da57eb03c19f96f3f1c843ea2a93dca3d243 -
- The message you pasted is not full. There should be something before the part you copied. Can you disclose the log from the beginning of the "enter to rails console" you say?
-
03:46 AM Revision e62586fd (git): test_thread.rb: count accurately
- * test/ruby/test_thread.rb (test_thread_timer_and_interrupt):
count only signal handling time accurately without setup and
cleanup time.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56485 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * test/lib/test/unit.rb (Test::Unit::Parallel#deal): update the
job status after a task finished, to show idling workers.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56484 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * complex.c (Init_Complex): undefine methods inherited from
Comparable, because Complex does not have <=> method.
[Bug #12866]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56483 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * class.c (rb_undef_methods_from): undefine methods defined in
super from klass.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56482 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * test/ruby/test_complex.rb, test/ruby/test_rational.rb: refine
assertions for descriptive messages.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56481 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Applied in changeset r56479.
----------
complex.c: undefine clamp
* complex.c (Init_Complex): undefine Complex#clamp, which does not
work like other Comparable methods, because Complex does not
have <=> method. patched by Tim Pete... - 01:19 AM Revision d2d44828 (git): * 2016-10-24
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56480 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * complex.c (Init_Complex): undefine Complex#clamp, which does not
work like other Comparable methods, because Complex does not
have <=> method. patched by Tim Peters <zomg.tim AT gmail.com>
in [ruby-core:77720]. [Bug #12866]
gi...
10/23/2016
-
- It looks like the same issue as [Bug #12292] rather than [Bug #12126], because #12126 should not be a cause of a segfault. (and net/http does not use SSLSocket#write_nonblock, does it?)
Please provide a reproducer if it is a different... -
- As commented at the GitHub PR, there doesn't seem to be anything wrong. `ssl_socket.set_params(ssl_version: :TLSv1)` should be equivalent to `ssl_socket.set_params; ssl_socket.ssl_version = :TLSv1`.
-
- The version string in OpenSSL::OPENSSL_VERSION is not the version running with but the OpenSSL version compiled with. Current versions of Ruby have OpenSSL::OPENSSL_LIBRARY_VERSION for the actual loaded version.
-
- I can't tell without a reproducer but the symptoms looks like similar to [Bug #12292].
-
- Sounds like [Bug #12292]. Please provide a reproducer if it does not fix.
-
- Indeed we should check for OPENSSL_NO_ENGINE instead of the existence of openssl/engine.h, and it's actually done in trunk (r55162).
However it is strange that this causes a runtime error. The existence of the header is checked by the... -
02:16 PM Feature #12867: Add ability to check validity of a URL
- I'd also love to work on this for my project over at the Fuss / Simply be http://www.thefuss.co.uk/simply-contact-number
-
- Hi!
Going through the URI module of Ruby, it was obvious that we can check the validity of a URI by using URI.regexp . However, I found out there is no way we can accurately check for a valid URL by using the URI module. Won't it be bet... -
- Yes, X509_STORE_{get,set}_ex_data() are implemented in ext/openssl/openssl_missing.c. In this specific case, applying r55074 that converted them into macros should fix (backport ticket: [Bug #12868]).
-
- Please backport r55074 ("openssl: register ex_data index for X509_STORE{_CTX,} respectively").
In Ruby <= 2.3, ext/openssl misused one ex_data index for both X509_STORE and X509_STORE_CTX, and it was working just by chance. This will ... -
08:05 AM Feature #12858: Supporting batch-requiring of files in ruby
- You realize [this gem](https://rubygems.org/gems/require_all/) does exactly that?
-
02:48 AM Bug #12862: Regular Expression Named Group Matching does not work with #{} (Similar? to issue #2778)
- Matthew Kerwin wrote:
> Leo Amigud wrote:
> ...
Thank you for the explanation!
