Activity
From 07/25/2020 to 07/31/2020
07/31/2020
-
10:51 PM Bug #15773: Net::HTTP doesn't try next IP address in case of timeout
- benlangfeld (Ben Langfeld) wrote in #note-4:
>
> ...
Minor point, I don't think it affects much, but RFC 3484 is obsolete and is replaced by RFC 6724 https://tools.ietf.org/html/rfc6724
Cheers -
06:46 PM Bug #15773: Net::HTTP doesn't try next IP address in case of timeout
- naruse (Yui NARUSE) wrote in #note-1:
> In general resolving DNS is done by libc (getaddrinfo) and Ruby just uses their result.
> ...
RFC3484 (https://www.ietf.org/rfc/rfc3484.txt) very clearly states:
```
Well-behaved applications... -
06:06 PM Feature #17097: `map_min`, `map_max`
- To put in context, consider that the expression might be much longer than `.length`.
Then repeating it is not elegant and is duplicated code. -
- I actually regularly wanted such functionality, but min_by/max_by do not return the result of the block but the element (which makes sense).
And `.map {}.min/max` works but is inefficient by generating an intermediate Array.
Maybe we sh... -
01:51 PM Feature #17097: `map_min`, `map_max`
- nobu (Nobuyoshi Nakada) wrote in #note-1:
> Then we'll need `map_` versions for all `Enumerable` methods.
Exactly, I don't see any good use case in that apart from being lazy and just using one method call. It seems to me that this k... -
12:37 PM Feature #17097: `map_min`, `map_max`
- Then we'll need `map_` versions for all `Enumerable` methods.
-
12:33 PM Feature #17097 (Open): `map_min`, `map_max`
- `min`, `min_by`, `max`, `max_by` return the element that leads to the minimum or the maximum value, but I think it is as, or even more, frequent that we are interested in the minimum or the maximum value itself rather than the element. F...
-
12:07 PM Revision f80020bc (git): [rubygems/rubygems] Read path binarily
- I believe this should fix a flaky test on Windows.
https://github.com/rubygems/rubygems/commit/a516b49a0a -
- https://github.com/rubygems/rubygems/commit/9c17f220af
-
- This restores 3.0 behavior and goes back to not formatting executables
by default on `gem update --system`. This is friendlier for jruby and
doesn't really affect OS packagers.
https://github.com/rubygems/rubygems/commit/1a504b651d -
12:07 PM Revision 0e2d8c48 (git): [rubygems/rubygems] unset XDG_* environmental variables for testing of rubygems.
- https://github.com/rubygems/rubygems/commit/539fd9a39a
-
- This reverts commit 694e6afee769ffb3168a564ee7d315af2a934993.
It was failed on Windows and I'm unsure why. Also, the check seems
brittle since it could fail potentially fail because of warnings outside
of our control.
https://github.co... -
- Since 1ccf0912a161d20e0c4a7b139fd76e8739a411ba, this method no longer
uses `Kernel.gem`, so this guard clause is now unnecessary.
https://github.com/rubygems/rubygems/commit/8a19e7401c -
- Since they changed their default to not generate documentation on `gem
install`, we're getting some intermittent test failures. Can be
reproduced with:
```
TRUFFLERUBYOPT="--experimental-options --testing-rubygems" TESTOPTS=--name="/^\(... -
- Catch up with that change in `gem update --system` tests.
https://github.com/rubygems/rubygems/commit/127ba14344 -
- https://github.com/rubygems/rubygems/commit/dca345441f
-
- https://github.com/rubygems/rubygems/commit/a54230093e
-
- If any error happens while verifying a package entry, it doesn't mean
that the package is corrupt. It could be a bug in rubygems, for example.
This in fact happened in CI and the current error doesn't make it easy
to troubleshoot the roo... -
- Fixes the `Gem::FilePermissionError` without the privilege.
Initialize `@plugins_dir` to the user gem directory, when
installing with `--user` option.
https://github.com/rubygems/rubygems/commit/21a71ac769 -
- Since `rubygems/util` is autoloaded.
https://github.com/rubygems/rubygems/commit/52a9f88b99 -
- https://github.com/rubygems/rubygems/commit/00ebf8c9f7
-
- The whole test suite actually passes on my system, but it has weird
errors in CI. Since I don't want to spend time on it, instead of running
the whole test suite, I'm just adding a bare test to check that `gem
list` works.
https://githu... -
- https://github.com/rubygems/rubygems/commit/41976ef3ec
-
- We patch `net-http-persistent` to not autoload `openssl`.
https://github.com/rubygems/rubygems/commit/757dec3cf2 -
- The `openssl` require when openssl not present was having the
side-effect the our custom require fallbacks would end up loading `Gem::Specification.stubs`.
