Activity
From 12/12/2020 to 12/18/2020
12/18/2020
-
11:32 PM Bug #17408 (Closed): Fiber.backtrace returns [] when unavailable, unlike Thread
- ```ruby
t = Thread.new{}
p t.backtrace
# => nil
f = Fiber.new{}
p f.backtrace
# => []
```
It seems inconsistent, is this intentional?
Ping @ioquatix -
- Maybe it is not the right time to ask, but why one need to do `require 'fiber'` before using `Fiber.current`?
For what I can see,
* it is this way since [their introduction](https://docs.ruby-lang.org/en/2.0.0/Fiber.html#method-c-cu... -
09:49 PM Feature #17378: Ractor#receive with filtering like other actor langauge
- ko1 (Koichi Sasada) wrote in #note-8:
> I have several concern about `Ractor#receive` proposed at #note-6.
> ...
I agree that is an issue. It would be much better if we had `NoMatchingPatternError#depth` #17406. We could intercept only... -
07:48 PM Feature #17378: Ractor#receive with filtering like other actor langauge
- I have several concern about `Ractor#receive` proposed at #note-6.
* `NoMatchingPatternError` can be raised in tasks unexpectedly (maybe by bug) and it can cause something strange behavior (like falsy on `receive_if`).
* 2nd paramete... -
- Could we have `NoMatchingPatternError#depth`, returning the number of `case...in...end` an exception has traversed?
```ruby
def show_depth
yield
rescue NoMatchingPatternError => e
puts "Depth: #{e.depth}"
raise
end
show... -
- Oh, sorry, I didn't see the `_ref` part. So you are proposing a different method. In my mind, they are all ways of sending a message, it's just a question of method to cross the Ractor frontier. move/copy/has-cross-border-visa 😅
Otherwi... -
- marcandre (Marc-Andre Lafortune) wrote in #note-2:
> ko1 (Koichi Sasada) wrote in #note-1:
> ...
not sure what you mean. I mean:
```ruby
def send_ref(obj)
raise unless Ractor.shareable?(obj)
send(obj)
end
```
> We have t... -
- ko1 (Koichi Sasada) wrote in #note-1:
> `#send_ref(message)` and raise an error if message is not a sharable?
There would be an easy way to deep-copy or deep-move an object, right? If you mean that the default should be `pass: :share`, ... -
- `#send_ref(message)` and raise an error if message is not a sharable?
-
- Yes, that looks good, that is what I was proposing in note 1 too 👍
-
- I don't think providing `#inspect` is good idea because any methods should respond to the moved object.
Otherwise there is possibility to ignore the accessing moved objects.
I prefer to show the moved location in the error message.
... -
- Great, thank you for the update.
One last concern with `experimental_everything`/`experimental_shareable` is for Ruby 3.1... If no longer experimental, we then accept `everything`. But if someone want to write code also compatible wit... -
- By seeing the list, about `experimental_everything`, `experimental_shareable` seems better naming...
-
- By discussion with Matz and Nobu, we choose "(2) raises an error for unshareable non-literals on runtime (my proposal)".
* on compile time: for `FOO = expr`
* case `FOO = literal_only`
* compile to => `FOO = Ruby... -
03:45 AM Feature #17397: `shareable_constant_value: literal` should check at runtime, not at parse time
- marcandre (Marc-Andre Lafortune) wrote:
> I think `shareable_constant_value: literal` is too strict because it has too crude checks at parse time.
This is true. Current restriction is very conservative not to break things (See #1727... -
- Applied in changeset commit:git|76e88480371469400346fca609efe67096813915.
----------
unfreeze Ractor::MovedObject
Matz prefers to unfreeze the class.
[Feature #17401] -
- Matz prefers to provide a free to shoot the foot.
-
- Matz prefers to unfreeze the class.
[Feature #17401] -
- Ractor.make_shareable(obj) tries to make obj a shareable object
by changing the attribute of obj and traversable objects from obj
(mainly freeze them).
"copy: true" option is more conservative approach by make deep
copied object and mak... -
- commit:c2fa024e0220aca6e2437e56df2abf1a2368cbdf
-
- https://bugs.ruby-lang.org/issues/17378
-
07:33 PM Misc #17399: Are endless methods experimental?
- I think the warning would be helpful if we change what is allowed after `=`, especially if we would forbid something that was previously allowed (e.g., to make the rule clearer).
If there is no warning, people will assume it's a stable ... -
06:35 AM Misc #17399: Are endless methods experimental?
- Additional note:
* I am not going to remove endless-def
* I don't think the warning is necessary. Since we are not going to remove it.
Matz.
-
- I agree with @eregon. We still need more investigation for issues like #17398
Matz. -
- Applied in changeset commit:git|cee02d754d76563635c1db90d2ab6c01f8492470.
----------
fix refinements/prepend bug
replaced method entry should be invalidated.
[Bug #17386] -
- replaced method entry should be invalidated.
[Bug #17386] -
-
- I've missed this was committed (https://github.com/ruby/ruby/pull/3824), the API seems fine.
Example usage:
```c
#ifdef HAVE_RB_EXT_RACTOR_SAFE
rb_ext_ractor_safe(true);
#endif
```
https://github.com/ruby/ruby/commit/3d31944129180
I ... -
06:25 PM Revision 0e79d4cd (git): Added tests for [Feature #17314]
- Partially reintroduce 34f06062174882a98ebef998c50ad8d4f7fc0f2e
-
- This reverts commit 41c208d4a463183fddca250026e5f1cd759d2604.
