Activity
From 11/11/2020 to 11/17/2020
11/17/2020
-
11:24 PM Feature #17326: Add Kernel#must! to the standard library
- jeremyevans0 (Jeremy Evans) wrote in #note-14:
> But the fact that there should be a high bar is something I would hope we all understand.
I hope that something I wrote didn't make you think I disagree there! I was just genuinely curiou... -
09:46 PM Feature #17326: Add Kernel#must! to the standard library
- jez (Jake Zimmerman) wrote in #note-13:
> jeremyevans0 (Jeremy Evans) wrote in #note-12:
> ...
The cost to adding a method to a core class is in the conceptual overhead, backwards compatibility, and need to support in perpetuity, since... -
- jeremyevans0 (Jeremy Evans) wrote in #note-12:
> the benefit of adding it is smaller than the cost of adding another method to Kernel
Can you speak more on the const of adding a method to Kernel? While I understand the costs of somethi... -
- This method is trivial to write in Ruby. While it can be useful even in codebases that do not use static typing, I think the benefit of adding it is smaller than the cost of adding another method to Kernel. I think it should probably b...
-
- I really don't want to add new syntax for something that can already be expressed in normal Ruby code.
I wanted to bump this suggestion from Ufuk:
ufuk (Ufuk Kayserilioglu) wrote in #note-5:
> As for the matching `must` method, I thin... -
03:44 PM Feature #17326: Add Kernel#must! to the standard library
- Just an aside note: not saying something against, or for the proposal, I can't help noticing that abandoned "method reference" idea solved at least (1) and (2) of original ticket, removing some of the necessity of constantly extending co...
-
03:20 PM Feature #17326: Add Kernel#must! to the standard library
- I don't think I can agree with the `|.` operator, but having the raise in a block would make a lot of sense to me.
task.or{raise}.mailing_params.or{raise}.fetch('template_context')
This could also be used to return a default va... -
06:35 AM Feature #17326: Add Kernel#must! to the standard library
- I like this syntax :
`task|.raise.mailing_params|.raise.fetch('template_context')`
As you said it's consistent with `&.` -
11:13 PM Revision 64926d50 (git): test/net/smtp - use TCPSocket when UNIXSocket unavailable
- 10:17 PM Revision 2ef3b979 (git): NEWS: Add --backtrace-limit option [ci skip]
-
09:58 PM Revision 898aff95 (git): Remove NEWS entry about taint deprecation warnings [ci skip]
- JunichiIto on GitHub correctly pointed out this is no longer
accurate due to the change to not display deprecation warnings by
default. -
08:30 PM Misc #17329 (Closed): Doc for pattern match?
- Ah, I confused both sites. Closing this then.
-
- > Who has access to do that?
If you mean access to ruby-doc.org, I believe it is one-person project. Last time I wanted something to be fixed there I've just contacted maintainer on Twitter: https://twitter.com/zverok/status/121051255... -
- Ohh, glad to see it's there, not updated.
There have been many improvements by @BurdetteLamar, we should definitely update the published doc as it is today.
Who has access to do that? -
- (As an aside note, I really believe Ruby needs some kind of "Documentation working group", the current state of things is errrrm sub-optimal.)
-
- Funny.
Here: https://docs.ruby-lang.org/en/master/doc/syntax/control_expressions_rdoc.html#label-case+Expression we have
> Since Ruby 2.7, case expressions also provide ....
I [added the docs](https://github.com/ruby/ruby/pull/2... -
- Is there an official doc for pattern matching?
I note that `case` in https://ruby-doc.org/core-2.7.2/doc/syntax/control_expressions_rdoc.html does not mention it, and `in` does not appear. -
- In #17323#note-5, ko1 mentioned there is a possibility to provide "fork" model. So I tried thinking if it could apply here.
We can imagine that accessing an auto-shareable constant
a) from non-main ractor: is made shareable
b) from ... - 06:53 PM Revision 0829f147 (git): * 2020-11-18 [ci skip]
-
06:52 PM Revision 5e3259ea (git): fix public interface
- To make some kind of Ractor related extensions, some functions
should be exposed.
* include/ruby/thread_native.h
* rb_native_mutex_*
* rb_native_cond_*
* include/ruby/ractor.h
* RB_OBJ_SHAREABLE_P(obj)
* rb_ractor_shareable_p(ob... -
06:37 PM Feature #17325: Adds Fiber#cancel, which forces a Fiber to break/return
- n.b. I consider this to be related, at least in spirit, to https://bugs.ruby-lang.org/issues/595. I consider this analogous to IO#open and IO#close. We ought to have a structured/standard way to "close" fibers that have been "opened". (T...