Co-authored-by: Alyssa Ross <hi@alyssa.is>
https://github.com/rubygems/rubygems... -
- https://github.com/rubygems/rubygems/commit/a763e539cd
-
- https://github.com/rubygems/rubygems/commit/054d57f74b
-
- https://github.com/rubygems/rubygems/commit/25912ce6c9
-
- No check is done for the other expectation and they are completely
symmetric as far as I can see.
https://github.com/rubygems/rubygems/commit/4de89e0718 -
- So it can be reused.
https://github.com/rubygems/rubygems/commit/b9fc6e40db -
- https://github.com/rubygems/rubygems/commit/babf943144
-
- https://github.com/rubygems/rubygems/commit/d81ce9e457
-
- https://github.com/rubygems/rubygems/commit/d92b94f3cf
-
- It will give more useful information.
https://github.com/rubygems/rubygems/commit/efcecb5af5 -
- The `rubygems/security` require already does this.
https://github.com/rubygems/rubygems/commit/bbb444b6f1 -
- It's not that slow.
https://github.com/rubygems/rubygems/commit/9b928a4503 -
- https://github.com/rubygems/rubygems/commit/1e2c3cf118
-
- GNU make in MSys is localized to use UTF-8 while Ruby's filesystem
encoding is set to OEM CodePage (e.g., CP932 in Japanese Edition),
the read output from the make has broken encoding and results in
"invalid byte sequence" errors. As `D... - 12:07 PM Revision 6c19b0ec (git): [rubygems/rubygems] Reword warning
- https://github.com/rubygems/rubygems/commit/cbd4abf8cf
- 12:07 PM Revision 00bdac18 (git): [rubygems/rubygems] Warn on duplicate dependency in a specification
- https://github.com/rubygems/rubygems/commit/af3e5f7883
-
- Co-authored-by: MSP-Greg <MSP-Greg@users.noreply.github.com>
https://github.com/rubygems/rubygems/commit/710b969b60 -
- https://github.com/rubygems/rubygems/commit/d722b8b578
-
- It seems like the most common case since it requires no tricks on our
CI environment.
Co-authored-by: MSP-Greg <MSP-Greg@users.noreply.github.com>
https://github.com/rubygems/rubygems/commit/751c475574 - 12:07 PM Revision 65a4b037 (git): [rubygems/rubygems] fix missing identation
- https://github.com/rubygems/rubygems/commit/2cca6714f3
-
12:07 PM Revision 56157636 (git): [rubygems/rubygems] Deduplicate the requirement operators in memory
- https://github.com/rubygems/rubygems/commit/9963d33cf2
-
12:07 PM Revision 1a935606 (git): [rubygems/rubygems] Fix Kernel#warn override to handle backtrace location with nil path
- It's very unlikely to hit this case, but it is possible, as
Thread::Backtrace::Location#path can return nil if the location is
a cfunc with no previous iseq. See location_path in vm_backtrace.c
in Ruby.
https://github.com/rubygems/ruby... -
- * See https://github.com/oracle/truffleruby/issues/2046
* `<internal:` is a common prefix also used by core Ruby files in CRuby.
* test_no_kernel_require_in_*warn_with_uplevel already test this.
* Unfortunately just skipping `<internal:`... -
- https://github.com/rubygems/rubygems/commit/694e6afee7
-
- https://github.com/rubygems/rubygems/commit/f4c4cddb68
-
- The `teardown` method doesn't call it either and I don't think it's
necessary.
https://github.com/rubygems/rubygems/commit/ca2a5d485d -
- https://github.com/rubygems/rubygems/commit/6e4bef758b
-
- https://github.com/rubygems/rubygems/commit/d1be8cdb3a
-
- It was added 10 years ago on a "146 additions, 170 deletions" commit
named "Deprecation removals and minor cleanup." that didn't explain much
other than that.
This TODO doesn't necessarily apply nowadays. I don't see how anyways.