Reintroduce 66090b9d10cdaed917b525225e59d1c19e399248. -
-
06:11 PM Revision 5944c4b3 (git): attr_reader, attr_writer, attr_accessor and attr methods returns array of symbols (#3935)
- Co-authored-by: Yusuke Endoh <mame@ruby-lang.org>
- 06:06 PM Revision f6641d73 (git): NEWS: Add an entry of StringScanner 3.0.0 [ci skip]
- 06:05 PM Revision b3b7f1e5 (git): NEWS: Add an entry of StringIO 3.0.0 [ci skip]
- 06:02 PM Revision f213a271 (git): NEWS: fiddle 1.0.4 [ci skip]
-
05:59 PM Bug #17405 (Closed): irb ---nomultiline gets exception when output contains some non-ascii characters
- When irb is called with the --nomultiline option (for example, when called from emacs), then if the output contains some special characters (such as the square root sign, or a single quotation mark), then irb gets an exception.
Withou... -
05:54 PM Revision 7e2dbbda (git): Use category: :experimental in warnings that are related to experimental features
- This adds rb_category_compile_warn in order to emit compiler warnings
with categories. Note that Ripper currently ignores the category
for these warnings, but by default it ignores the warnings completely,
so this shouldn't matter. -
- Also document that both :deprecated and :experimental are supported
:category option values.
The locations where warnings were marked as deprecation warnings
was previously reviewed by shyouhei.
Comment a couple locations where depreca... -
- Use this to simplify rb_warning_category_from_name.
This also adds support for using the :experimental category in
Kernel#warn and Warning.warn. -
- Since we decided to only allowing specific warning categories,
there is no reason to have an API that accepts a general string,
as it is more error-prone. Switch to only allowing the specific
warning categories.
As rb_category_warn{,in... - 05:44 PM Revision 7b06085c (git): [bigdecimal] Fix test for d5ab8e8562
- 05:29 PM Revision d5ab8e85 (git): [bigdecimal] Use rb_undef_alloc_func to undefine allocate
-
05:12 PM Revision c2a5594e (git): [ruby/reline] Use cached prompt list when just moved the cursor
- https://github.com/ruby/reline/commit/cfe619460b
-
- This closes ruby/reline#106.
https://github.com/ruby/reline/commit/2549a52e15 -
- https://github.com/ruby/reline/commit/3c74beac65
-
- https://github.com/ruby/reline/commit/d208874152
-
- https://github.com/ruby/irb/commit/babb122a48
https://github.com/ruby/reline/commit/e6dbcb3b42 -
05:05 PM Bug #17345: ripper: nothing raised when assigning to keyword variables
- > Now, the '#tok' on some error events are `Elem` instead of string.
pull request: https://github.com/ruby/ruby/pull/3936
-
-
-
03:34 PM Revision 19a98a87 (git): Fixed not to make non-literal expression shareable [Feature #17273]
- Non-literal expression which is not a part of a literal expression
is not a subject of `shareable_literal_value: literal`. - 03:04 PM Revision 6945597f (git): * 2020-12-19 [ci skip]
-
-
02:37 PM Feature #17339 (Assigned): Semantic grouping with BigDecimal#to_s
-
- This no longer occurs with the latest version of bigdecimal.
- 01:00 PM Revision c71934f8 (git): [stringio] Version 3.0.0
- https://github.com/ruby/stringio/commit/831be01071
- 01:00 PM Revision e1b5289a (git): [stringio] Add test-unit in the development dependencies
- https://github.com/ruby/stringio/commit/de010fc0e9
- 01:00 PM Revision 3d319441 (git): [stringio] Make stringio Ractor safe
- https://github.com/ruby/stringio/commit/ee3fec7512
https://github.com/ruby/stringio/commit/18dcd045ef
https://github.com/ruby/stringio/commit/18dcd045ef - 12:58 PM Revision 14ca7f63 (git): [openssl] Fix dependencies
-
-
- ossl_bn_ctx is C's global variable and it should be ractor-local
to make it ractor-safe. -
-
-
- gc_finalize_deferred() runs finalizers and it accesses objspace,
so it need to sync. -
-
-
- This reverts commit 81739ad4fdfcc86a769056fec352f27c686fba1b.
-
- This reverts commit 66090b9d10cdaed917b525225e59d1c19e399248.
-
- This reverts commit e042e8460bb9a63c05f938d51e8c7c5345a6f3a4.
-
- This reverts commit 34f06062174882a98ebef998c50ad8d4f7fc0f2e.
-
- solnic (Piotr Solnica) wrote in #note-18:
> marcandre (Marc-Andre Lafortune) wrote in #note-17:
> ...
It looks like you are replying to me, but of course this is Matz' decision. That said I feel that the current name is good, especiall... -
06:25 AM Feature #16274: Transform hash keys by a hash
- marcandre (Marc-Andre Lafortune) wrote in #note-17:
> Closing, since this has been committed by Nobu 🎉 (in b25e27277dc39)
> ...
This is a wonderful addition but is there a chance you'd reconsider it? Instead of making `transform_keys` ... -
-
05:59 AM Revision ffc8a72f (git): Sort URLs by issue numbers [ci skip]
-
05:36 AM Revision f0ec923c (git): Remove --jit-save-temps from MJIT
- It's probably not helping the backtrace.