Leo -
- * include/ruby/encoding.h: include "ruby/ruby.h" explicitly for
enum ruby_fl_type and VALUE.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56478 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 02:03 AM Revision 49432c84 (git): * 2016-10-23
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56477 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * test/ruby/test_refinement.rb (test_symbol_proc): a Symbol is
converted to a Proc in the caller's context. [Feature #9451]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56476 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
01:56 AM Bug #12866 (Closed): Complex should not respond to clamp
- `Comparable#clamp` has been added, but doesn't work in `Complex`. It should be undefined on `Complex` like other `Comparable` methods are.
10/22/2016
-
- Leo Amigud wrote:
>
> ...
It's what Hans quoted and said:
"When named capture groups are used **with a literal regexp** *[i.e. not an interpolated regexp]* ... the captured text is also assigned to local variables ..."
`/(?<doll... -
- Hans Mackowiak wrote:
> it did took me a while to understand what you mean, because the regexp are displayed wrong.
> ...
Sorry, not sure what you mean.
As a matter of fact i can use regexp like:
irb(main):013:0> numbers = '\d+'
... -
03:51 PM Bug #12862: Regular Expression Named Group Matching does not work with #{} (Similar? to issue #2778)
- it did took me a while to understand what you mean, because the regexp are displayed wrong.
"When named capture groups are used with a literal regexp on the left-hand side of an expression and the =~ operator, the captured text is als... -
- Loading development environment (Rails 4.2.5)
```
irb(main):001:0> /\$(?<dollars>\d+)\.(?<cents>\d+)/ =~ "$3.67"
=> 0
irb(main):002:0> dollars
=> "3"
```
BUT:
```
irb(main):001:0> numbers = '\d+'
=> "\\d+"
irb(main):002:... -
- * parse.y (reg_compile_gen): always append error message to the
error buffer.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56475 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Applied in changeset r56474.
----------
numeric.c: fix up r55891
* numeric.c (num_funcall1): check recursion by inverse pair, to
fix fake infinite recursion. [ruby-core:77713] [Bug #12864] -
- If you have a class including `Comparable` and expecting to be compared to `Integer`, equality test doesn't work anymore. This worked in 2.3.1.
Example:
~~~ruby
class MyInteger
include Comparable
def initialize(i)
... -
- * numeric.c (num_funcall1): check recursion by inverse pair, to
fix fake infinite recursion. [ruby-core:77713] [Bug #12864]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56474 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
-
12:24 PM Bug #12865 (Closed): openssl: segmentation fault when nil is passed as key
- This bug occurs both on Gentoo and Ubuntu 16.04.
Steps to reproduce on ruby 2.3.1:
irb(main):001:0> require 'openssl'
=> true
irb(main):002:0> OpenSSL::PKey::RSA.new.public_encrypt("")
(irb):2: [BUG] Segmentation fault at 0x0000... -
- Applied in changeset r56473.
----------
hash.c: fix Hash#compact! return value
* hash.c (rb_hash_compact_bang): should return nil if no elements
is deleted. [ruby-core:77709] [Bug #12863] -
09:10 AM Bug #12863: Hash#compact! behavior different from the Active Support Hash#compact!
- Hm, looks like returning `nil` (if nothing was dropped) is standard behavior of [Hash#reject!](https://ruby-doc.org/core-1.9.3/Hash.html#method-i-reject-21). Weird.
-
- What is the reason of returning `nil` in that case? I believe in standard containers all bang-versions always return self?
-
08:23 AM Bug #12863 (Closed): Hash#compact! behavior different from the Active Support Hash#compact!
- Hash#compact and Hash#compact! were added in https://bugs.ruby-lang.org/issues/11818 but the Hash#compact! is different from Active Support. Please check following snippet:
~~~ ruby
# Active Support
>> hash = { a: true, b: false, c:... -
- * hash.c (rb_hash_compact_bang): should return nil if no elements
is deleted. [ruby-core:77709] [Bug #12863]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56473 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Applied in changeset r56472.