-
- Like I wrote above, I think I'd like to change my PR so that `terminated_fiber.cancel` returns `nil` or `false` instead of raising `FiberError`. That way any fiber on the current thread, except for the root fiber, can be canceled without...
-
- one more update to alternatives
-
- added more to alternatives
-
- Added some other alternatives to the description.
-
- Thanks for taking a look at this, Benoit. I agree it's not obvious why this is necessary with Fiber#raise, so I'll try to explain my reasoning in more detail:
Yes, a library (e.g. `async`) could write a "suspend" function that wraps r... -
- > But even with Enumerator the recursive aspect still represents a compatibility problem.
I am not sure about its severity, though. I mean, Universe is big and sure somewhere in it there should be a code which has an *array of enumera... -
- I understand the thinking behind #flatten; if `ary.flatten` is possible then why not `ary.to_enum.flatten`? It should be isomorphic. But even with Enumerator the recursive aspect still represents a compatibility problem. So as long as th...
-
- @mame @Dan0042 Oh, you are right, starting to think from `Enumerable::Lazy` perspective I've missed a huge incompatibility introduced by `flatten`.
@mame I'll split into several proposals+patches: `Enumerable#compact`, and, I am start... -
- Eregon (Benoit Daloze) wrote in #note-2:
> jeremyevans0 (Jeremy Evans) wrote in #note-1:
> ...
Yes, the same is possible with `Ripper.sexp/sexp_raw`. I've updated the pull request to handle those as well. -
10:10 AM Feature #17276: Ripper stops tokenizing after keyword as a method parameter
- jeremyevans0 (Jeremy Evans) wrote in #note-1:
> Maybe we could support keyword arguments in `Ripper.lex` and `Ripper.tokenize` to raise SyntaxError for errors? Here's a pull request for that approach: https://github.com/ruby/ruby/pull/... -
- Ripper records errors, but `Ripper.tokenize` and `Ripper.lex` cannot return them. Here's how you can handle errors with Ripper (for tokenize, lex is similar):
```ruby
require 'ripper'
r = Ripper::Lexer.new('def req(true) end', 'a',... -
02:24 PM Bug #17197: Some Hash methods still have arity 2 instead of 1
- Ah, I was wrong.
mame (Yusuke Endoh) wrote in #note-6:
> IMO, this is a bug since 1.9.0 because `Hash#select` looks like a faster version of `Enumerable#select`, so it should behave as possible as like `Enumerable#select`.
Hash#select ... -
- Eregon (Benoit Daloze) wrote in #note-8:
> Check if the block uses arity 1 and warn that it should instead use `|k,v|` or `|k,|`?
Yes. It seems obvious, am I missing something? I'm aware that `proc{ |k,| }.arity == 1` (imho a bug) bu... -
12:55 PM Feature #17330: Object#non
- It reminded me https://bugs.ruby-lang.org/issues/12075#change-57152.
-
- (As always "with core" method proposals, I don't expect quick success, but hope for a fruitful discussion)
### Reasons:
Ruby always tried to be very chainability-friendly. Recently, with introduction of `.then` and `=>`, even more ... -
- Maybe, `uname -m` (and tool/config.guess) returns "arm", then `RUBY_UNIVERSAL_ARCH` replaces target_cpu only?
-
- As I have no access to Apple silicon machines, can anyone share config.log file on that platform?
-
- This seems a clear bug, at least `host_cpu` and `build_cpu`.
`arm` means 32-bit and is already used e.g. on Linux to mean that.
So it should be `arm64` or `aarch64`.
`target_cpu` is already `arm64` interestingly.
@nobu Do you kno... -
- https://github.com/ruby/ruby/pull/3777
-
10:25 AM Feature #17314: Provide a way to declare visibility of attributes defined by attr* methods in a single expression
- phluid61 (Matthew Kerwin) wrote in #note-9:
> ``` ruby
> ...
This is exactly why I created this feature request. Improved `attr_accessor` returns array of defined methods and `private/protected/public` receives single array as argument. -
- marcandre (Marc-Andre Lafortune) wrote in #note-8:
> I would also like `alias_method x, y` to return `x` so we could write `protected alias_method :foo, :bar`
Additionaly both activesupport's `delegate` and ruby's `def_delegators` me... -
07:55 AM Feature #17314: Provide a way to declare visibility of attributes defined by attr* methods in a single expression
- Dan0042 (Daniel DeLorme) wrote in #note-7:
> +1
> ...
``` ruby
attr_accessor :x, :y #=> ?
```
-
- I would also like `alias_method x, y` to return `x` so we could write `protected alias_method :foo, :bar`
-
09:05 AM Revision 0683912d (git): Skip tests related TLS with Windows platform.