- 05:25 AM Revision b5de66e1 (git): [strscan] Fix license comment and files
- https://github.com/ruby/strscan/commit/a999f2c6d1
- 05:25 AM Revision 53709639 (git): [strscan] Version 3.0.0
- https://github.com/ruby/strscan/commit/08645e4e77
- 05:25 AM Revision 985f0af2 (git): [strscan] Make strscan Ractor safe (#17)
- * Make strscan Ractor safe
* Add test-unit in the development dependencies
https://github.com/ruby/strscan/commit/3c93c2bebe -
- There is a case to call this function without VM lock acquiring.
-
05:11 AM Revision 8cba3b35 (git): Bump version RubyGems and Bundler in NEWS
-
-
-
- It has not been warned unless `$VERBOSE` yet.
We may need wait one more release if we obey the previous rule.
But deprecations aren't warned even if `$VERBOSE` since 2.7.2, I think we can reconsider that rule. -
- Applied in changeset commit:git|34f06062174882a98ebef998c50ad8d4f7fc0f2e.
----------
Added tests for [Feature #17314] -
-
-
-
-
- Deleted since fc3c60f6081d85f6274986a7a08b59db1515fcb5
-
- The following patch allows `def foo() = puts "bar"`. It brings no parser conflict.
https://gist.github.com/mame/0773bf3938e046e2b608de5fb2a826c8
However, it is not perfect. `private def foo() = puts "foo"` does not parse.
`private... -
- Only one warning is shown for the same Regexp object, so create
different objects to support repeating tests.
http://ci.rvm.jp/results/trunk-repeat20@phosphorus-docker/3290658
12/17/2020
-
- To debug this issue:
https://rubyci.org/logs/rubyci.s3.amazonaws.com/solaris10-gcc/ruby-master/log/20201217T220004Z.fail.html.gz -
09:28 PM Misc #16443 (Closed): GC.compact is not documented
- Applied in changeset commit:git|895b10ad020a788566c6e35431d07e6121e38c87.
----------
Add documentation about GC.compact
[Misc #16443][ruby-core:96395] -
- Hi Victor. I added some documentation in 895b10ad020a788566c6e35431d07e6121e38c87. Please feel free to add more details, but I don't want to describe how the internals work since they may change.
Thanks. -
- [Misc #16443][ruby-core:96395]
-
- I think `except` should always return a Hash, since Matz prefers returning [the base class instead of subclass](https://bugs.ruby-lang.org/issues/6087#note-12) and we're changing a bunch of String methods too.
I will merge https://git... -
- I'd like to `ractor.send(message)` and express that `message` should be shareable. Currently I'm given two choices: `move: true` and `move: false` / nothing, neither of which have an effect if my `message` is shareable, and neither of wh...
-
- 04:20 PM Revision f7a6b460 (git): * 2020-12-18 [ci skip]
-
- gc_verify_internal_consistency() accesses all slots (objects) so
all ractors should stop before starting this function. -
04:16 PM Feature #17398: SyntaxError in endless method
- Eregon (Benoit Daloze) wrote in #note-3:
> So it's probably going to be very difficult to explain the actual condition, other than showing specific examples.
Endless methods definitions don’t support poetry mode? -
- Conceptually, according to the typical definition in computer science, both `puts("bar")` and `puts "bar"` are expressions (i.e., they return a value, and if it was some other method than `puts` it would also not always be `nil`).
It mi... -
- @mame Hmm, haven't thought about it from this perspective... Can you please explain a bit? As far as I can see, in, say, assignment context it behaves like an expression:
```ruby
result = puts 'foo'
# prints "foo", result = nil
```... -
- The body of an endless method must be an expression (called "arg" in the syntax rules of parse.y). `puts("bar")` is an expression, but `puts "bar"` is a statement (called "command" in the syntax rules).
I think it could be a bit confu... -
- I think a warning would better, especially since there were quite a few changes regarding what was accepted after the `=`.
Also, they purely seem redundant to me and I guess to others, so maybe they might be removed one day like the p... -
- follow up to 9908177857a28633d6279c43a1ad4dfedcb98596
-
- They are warned as deprecated since 2.4.
As deprecated warnings are not shown by default since 2.7.2, I think we should clean up longstanding warnings.
https://github.com/ruby/ruby/pull/3927 - 11:24 AM Revision 980f994b (git): [ruby/irb] Change context-mode's default to new mode 4.
- This new mode uses a copy of the TOPLEVEL_BINDING. This is compatible with refinements (contrary to mode 3), while keeping nested IRB sessions separate
https://github.com/ruby/irb/commit/25c731cb2f -
- Requested in Ruby Bug 17377
https://github.com/ruby/irb/commit/698f6eb34a -
- https://github.com/ruby/irb/commit/22717844c0
-
- https://github.com/ruby/irb/commit/ce373417f3
-
- * Fixes #38
https://github.com/ruby/irb/commit/905fb8e52e - 11:20 AM Revision 422e2c72 (git): [ruby/reline] Use plus operator to create unfrozen string
- https://github.com/ruby/reline/commit/ca274b0d37
-
- https://github.com/ruby/reline/commit/2a97ca9362
-
- https://github.com/ruby/reline/commit/dff30a2d44
-
- https://github.com/ruby/reline/commit/e83a3de9ed
-
- Instead of suppressing all warnings wholly in each test scripts by
setting `$VERBOSE` to `nil` in `setup` methods. - 11:04 AM Revision d597d7a8 (git): Fixing glitchy kill ring behavior in irb
- See https://github.com/ruby/irb/issues/85 for details
-
- objspace->obj_to_id_tbl is a shared table so we need to synchronize
it to access. -
- mark needs barrier (stop other ractors), but other GC events don't need
barriers (maybe...). -
- Current synchronization is too much on write barriers.