----------
configure.in: fallback DLDFLAGS to LDFLAGS
* configure.in (DLDFLAGS): fallback to LDFLAGS.
[ruby-core:72444] [Bug #11863]
* configure.in (LIBRUBY_DLDFLAGS): fallback to DLDFLAGS. -
- Applied in changeset r56472.
----------
configure.in: fallback DLDFLAGS to LDFLAGS
* configure.in (DLDFLAGS): fallback to LDFLAGS.
[ruby-core:72444] [Bug #11863]
* configure.in (LIBRUBY_DLDFLAGS): fallback to DLDFLAGS. -
- * configure.in (DLDFLAGS): fallback to LDFLAGS.
[ruby-core:72444] [Bug #11863]
* configure.in (LIBRUBY_DLDFLAGS): fallback to DLDFLAGS.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56472 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * configure.in (RUBY_APPEND_OPTION, RUBY_PREPEND_OPTION): expand
the option to be appended/prepended when matching, as well as
RUBY_APPEND_OPTIONS and RUBY_PREPEND_OPTIONS.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56471 b2... - 12:53 AM Revision f0f3f225 (git): * 2016-10-22
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56470 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * compile.c (setup_args): duplicate splatting array if more
arguments present to obey left-to-right execution order.
[ruby-core:77701] [Bug# 12860]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56469 b2dd03c8-39d4-4d8f-98ff-823f...
10/21/2016
-
10:31 PM Feature #12664: Multiline pretty-printing of multiline strings
- Thanks!
-
09:30 PM Bug #12548: Rounding modes inconsistency between round versus sprintf
- Benoit says RubySpec PR #322 looks good. Just waiting for us to commit round-to-nearest-even mode to MRI so he can merge it. I've used [Nobu's branch](https://github.com/nobu/ruby/tree/round-to-even) locally, including to create PR #322,...
-
- ```ruby
class Class1
def self.foo
'foo'
end
def self.method1
'method1'
end
end
class Class2 < Class1
def self.foo
bar do
super()
end
end
def self.bar(&block)
a = block.()
def... -
02:44 PM Bug #12860: Splatting an argument does not obey left-to-right execution order
- It is worth mentioning that if the splatted value is not an array, we can see that `to_a` does get called before `shift`. That makes this even more unintuitive, since the calls happen in proper order but the assembly of the args list hap...
-
- This behavior appears to go back as far as Ruby 1.9.3. Ruby 1.8.7 and earlier could not have regular arguments after a splat.
-
- Rails PR (which unpacks via multi-assign to avoid this bug) filed here: https://github.com/rails/rails/pull/26854
-
- Ruby evaluates arguments left to right, but it does not appear to handle construction of the eventual argument list from left to right.
Take this example:
```ruby
def foo(*args)
p args
end
ary = [1,2]
foo(*ary, ary.shift... -
12:50 PM Bug #12859 (Third Party's Issue): Bug on Rails Console?
- I create a new project and use scaffold.
So after i create my model, view and controller i enter to rails console
```ruby
p = Property.new
```
And i get all of this errors:
(snip)
-
- Ping? What is the status here?. Now I noticed that the LDFLAGS are not used during build of Fedora package, so Ruby is not properly hardened as it should be. That is unexpected :/
-
- According to Fedora OpenSSL maintainer, there seems to be conflict in X509_STORE_set_ex_data symbol. The symbols which are from latest OpenSSL should not be defined locally with the same name (unless they are static).
-
- Ok, so now we have openssl-1.1.0b-3.fc26 as well as compat-openssl10-1.0.2j-5.fc26 (including -devel subpackage) on Fedora Rawhide, so it should be possible to build older Ruby, but unfortunately, some other libraries fail, since they lo...
-
09:04 AM Bug #11929: No programatic way to check ability to dup/clone an object
- Mike, your last proposal is simple, beautiful and smart.