-
-
- I bisected the fix to commit:a569bc09e25a2ba813d0bec1228d9ff65330a3db. We definitely don't want the entire commit. Attached is a minimal subset of the commit that doesn't introduce a new external function, in case @nagachika would like...
- 12:41 AM Revision fcc88da5 (git): configure.ac: fix for upcoming autoconf-2.70
- The failure initially noticed on `autoconf-2.69d` (soon to become 2.70):
```
$ ./configure
./configure: line 8720: syntax error near unexpected token `fi'
./configure: line 8720: `fi'
```
Before the change generated `./configure ` snip... -
12:39 AM Feature #13381 (Closed): [PATCH] Expose rb_fstring and its family to C extensions
- Applied in changeset commit:git|ef19fb111a8c8bf1a71d46e6fcf34b227e086845.
----------
Expose the rb_interned_str_* family of functions
Fixes [Feature #13381] -
- Fixes [Feature #13381]
-
- You can run it as `ruby run.rb`.
11/16/2020
-
10:41 PM Revision 68ffc8db (git): Set allocator on class creation
- Allocating an instance of a class uses the allocator for the class. When
the class has no allocator set, Ruby looks for it in the super class
(see rb_get_alloc_func()).
It's uncommon for classes created from Ruby code to ever have an
al... -
- Before this commit, `clone` gave different results depending on whether the original object
had an attached singleton class or not.
Consider the following setup:
```
class Foo; end
Foo.singleton_class.define_method(:foo) {}
obj = Foo.n... -
- @Dan0042 How would we warn that case? Check if the block uses arity 1 and warn that it should instead use `|k,v|` or `|k,|`?
I think we should fix `Hash#map` at least, that's also a clear optimization bug like `Hash#each`, and it forc... -
- So when you have a bug that is troublesome to fix because it may result in large incompatibility... rather than "just fix it" and wait to see if other people's code explodes, wouldn't the Industry-Standard Best Practice™ here be to just ...
-
- This story is rather different from `Hash#each` in #14015.
`Hash#each` has yielded an array that has a key and a value since a long time ago:
```
# 2.7
{ a: 1 }.each {|a| p a } #=> [:a, 1]
```
BTW, `Hash#each` has an optimization to s... -
- Applied in changeset commit:git|084e7e31b257f6ac859769553b4c1c6ca2930152.
----------
remain enabled and line specified trace points
If two or more tracepoints enabled with the same target and with
different target lines, the only last ... -
- If two or more tracepoints enabled with the same target and with
different target lines, the only last line is activated.
This patch fixes this issue by remaining existing trace instructions.
[Bug #17302] -
08:14 PM Feature #16786: Light-weight scheduler for improved concurrency.
- I am happy with `Fiber.set_scheduler`. I'm also happy with `Fiber.scheduler=` (canonical Ruby style) but I accept and agree with Matz's reasoning behind the former.
I originally preferred `Thread.current.scheduler=` because it seemed lo... - 05:38 PM Revision 1271782f (git): * 2020-11-17 [ci skip]
- 05:38 PM Revision 1ee8d4b0 (git): Fix typo on Proc docs
-
- This has the potential to introduce security issues. Avoiding security issues by not automatically loading files in the current directory is the reason `.` was removed from the default `$LOAD_PATH` in Ruby 1.9.
-
02:37 PM Feature #17328 (Rejected): Extend `un.rb` to be usable by everyone
- I've opened the PR here: https://github.com/ruby/ruby/pull/3771, although if you need me to submit a patch just let me know.
Basically I want to be able to use `ruby -run` for more than just running a http server, and thought it could r... -
- +1
Ever since ruby 2.1 when `def` was improved to return a Symbol, I've always wondered why the same was not done for `attr_*` methods. -
01:30 PM Misc #16778: Should we stop vendoring default gems code?
- Not that I'm no longer interested, but it sounds like it'd be wasted work.
Even if the reason for rejection doesn't make sense, I think it can be read between lines that the ruby-core team is not interested in this. -
-
12:53 PM Feature #17326: Add Kernel#must! to the standard library
- Since this feature is reminiscent of the safe navigation operator `&.`, I think it would be good if we can make the notation similar to the latter notation or use some variant of it. I suggest to make the following two changes to the cur...
-
12:25 PM Feature #17326: Add Kernel#must! to the standard library
- Given that Ruby 3 will ship with RBS which will be supplanted with a type-checker based on RBS (Steep) soon, I expect more people to experiment with types adoption in their codebases. Our early adoption of type checking at Shopify has sh...
-
- I like this feature as a tiny tool for debugging and assertion.