-
-
-
- Some tunings.
* add `inline` for vm_sendish()
* pass enum instead of func ptr to vm_sendish()
* reorder initial order of `calling` struct.
* add ALWAYS_INLINE for vm_search_method_fastpath()
* call vm_search_method_fastpath() from vm_sen... -
- ```
$ benchmark-driver -v --rbenv 'before --jit;after --jit' --repeat-count=12 --alternate --output=all benchmark.yml
before --jit: ruby 3.0.0dev (2020-12-17T06:17:46Z master 3b4d698e0b) +JIT [x86_64-linux]
after --jit: ruby 3.0.0dev (20... -
06:56 AM Bug #17400: Incorrect character downcase for Greek Sigma
- I have to acknowledge that I 'cut some corners'. It's essentially table 3.17 on p. 151/2 of the Unicode Standard (see https://www.unicode.org/versions/Unicode13.0.0/ch03.pdf).
The problem from the implementation side is that it requir... -
04:26 AM Bug #17400: Incorrect character downcase for Greek Sigma
- Java has complicated opinions as well:
```
class Main {
public static void main(String args[]) {
System.out.println("Σ".toLowerCase());
System.out.println("ΣΣs".toLowerCase());
System.out.println("ΣΣ".toLowerCase... -
- If the word has a single letter (i.e., `"Σ"`), toLowerCase returns `"σ"` instead of `"ς"` even though the letter is at the end of the word. The condition seems more complex.
```
$ node
> "Σ".toLowerCase()
'σ'
``` -
- Yes everybody wants ruby to handle it "correctly". The problem right now is the lack of concrete definition of "correct" here; especially we need a definition of a word boundary.
-
- Oops, my understanding seemed to be wrong. Please forget my previous comment. When the next letter is an apostrophe, the further next letter seems to determine, but I don't understand this behavior from the code of v8. I leave it to an e...
-
- Prior art here is:
https://github.com/elixir-lang/elixir/issues/6437
https://github.com/elixir-lang/elixir/pull/6990/files
Rust
https://github.com/rust-lang/rust/issues/26035
Golang
https://github.com/golang/text/blob/master/cases... -
- Just FYI: I found the special handling code of Greek Sigma in v8. Looks like that it checks if the next character (if any) is a letter or not.
https://github.com/v8/v8/blob/4b9b23521e6fd42373ebbcb20ebe03bf445494f9/src/unicode.cc#L177-... -
- I guess the ultimate reason why this is not implemented in ruby is the corresponding case mapping is commented out in https://unicode.org/Public/UNIDATA/SpecialCasing.txt
So strictly speaking we just follow what Unicode says but well,... -
- We probably don't need to move it when an insn is leaf...
-
-
- MaxLap (Maxime Lapointe) wrote in #note-19:
> matz (Yukihiro Matsumoto) wrote in #note-16:
> ...
Matz says it's rare.
I'm not sure if that is true, but so far nobody shows any evidence about its rarity. -
-
- Are these backward compatibilities still needed?
I think no.
```
$ RUBY_FREE_MIN=1 RUBY_HEAP_MIN_SLOTS=1 ruby2.1 -e0
ruby2.1: warning: RUBY_FREE_MIN is obsolete. Use RUBY_GC_HEAP_FREE_SLOTS instead.
ruby2.1: warning: RUBY_HEAP_MIN... - 04:22 AM Revision ccf9da07 (git): fix doc typo
- s/it's/its
-
- nobu (Nobuyoshi Nakada) wrote in #note-5:
> Your proposal seems `shareable_constant_value: everything` (it's prefixed with `experimental_` now though).
No, that value deep-freezes implicitly all constants. Non-literals are not frozen... -
- Your proposal seems `shareable_constant_value: everything` (it's prefixed with `experimental_` now though).
-
-
- If `Ractor::MovedObject` can be not frozen that would be more helpful than currently. For example, to implement a helpful `inspect` in pure Ruby 3.0 like in #17393
It is currently possible to go around the fact that it is frozen with ... -
- vo.x (Vit Ondruch) wrote in #note-15:
> Should I open separate ticket for the revert and keep this ticket open for continuing the "whole principle" discussion?
Don’t worry, I have reverted that part already. Use this ticket for the ...
12/16/2020
-
11:47 PM Bug #17400 (Assigned): Incorrect character downcase for Greek Sigma
- An issue caused by this bug was first reported at Discourse support community at https://meta.discourse.org/t/unicode-username-results-in-error-loading-profile-page/173182?u=falco.
The issue is that in Greek, there are two ways to dow... -
- @tenderlovemaking? @ko1? Anybody? It is still this way a week before 3.0?
Can somebody at least provide some draft explanations, or should I dig the sources more?.. -
- Discussed with ko1, who said he is afraid that unfreezing the class might create issues, as implementation is quite special. Possible implementation of `#inspect` for Ruby 3.1
Two things to keep in mind:
- usecase for unfreezing the ... -
- They are marked as such in NEWS, but an attempt to define one doesn't produce a warning (unlike other features marked this way, like one-line pattern-matching or find patterns)
-
- After discussion with ko1, the following options came up:
(1) raises an error for non-literals on parse time (current)
(2) raises an error for unshareable non-literals on runtime (my proposal)
(3) just ignore non-literals
(4) = (3)... -
- Ideally, the following would also be accepted:
```ruby
# shareable_constant_value: literal
Map = {
int: Integer,
str: String,
# ...
}
Map.frozen # => true
```
Literals that are arguments to method calls should not b... -
- I think `shareable_constant_value: literal` is too strict because it has too crude checks at parse time.
