Activity
From 06/21/2019 to 06/27/2019
06/27/2019
-
11:59 PM Bug #11628 (Closed): Net::SMTPServerBusy is not behaving according to documentation
- Applied in changeset commit:git|14b0c94ac38c40459d908fd4c5778efa60e4689f.
----------
Fix documentation for Net::SMTPServerBusy
Fixes [Bug #11628] -
- Fixes [Bug #11628]
-
11:39 PM Bug #11659 (Rejected): Strange behavior setting previously-undefined local variables with a statement modifier
-
-
- `Array#grep` is actually `Enumerable#grep`:
```ruby
Array.instance_method(:grep).owner
=> Enumerable
```
If I had to guess, the cause of the difference is that `Array#each` is implemented in C, and `Test#each` is implemented in ... -
-
- I've made a couple minor tweaks to the pull request: https://github.com/ruby/ruby/pull/2205. Travis and AppVeyvor tests still pass with it. I plan to commit this in a few days, and if there is any fallout, I will address that by fallin...
-
-
- This was fixed in Ruby 2.6:
```
$ ruby26 -e 'File.read "|/bin/ls"'
Traceback (most recent call last):
1: from -e:1:in `<main>'
-e:1:in `read': No such file or directory @ rb_sysopen - |/bin/ls (Errno::ENOENT)
``` -
-
- This isn't a bug. `DateTime.parse` assume the string you are passing in has some date related information in it, and tries the best it can to find something. In this case, it guesses that the first four digits are the year and the last...
-
- This crash occurs in a 3rd party C extension (EventMachine), probably inside this function: https://github.com/eventmachine/eventmachine/blob/v1.0.9/ext/rubymain.cpp#L775. It appears to be the same issue as https://github.com/eventmachi...
-
-
-
- Given that `singleton_method_added` is called after the method is added and not before, I think the current behavior makes sense and do not think it is the bug. The same behavior exists for `method_added`:
```ruby
module A
extend... -
-
09:16 PM Feature #15963 (Closed): Add *_start and *_running methods to VM_COLLECT_USAGE_DETAILS API
- I want to count how many VM instructions get executed for an arbitrary block of ruby code with `VM_COLLECT_USAGE_DETAILS` enabled. However, the current API enables VM usage detail collection immediately and only exposes methods to turn c...
-
08:50 PM Bug #15791 (Closed): Clarify reason for RbConfig's ruby_version not reflecting "teeny" value
- Ok, if this is supposed to map the the ABI version, then I guess we'll leave it as is. It doesn't really have the same meaning on other impls, and I agree with Vit that it shouldn't be used for other purposes (like it seems to be used in...
-
07:21 PM Revision a2091c30 (git): Update to ruby/spec@94d98ff
- 07:04 PM Revision 632a9722 (git): * 2019-06-28
-
-
-
- Actually, the `should_not` above indeed doesn't make much sense, because it relies on `nan_value` returning different NaNs, which is probably platform-specific.
I added a spec for this case: `[Float::NAN].should == [Float::NAN]`. -
- It makes sense in the context of #1720.
I.e., all Ruby implementations behave that way, because `Float::NAN.equal? Float::NAN` must be true, and Array#== uses an #equal? + a #== check, not just #==. I want to restore the spec and add ... -
05:43 PM Bug #15933: OpenURI: Assign default charset for HTTPS as well as HTTP
- [HowToContribute](https://bugs.ruby-lang.org/projects/ruby/wiki/HowToContribute) suggests that I can ping this ticket if it looks like it's been missed - there was some good discussion for a couple of days and then nothing since.
It's... -
04:34 PM Feature #15961: Always warn for URI.{,un}{escape,encode}
- I am not sure if such a verbose message would make sense, even more so as
long as the ruby user has no trivial way to silence a particular warning
(without e. g. the $VERBOSE trick to assign to a new value, make a change,
then re-assi... -
01:32 PM Revision 310a42e8 (git): Explain rubyspec Travis job in job name
- Eregon did not want https://github.com/ruby/ruby/commit/7b9dd379bfe4d597acfe4c26365feff61dabd488
because people only see Travis job name and don't see .travis.yml.
I believe the new job name is simple and informative enough to explain
w... -
11:31 AM Revision 1f7cb4be (git): Omit `EXPR_` prefix to show lex_state
-
10:28 AM Feature #15931: encoding for CESU-8
- naruse (Yui NARUSE) wrote:
> 文字コードの混乱という懸念は10年前ならあったでしょうが、もはやUTF-8が最高にして唯一の文字コードであることは誰の目にも明らかです。
> ...