TODO n... -
- Profiling a simple `ruby -e '1'` I see:
```
==================================
Mode: wall(10)
Samples: 3414 (55.10% miss rate)
GC: 856 (25.07%)
==================================
TOTAL (pct) SAMPLES (pct) FRAME
... -
- 12:07 PM Revision 734c5276 (git): [rubygems/rubygems] Simplify digest name selection and use SHA256
- The previous commit introduces the Gem::Security.create_digest method, allowing to:
- decouple algorithm choice from implementation (OpenSSL or Ruby built-in)
- untangle the SHA512 fallback for TarWriter from the generic hashing digest c... - 12:07 PM Revision 8161cf85 (git): Stop using deprecated OpenSSL::Digest constants
- 12:07 PM Revision e7b6e0ff (git): [rubygems/rubygems] remove explicit require(s) for auto-loaded constant(s)
- the Gem module's auto-loads will handle loading these as needed,
this started as a redundancy found in *rubygems.rb* which had:
`autoload :Specification, 'rubygems/specification'` as well as
`require 'rubygems/specification'`
https://g... - 12:07 PM Revision c3b480b7 (git): [rubygems/rubygems] Test files should not be included in spec.files
- https://github.com/rubygems/rubygems/commit/0c0760b734
-
- To normalize the code style with `bundler`.
-
09:26 AM Bug #13670: [BUG] Bus Error at 0xefce7b (armv7l) (ruby 2.3.4p301)
- This was resolved in Ruby 2.6+, wasn't it? This ticket [1] references following commits which resolves the issue:
https://github.com/ruby/ruby/commit/58a2084483ce8baaf90d7b1cb00e3fa9570fbc79
https://github.com/ruby/ruby/commit/b6185e... -
08:31 AM Bug #17093: attr_accessor works strange
- are you try this solution?
```
if type.nil?
type = 'default'
end
```
just test it ) -
-
07:34 AM Bug #17093: attr_accessor works strange
- This:
```ruby
if type.nil?
type = 'default'
end
```
is different from this:
```ruby
type = 'default' if type.nil?
```
because of the lexical reading, it reads the setting of local variable first and now assumes that type is local va... -
- you doesnt see "`if type.nil?`" its local variable?
`p type.nil?` return false
next "`if type.nil?`" return true and "`type = 'default'`" will work
why? -
- `type` before and after `type =` are different things in Ruby.
-
- its copy of incorrectly closed https://bugs.ruby-lang.org/issues/17093
```
require 'rubygems'
class A
def initialize(type:)
@type = type
end
def b
p type
p type.nil?
type = 'default' if type.nil?
... -
- As it can be incompatible only when passing a lambda, I don't think it is a serious problem.
-
05:44 AM Bug #17094 (Closed): PTY methods with blocks
- Some methods on PTY yield one array if a block is given, but the RDoc says it passes two arguments to the block.
https://github.com/ruby/ruby/blob/master/ext/pty/pty.c#L529
``` c
return rb_ensure(rb_yield, assoc, pty_close_pty, asso... -
06:25 AM Bug #17095 (Closed): Backport c0b93267497d47858ea9539ef863be321bc1ddbf
- This is a ticket for backport management.
See https://github.com/ruby/ruby/pull/3364 too.
-
- I found a word [traject], means to transport, transmit, or transpose.
It may (or may not) imply [trajectory]...
[traject]: https://www.dictionary.com/browse/traject
[trajectory]: https://www.dictionary.com/browse/trajectory -
04:03 AM Bug #16654 (Open): Segfault in rb_str_hash
07/30/2020
-
10:41 PM Bug #17093 (Closed): attr_accessor works strange
- In Ruby `type = 'default'` only sets a local variable named `type`. It does not call the `type=` method on self, you need `self.type = 'default'` for that.
See https://ruby-doc.com/docs/ProgrammingRuby/html/tut_expressions.html, spec... -
- seriously?
attr_accesor doesnot provide value? -
- Do you expect the `type = 'default'` in `b` to set the @type instance variable to `'default'`? If so, that doesn't work, you need to use `self.type =` instead of `type =`. If not, can you provide an explanation of what behavior you are...
-
- ```ruby
require 'rubygems'
class A
def initialize(type:)
@type = type
end
def b
p type
p type.nil?
type = 'default' if type.nil?
type
end
private
attr_accessor :type
end
RSpec.desc... -
- I get `OK` for all three test files uploaded. I tested Ruby 2.0, 2.2, and 2.7 and got the same result with each. So if this was a problem, it appears to have been fixed.
-
- Applied in changeset commit:git|a6bfc951aa9c48e2a4608a9ae84e2e3b4fd9b390.
----------
Document Array#flatten{,!} accept explicit nil argument [ci skip]
Fixes [Bug #10475] -
- Fixes [Bug #10475]
-
07:16 PM Feature #17016: Enumerable#accumulate
- I'd like to sum up where we're at with this discussion. Let me know if you disagree with my interpretation here!
1. There is some support for the idea of adding this method.
2. We should avoid changing the behavior of `#inject` on la... -
- I think the need for this is handled by `Net::HTTP#ipaddr=`, which was added in Ruby 2.4.