I wish the following code would parse and run:
```ruby
# shareable_constant_value: literal
class Foo < RuntimeError
end
... -
- This works:
```ruby
def foo() = puts("bar")
```
This does not:
```ruby
def foo() = puts "bar"
# ^ syntax error, unexpected string literal, expecting `do' or '{' or '('
```
Is this intentional or accidental? No... -
- Enabled this flag, maybe accidentally.
-
- rb_enc_from_index(index) doesn't need locking if index specify
ASCII/US_ASCCII/UTF-8.
rb_enc_from_index() is called frequently so it has impact.
user system total real
r_parallel/miniruby ... -
-
- Closing, this has been implemented by Nobu
-
- Thanks for opening this issue.
Ruby 3.0 no longer freezes dynamic strings.
-
04:18 PM Bug #17396 (Closed): Missing .gemspec files
- I am missing every .gemspec file, which used to be installed by bundled gems. IOW previously there were installed files such as rexml.gemspec and rss.gemspec. This is presumably caused by commit:91f831ac17c8ff58d536ce1fa320294bb477d2ae. ...
-
- Closing this, mainly for performance reason; best use `Matrix#minor`.
-
- This has (finally) been fixed, thanks for opening the issue.
-
- Closing for lack of a viable solution
- 03:42 PM Revision d5929b39 (git): Make Ractor#receive_if private
-
- ractor_sleep() can remain wait.status by interrupt, so that this
patch handles more correctly.
This patch fixed this kind of assertion failures:
Assertion Failed: ../src/ractor.c:1332:ractor_yield_atexit:cr->sync.wait.status == wait... - 03:31 PM Revision 84c9ebd6 (git): * 2020-12-17 [ci skip]
-
- define HAVE_RB_EXT_RACTOR_SAFE to check rb_ext_ractor_safe() is
available or not. -
- * Inline getconstant on JIT
* Support USE_MJIT=0 -
-
- [Bug #17345]
-
- Applied in changeset commit:git|47328ad217ecaf240f0faedac89723dcd6a917ac.
----------
Ripper: Fixed erred token on wrong alias [Bug #17345] -
-
- Sharing error messages between ruby core and ripper is fine, thank you!
Now, the '#tok' on some error events are `Elem` instead of string.
```
$ ./ruby -rripper -ve 'pp Ripper::Lexer.new("alias $x $1").scan'
ruby 3.0.0dev (2020-12-... -
-
-
- Applied in changeset commit:git|a9a7f4d8b8ec30abc7a47ce700edc7209ae12279.
----------
Ractor#receive_if to receive only matched messages
Instead of Ractor.receive, Ractor.receive_if can provide a pattern
by a block and you can choose th... -
- Instead of Ractor.receive, Ractor.receive_if can provide a pattern
by a block and you can choose the receiving message.
[Feature #17378] -
-
- rb_w32_pipe_exec was removed in 1.9.3, at
commit:7ac32d7a16734ea66de15319bcff2fd429abae7f. -
-
-
- to avoid SEGV like
http://ci.rvm.jp/results/trunk-mjit@phosphorus-docker/3289588
by a race condition between mjit_recompile and compation around active_units -
- Ractor#receive can be called by the another Ractors using send,
so making this method completely same as `Ractor.receive` even if
the ractor is specified by the receiver (OO term :p). -
- 06:32 AM Revision 19157a3c (git): vm debug: dump registers on ARM too. (#3900)
- * vm debug: dump registers on ARM too.
-
- specify inline/noinline code for is_attr condition.
- 04:43 AM Revision 1bafb3cb (git): [memory_view] Make MemoryView API Ractor-safe (#3911)
- * memory_view.c: make Ractor-safe
* test/ruby/test_memory_view.rb: Add test_ractor
* memory_view: fix typo
* memory_view.c: Use st_update in unregster_exported_object
* memory_view: update dependency -
-
- * make rb_init_iv_list() simple
* introduce vm_setivar_slowpath() for cache miss cases
../clean/miniruby is 647ee6f091.
Calculating -------------------------------------
./miniruby ../clean/miniruby ../ruby_2_7/... -
-
-
-
- * vm_sync_lock
* vm_sync_lock_enter
* vm_sync_lock_enter_nb
* vm_sync_lock_enter_cr
* vm_sync_barrier -
-
12/15/2020
-
- Applied in changeset commit:git|ccc828f49956424b2548e32cb3bc3a78e16e207b.
----------
configure.ac: avoid squashing CXX=g++
We are discussing this issue at [Bug #17337] but in the meantime, leave
this questionable autoconf glitch as-is ... -
- We are discussing this issue at [Bug #17337] but in the meantime, leave
this questionable autoconf glitch as-is to save sassc and eventmachine. -
- Closing, since this has been committed by Nobu 🎉 (in b25e27277dc39)
I just added doc (in 8558d5e48) - 10:09 PM Revision 8558d5e4 (git): Document Hash#transform_keys with hash. Amend NEWS [DOC] [ci skip]
-
05:55 PM Bug #17030 (Closed): Enumerable#grep{_v} should be optimized for Regexp - Applied in changeset commit:git|d5f0d338c7b5d3d64929b51d29061d369550e8c4.