勇気づけありがとうございます。書き込みの一つの目的は、CESU-8 の立場をはっきりさせることでした。
> 「RubyとJavaのブリッジ」と書かれていますが、リンク先を見るとこれはJNI (Java Native Interface) の話で、要... -
-
-
- naruse (Yui NARUSE) wrote:
> Note that an incompatibility which is caused by the change of string encoding is `String#<<(integer)`.
> ...
If I understand this correctly, the proposal is to change the encoding of Symbols from ASCII to U... -
-
07:03 AM Revision 2fd03fd7 (git): EXPR_LABEL also indicates non-continuation line
- Example:
[
1, # this is not continuation line
2
] -
- Applied in changeset commit:git|097554855c1a9404f3220bcc5c1906cc62399b59.
----------
Fix ripper fatal
* parse.y (parser_yylex): return END_OF_INPUT at unterminated here
document instead of an error. [Bug #15962] -
- It was my mistake and could reproduce. This is Ripper's matter.
-
- I can't reproduce it. Please show me a commit what you tried, and your irbrc.
-
02:39 AM Bug #15962 (Assigned): irb immediately crashes when you type <<' or <<"
-
02:22 AM Bug #15962 (Closed): irb immediately crashes when you type <<' or <<"
- irb crashes right after inputting `<<'` or `<<"` from the top of a line.
The crash happens without inputting Enter key.
This is the log:
```
irb(main):001:0* <<Traceback (most recent call last):
35: from /home/hasumi/o... -
- * parse.y (parser_yylex): return END_OF_INPUT at unterminated here
document instead of an error. [Bug #15962] -
06:23 AM Feature #15865: `<expr> in <pattern>` expression
- mame (Yusuke Endoh) wrote:
> I'd like to avoid to discuss a common topic between one-line and `case/in` in this particular ticket.
Agreed; the gist discusses the entirety of the feature-space, but the conversation in this ticket shou... -
02:47 AM Feature #15865: `<expr> in <pattern>` expression
- Thank you for your comment.
This ticket discusses one-line pattern matching: `<expr> in <pattern>` in my proposal. This proposal is based on the `case/in` pattern-matching statement. I'd like to avoid to discuss a common topic betw... -
- As suggested by Yusuke on Twitter, I'm posting a link to my own personal "wishlist" around pattern matching. I'm happy to discuss any points that might benefit from clarification.
https://gist.github.com/pvande/822a1aba02e5347c39e8e0a... -
-
02:18 AM Revision c86a9e65 (git): bc mode is already removed [ci skip]
-
-
- As suggested by Yusuke on Twitter, I'm posting a link to my own personal "wishlist" around pattern matching. I'm happy to discuss any points that might benefit from clarification.
https://gist.github.com/pvande/822a1aba02e5347c39e8e0...
06/26/2019
-
11:50 PM Feature #15931: encoding for CESU-8
- duerst (Martin Dürst) wrote:
> 念のために言っておきますが、https://www.unicode.org/reports/tr26/tr26-4.html に「It is not intended nor recommended as an encoding used for open information exchange.」と書いてあります。くれぐれもよくご注意ください。
> ...
「RubyとJavaのブリッジ」と書かれてい... -
- 念のために言っておきますが、https://www.unicode.org/reports/tr26/tr26-4.html に「It is not intended nor recommended as an encoding used for open information exchange.」と書いてあります。くれぐれもよくご注意ください。
さらに、https://www.unicode.org/reports/tr26/ に「CESU-8 documen... -
- I don't think this is a bug in the code, but I do think we should improve the documentation for valid_date. I've added a pull request to do so: https://github.com/ruby/date/pull/7.
-
- When using native fibers, do not load ucontext, as it isn't needed.
-
- This isn't a bug. `undef_method` and `remove_method` mean two different things in Ruby:
* `remove_method`: Remove the method with this name from the module/class method table
* `undef_method`: Add an entry to the module/class method... -
- I can't replicate the `NoMethodError` you are receiving on any ruby version:
```
$ ruby18 -ruri -e 'p URI.parse("mailto:?subject=hi")'
#<URI::MailTo:0xab543e6f5c0 URL:mailto:?subject=hi>
$ ruby19 -ruri -e 'p URI.parse("mailto:?subj... -
08:52 PM Feature #15961: Always warn for URI.{,un}{escape,encode}
- I am not sure, but can this warning be more helpful? Like method's documentation, which says:
> This method is obsolete and should not be used. Instead, use CGI.escape, URI.encode_www_form or URI.encode_www_form_component depending on... -
- These methods have both had a verbose mode warning for almost 10 years. If we ever plan on removing them, we should switch from a verbose mode warning to a regular warning. If we don't, people will still use them, and run into their pr...