-
-
-
- Support for this was implemented in commit:3556a834a2847e52162d1d3302d4c64390df1694 and commit:41582d5866ae34c57094f70f95c3d31f4a1fa4ff.
-
-
-
-
06:16 PM Feature #17055: Allow suppressing uninitialized instance variable and method redefined verbose mode warnings
- Another note:
> In fact I wouldn't be surprised if on other Ruby implementations initializing led to better performance (e.g., no need to grow the ivar storage dynamically later, or change the shape/hidden class).
JRuby will attemp... -
- Some JRuby perspective...
Personally, I have never been a big fan of the warning, but I don't have a strong opinion one way or another. I am not surprised that avoiding initialization is a faster on MRI because there's quite a few cycle... - 06:11 PM Revision 75335199 (git): NODE_MATCH needs reference updating
-
-
- 03:59 PM Revision b933b43b (git): * 2020-07-31 [ci skip]
-
- 02:07 PM Revision 352895b7 (git): [ruby/racc] Return `nil` for all syntax errors
- https://github.com/ruby/racc/commit/51817ce0f6
-
01:55 PM Bug #17092 (Closed): Array#flatten with finite depth should flatten recursive arrays - Applied in changeset commit:git|1b1ea7b3bc9484e6e59d716fce2965a2f39d1e3d.
----------
Fix Array#flatten for recursive array when given positive depth [Bug #17092] -
12:29 PM Bug #17092 (Closed): Array#flatten with finite depth should flatten recursive arrays
- Recursive arrays can not be flattened currently:
```ruby
a = []; a << a
a.flatten(2) # => tried to flatten recursive array
```
The only valid reason to raise an error for recursive arrays is for flatten with no argument (or nega... - 01:53 PM Revision 1b1ea7b3 (git): Fix Array#flatten for recursive array when given positive depth [Bug #17092]
-
-
-
-
- "requiring version.rb" strategy has some issues.
- cannot work when cross-compiling
- often introduces wrong namespace
- must know the superclasses
- costs at each runtime than at build-time
etc. -
09:34 AM Revision cfbae7da (git): [skip-ci] Clarification for dup vs clone docs
- Both clone & dup returns a new object when executed
on the documentation looks like they are returning the
same object cloned or dup'ed which is true for method
as extend, but not for the above mentioned. - 09:28 AM Revision 265968d6 (git): Apply timeout-scale to test_nogvl_poll.
-
01:39 AM Feature #16345: Don't emit deprecation warnings by default.
- marcandre (Marc-Andre Lafortune) wrote in #note-49:
> If that's too much warnings, then the breaking change should happen more slowly: first version with opt-in deprecation warnings, then version with opt-out deprecations, then breaking ...
07/29/2020
-
10:25 PM Revision 35e5b8fb (git): Enhanced RDoc for Array (#3372)
-
- Eregon (Benoit Daloze) wrote in #note-4:
> jeremyevans0 (Jeremy Evans) wrote:
> ...
Yes. When you initialize an instance variable to nil, it slows things down, and there is no benefit because the trying to access an uninitialized inst... -
- Also callbacks seems a very odd way to handle this, if we really want methods to suppress warnings for specific methods/ivars, let's do it proactively like:
```ruby
ignore_warning_method_redefinition :foo
def foo
...
end
ignore... -
- jeremyevans0 (Jeremy Evans) wrote:
> Not initializing instance variables to nil can be much better for performance
Why is that? Because just writing extra instance variables in `initialize` is much slower in MRI?
It's already an all... -
- I have no particular opinion on the instance variable part, except that it makes me think of [this request of reporting on instance variable typos](https://discuss.rubyonrails.org/t/typos-in-instance-variable-names-can-give-confusing-err...
-
06:54 AM Feature #17055: Allow suppressing uninitialized instance variable and method redefined verbose mode warnings
- Rails also want a way to suppress method redefinition warnings.
For now it uses the alias hack.
https://github.com/rails/rails/blob/b2eb1d1c55a59fee1e6c4cba7030d8ceb524267c/activesupport/lib/active_support/core_ext/module/redefine_me... -
02:26 AM Feature #17055: Allow suppressing uninitialized instance variable and method redefined verbose mode warnings
- Not against the feature itself, but JFIY you can suppress method redefinition warnings by something like:
```ruby
def foo
end
a = 128.times.map do
Thread.start do
alias foo foo
def foo
Thread.current
end
... -
-
- Applied in changeset commit:git|241244739f2b721ac7aa0961bd90d904c5e3fff6.
----------
Fix arity of Hash#to_proc [Bug #12671] - 04:10 PM Revision 478dc67c (git): * 2020-07-30 [ci skip]
-
-
- Actually, this was already fixed in commit:9fc25811d59422355fc3a0ed0f28155e4e47267f.