When I encounter `undefined method 'foo' for nil:NilClass` on `x.foo`, I insert `Kernel#p` to many places, such as `x = expr` to `x = p(expr)`, to try identifying where the... -
- There is `not_nil!` in Crystal https://github.com/crystal-lang/crystal/blob/master/src/object.cr#L223
-
12:31 AM Feature #17326: Add Kernel#must! to the standard library
- - Yes I think we need this feature (except the name)
- Re: naming, the rule of a bang method is:
> A bang method cannot exist alone. If there is `#must!`, there shall also be `#must` which is a "safer" variant of it.
... -
- Agreed with @byroot (actually I was going to note the same, he beat me to it).
I would also like to note that different from `Set`, for example, the order of items in the supplied parameter matters in the `Queue` case. Even though th... -
- I agree that the constructor should take an enumerable rather than variadic arguments, as it would be consistent with `Set.new([1, 2, 3])`, and `Array.new([1, 2, 3])`
-
-
08:44 AM Revision bcd29fed (git): Update TypeProf to 0.5.1
-
-
- I think `Enumerable#compact` is trivial and maybe useful.
In regard to `Enumrable#flatten`, I agree with @Dan0042's concern. I think that it should flatten only `Array` elements, but it might look unnatural.
I'm unsure if `Enumerable#p... -
- Dan0042 (Daniel DeLorme) wrote in #note-4:
> The problem with the example above is that it's too magical to have a variable or constant return an object different from what was assigned.
Absolutely. By ractor's design, there is a pos... -
02:01 AM Revision 0ba096df (git): Fix a link [ci skip]
-
11/15/2020
-
- Applied in changeset commit:git|f5bb9115a7f69c32089b9b970933c3507fb9f82b.
----------
Use more specific warning for ambiguous slash
Fixes [Bug #17124] -
- I think your wording is a big improvement 👍
I think you've paid the cost already with your diff, we should reap the benefit by committing it! -
- marcandre (Marc-Andre Lafortune) wrote in #note-3:
> I hadn't thought of `method /a /x` or `method /a /<newline>other_method`, indeed.
> ...
No, the warning is used for `+` and `-` in addition to `/`.
> If so, could we reword the wa... -
- Fixes [Bug #17124]
-
- This sounds like Thread#kill but for Fiber.
However, Fiber is cooperative so it seems this shouldn't be needed.
What's a concrete issue with Fiber#raise? It seems redundant with it to me.
If the application swallows Exception I thin... -
- I've updated the description to 1) better describe the motivation and 2) match the PR's current behavior when a Fiber is canceled multiple times.
-
- updated description and github PR
-
- That all makes sense to me.
-
10:13 PM Feature #17327: The Queue constructor should take an initial set of items
- I'm not worried about `Queue.new(worklist)` or `Queue.new(*worklist)`, so that's fine if more people feel that way. I think the key thing is conciseness in text source code, and also avoiding needing to synchronise while adding each indi...
-
- What about a way to bulk add items, e.g. `q.concat` or whatever is the same for Array, and maybe it would be best to have the first argument as an array, e.g. `Queue.new(worklist)`? I think it's more expensive to expand it in CRuby when ...
-
- https://github.com/ruby/ruby/pull/3768
-
- I often create a `Queue` and then process it with a set of concurrent workers in threads. I end up writing:
```ruby
q = Queue.new
worklist.each do |work|
q.push work
end
```
I'd rather be able to write
```ruby
q = Queue.... - 10:45 PM Revision 45fd53e0 (git): * 2020-11-16 [ci skip]
- 10:44 PM Revision fd46ff9d (git): NEWS: merge Range and Regexp being frozen [doc]
-
-
-
- 01:04 AM Revision cd50ff80 (git): * 2020-11-15 [ci skip]
-
11/14/2020
-
- # Abstract
We should add a method `Kernel#must!` (name TBD) which raises if `self` is `nil` and returns `self` otherwise.
# Background
Ruby 3 introduces type annotations for the standard library.
Type checkers consume these a... -
- I also created a github PR at https://github.com/ruby/ruby/pull/3766.
-
- This is my first big foray into the ruby C API, so I hope it makes sense what I was trying to do. Did I handle `vm_fiber_break` and `fiber_start` appropriately? Is there is somewhere more appropriate to save the cancel reason than a new ...
-
Calling `Fiber#cancel` will force a fiber to return, skipping rescue and catch blocks but running all ensure blocks. It behaves as if a `break` or `return` were used to jump from the last suspension point to the top frame of the fiber....-
- @chrisseaton I do!
Everything that is good for the community is good by me. -
- @zverok are you making available the code in https://github.com/zverok/enumerator_generate available under the same licence as Ruby? We'd like to just run that proof-of-concept code as-is in TruffleRuby (it still passes all the specs.)
-
-
01:41 AM Bug #17318: Raising float to the power of other issue
- That depends on your purpose or background of the problem. In non-math situations, `power(a, b)` should not return any ordinary number for given `a`<0 where `power(a, b)` stands for "`a` to the power of `b`" over real, just identical to ...