----------
Optimize `Enumerable#grep{_v}`
[Bug #17030] - 05:54 PM Revision d5f0d338 (git): Optimize `Enumerable#grep{_v}`
- [Bug #17030]
-
- Applied in changeset commit:git|a039dc018ccf34e217f767eac500b9400c58f069.
----------
[DOC] Described "numeric representation" more precisely [ci skip]
[Bug #17395] -
02:22 PM Bug #17395 (Closed): Kernel::Integer does not raise ArgumentError for strings with whitespace padding
- According to the Kernel::Integer docs, ArgumentError should be raised if the argument is a String and doesn't conform with numerical representation. However, it appears that Kernel::Integer does not raise ArgumentError when a string cont...
-
- [Bug #17395]
- 03:04 PM Revision efc6a4e5 (git): * 2020-12-16 [ci skip]
-
-
02:13 PM Revision 8f6cb5b7 (git): TEST: multiarch support for Haiku
-
-
-
- 12:00 PM Revision 2e436982 (git): Fix typo in NEWS.md
-
-
-
- 556a7285080c1344c75bb93a333c9bfc5d631c61 was not good maybe because it
wasn't using list_for_each_safe. If list_for_each_safe is safe for
list_del for any nodes (is that true?), this should be fine. -
- BTW, `rb_category_warn` and `rb_category_warning` take the category as a string instead of `rb_warning_category_t`?
I was going to move it under "include/" when exposing functions that use it... -
12:58 AM Bug #17388: Doesn't Warning.warn support `category: :experimental` in Ruby 3.0.0?
- jeremyevans0 (Jeremy Evans) wrote in #note-2:
> If we are going to add support for an :experimental category, we should probably change the cases that respect `Warning[:experimental] = false` to use the :experimental category when emitt... -
- 06:17 AM Bug #17264 (Closed): BigDecimal exponentiation cannot be used with #** method
- Applied in changeset commit:git|a86c147579745859ea064ec22b2901a7ac7e4abf.
----------
Import bigdecimal 2.0.2 (#3905)
* remove duplicated include
* Make BigDecimal#round with argument < 1 return Integer
Fixes [Bug #12780]
* Use a hig... - 06:17 AM Revision a86c1475 (git): Import bigdecimal 2.0.2 (#3905)
- * remove duplicated include
* Make BigDecimal#round with argument < 1 return Integer
Fixes [Bug #12780]
* Use a higher default precision for BigDecimal#power and #**
When a fractional power is given, increase the precision if the
pre... -
- and fix inconsistent indentation in mjit_compile.inc.erb
-
- add cc_found_in_ccs (renamed from cc_found_ccs), cc_not_found_in_ccs,
call0_public, call0_other debug counters to measure more details.
also it contains several modification. -
- `cd` is passed to method call functions to method invocation
functions, but `cd` can be manipulated by other ractors simultaneously
so it contains thread-safety issue.
To solve this issue, this patch stores `ci` and found `cc` to `calli... -
03:38 AM Revision 40b7358e (git): Skip defined check in NODE_OP_ASGN_OR with ivar
- Previously we would add code to check if an ivar was defined when using
`@foo ||= 123`, which was slower than `@foo || (@foo = 123)` when `@foo`
was already defined.
Recently 01b7d5acc702df22d306ae95f1a9c3096e63e624 made it so that
acce... -
-
-
- I disagree with the proposal for a few reasons:
* It would introduce a huge compatibility breakage. We cannot accept that.
* I believe Lisp-2 behavior is more natural for OOP (not much for FP, I admit)
* It wouldn't improve performa... -
01:01 AM Revision c59b9a8c (git): Move docs for Integer#bit_length [ci skip]
-
- This was on top of `Init_Exception()`.
12/14/2020
-
-
-
- I'm using `
-
11:04 PM Bug #17394: TCPServer is not thread safe on win32
- I've added a screenshot from windows virtual machine on mingw64 env.
-
- PS This issue is not 100% reproducible, please try to run script several times, thank you.
-
- Hello, it looks like win32 version of TCPServer is not thread safe. I've extracted the following code from project:
``` ruby
require "socket"
require "parallel"
::Parallel.each(1..2, :in_threads => 2) do
::TCPServer.open 0 do ... -
-
- Maybe refining the message from `method_missing` is the best and simplest. (giving info that object was moved, and where).
I would also unfreeze the class unless there's a good reason not to. -
- It could be helpful to define `Ractor::Moved#inspect` and output the source location of when the data was moved. If preferred, it could raise an error with this information:
```
x = []
Ractor.new{ receive }.send(x, move: true)
p x ... -
08:47 PM Feature #17392: Is there any plan to unify the namespace after ruby3
- > but few people define variables and methods as the same name, right?
I don't know - do you? Maybe we should measure it?
> ...
I don't know if it's better and does lead to less confusion - does it? Again maybe we should measure it... -
08:30 PM Feature #17392 (Rejected): Is there any plan to unify the namespace after ruby3
Hello.Currently, methods and variables in ruby are separated (lisp-2 semantics), but few people define variables and methods as the same name, right?
Although some people may do this, should we unify the namespace for the better devel...-
08:45 PM Bug #17388: Doesn't Warning.warn support `category: :experimental` in Ruby 3.0.0?
- zverok (Victor Shepelev) wrote in #note-1:
> In addition:
> ...