-
- Looks like this changed between Ruby 2.2 and 2.3. I'll update the documentation to remove the line, as I think it's an implementation detail (that is now wrong).
-
03:18 PM Bug #17058 (Closed): Array#delete_if doesn't change array instantly
- According to the documentation: https://ruby-doc.org/core-2.7.1/Array.html#method-i-delete_if
"The array is changed instantly every time the block is called, not after the iteration is over."
I've tried printing the array while itera... -
03:33 PM Feature #17059 (Rejected): epoll as the backend of IO.select on Linux
- Current Ruby's `IO.select` method calls POSIX `select` API directly. With the new non-blocking scheduler, this may be the bottleneck of the I/O scheduling. For keeping the backward compatibilty of the current `IO.select` methods, a propo...
-
-
01:57 PM Bug #17057 (Closed): CGI different behaviour with C vs Ruby version
- ```
irb(main):001:0> require 'cgi' # don't load a native extension
irb(main):002:0> CGI.unescapeHTML("\xFF&")
=> ArgumentError (invalid byte sequence in UTF-8)
```
there's a fix on upstream https://github.com/ruby/cgi/pull/2 -
-
-
-
- On btest, stderr messages are not displayed if core files are
generated. There is no reason to skip it, so this patch display
stderr and check core files. -
-
- >Also, I am trying to learn from sawa on the issue tracker here, making useful suggestions. :)
If I have provided here anything that is worth your learning from, the first thing among them is probably to try to write things concisely. -
12:14 AM Feature #15580: Proposal: method addition to class String called .indices ( String#indices )
- duerst (Martin Dürst) wrote in #note-1:
> Just a quick question: Should the results include overlaps or not? I.e. is it
> ...
Good question! For what it's worth, it looks like the [`String#index_all` version](https://github.com/rubyworks... -
- See also #6596, which proposes adding a similar method to `Array` (`Array#indexes`).
I don't know which term ("indices" or "indexes") is preferred. (They [are both widely used plural forms](https://grammarist.com/usage/indexes-indices... -
- TylerRick (Tyler Rick) wrote in #note-23:
> What can I do to help move this proposal forward?
Provide a use case.
07/28/2020
-
11:59 PM Bug #15773: Net::HTTP doesn't try next IP address in case of timeout
- I've encountered this issue as well. My workaround - in case someone else has the same problem:
```ruby
require 'net/http'
require 'resolv'
# ...
url = URI.parse(url) unless url.is_a?(URI)
addrs = Resolv::DNS.new.getaddresses(url.host... -
- I have a use case of finding the first matching line within a given section in a file. After finding the line number of the start of the section, I want to find the first match after that line.
My workaround for now is to use `with_in... -
11:45 PM Bug #17048: Calling initialize_copy on live modules leads to crashes
- Thank you for the code, @nobu . I think with your branch we could even keep `.allocate`, though people wouldn't be able to do much with it.
As long as no one is able to call `initialize_copy` after children (iclasses) exist, it's fine.
I... -
11:34 PM Feature #17016: Enumerable#accumulate
- RubyBugs (A Nonymous) wrote in #note-27:
> To return to naming for a moment. Did I understand an interest may exist in naming that reflects a connection or duality with `inject`?
I guess that wouldn't be a bad idea.
> ...
My most ... -
05:42 PM Feature #17016: Enumerable#accumulate
- To return to naming for a moment. Did I understand an interest may exist in naming that reflects a connection or duality with `inject`?
If so, might the **scan** operation find a better Ruby name in **interject**?
The idea would be to ... -
- I'm not very good at writing C, but here is a reference implementation in Ruby (plus unit tests) in case it's helpful: https://github.com/rubyworks/facets/pull/294/files
This version adds `indexes` (aliased as `index_all`) to `Array`,... -
- I would really like to see this included in Ruby. I have wished for this method several times in the past, and again today.
What can I do to help move this proposal forward? -
- * [Feature #17055] Allow suppressing uninitialized instance variable and method redefined verbose mode warnings (jeremyevans0)
* This keeps the default behavior the same as before, but allows opt-in to supress the warnings.
* This ... -
- These two verbose mode warnings are both fairly common and have good reasons why you would not want to warn about them in specific cases. Not initializing instance variables to nil can be much better for performance, and redefining meth...
-
- @matz considered this during the March 2020 developers meeting, but has not yet made a decision on it.
-
- @matz considered this during the September 2019 and December 2019 developers meetings, but has not yet made a decision on it.