11/13/2020
-
11:26 PM Feature #17100: Ractor: a proposal for a new concurrent abstraction without thread-safety issues
- `send` is a term traditionally used in the actor context. We'd like to experiment how it works well (or bad) for Rubu3.0.
Matz.
-
- Thank you for the detailed explanation, that makes sense to me.
I am sure it will result in some confusion, but hopefully it will be rare in practice.
ko1 (Koichi Sasada) wrote in #note-45:
> * Matz agreed that `__send__` is ugly (*2), ... -
- # `Ractor#send` naming issue
Recent weeks I'm thinking about the name of `Ractor#send`.
Matz and I discussed possibilities long time (this discussions and considerations stop my developments), and we agree to remain the name `Ractor#... -
- deXterbed (Manoj Mishra) wrote in #note-3:
>
> ...
Don't raise a negative number to a non-Integer power. That's just maths. -
- deXterbed (Manoj Mishra) wrote in #note-3:
> ```
> ...
I don't think that this should be the default behavior because `raise(-1, 2)` returns `-1` but I expect the square of `-1` to be `1`. -
- Also see #16677 w/r edge cases.
-0.4**0.9 #=> -0.4383832905540869
-0.4.to_f**0.9 #=> (-0.416927285116376+0.13546788683122327i)
-(0.4).to_f**0.9 #=> -0.4383832905540869
-
03:43 PM Bug #17318: Raising float to the power of other issue
- mame (Yusuke Endoh) wrote in #note-2:
> It is due to precedence of operators unary minus and `**`.
> ...
Thanks for the explanation Yusuke Endoh,
that still raises the question, how exactly am i supposed to calculate x**y without ending... -
08:38 PM Feature #17315: Hash #transform
- to_h works for me even though it is unintuitive a name for monadic mapping a hash. I wish we had a common name function name in Enumerable for monadic mapping that would return the same Class that we map over.
-
- Would it be possible to somehow have ractor-local variables that are automatically dereferenced instead of having to append `.value` everywhere?
I'm thinking of cases like this where a class-level mutable constant (or classvar) makes ... -
- Eregon (Benoit Daloze) wrote in #note-6:
> I think `rand`, etc should use a thread-local `Random` instance (and thread-local implies Ractor-local, of course).
> ...
Bypassing the need for synchronization is a very good idea.
> Is there ... -
- Dan0042 (Daniel DeLorme) wrote in #note-71:
> Fiber.schedulers[Thread.current] = sch
It seems weird to me to use a Thread as a key for some state on "Fiber".
I think there is no point to hide the fact this state is per-thread.... -
- Would this be too crazy?
Fiber.schedulers #=> Hash
Fiber.schedulers.default = sch
Fiber.schedulers[Thread.current] = sch -
- Sorry for the side discussion.
I think this is the important point to discuss:
Eregon (Benoit Daloze) wrote in #note-6:
> I think (2) is a better solution.
> ...
I think in some actor models `send` is considered safe and never ra... -
- ko1 (Koichi Sasada) wrote in #note-9:
> > It is convenient to Ractor.new { ... }.take, but we could have Ractor#join for that purpose, and that would also work more reliably, independent of intermediate Ractor.yield calls.
> ...
I was ... -
- zofrex (James Sanderson) wrote in #note-6:
> I think that people reading "raw_seed" are likely to think that either this returns the seed for the RNG or perhaps that this method is only suitable for seeding another RNG, rather than usin... -
01:20 PM Misc #17319: Rename Random.urandom to os_random and document random data sources
- I don't think `raw_seed` is name that does a good job of communicating what this method does and what it can be used for. I know this method is *used* to seed the OpenSSL random number generator, but its functionality is more broad than ...
-
- How about `Random.raw_seed(size)`?
-
- rb_ivar_get() can issue an uninitialized ivar warning. We never want to
issue warnings about hidden ivars as they are not actionable for users. -
- What would be the interaction between Array#flatten and Enumerable#flatten ?
It's a big compatibility problem if flatten recursively applies to any Enumerable object within an array.
```ruby
x = Struct.new(:a).new(1) #=> #<struct a... -
- A comment about naming... my first impression was that a "basket" is a container with multiple objects to transmit between Ractors. Something like
```ruby
basket = Ractor::Basket.new
basket.to_move << a
basket.to_copy << b
basket.... -
- Applied in changeset commit:git|ce9beb9d20187c861462282460b998684759e5e7.
----------
Improve error message when subclassing non-Class
Fixes [Bug #14726] - 03:06 PM Revision 1bd98ee2 (git): * 2020-11-14 [ci skip]
-
- Fixes [Bug #14726]
-
-
-
05:19 AM Bug #17324: You may have encountered a bug in the Ruby interpreter or extension libraries.