-
- I don't think this is a bug. It makes sense for `IO#wait_writable` to raise an exception, because the specific `IO` instance will never be writable after `close_write`. As the documentation for `IO.select` indicates, it calls `select(2...
-
- I do not think this is a bug. DateTime has always defaulted to UTC, not just in `DateTime.parse`:
```ruby
DateTime.new.zone
# => "+00:00"
DateTime.strptime('2009', '%Y').zone
# => "+00:00"
DateTime.jd(2458000)
# => "+00:00"... -
-
-
- Fixed in commit:2743c491a7d7b4e015b9e0a2b114856371bdd636.
-
- Fixed by commit:3666ceb729998a6149d335151517f56aef2eba74 and commit:4c6eb271ddecc8534d10449ec2ca40df6bea2e58.
-
- v =
3 # auto indent - 04:21 PM Revision df3c9471 (git): * 2019-06-27
-
-
- OpenBSD does not support ucontext.
There may be a more generic way to handle this, using the result
of AC_CHECK_HEADERS(ucontext.h). -
-
03:00 AM Bug #15960 (Closed): rb_fstring should resize capacity [patch]
- https://github.com/ruby/ruby/pull/2256
When a string is `#frozen`, it's capacity is resized to fit (if it is much larger), since we know it will no longer be mutated we'd like to save the space.
> puts ObjectSpace.dump(String.n... -
-
- 10:02 AM Revision acb67472 (git): Restore updated implementation of arm32 coroutine code, but prefer ucontext.
-
-
- This reverts commit 6df1814c08df93bbc0b3e7a73649bcf82e126064.
It caused a SEGV again:
https://rubyci.org/logs/rubyci.s3.amazonaws.com/scw-9d6766/ruby-master/log/20190626T051707Z.fail.html.gz - 08:19 AM Revision 6c6bf9ff (git): Add `ucontext` coroutine implementation for generic fallback.
-
- This test invokes GC.start 100 times, which takes approx. six minutes in
Solaris. This change runs the test in a separated process, which makes
GC.start faster. -
- When a string is #frozen, it's capacity is resized to fit (if it is much
larger), since we know it will no longer be mutated.
> puts ObjectSpace.dump(String.new("a"*30, capacity: 1000))
{"type":"STRING", "class":"0x7feaf00b7bf0"... -
-
- Retry of 518adcca0a and dbe232e24e
-
- This isn't a bug, this is a request for a feature that doesn't currently exist, which is the ability to connect to a server with an SSL certificate that doesn't match the address you are using to connect to the server. While this is in ...
-
-
- Before this, tool/rbinstall would try to remove a gem file that
it may not have access too.
Fixes issue introduced in f550da512cefbe0283106ca839c4836a98efaf3a. -
-
-
01:30 AM Bug #12764: rubygems with multi default gem problem
- @jeremyevans0
+1, Can you commit it? Thanks. -
06/25/2019
-
-
10:06 PM Feature #15940: Coerce symbols internal fstrings in UTF8 rather than ASCII to better share memory with string literals
- @naruse Interesting, I actually had no idea you could `String#<<(integer)`.
In my humble opinion, integer shifting on string returned by `Symbol#to_s` is quite specific, and is unlikely to be common in the wild.
Additionally UTF8 ... -
- Note that an incompatibility which is caused by the change of string encoding is `String#<<(integer)`.
Maybe String#<<(n) should be deprecated if n > 127 and explicitly specify the encoding argument. -
- As pointed out by nobu, "defined" should be used instead of "assigned".
-
- This change broke installation when using `--dest-dir`, if the gem file without `--dest-dir` exists and the user performing the install does not have write access to the containing directory. This affects building a package for the Open...
-
- nobu (Nobuyoshi Nakada) wrote:
> "local variables assigned inside the call to +eval+" sounds like a just assignment to a local variable defined outside, doesn't it?