`Warning.[]` does not use a hash, it uses an if/else construction. So there wasn't an existing hash I could use, I had to create a new hash.
I'm not opposed to support... -
- In addition:
```ruby
warn('foo', category: :deprecated) # works
warn('foo', category: :experimental)
# ArgumentError (invalid warning category used: experimental)
```
It all comes down to this commit: https://github.com/ruby/ru... -
08:19 PM Feature #16986: Anonymous Struct literal
- A slick syntax like `${a: 0, b: 0}` would encourage many to use it when they otherwise might turn to a Hash literal. Performance aside, a dollar sign Struct literal looks relatively lovely.
-
- Applied in changeset commit:git|c183288754fdad17e627c4182de599d965e99405.
----------
Remove references to Random::DEFAULT in the documentation
* Random::DEFAULT is no longer a Random instance, and referencing it is
more confusing tha... -
- * Closes [Feature #17351].
-
- * Random::DEFAULT is no longer a Random instance, and referencing it is
more confusing than helpful. Related to [Feature #17351] -
- no6v (Nobuhiro IMAI) wrote in #note-2:
> Thanks for the quick fix, however `*_error` parser events except `parse_error` seem not to set an error message properly.
> ...
Won't to fix this?
This breaks `irb --colorize` which uses `Rippe... -
- 03:05 PM Revision 7ca82ff8 (git): * 2020-12-15 [ci skip]
-
- RDoc::Markdown requires all block elements nested under bullet
list to be indended deeper than the first column of the list. -
- It seems that nested points need four spaces
-
- shyouhei (Shyouhei Urabe) wrote in #note-14:
> But in the meamtine we could revert this particular C++ compiler detection business to help eventmachine & sassc. Thoughts?
+1
Should I open separate ticket for the revert and keep t... -
- We need to discuss the "whole principle" part. But in the meamtine we could revert this particular C++ compiler detection business to help eventmachine & sassc. Thoughts?
-
02:30 PM Feature #16697: Hash.ruby2_keywords_hash?(value) should support any object
- @mame Thanks for clarifying! Ok let's see :)
-
02:17 PM Bug #17385: Test failures on gcc 11
- > > Applied in changeset git|68af5235292f82145c599038b4bff7d9225e62d7.
> ...
I confirmed the commit `68af5235292f82145c599038b4bff7d9225e62d7` worked with gcc-11 on Fedora 34 now! Thanks.
-
- > JFYI tests for various compiles in our CI suite currently stick to Ubuntu bionic (18.04) because this was the last release which shipped gcc 4.x. OTOH gcc 11 on Ubuntu seems hirsute (21.04) exclusive for now.
I see. Thanks for the i... -
- jeremyevans0 (Jeremy Evans) wrote in #note-17:
> nobu (Nobuyoshi Nakada) wrote in #note-16:
> ...
I don't claim we should switch to 2.70 (at least now), but we'll need to be 2.70-ready.
It may be necessary when the platform used by re... -
- nobu (Nobuyoshi Nakada) wrote in #note-16:
> shyouhei (Shyouhei Urabe) wrote in #note-15:
> ...
What are the advantages of switching to autoconf 2.70 compared to sticking with 2.69? OpenBSD uses gcc 4.2 on many older platforms where c... -
- shyouhei (Shyouhei Urabe) wrote in #note-15:
> JFYI tests for various compiles in our CI suite currently stick to Ubuntu bionic (18.04) because this was the last release which shipped gcc 4.x. OTOH gcc 11 on Ubuntu seems hirsute (21.04... -
- jaruga (Jun Aruga) wrote in #note-14:
> By the way, if you want to try gcc-11 on Ubuntu, I found gcc-11 on ubuntu-toolchain-r repo.
> ...
JFYI tests for various compiles in our CI suite currently stick to Ubuntu bionic (18.04) because ... -
-
-
-
-
-
-
-
-
-
-
-
-
-
- It does shallow freeze only for now.
-
-
- mc_inline_miss_same_def is added to check same method or not.
Also the mc_inline_miss_same_cc calculation was fixed. - 09:26 AM Revision e889c025 (git): Add information to Caveats section in NEWS.md
-
- ccs_not_found to count not found in ccs table.
-
-
- It's not working
http://ci.rvm.jp/results/trunk-mjit@phosphorus-docker/3288206. I'm
debugging why. -
-
-
- rb_method_boundp (method_boundp) searches method_entry, but this
search did not use pCMC, so change to use it. -
-
- counts for invalidating negative cache.
-
- pCMC doesn't have negative method cache so this patch implements it.
-
- ccs is created with default 4 capa, but it is too much, so start
with 0 and extend to 1, 2, 4, 8, ... -
12/13/2020
- 11:17 PM Revision 66756c6d (git): * 2020-12-14 [ci skip]
- 11:17 PM Revision ea6856dc (git): Remove "2.8" from NEWS.md
-
- `File.exists?` has been deprecated since Ruby 2.1.0. I think the release of Ruby 3.0.0 might be a good time to remove it.
-
- **GitHub PR: https://github.com/ruby/ruby/pull/3895**
Copying from PR descriptioin:
Currently, ractors documented only by [doc/ractor.md](https://docs.ruby-lang.org/en/master/doc/ractor_md.html), which has a flavor of a design/disc... -
01:58 PM Feature #14912 (Closed): Introduce pattern matching syntax
-
- So another option would be to move the `timeout` stdlib to core, which could be interesting (can be better optimized, avoid an extra Ruby thread, etc).