```
1.dup #=> 1
``` -
- * test/ruby/test_dir.rb (test_glob_gc_for_fd): needs
RLIMIT_NOFILE.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56468 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * dir.c (do_opendir): retry after GC when the limit for open file
descriptors reached.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56467 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * dir.c (rb_dir_s_empty_p): use rb_gc_for_fd for the condition to
invoke GC by errno.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56466 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * dir.c (to_be_ignored): make an inline function to get rid of
multiple errno function calls.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56465 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * dir.c (dir_initialize, dir_read): prefer rb_syserr_fail over
rb_sys_fail.
* io.c (ruby_dup, rb_sysopen): ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56464 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
07:12 AM Feature #12858 (Open): Supporting batch-requiring of files in ruby
- Hello. This is a very short proposal, unlike some of my previous
ones; and it also is hopefully possibly better than some of my
other proposals before.
Could we have a way to batch-require .rb files from a directory?
Right now we... -
- * ruby.c (open_load_file): retry after GC when the limit for open
file descriptors reached.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56463 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * test/ruby/test_require.rb (test_loading_fifo_fd_leak): invoke GC
only when FDs exhausted to reduce the test time.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56462 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 02:34 AM Revision 81e687d2 (git): * 2016-10-21
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56461 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * string.c (rb_str_sub, rb_str_gsub): [DOC] 'backlash' should read
'backslash'. [Fix GH-1461]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56460 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
-
- It's said that it is an Apple's bug.
Try to install sqlite3 by HomeBrew, MacPorts, or etc.
http://stackoverflow.com/questions/39812707/segmentation-fault-with-rails-after-upgrading-to-os-sierra-possibly-related-to
https://bugs.pytho...
10/20/2016
-
09:08 PM Bug #11929: No programatic way to check ability to dup/clone an object
- Martin Dürst wrote:
> I clearly prefer the last proposal (fail silently).
> ...
I very much agree. -
09:11 AM Bug #11929: No programatic way to check ability to dup/clone an object
- I clearly prefer the last proposal (fail silently).
That would make for a much more unified, streamlined protocol, avoiding needless exposition of internals. It would do exactly what dup (and clone) are described to do, namely (preten... -
04:03 AM Bug #11929: No programatic way to check ability to dup/clone an object
- *D.E.*, while you're technically not wrong, IMHO, the need to **`undef`** the method is only abated in the strictest of senses: the sanity check averts what might otherwise be a segfault or memory leak or some other aberration, and makes...
-
- Benoit Daloze says that RubySpec should test previous rounding behavior for Ruby before 2.4.0, and test round-even for 2.4.0+. That makes sense to me. Opened RubySpec PR #322 (https://github.com/ruby/spec/pull/322) to reflect this.
Wh... -
07:10 PM Bug #12857 (Third Party's Issue): macOS Sierra sqlite3 bug
- ```
Hughs-MacBook-Pro:travel hughtures$
```
-
- Jeremy Evans wrote:
> While this was a behavior change between 2.1 and 2.2, I'm not sure I would consider it a regression.
> ...
The main reason I consider it a bug is that it contradicts the very basic intuition that replacing a liter... -
- While this was a behavior change between 2.1 and 2.2, I'm not sure I would consider it a regression. It seems unlikely anyone who uses compare_by_identity hashes would also be using string literals and want string literals keys to have ...
-
- This seems a regression since 2.2.
I would guess it's due to some optimization for having a string literal between []=.
That optimization should not trigger for compare_by_identity hashes, so both cases below are consistent.
~~~ruby... -
05:00 PM Bug #12856 (Rejected): Help Please
- This is for bugs with the Ruby language.