-
- @matz considered this during the September 2019 and December 2019 developers meetings, but has not yet made a decision on it.
-
-
- sdbm was removed at commit:343ad9eff5b8e4c776c1bf193fc125f9ba1cd612.
-
-
- Applied in changeset commit:git|520a734ad9c7348f4e4858ee24640f42c88fd389.
----------
Fix Time#ceil when result should be the same as the receiver
Fixes [Bug #17025] -
- Fixes [Bug #17025]
- 07:06 PM Revision 1027520f (git): * 2020-07-29 [ci skip]
-
-
- Proposed fix (simple as it gets): https://github.com/ruby/ruby/pull/3370
-
- Looking at the commit, should Hash#to_proc use `rb_func_lambda_new()` instead of `rb_func_proc_new()`? (and then `rb_func_proc_new` is unused)
Also it seems confusing that `rb_func_proc_new()` creates a lambda, so using `rb_func_lambda_n... -
10:51 AM Revision 946cd6c5 (git): Use https instead of http
-
-
06:55 AM Bug #16921: s390x: random test failures for timeout or segmentation fault
- Can we apply [the patch](https://bugs.ruby-lang.org/attachments/8500) mentioned at https://bugs.ruby-lang.org/issues/16492#note-11 to the following tests using `assert_in_out_err` in those too?
* TestProcess#test_status_quit
* TestRu... -
- On case-insensitive filesystem, generated Net/FTP.html is
overwritten by Net/Ftp.html.
https://github.com/ruby/net-ftp/commit/14a6ff5134 -
- This ticket was discussed at the dev-meeting; there is no objection, and matz accepted. I've just merged my PR.
-
- ... in addition to normal iseq frames.
It is sometimes useful to point the bottleneck more precisely. - 03:14 AM Revision e4b63202 (git): Remove unnecessary word from comment
-
03:13 AM Revision 09e9f7cf (git): [DOC] Clarify ObjectSpace return values are in bytes [ci skip]
- 03:11 AM Revision db0eab1c (git): Remove `EnvUtil.rubyexec`, because it is not used anywhere.
-
- Deprecation warnings for things that will break in the next version should be on by default (that policy being for Ruby or for gems). If that's too much warnings, then the breaking change should happen more slowly: first version with opt...
07/27/2020
-
- I'm not sure if this is the right way.
I agree that a deprecation warning is annoying for end users. I also understand that application developers don't want them to be nervous.
However, disabling the warnings by default does not so... -
- Eregon (Benoit Daloze) wrote in #note-4:
> Thanks for the fix!
> ...
It's a consequence of the fix (skips the currently activated refinement during super), but I wouldn't say it is intended. Ideally, the fix would only affect looping ... -
- Thanks for the fix!
BTW this changes the behavior on a new spec, is that intended? (result is [:A, :C] instead of [:A, :LOCAL, :C] on < 2.8)
https://github.com/ruby/spec/commit/b0da11b52560860e844470d145acee0ff4d4acea?w=1 -
- Applied in changeset commit:git|eeef16e190cdabc2ba474622720f8e3df7bac43b.
----------
Prevent SystemStackError when calling super in module with activated refinement
Without this, if a refinement defines a method that calls super and
in... -
- Is there a reason why the arity was not fixed to `1` (instead of `-1`)?
-
-
-
07:40 PM Revision 35ba2783 (git): Use a linked list to eliminate imemo tmp bufs for managing local tables
- This patch changes local table memory to be managed by a linked list
rather than via the garbage collector. It reduces allocations from the
GC and also fixes a use-after-free bug in the concurrent-with-sweep
compactor I'm working on. -
07:36 PM Misc #17053: RDoc for Hash Keys
- Thanks, @marcandre. Your first update above makes things much clearer to me. I'll continue studying this.
-
- burdettelamar@yahoo.com (Burdette Lamar) wrote:
> My view has been this: This is API reference documentation. Ruby/ruby should have *the reference documentation*, and therefore should omit nothing.
This is a very ambitious goal I'm... -
- Let me first repeat that I am very grateful of all efforts to improve our documentation.
Documentation can be difficult because it requires deep understanding of what needs to be documented and good experience with what is actually usef... -
- Should have linked to the source of the remarks: https://github.com/ruby/ruby/pull/3139#issuecomment-663281047
-
- Where is the source of the quote? Neither of the links seems to lead to it.
-
- marcandre [comments](https://github.com/ruby/ruby/pull/3139#issuecomment-663281047) on my pull request regarding [documentation of Hash in Rdoc](https://docs.ruby-lang.org/en/master/Hash.html#class-Hash-label-Hash+Keys):
> The only th... -
- Sorry, looks like I misread.