It could definitely be interpreted that way. I'll reword it to make... -
- "local variables assigned inside the call to +eval+" sounds like a just assignment to a local variable defined outside, doesn't it?
-
- Applied in changeset commit:git|0bd5f846df2f6268f653773e0cd4a20e2a944646.
----------
Document local variable interactions with eval
Fixes [Bug #13337] -
- jeremyevans0 (Jeremy Evans) wrote:
> Attached is a documentation patch.
Seems fine.
A question:
> ...
Isn't "set" vague a little, between "defined" an "assigned"?
-
- I believe that this is a duplicate of #10933. Please try a newer version of Ruby and let us know if it reproduces.
-
- Ruby 2.2.1 is no longer supported. Can you try with Ruby 2.5.5 or 2.6.3 and post back with the results?
-
04:49 PM Bug #15959 (Closed): Bug in the Ruby interpreter or extension libraries
- ``` ruby
[NOTE]
You may have encountered a bug in the Ruby interpreter or extension libraries.
Bug reports are welcome.
For details: http://www.ruby-lang.org/bugreport.html
bash: line 203: 20745 ... - 04:53 PM Revision 0aa8c3d8 (git): * 2019-06-26
- 04:53 PM Revision dd57442f (git): * remove trailing spaces.
-
- Fixes [Bug #13337]
-
- 518adcca0a: "Try using arm32 implementation for fibers."
dbe232e24e: "Order of arguments might be incorrect in arm32 coroutine implementation."
It seems to cause SEGV consistently on Ubuntu armv7l eabihf:
https://rubyci.org/logs/rubyci... -
- [Feature #15958] Time#inspect with frac
-
- When Matz changed Time#to_s format in 2006, frac part of Time string is dropped because it is considered not useful.
[ruby-dev:29440]
But recently we encounters some troubles the comparison of Time objects whose frac parts are differ... -
-
-
- Still timeout occurs. Retry for 5dd8fdd3f328f741fae4abba00c478e8a51d2a7e.
https://rubyci.org/logs/rubyci.s3.amazonaws.com/unstable11x/ruby-master/log/20190625T032405Z.fail.html.gz -
-
-
-
-
-
- 06:56 AM Revision 03dac55f (git): Dump Hash ivars before elements
- When the Hash is revived, the ivars will be set before elements. This
is for the case when the hash setter depends on ivars set on the hash. - 06:56 AM Revision 0016edbe (git): Add test to demonstrate issue deserializing hash with ivar
- Currently the elements of a hash are revived before any ivar values.
This causes an issue when the `[]=` method references an instance
variable. - 06:56 AM Revision a53ab897 (git): Round trip exception backtraces
- This commit allows exception backtraces to round trip
- 06:56 AM Revision f770a5be (git): Refactor exception dumping
- 06:56 AM Revision 4bd19094 (git): Dumping the backtrace never worked, so remove it
- Dumping the backtrace for exceptions never worked, so lets remove it.
-
- IO.seek on a pipe on Windows raises Errno::EINVAL instead of
Errno::ESPIPE.
Fixes Ruby Bug #15199 -
06:56 AM Revision e572ff2f (git): Require version file relatively
-
-
-
05:07 AM Bug #15957 (Closed): Splatting nil as **kwargs
- Splatting nil as *array is ignored silently.
```ruby
[1, *nil]
# => [1, 2]
```
But splatting nil as **kwargs raises TypeError.
```ruby
{foo: 1, **nil}
# => no implicit conversion of nil into Hash (TypeError)
```
Is this int... - 05:04 AM Revision dbe232e2 (git): Order of arguments might be incorrect in arm32 coroutine implementation.
-
- ```
$ ruby -e 'p({*nil})'
-e:1: syntax error, unexpected *, expecting '}'
p({*nil})
-e:1: Can't assign to nil
p({*nil})
```
First syntax error is intended.
But second error message `Can't assign to nil` is confusing.
This ha... -
- Pull request merged: https://github.com/ruby/scanf/commit/50f3b58ea0b56df436025e77669413622d544fbf
-
06/24/2019
-
- I do not think this is a bug. `Pathname#rename` operates exactly like `File.rename`, which like `rename(2)`, require that the source and destination be on the same file system.