You can ask for help on the ruby-talk mailing list (http://lists.ruby-lang.org/cgi-bin/mailman/listinfo/ruby-talk ), StackOverflow.com, ... -
02:54 PM Bug #12856 (Rejected): Help Please
- Hello,
I downloaded the ruby program for Windows OS. Whenever I write a program in the notepad and I save it as a ruby program following instructions from my computer studies course, the program opens for a few seconds and before the co... -
01:18 PM Revision 6f22fc6b (git): fix hexdigest [ci skip]
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56459 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
08:10 AM Bug #6785 (Feedback): Documentation for RubyVM::InstructionSequence
-
-
- [Misc #11960][ruby-core:72733][ci skip]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56458 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * test/ruby/test_proc.rb (test_curry): split.
* test/ruby/test_proc.rb (test_curry_passed_block): simplify the
assertion.
* test/ruby/test_proc.rb (test_curry_with_trace): run all curry
tests.
git-svn-id: svn+ssh://ci.ruby-lang.or... -
- Applied in changeset r56455.
----------
* ext/digest/digest.c: Add documentation for Digest.
[Feature #10452][ruby-core:66001][ci skip]
* remove HMAC from list of digest algorithms,
* add MD5 in list of digest algorithms,
* add ... - 07:57 AM Revision 10035cb4 (git): * remove trailing spaces.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56456 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- [Feature #10452][ruby-core:66001][ci skip]
* remove HMAC from list of digest algorithms,
* add MD5 in list of digest algorithms,
* add information about writing a C digest implementation using Digest::Base,
* add documentation fo... -
-
- open-uri#open option. [Misc #11329][ruby-core:69868][ci skip]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56454 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
05:45 AM Feature #12854: Proc#curry should return an instance of the class, not Proc
- Another (albeit, poor/strange) example:
https://gist.github.com/tenderlove/ca673b3ce18460890cfd18e09bb1657c
-
- ~~~ ruby
class ChainedProc < Proc
# ... this stuff here is the use-case
end
~~~
-
- What's the rationale or the use-case?
Curried proc is not a subset of the original proc, I think. -
02:42 AM Bug #12833: incorrect iso8601 parsing of YYYY-MM format
- I have a pull request at https://github.com/ruby/ruby/pull/1463. I can make a patch if that is preferable (though this is a very small change).
Since this ticket is marked duplicate, should my patch go on the ticket that this one dup...
10/19/2016
-
- ~~~ ruby
class ChainedProc < Proc
end
ChainedProc.new { |x, y, z| 42 }.curry.class # => Proc
~~~
- 10:44 PM Revision 38689e4b (git): * 2016-10-20
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56453 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * NEWS: fix typos "Refinments" as "Refinements".
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56452 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56451 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- Applied in changeset r56450.
----------
vm_insnhelper.c: refinements with send
* vm_insnhelper.c (vm_call_opt_send): enable refinements with
`Kernel#send` and `BasicObject#__send__`. [Feature #11476] -
- * vm_insnhelper.c (vm_call_opt_send): enable refinements with
Kernel#send and BasicObject#__send__. [Feature #11476]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56450 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Thank you reporting. The crash log you attached indicates your process dies inside of libsqlite3.dylib.
-
06:20 AM Bug #12853 (Third Party's Issue): Error using rails c on Mac OS X 10.12 Sierra
- Was able to initiate rails c; upon executing User.count (SQL statement), received error. Have attached the log file that was generated during my error (same or similar to the error reported in ticket 12844 by F. Schiller).
Includes: ... -
- * basictest/runner.rb: do not clobber the option by --run-opt with
RUBYOPT. reported by Allen Hewes <rallenh AT hotmail.com>.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56449 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
10/18/2016
-
- Olivier Lacan wrote:
>
> ...
You keep using the word "URI" to refer to these data objects, and by specification a URI data object cannot contain non-ASCII characters (and even some ASCII characters are forbidden.) If we agree that `"h... -
09:57 PM Bug #12852: URI.parse can't handle non-ascii URIs
- Matthew Kerwin wrote:
> The rails snippet you linked is part of a HTML form. A web browser displaying and submitting that form would interpret the `✓` entity as U+2713 CHECK MARK, yes, but it would percent-encode it as `%E2%9C%93... -
- As a point of order, there's no such thing as a "non-ASCII URI"*; that would be an [IRI](https://tools.ietf.org/html/rfc3987).