- gundamseedw (J Wong) wrote in #note-1:
> gundamseedw (J Wong) wrote:
> ...
jeremyevans0 (Jeremy Evans) wrote in #note-2:
> The crash information includes:
> ...
It's indeed postgres related. Thanks for the clue!
adding the below in data... -
-
- I've submitted my patch as a pull request: https://github.com/ruby/resolv/pull/1
-
- I've submitted the patch as a pull request, with a test: https://github.com/ruby/open-uri/pull/1
-
- Because the test for this change was still broken.
- 02:38 AM Revision c046cc1d (git): * 2020-11-13 [ci skip]
- 02:36 AM Revision 37e5b367 (git): [ruby/webrick] add mime type of extention .mjs
- https://github.com/ruby/webrick/commit/45d68f9eba
-
- RFC 7230 section 3.3.3 allows for this.
Fixes #30
https://github.com/ruby/webrick/commit/069e9b1908
11/12/2020
-
- In JRuby and in TruffleRuby, Random instances are thread-safe (i.e., they use synchronization internally).
Also, `rand`, etc, use a per-Thread Random instance to avoid needless contention when calling `rand` in multiple threads concurren... -
- It sounds rather intuitive to me that `rand` etc. are not sharable among ractors. It is their nature to mutate process-global state of executions.
If a programmer wants randomness inside of a ractor they needs special care to separat... -
- > Maybe Ractor#[] (like Thread#[]) ?
yes, it is one option (but I don't like this, make a new ticket soon about it with reasons).
However, if we choose `Ractor#[]`, `Ractor.current[:DEFAULT_RANDOM]` seems not good idea (not imposs... -
- ko1 (Koichi Sasada) wrote in #note-2:
> > Why not introduce general-purpose per-ractor variables first, instead of magical method that interacts with the crrent ractor behind-the-scene.
> ...
Both :)
But I understand that new syntax nee... -
- > Why not introduce general-purpose per-ractor variables first, instead of magical method that interacts with the crrent ractor behind-the-scene.
Do you mean new syntax or method?
Just now i'm writing the per-ractor storage class (... -
- Why not introduce general-purpose per-ractor variables first, instead of magical method that interacts with the crrent ractor behind-the-scene.
-
- `Random::DEFAULT` a default random generator used by `rand`, `srand`, `Array#shuffle` without a given random generator, and so on.
Random generators are not thread-safe, so they are not ractor safe, and they are not shareable.
So a p... -
- The crash information includes:
```
Crashed Thread: 4 thread_pool.rb* Dispatch queue: com.apple.security.keychain-cache-queue
```
From what I can find, this can be caused by attempting to access the Mac OS X keychain aft... -
- gundamseedw (J Wong) wrote:
> i've attached the crash report log.
> ...
im using macbook pro 2019. mac os catalina 10.15.7 -
- i've attached the crash report log.
im using
Rails 5.2.4.4
Puma 5.0.4
Ruby 2.5.8 -
- marcandre (Marc-Andre Lafortune) wrote in #note-2:
> - deep-copying the cache is probably faster than having to recalculate part of it
> ...
it depends on the problem. per-ractor cache is faster to access because there are no synchroni... -
- I'm curious for better use cases.
The above example is not very convincing:
- there's no much use in sending this object to a Ractor to start with
- deep-copying the cache is probably faster than having to recalculate part of it
- more... -
- Another advantages compare with current API is,
* we don't need to care about variable name.
* we can make ractor-local constants and instance variables like that:
```ruby
class Fib
attr_reader :cache
def initialize
... -
- Ruby supports thread and fiber local storage:
* `Thread#[sym]` provides *Fiber* local storage
* `Thread#thread_variable_get(sym)`
These APIs can access other threads/fibers like that:
```ruby
th = Thread.new{
Thread.current... -
- https://rubyci.org/logs/rubyci.s3.amazonaws.com/rhel8/ruby-master/log/20201112T123004Z.fail.html.gz
```
1)
File.utime allows Time instances in the far future to set mtime and
atime (but some filesystems limit it up to 2446-05-10) FAILED
... -
12:35 PM Revision 028d52bb (git): [ruby/date] Numeric already includes Comparable
- https://github.com/ruby/date/commit/f6140df0ad
-
- https://github.com/ruby/date/commit/f08175e34d
-
- https://github.com/ruby/date/commit/d20380fc55
-
- https://github.com/ruby/date/commit/3c002b1daa
-
12:34 PM Revision 0fdcb947 (git): [ruby/date] [DOC] declate DateTime class is deprecated
- https://github.com/ruby/date/commit/58ca6e6a3e
-
-
-
-
10:16 AM Bug #17320 (Closed): Is 'make realclean' supposed to work?