`FileUtils.mv` tries to use `File.rename`, and if it fail... -
- This appears to have been fixed between 2.6 and 2.7.0-preview1:
```
$ irb26
irb(main):001:0> %q_abc\_def\_ghi_
irb(main):002:1> ^C
irb(main):002:0>
$ irb27
irb(main):001:0> %q_abc\_def\_ghi_
=> "abc_def_ghi"
irb(main):002:0>... -
- Fixed by commit:67723c1e4673253b2f4a2c7204ccab9d0daaaeb9.
-
- This crash appears to occur in the `di_binding` C function in the debug_inspector C extension:
https://github.com/banister/binding_of_caller/blob/v0.7.2/lib/binding_of_caller/mri2.rb#L25
https://github.com/banister/debug_inspector/bl... -
- This appears to have been fixed between 2.5 and 2.6:
```
$ irb25
irb(main):001:0> binding.irb
irb(main):001:0> 1+3
=> 4
irb(main):002:0> exit
=> 0
irb(main):002:0> 1+4
=> 5
irb(main):003:0> _
=> 4
$ irb26
irb(main):001:0... -
-
-
- From my testing, it looks like Ruby 2.2 was the first release to stop supporting nested constant lookup in `delegate`:
```
$ ruby21 -rforwardable -e 'class M; ENUMERATOR = 1; extend Forwardable; delegate [:to_s] => :ENUMERATOR; end; ... -
- Fixed: https://github.com/ruby/rdoc/issues/452
-
09:11 PM Bug #13337: Eval and Later Defined Local Variables
- This is as specified here
https://github.com/ruby/spec/blob/db74bf3313b4d49650417ba91b4e2d67cc253dc5/core/kernel/eval_spec.rb#L31-L68 -
- ko1 (Koichi Sasada) wrote:
> This is specification.
> ...
Attached is a documentation patch. However, a committer with more knowledge of the internals should probably check to see if it is technically accurate. -
- This crash happens inside a 3rd party C extension (pg): https://github.com/ged/ruby-pg/blob/e399861022f7c8c5419bfc7ca486b2c44caa4030/ext/pg_connection.c#L2487-L2523
If this issue is still occurring for you, you should probably report ... -
- Ruby has dropped support for IA64.
-
- `{Date,DateTime}.parse` use a loose parsing. They assume that somewhere in the string, there is something representing a date or time, and do their best to find it. Only if they can find nothing related to a date or time do they raise ...
-
-
-
-
- valerauko (Balint Erdos) wrote:
> I still can't consider this correct behavior. Aren't the multi-line and the single-line ways of writing `if` supposed to be equivalent?
No. This is documented in `doc/syntax/control_expressions.rdoc... -
-
- I'm not sure if this is a bug or not. However, considering that irb ignores permission issues on `.irbrc` (not intentionally I think), I suppose it makes sense to also ignore permission issues on the history file. I added a pull reques...
-
05:49 PM Feature #15955 (Closed): UnboundMethod#apply
- I'd love a way to apply an UnboundMethod to a receiver and list of args without having to first `bind` it. I've ended up using `UnboundMethod`s in some hot paths in my application due to our metaprogramming idioms, and the allocation fro...
-
-
- Fixed by commit:027cf467a4d093275b9c18fe694337765ff00213.
-
- So it looks like there was a behavior change between Ruby 2.3 and 2.4, in that `Integer#fdiv` with a BigDecimal argument used to return BigDecimal instead of Float. However, I consider that change a bug fix, as the documentation for `fd...
-
- I tried this example with the current master branch using OpenBSD's use-after-free checking (with Clang 7), and it raised a SystemStackError (as expected due to the infinite recursion), but it didn't crash. Can you replicate the use-aft...
-
- I believe the fact that sets are currently insertion ordered due to the use of hash is an implementation detail, and we shouldn't document it.
- 03:19 PM Revision e09afd15 (git): * 2019-06-25
-
-
11:27 AM Feature #15954: 簡単にマルチスレッドを一度に合流させる
- matz (Yukihiro Matsumoto) wrote:
> ライブラリでスレッドを作ったりしている可能性もあるので「自分以外全部join」はマズいんじゃないかなあと思います。
> ...
確かにそうでした。現在のファイルで作られたスレッドなどの制限を付ければいいのかな。良さそうな考えが浮かんだらまた出直します。 -
11:17 AM Feature #15954 (Rejected): 簡単にマルチスレッドを一度に合流させる
- ライブラリでスレッドを作ったりしている可能性もあるので「自分以外全部join」はマズいんじゃないかなあと思います。
Since required libraries may create threads inside, joining other threads altogether could cause problems.