The rails snippet you linked is part of a HTML form. A web browser displaying and submitting that form woul... -
- Given a return URL path like: `/search?utf8=\u{2713}&q=foo`, `URI.parse` raises the following exception:
```ruby
URI.parse "/search?utf8=\u{2713}&q=foo"
URI::InvalidURIError: URI must be ascii only "/search?utf8=\u{2713}&q=foo"
``... -
- This isn't a bug. String#gsub! returns nil if no changes are made. From the documentation for String#gsub!: "Performs the substitutions of String#gsub in place, returning str, or nil if no substitutions were performed."
-
05:21 PM Bug #12851 (Rejected): string.gsub!(/\W/, '').downcase! returns undefined method in some (listed) cases
- Hi! Found this strange issue with gsub! and downcase! methods used together for a string. Example code:
~~~ ruby
def palindrome? (str)
str.gsub!(/\W/, '').downcase!
str == str.reverse
end
puts palindrome?("Madam, I'm Adam!"... -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56448 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- Applied in changeset r56446.
----------
addr2line.c: boundary checks
* addr2line.c (parse_debug_line_cu): boundary checks for
compressed debug sections. [ruby-dev:49840] [Bug #12850] -
- `--enable-shared`のときしか有効にしていないからでしょう。
シグナルハンドラでzlib呼び出すのはかなり怖いので、revertしますかね。 -
- http://rubyci.org/ の Ubuntu 16.04 x86_64 では通っているように見えるので、違いを調べてみたところ、`--enable-shared` を外すと手元でも大丈夫になりました。
-
- Ubuntu 16.04.1 LTS の環境で、r56445 の変更の影響で SEGV のテストの `C level backtrace information` の表示で SEGV します。
```
% gcc -v
Using built-in specs.
COLLECT_GCC=gcc
COLLECT_LTO_WRAPPER=/usr/lib/gcc/x86_64-linux-gnu/5/lto-wrapper
Target: x86_64-li... - 03:09 PM Revision 40517422 (git): * 2016-10-19
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56447 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * addr2line.c (parse_debug_line_cu): boundary checks for
compressed debug sections. [ruby-dev:49840] [Bug #12850]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56446 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- I'm going to prepare FreeBSD 11 environment for this issue at this weekend.
-
- Added a comment at the GitHub issue. It sounds weird, because its existence of SSLv2_method should be checked by ext/openssl/extconf.rb in all stable branches including 2.3.
-
- I got build failure report for openssl build on Ruby 2.3.1
see. https://github.com/rbenv/ruby-build/issues/1009
It caused that FreeBSD 11 removed `SSLv2_method`.
Can we pick patch from r55162 for Ruby 2.3.1 ? -
- * configure.in (DLDFLAGS): append --compress-debug-sections=zlib
if available, which reduces the size of LIBRUBY_SO by half or
more.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56445 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * defs/separated_version.mk (RUBY_VERSION_DLDFLAGS): move linux
specific option.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56444 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 06:35 AM Revision fed06646 (git): * 2016-10-18
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56443 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * defs/separated_version.mk: FOR DEVELEPERS ONLY.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56442 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
05:44 AM Bug #12847: Documentation of instance method Thread.abort_on_exception= is wrong
- Hah sorry about that, I completely misread the sentence. I would suggest changing the docs to be a bit more clear that this is different than `Thread.abort_on_exception=` (though it is logical). Maybe something like:
~~~
When set to ... -
- It is the instance-level setting but terminates the whole process.
-
03:55 AM Feature #9108: Hash sub-selections
- Indeed, I'm still hanging out for this. Seems like such a common thing I run into, and I'm always surprised this functionality isn't built in. `Hash#select { |k,v| {...}.include?(k) }` is very verbose.
-