-
- nobu (Nobuyoshi Nakada) wrote in #note-11:
> It is just one line to built-in `Timeout::Error`.
> ...
I think that would be confusing, if `Timeout::Error` is in core, and so a `Timeout` module is always defined, and yet `Timeout.timeout... -
- https://github.com/ruby/ruby/pull/3795 is merged, closing.
IMHO this is not a bug but additional behavior, I don't think it makes sense to backport. -
-
- ```
ruby -e 'puts readlines.sort_by {[_1[%r[(https?://.*?)(?:/\d+)?$],1], _1[/\d+$/].to_i]}'
``` - 09:18 AM Revision a3adb10a (git): Add NEWS about open-uri and Kernel#open
- There might be a lot of codes using redefined Kernel#open via open-uri, so it's worth mentioning that.
-
-
-
- I am pretty positive. I will discuss it with Matz after I finish to implement.
-
- Since UNALIGNED_MEMBER_ACCESS assigns to an intermediate variable,
it can cause unused-value warnings. -
- Applied in changeset commit:git|88f3ce12d32ffbef983b0950743c20253ea2d0c6.
----------
Reintroduce `expr in pat` [Feature #17371] -
-
- https://github.com/ruby/dev-meeting-log/blob/master/DevelopersMeeting20201210Japan.md#feature-17371-reintroduce-expr-in-pat-ktsj
-
- to see a C backtrace.
-
- Because Ruby often fails to dump a C backtrace.
-
- jeremyevans0 (Jeremy Evans) wrote in #note-4:
> jnchito (Junichi Ito) wrote in #note-3:
> ...
Thank you for your comment. As for this issue, I should fix the Japanese document, so I submitted PR: https://github.com/rurema/doctree/pull/... -
- jnchito (Junichi Ito) wrote in #note-3:
> Therefore, the compatibility is broken (the parameter `message` should be `["my warning\n"]`). Is it a bug or not?
It is only a bug in the Japanese documentation. Unlike `Kernel#warn`, `Warn...
12/12/2020
-
- Eregon (Benoit Daloze) wrote in #note-2:
> The English documentation seems fine:
> ...
Thank you for information. I noticed the first parameter can be string and array according to the method signature.
```ruby
module Warning
de... -
- The English documentation seems fine:
https://docs.ruby-lang.org/en/master/Warning.html#method-i-warn
So I guess the Japanese docs should be updated. -
- The sample code in https://docs.ruby-lang.org/ja/latest/method/Warning/s/warn.html does not work in Ruby 3.0.0. Is it intended behavior? I'm worrying about its compatibility.
``` ruby
warn "hoge"
module Warning
def self.warn(*m... -
- **GitHub PR: https://github.com/ruby/ruby/pull/3891**
Copying from its description:
I propose new documentation approach for new features in Ruby 3.0: non-blocking Fiber and the scheduler.
Right now, the documentation is in a co... - 04:30 PM Revision 92dfe9ae (git): * 2020-12-13 [ci skip]
-
- Before 9189cf5793cd527a86b711d15d5fd0633ec082e1 the result of
`m4_version_compare` was compared to -1, however the `$2` of
`m4_version_prereq` has different meaning and is expanded when
the required version met. -
- > Applied in changeset git|68af5235292f82145c599038b4bff7d9225e62d7.
Thanks! I will try the patch on Monday.
By the way, if you want to try gcc-11 on Ubuntu, I found gcc-11 on ubuntu-toolchain-r repo.
https://launchpad.net/~ubuntu... -
- Applied in changeset commit:git|68af5235292f82145c599038b4bff7d9225e62d7.
----------
Added parentheses to silence sizeof-array-div warnings [Bug #17385]
As well as 2366c681166a1dab95de6b9ca8ffcaae18aadd39. -
- jaruga (Jun Aruga) wrote:
> ```
> ...
I've missed the same note. -
- jaruga (Jun Aruga) wrote in #note-9:
> ```
> ...
As this call trace is same as the before, the patch doesn't seem to be applied. -
-
-
-
- I wrote the following script:
```ruby
module Warning
def self.warn(message, category: nil)
puts "category=#{category}"
super
end
end
lambda(&:foo).lambda?
Ractor.new{}
```
Then, run it:
```
$ ruby -w ~/Desk... -
- Use regular `AC_CHECK_MEMBERS` instead of:
* `AC_STRUCT_ST_BLKSIZE`
* `AC_STRUCT_ST_BLOCKS`
* `AC_STRUCT_ST_RDEV` -
-
- Now we require C99, these features available of course.
* prototypes
* stdarg prototypes
* token pasting
* stringization
* string literal concatenation -
- hoping to improve C backtrace outputs
-
- As well as 2366c681166a1dab95de6b9ca8ffcaae18aadd39.
-
05:12 AM Revision 52bca75c (git): Remove unused function declarations
-
- The almost same issue #13663 has been closed.
-
- #13663 でほぼ同様のチケットが close されたのでこちらも閉じます。
-
- To fix
http://ci.rvm.jp/results/trunk-mjit-wait@phosphorus-docker/3286265 -
-
- jbeschi (jacopo beschi) wrote in #note-9:
> Anyways I'd wait for the devs meeting outcome and see if we can move this forward.
My comment #7 is the meeting outcome. Some committers including matz discussed this issue, but we were not... -
- At the December 2020 development meeting, @matz agreed the current behavior is expected and this is not a bug.