What do you think?
Matz.
-
- 現在いるスレッド以外の全てのスレッドを現在のスレッドに合流させたいときは多いと思います。
その場合、一つの方法は、スレッドを生成するときにそれらを配列に蓄えておき、あとでその配列の各要素のスレッドに対して`join`をすることですが、そのためにわざわざスレッドを配列に蓄えておかなければならないのは、若干手間で無駄かなと思います。
それを回避する方法として、私の思いつく限りで単純な方法は次のようなものです。
```ruby
Thread.list.ea... -
- 説明をありがとうこざいます。バクでないことは分かったので、閉じて下さい。
-
- * [Feature #15950]Allow negative length in `Array#[]`, `Array#[]=`, `Array#slice`, `Array#slice!`, `String#[]`, `String#[]=`, `String#slice`, `String#slice!`
* 長さの負の値を許し、その場合に左向きに長さを測る提案です。
* 文字列の右端の定められた長さを取り出すときなどに便利だと思います。
... - 07:09 AM Revision 518adcca (git): Try using arm32 implementation for fibers.
-
-
04:52 AM Bug #15953 (Closed): Wrong heap size given to ruby_sized_xfree when freeing shared root arrays
- Applied in changeset commit:git|97b4fe2ff6bffe8728952236f2027aa6242f8f40.
----------
array.c: Wrong heap size given to ruby_sized_xfree when freeing shared roots
Fixes [Bug #15953]
Closes: https://github.com/ruby/ruby/pull/2253 -
- Fixes [Bug #15953]
Closes: https://github.com/ruby/ruby/pull/2253 -
- Applied in changeset commit:git|7f64a0b4db7ee27a04579236950d88301c7bcabb.
----------
Add new encoding CESU-8 [Feature #15931] -
- 03:13 AM Revision 0e6396d2 (git): Try using arm64 implementation for fibers.
- 03:13 AM Revision abdbfeba (git): Use spaces for indentation in `configure.ac`.
-
- IMO, literals are purely syntactic constructs. That doesn't mean they always behave the same as objects created by other means, ex: when calling methods, but as far as the definition of a literal, it's just a syntactic thing. Perhaps tha...
- 02:25 AM Revision 5d4dfcd8 (git): * remove trailing spaces.
- 02:24 AM Revision a071bed7 (git): Print warning if not using native coroutine.
-
01:53 AM Bug #14968: [PATCH] io.c: make all pipes nonblocking by default
- samuel@oriontransfer.net wrote:
> Eric do you mind explaining where we got to with this and how
> to move forward?
I don't think this patch is doable given potential incompatibilities
(and I fixed the problem with timer-thread bef... - 01:24 AM Revision 0b36c90e (git): * 2019-06-24
-
01:12 AM Revision edbad4ad (git): httpstatus.rb 100% coverage
-
01:11 AM Revision c6229e7c (git): Adds WebAssembly MIME type
-
- Remove extraneous spaces after the status code that is
non-compliant with RFC, i.e `HTTP 200 OK `, to unnecessary
confusion for WEBrick users, by a risk that WEBrick instances in
the wild will have server responses flagged as suspicious ...
06/23/2019
-
- PR here: https://github.com/ruby/ruby/pull/2253
I ran the `make test-all` suite with `#define ARRAY_DEBUG 1`, no failed assertions (I added a new assertion in a macro).
Thanks, -
- In `ary_heap_free`, the macro `ARY_HEAP_CAPA` is used to pass on to `rb_sized_xfree`. However, `ARY_HEAP_CAPA` uses the `aux.capa` field,
which is actually the number of shared strings that are using the root array, if a shared root is ... -
-
- * spec/ruby/library/socket/unixsocket/recvfrom_spec.rb: do not
lengthen UNIX socket path, which is very stricted. -
- 01:28 PM Revision 096a45d3 (git): Update NEWS.
- - Add note regarding IA64 removal.
- Add note about Thread allocation performance improvements. -
- I guess @eregon came from #15950. Will `ary[-3..]` be as efficient as `ary[-3, 3]` by freezing and deduping a literal `(-3..)`? Looks good if we can confirm it by an experiment.
Some thoughts:
* Even if a range is frozen, `("a"..... -
- Bug fix in commit ec8e5f5aa64e2a [Bug #15952] disabled an
optimization in this function.