-
-
05:00 PM Feature #17054: Some NilClass methods are faster if implemented in Ruby
- Patch was created.
https://github.com/ruby/ruby/pull/3366 -
- Looking at these pull requests, I wondered "Is it faster to implement some NilClass methods in Ruby?"
* https://github.com/ruby/ruby/pull/3264
* https://github.com/ruby/ruby/pull/3277
So I tried implementing some NilClass methods ... - 03:18 PM Revision c5ae79d7 (git): * 2020-07-28 [ci skip]
-
- Without this, if a refinement defines a method that calls super and
includes a module with a module that calls super and has a activated
refinement at the point super is called, the module method super call
will end up calling back into ... -
-
- Actually, the platform might not matter. It might fail just randomly, because now I have caught the same issue also on x86_64.
-
- Fedora aims to enable link time optimization (LTO) of packages in next release. The specific changes in configuration options are available here [1]. Since that time, I observe following errors [2] at least on {aarch64,ppc64le} (and poss...
-
- After a lot of testing, I think that the test should not fail with the timeout, because the input for the asserts is collected quite early and the system just does not have time to finish the threads and the child process. Since we don't...
- 06:25 AM Revision cb512792 (git): * 2020-07-27 [ci skip]
-
06:24 AM Revision c4f8095e (git): Use typewriter notation for var[:sym] correctly
07/26/2020
-
-
06:33 PM Bug #17051: Crash when attempting to load schema in Rails application
- I have confirmed you are correct. After switching back to my original installation of mysql, all is good.
-
- nobu (Nobuyoshi Nakada) wrote in #note-10:
> Tried it.
> ...
Right, removing `Module.allocate` is one way since `dup`/`clone` uses the alloc function directly (and does not call `allocate`).
I think calling `Module.allocate` in user... -
- Tried it.
https://github.com/nobu/ruby/tree/uninitialized-module - 12:24 PM Revision a8e5f785 (git): * 2020-07-26 [ci skip]
-
-
- ruby_2_7 05c4c88767b54808cd1459a1d6da11179c422cfa merged revision(s) 1b4d406e3a04032b6d01e92b6d184a16945c6ac3.
-
- ruby_2_7 05c4c88767b54808cd1459a1d6da11179c422cfa merged revision(s) 1b4d406e3a04032b6d01e92b6d184a16945c6ac3.
-
- Hash#transform_values should return a plain new Hash
[Bug #16498] -
- We are discussing on the deprecated warning category feature in 2.7.2 on #17000 too.
As these are keyword arguments warnings, these will be suppressed by default in 2.7.2 anyway.
-
- ruby_2_7 61c6d433060881e952140d2154c06f8c9803dc8a.
-
- Classes made from Struct should have default new singleton method.
Co-authored-by: Yusuke Endoh mame@ruby-lang.org
Co-authored-by: John Hawthorn john@hawthorn.email
Co-authored-by: Adam Hess HParker@github.com
... -
- ruby_2_7 43cc6997c59d2fbe3b91e91ee52faaa8749e9349 merged revision(s) a0bc3f2a1c2c98f225612101cb4e1ea1a6813546,75a0447c15a7ab017bd4240c2a9cc69b134b80b9,f1699314147bad2cf5550cc582693424fdbc2510.
-
- Suppress C4267 "possible loss of data" warnings
Suppress C4267 "possible loss of data" warnings
Just cast down explicitly.
win32ole: separate global variable declarations and definitions
https://gcc.gn...
07/25/2020
-
11:27 PM Misc #17050: profiler gem
- I was able to profile 200+ million* allocations with my fork which was impossible with the current implementations. Right now my current plan is to upstream these changes to `memory_profiler`. (*sorry, incorrectly said billions previousl...
-
- @ioquatrix I’m curious. What’s the difference between your memory-profiler and the existing memory_profiler?
For context I’ve been working on https://github.com/shopify/heap-profiler for the last week, and I’m also very interested in ... -
- In the `profile` stdlib, both files are used: `-rprofile` starts the profiler immediately, while `profiler.rb` is like a library:
https://github.com/ruby/ruby/blob/ruby_2_6/lib/profile.rb
So it seems to me not advisable to reuse thos... -
- I agree all your points. I already have working `profiler-memory` to a certain extent.
-
- I'd recommend starting with something specific like `profiler-memory`.
`profiler` is very generic but therefore also unclear and will make people think it's a stdlib-like profiler, but I guess it will not be.
I also think it's too ... -
- My initial goal is to expose some general data structures for efficiently capturing large memory profiles on top of `ObjectSpace` and `GC` modules, similar to how memory_profiler gem works. I don't mind making separate gem, but it seems ...