- Of course `configure` must be execute prior trying to run `make something`. Oh my. Sorry for the noise.
-
- I hear what you're saying about communicating that it is not coming from `/dev/random`. It's hard to communicate all of that in a method name, because it will be a different source on different OSes or OS versions.
> If we rename the me... -
- Also until very recently (until version 5.6), Linux kernel has had distinct /dev/random and /dev/urandom. Some people argued the difference between those two are important. If we rename the method to `os_random`, that loses information...
-
- +1 for doc update. Naming wise, ruby is rather POSIX-centric. For instance it has IO.select which works on Windows as well.
- 08:00 AM Revision 98de98fb (git): * 2020-11-12 [ci skip]
-
- A fix to 54fb8fb62a30c7b60ab6443a62821f6f8bc479c4
-
- Partially revert 54fb8fb62a30c7b60ab6443a62821f6f8bc479c4
-
- marcandre (Marc-Andre Lafortune) wrote in #note-5:
> > What about allowing Kernel#instance_variable_set to take a block instead of the second argument, in which case the assignment should be done only when the instance variable is not d... -
- Ractor is designed to manage blocking operations by
* receive
* yield and take
and they can be multiplex with `Ractor.select`. I don't want to introduce more.
I think yielder/taker can communicate on the protocol which contains... -
- > It is convenient to Ractor.new { ... }.take, but we could have Ractor#join for that purpose, and that would also work more reliably, independent of intermediate Ractor.yield calls.
When do you need `Ractor#join`?
11/11/2020
-
- I had a call with @ioquatix .
He doesn't like `Thread.current.fiber_scheduler=` (mixing too many things).
He says `Fiber.set_scheduler` or `Thread.current.scheduler=` are the best options in his view.
The method names `Fiber.set_schedul... -
## Problem
---
Running the following reproduction script:
```ruby
class Foo
def self.foo; end
end
def report(klass, name)
puts " #{name}.instance_methods(false): #{klass.instance_methods(false)}"
end
def clone_and_co...-
- @hsbt Was it extracted to a gem?
BTW:
```
$ gem-codesearch "require.+Win32API" | wc -l
339
$ gem-codesearch -i "require.+win32api" | wc -l
355
``` -
-
- Eregon (Benoit Daloze) wrote in #note-7:
> Also, it seems a big issue if exceptions in Ractor are silently ignored.
> ...
Sorry, I should have checked before, that's already the case. -
- Also, it seems a big issue if exceptions in Ractor are silently ignored.
If we can't `Ractor.yield` the exception, then I think we should print it much like `Thread.report_on_exception`.
Not having `Ractor#close_outgoing` seems to so... -
- I think (2) is a better solution.
A Ractor should always be able to send messages while it's alive.
Another thought: maybe `Ractor.new { 42 }` should not automatically Ractor.yield the result?
Because `Ractor.new {}` does not wait f... -
- Applied in changeset commit:git|deed21bb08170431891b65fda26f4a3557c9ffd4.
----------
ignore yield_atexit if outgoing port is closed
If outgoing_port is closed, Ractor.yield never successes.
[Bug #17310] -
- At least, I merged (1) patch.
`Ractor#kill` is not acceptable to avoid non-deterministic behavior like introduced by `Thread#kill`.
----
I'm not sure we need `Ractor#close_outgoing`.
One possibility is to make a ractor detached (indep... -
- > What about allowing Kernel#instance_variable_set to take a block instead of the second argument, in which case the assignment should be done only when the instance variable is not defined?
I would like `Kernel#instance_variable_get`... -
- marcandre (Marc-Andre Lafortune) wrote in #note-3:
> I just released a small gem to deal with memoization
Looks interesting. -
- Memoization is tricky, not just for `nil`/`false` values. What about freezing that object? What about calling `Ractor.make_shareable` on it?
I just released a small gem to deal with memoization that:
- works with `nil`/`false` result... -
- I have seen so many attempts to memoize a value in the form:
```ruby
@foo ||= some_heavy_calculation(...)
```
improperly, i.e., even when the value can potentially be falsy. This practice is wide spread, and since in most cases m... -
06:54 PM Feature #17307: A way to mark C extensions as thread-safe, Ractor-safe, or unsafe
- I talked this over with Eregon yesterday in the context of doing this work in TruffleRuby, and we came to the conclusion that there is a real advantage in being able to selectively mark methods and procedures as thread safe or not. For e...
-
- Sorry, hit the enter. Prematurealy.
The DeveloperHowto [1] mentions `make realclean` to remove everything pregenerated. Trying that, it does not work:
~~~
$ make realclean
make: *** No rule to make target 'realclean'. Stop.