Closes: https://github.com/ruby/ruby/pull/2252
06/22/2019
-
- Applied in changeset commit:git|ec8e5f5aa64e2a54cf1e303f2b012c98e8d521ba.
----------
array.c: always check frozenness in Array#unshift. Fixes [Bug #15952]
Closes: https://github.com/ruby/ruby/pull/2251 -
- PR here: https://github.com/ruby/ruby/pull/2251
-
- This should raise a `FrozenError`, but doesn't:
```ruby
a = [1] * 100
b = a[4..-1]
a.replace([1])
b.freeze
b.unshift("a") # no error
```
There is a code path in `ary_ensure_room_for_unshift` where the frozen check is skipped.... -
- Closes: https://github.com/ruby/ruby/pull/2251
-
- [Bug #14194]
-
- [Bug #14194]
-
10:37 PM Bug #14194 (Closed): --enable-frozen-string-literal ruby runner.rb cgi
- Applied in changeset commit:git|2ad7a7f801c07f18150116b819530840eeefebf8.
----------
Get rid of error with frozen string literal
[Bug #14194] -
- FYI,
Just ran ruby-loco with frozen string. Both test-all & spec are run parallel, and both crashed.
test-all failures/errors before crash:
```
ruby 2.7.0dev (2019-06-22T14:38:07Z master f738eeabc2) [x64-mingw32]
CRASHED?
... -
- nobu (Nobuyoshi Nakada) wrote:
> It doesn’t seem harmful, and tests should be independent from the condition.
I'm unsure, it forces existing code to add `# frozen_string_literal: false`, which seems not valuable if anyway the default... -
- sorry, **should** be independent
-
- It doesn’t seem harmful, and that test should be independent from the condition.
-
- @jeremyevans0
Sorry for the tardy response, I had hoped to run a CI build with the flag set, but busy with other things...
I was unaware that the decision had been made re Ruby 3. If you have any links to discussions, etc, I'd be ... -
- [Bug #14194]
-
-
- Applied in changeset commit:git|93843830198ba436e2ea21a60a11758d47cf521b.
----------
Module#constant_source_location [Feature #10771] - 04:47 PM Revision c5c34863 (git): * expand tabs.
-
- [Feature #10771]
-
-
-
-
- Applied in changeset commit:git|c033dc3073839e3578f1ba25d53b837974b56474.
----------
Fix issue with Array#rindex when rb_equal modifies receiver array
Fixes [Bug #15951]
Closes: https://github.com/ruby/ruby/pull/2250 -
- PR here: https://github.com/ruby/ruby/pull/2250
-
- Hi, this causes out of bounds RARRAY_AREF call in `rb_ary_rindex` in trunk branch, and results in a segmentation fault on my system:
```ruby
o = Object.new
def o.==(other)
other.replace([]) if Array === other
false
end
a... -
- Fixes [Bug #15951]
Closes: https://github.com/ruby/ruby/pull/2250 -
- Applied in changeset commit:git|f3c81b4e90ec492382e299573f2c3ac272adbb5f.
----------
Frozen objects in WeakMap
* gc.c (wmap_aset): bypass check for frozen and allow frozen
object in WeakMap. [Bug #13498] - 03:37 PM Revision fd9ac1e7 (git): * 2019-06-23
-
- * gc.c (wmap_aset): bypass check for frozen and allow frozen
object in WeakMap. [Bug #13498] -
- * test/lib/minitest/unit.rb (MiniTest::Assertions.diff): use
`EXEEXT` configured value instead of switching by hard coded
`host_os`. -
- I think it would make sense to freeze Range literals.
Adding methods to Range might be reasonable, but singleton methods, I would think much less so. -
-
- sawa (Tsuyoshi Sawada) wrote:
> And needless to say, this proposal would be irrelevant to Ruby beginners who insist on needlessly creating a range object as in `"abcdefg"[0...3]` to take the first three characters of the string, when y... -
- shevegen (Robert A. Heiler) wrote:
> ```ruby
> ...
It is the same idea as negative index. If you did not have the concept of cycling, then you would be surprised with negative index too. You would need to say "0 lets me assume to start... -
- sawa (Tsuyoshi Sawada) wrote:
> But to take the last n characters, we need to use n in two arguments: in the index (in negative form) in addition to the length:
> ...