-
- Can you tell us shortly what you plan to use that gem/namespace for? If it's something that can (in some reasonable way) replace the (former?) profile/profiler functionality, then please go for it. At one point, I was planning to work on...
-
- Also I'd be happy to receive/resuse the `profile` namespace too but either one is fine and the `profile` gem actually has an interesting implementation.
-
- I would like to reuse profiler gem namespace:
https://rubygems.org/gems/profiler/versions
It currently conflicts with `profile/lib/profiler.rb` however this is no longer part of stdlib.
cc @ko1 -
- I think you are correct Alan. During my rebuild of the laptop, I inadvertently installed v 8 instead of 5.6. What threw me off was the other Rails app working without issue, using the same ruby version, mysql2 version, and Mysql database...
-
- > Thank you for responding Alan. Is 2.5.8 sufficient?
Actually, you seem to be running into a known problem in the mysql2 gem.
Please see https://github.com/brianmario/mysql2/issues/1036 and https://github.com/brianmario/mysql2/issue... -
- Thank you for responding Alan. Is 2.5.8 sufficient?
-
- Support for the 2.4 series has [ended](https://www.ruby-lang.org/en/news/2020/04/05/support-of-ruby-2-4-has-ended/). Could you see if upgrading to a newer version of Ruby solves the crash?
-
- I am experiencing a crash when running `rake db:schema:load` or `rake db:migrate` in a Rails 4.2.9/Mysql2 0.4.0 application.
I don't believe it is a specific issue with a ruby version, as I have tried with Rubies 2.3.1, 2.3.8, and 2.... -
- > How about having an initialized flag that's set by #initialize and #initialize_copy and checked in both of these methods if we want a quick fix?
That doesn't work because you can trigger the bug without ever calling `initialize` on th... -
- Should we rather design a good way to allow copying but yet not have to deal with uninitialized state?
Right now the only well-defined protocols for copying are
* `dup = allocate, copy @ivars, initialize_dup (which calls initialize_c... -
- I agree with @alanwu, that it won't be worth.
```diff
diff --git c/class.c i/class.c
index 6835d2d7289..f7a56601634 100644
--- c/class.c
+++ i/class.c
@@ -354,6 +354,13 @@ static void ensure_origin(VALUE klass);
VALUE
rb_mod_... -
- ruby_2_7 578bacc471fa3fa14f8607fe67adccce21e81657 merged revision(s) 5e81e8675a020ecd493620a4ff38db8fcf4b972a.
-
- Ignore exceptions when closing data connections [Bug #16780]
Patch by koshigoe (Masataka SUZUKI). Thanks! -
10:12 AM Feature #17016: Enumerable#accumulate
- y.annikov (Yakov Annikov) wrote in #note-22:
> I've created a PR with the implementation of `scan` and `lazy.scan` methods https://github.com/ruby/ruby/pull/3358
> ...
I renamed it to `reflect` for a while though I like `scan` more.
Her... -
- nobu (Nobuyoshi Nakada) wrote in #note-23:
> RubyBugs (A Nonymous) wrote in #note-19:
> ...
I think this is way too generic. In the same vein, we could call it "return", because it returns all intermediate states.
Also, I think "proje... -
- parker (Parker Finch) wrote in #note-21:
> nobu (Nobuyoshi Nakada) wrote in #note-20:
> ...
I fully agree. Lazy variants of methods should only differ with respect to laziness, not anything else.
-
- ruby_2_7 12391cdbe11f3fbeb3b04c30b944139e5bbd9ea6 merged revision(s) cbe4f75ef802f13d05f94e42274b65a062bd3666.
-
- Fix rb_profile_frame_classpath to handle module singletons
Right now `SomeClass.method` is properly named, but `SomeModule.method`
is displayed as `#<Module:0x000055eb5d95adc8>.method` which makes
profiling annoy... -
- ruby_2_7 be359442cb07c25fdc769431ffed3249b9016829 merged revision(s) 71c166e11e761dcaaa943f9b94da38f86ada1bdb.
-
- Fixed argument forwarding in reserved word method [Bug #16854]
-
- ruby_2_7 89f06ce8b8a887f12b53ea190d79a58e98b59008 merged revision(s) 08529a61153e5c40f57a65272211357511d6e6db.
-
- Compare environment variable names in those manor [Bug #16798]
-
- ruby_2_7 2e9626fddd168bd12352b4f5dc3412c6a33ce44e merged revision(s) 2f1895fa15386fb3cdc91c5604171290828b9da8.
-
- Fixed formatted substring expansion [Bug #16767]