~~~... -
- https://github.com/ruby/ruby/blob/fa3670e6e48a8553ad1f37bbfbd112911da497d1/common.mk#L649
-
- I didn't suggest anything, as I don't see a good name, and I don't see the appeal of this method:
- it can not be more efficient than `update` + `to_h`
- there's not a frequent usecase that I know of
- seems like avoidable mutation
-... -
- marcandre (Marc-Andre Lafortune) wrote in #note-2:
> I'll reopen this, as you are asking for a mutating version of `to_h`.
It might be strange to have `transform!` as the destructive counterpart to `to_h`. Perhaps, the naming should be ... -
- [PR here](https://github.com/ruby/ruby/pull/3760)
-
- SecureRandom gets randomness from `Random.urandom`, which is a confusing name because `urandom` only uses /dev/urandom in some circumstances. On reading the `secure_random.rb` code this morning I got very confused how it was supporting "...
-
- It is due to precedence of operators unary minus and `**`.
`x**y` calculates `(-0.4790529833050308)**0.9918032786885246`, and `-0.4790529833050308**0.9918032786885246` calculates `-(0.4790529833050308**0.9918032786885246)`.
```
ir... -
- Formatting replaced the double * with a single * i put in the snippet above, please ignore
-
- Raising a negative float to another float results in a complex number. Interestingly, doing the same thing without using variables works fine!
Sample Snippet:
$ irb
2.6.3 :001 > x=-0.4790529833050308
=> -0.4790529833050308
2.6.... -
- >> The idea behind private attribute methods (which I've seen used in a lot of codebases, and use myself) is uniformity of the call-sequence of different values. I believe Avdi Grimm had a nice article on this point of view (will try to ...
-
- sawa (Tsuyoshi Sawada) wrote in #note-2:
> Private attribute methods defeat the purpose of attribute methods. If you want to access them within the class, you can directly access the instance variables.
> ...
What kind of resource, cpu? ... -
- > Private attribute methods defeat the purpose of attribute methods. If you want to access them within the class, you can directly access the instance variables.
The idea behind private attribute methods (which I've seen used in a lot... -
- > Private attribute methods defeat the purpose of attribute methods
I should have mentioned it, but it is more useful for `protected` than `private`. -
- Private attribute methods defeat the purpose of attribute methods. If you want to access them within the class, you can directly access the instance variables.
>[I]t's convenient to see what kind of attributes [a] class has
For tha... -
- Actually, situation with documentation sites is kind of weird now (to say the least).
* "Official" docs.ruby-lang.org always renders correctly, but other than that is not very usable (because of mixing all the lang and standard librar... -
02:11 PM Bug #17317: In 2.7.2, Hash#except doesn't seem to exist, but is in the documentation - Thanks, and apologies for raising this in the wrong place! I've emailed the site owner.
-
- If I recall correctly, we have received some issue reports about ruby-doc.org maybe because the site looks official to many people. The site is undoubtedly valuable to Ruby ecosystem, but the confusion is a bit unfortunate. I wonder if w...
-
- Thanks for the report, but ruby-doc.org is a third-party project that the ruby-core team is not maintaining. Could you contact on the admin of the site?
- 10:54 AM Bug #17317 (Third Party's Issue): In 2.7.2, Hash#except doesn't seem to exist, but is in the documentation
- According to https://ruby-doc.org/core-2.7.2/Hash.html#method-i-except, Hash#except is available in 2.7.2, but it doesn't appear to actually be there.
Run the following on the command line:
```
ruby -v -e 'h={a:1,b:2}; puts h, h.exc... -
- close_incoming by antoher ractor means there is no other messages
will be sent to the ractor, so Ractor.receive will block forever,
and it should raise and stop.
close_outgoing by antoher ractor means, ... I don't have good idea
to use ... -
- If outgoing_port is closed, Ractor.yield never successes.
[Bug #17310] -
- If a terminating ractor has child threads, then kill all child
threads. -
- The timer function used on windows system set timer interrupt
flag of current main ractor's executing ec and thread can detect
the end of time slice. However, to set all ec->interrupt_flag for
all running ractors, it is requires to synch... -
- chucke (Tiago Cardoso) wrote in #note-5:
> I wonder why `uri` can't do the same. [punycode IDN domains are a standard](https://tools.ietf.org/html/rfc3492), and although `uri` doesn't support it, it could at least return a URI object wi... -
12:29 AM Misc #17309: URI.escape being deprecated, yet there is no replacement
- I may not be in position of the full details of the decision, and what is exactly the full of the deprecated URI.escape . I do have seen previous discussions on what is a URI, what's stated in the RFC, and what's the scope of the `uri` l...
-
-
-