Why not:
```ruby
"abcdefgh"[-3..-1] # => "fgh"
```
I would t... -
- Hmm. I do not necessarily doubt that it may be useful (at the least for
some users), but that also actually surprised me. For example:
```ruby
"abcdefgh"[0, -3] # => "fgh"
```
I guess my brain got confused here, since the 0 wou... -
- Negative **offset** is allowed, but negative **length** has never been allowed, IIRC.
-
- To take the first n characters of a string, using `[]` is straightforward:
```ruby
"abcdefgh"[0, 3] # => "abc"
```
But to take the last n characters, we need to use n in two arguments: in the index (in negative form) in addition ... -
- * test/lib/test/unit/assertions.rb (assert_join_threads): kill and
show timed out threads. -
04:57 AM Feature #9948 (Closed): Safely insert an object into an array
-
- I withdraw this proposal. Please close it.
-
01:56 AM Bug #15820: Backport 5e23b1138f1 (Fix potential memory leak)
- ruby_2_6 r67712 merged revision(s) 5e23b1138f16af0defb184d7deeffadfd2ce3c04.
-
- Fix potential memory leak
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_6@67712 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * test/lib/minitest/unit.rb (MiniTest::Assertions#mu_pp_for_diff):
do not expand escaped backslash followed by 'n'.
06/21/2019
-
11:52 PM Feature #15947: ENV.merge!
- Thank you!
-
- Applied in changeset commit:git|d01fd821875b989affc36e54e98f5dd67f47062e.
----------
Alias ENV.merge! as ENV.update
[Feature #15947]
Closes: https://github.com/ruby/ruby/pull/2246 -
- This looks good to me, I'll merge it.
Thank you for the PR. -
- ENV basically has hash like methods.
So I would like to apply an alias `ENV.merge!` that behavior is same as `ENV.update`.
github: https://github.com/ruby/ruby/pull/2246 -
- I agree this is a bug that should be fixed. Attached is a patch that implements support for %U/%W/%u/%w in `Time#strptime`.
-
- Ruby 2.6.0 started honoring refinements in `Kernel#public_send`.
-
- [Feature #15947]
Closes: https://github.com/ruby/ruby/pull/2246 -
-
-
-
- v =
if true # starting token at head of 2nd line
3
end # closing token -
- if true
3; end # this isn't auto-indented - 03:03 PM Revision d1fa0f61 (git): * 2019-06-22
-
-
-
- Applied in changeset commit:git|8f51da5d41f0642d5a971e4223d1ba14643c6398.
----------
Get rid of undefined behavior
* string.c (rb_str_sub_bang): str and repl can be same.
[Bug #15946] -
- Thank you.
`s3` doesn't seem to need to share `s`. -
- Hi, I've found an overlapping memcpy in an odd situation:
```ruby
s = "hello there this is a long string and cant be embedded"
s2 = s.dup # create shared string pointing to s
s3 = s.slice(1, 100) # create shared string pointing to ... - 01:48 PM Revision 551ef274 (git): * expand tabs.
-
- * string.c (rb_str_sub_bang): str and repl can be same.
[Bug #15946] -
- v =
if true
3
end # this "end" is auto-indented correctly -
-
-
- Applied in changeset commit:git|702cf3ec9051914c8ef5a353580e0e74f7ec7005.
----------
Enumerator::Lazy should support filter_map
Fixes [Bug #15949] -
- filter_map has been added to Enumerable, but it's not supported by Enumerator::Lazy.
```
p (1..).lazy.filter_map { |i| i.even? && i.to_s }.first # never return
```
- 08:43 AM Revision 432b28af (git): * expand tabs.
-
- Fixes [Bug #15949]
-
-
- Applied in changeset commit:git|fb85a428605265a8fd449b0702a4dd88cb6f3b20.
----------
Add an optional `inherit` argument to Module#autoload?
[Feature #15777]
Closes: https://github.com/ruby/ruby/pull/2173 -
- The new patch looks fine to me.
@nobu I'll let you merge it since you did most of the review. - 08:29 AM Revision c997f7da (git): * expand tabs.
-
-
- [Feature #15777]
Closes: https://github.com/ruby/ruby/pull/2173 -
-
- Follow https://bugs.ruby-lang.org/issues/15111
For consistency, I'd like ENV.update takes multiple hashes.
github: https://github.com/ruby/ruby/pull/2248 - 07:31 AM Revision 0174285e (git): * expand tabs.
-
-
06:29 AM Revision 4149fd60 (git): Fix small typo on block_scanf example [ci-skip]
- Closes: https://github.com/ruby/ruby/pull/2247