Activity
From 05/02/2019 to 05/08/2019
05/08/2019
-
11:54 PM Revision 10723dd6 (git): dryrun option is for `push`, not `git`
-
10:56 PM Revision dc405eb7 (git): Pin finalizer table
- Objects in the finalizer table stay pinned for now. In some cases, the
key could move which would cause a miss when removing the object from
the table (leading to a T_MOVED reference staying in the table). - 10:26 PM Revision 4ff0911c (git): * 2019-05-09
- 10:26 PM Revision 8b12db6e (git): * expand tabs.
-
- `obj_info` will look at references of objects in some cases (for example
it will try to access path information on ISeq objects). But during the
sweep phase, if the referenced object is collected before `obj_info` is
called, then it cou... -
05:37 PM Bug #15838 (Third Party's Issue): IOError: uninitialized stream
- - **What did you do?**
I ran the command `/Users/joeychou/.rbenv/versions/2.3.0/bin/bundle install`
- **What did you expect to happen?**
I expected Bundler to normally install all the gems.
- **What happened instead... -
03:04 PM Bug #15835: Path traversal symlink - WEBrick
- While I agree with naruse, it may be worthwhile to mention this briefly at e. g.
https://ruby-doc.org/stdlib/libdoc/webrick/rdoc/WEBrick.html - it could still surprise
users so it could be useful to mention it; perhaps at the section ... -
02:21 PM Feature #15836: [Proposal] Make Module#name and Symbol#to_s return their internal fstrings
- @byroot thanks for sharing the benchmark!
IMO module names should be frozen
even if it's easy to fix, this change could definitely break existing apps that depend on it, see
https://github.com/jruby/jruby/issues/5229 -
11:30 AM Feature #15836: [Proposal] Make Module#name and Symbol#to_s return their internal fstrings
- @mame re benchmark
So I decided to run this against redmine boot, using this branch: https://github.com/redmine/redmine/compare/master...byroot:boot-benchmark
Eager loading is enabled so that the entire codebase is loaded, and it u... -
- > Is HashWithIndifferentAccess the main rationale behind this request?
No. It's simply the poster child of how common `Symbol#to_s` is in code bases.
I shouldn't have mentioned `HashWithIndifferentAccess` because clearly lots of pe... -
03:15 AM Feature #15836: [Proposal] Make Module#name and Symbol#to_s return their internal fstrings
- Is `HashWithIndifferentAccess` the main rationale behind this request?
I have doubts about the usefulness of `HashWithIndifferentAccess` today, now that Rails has protected parameters.
Moreover, now that symbols are garbage collect... -
- I was about to write a long reply, but I think it would become too long, so just a shorter comment.
I use something like this in a "base" gem that I tend to use. Example:
class Foo
class Bar
NAMESPACE = inspect
... -
03:49 AM Feature #15837 (Assigned): Module#name_components
- I sometimes wrote the expression like below to extract the components of class path:
```ruby
klass.name.split('::').last
```
Similar expressions can be found in ruby-trunk:
```
mrkn@mrkn-devel:~/src/github.com/ruby/ruby$ git ... -
08:30 AM Revision a95ca6d5 (git): Trim MJIT output from TestHideSkip
- to prevent failure like
https://app.wercker.com/ruby/ruby/runs/mjit-test1/5cd28aa6ab79a30008ee819b?step=5cd28b2403f44600070db083
with --jit-verbose=1. -
-
- [EXPERIMENTAL]
-
-
07:46 AM Revision e8e415b5 (git): Add workaround for `Permission denied` of `cp`
- see r67347
recent log: https://travis-ci.org/ruby/ruby/jobs/529640417 -
-
- for debugging failure like
https://app.wercker.com/ruby/ruby/runs/mjit-test1/5cc98936ab79a30008eb86fa?step=5cc990d73d81fb0007bb04c3
https://app.wercker.com/ruby/ruby/runs/mjit-test1/5cd18efa23fcb70008ddfd45?step=5cd1908603f4460007076c5a -
-
- Fixing 7d805e67f3275aef066d77aa9c32bef715c362ed
-
- Recently `ruby/spec on Ruby 2.4` seems stable.
- 02:18 AM Revision 6ec43d3d (git): * 2019-05-08
-
05/07/2019
-
- > Matz said that but it will not be for ruby 3.0 at the least.
I assumed it was due for 3.0, but good to know it isn't.
> ...
I contribute to many gems, and from what I can see `# frozen_string_literal: true` is extremely common. A... -
- > frozen_string_literalAFAIK should become the default some day
Matz said that but it will not be for ruby 3.0 at the least.
In my own code bases I am using frozen strings a lot, through the shebang; either
"# frozen_string_litera... -
- @mame very good point. I'll try to run our app against that patch tomorrow.
-
03:36 PM Feature #15836: [Proposal] Make Module#name and Symbol#to_s return their internal fstrings
- > how is the GC handling in case the Symbol was a dynamic generated one?
It copies it - strings and symbols aren’t the same thing so it couldn’t jus return the symbol again. -
02:54 PM Feature #15836: [Proposal] Make Module#name and Symbol#to_s return their internal fstrings
- is the String returned by rb_sym2str not always frozen?
how is the GC handling in case the Symbol was a dynamic generated one?
wouldn't that remove the string that got returned from rb_sym2str too? -
02:42 PM Feature #15836 (Feedback): [Proposal] Make Module#name and Symbol#to_s return their internal fstrings
- Could you show us a benchmark, especially non-micro one? I believe that it is definitely required to discuss this proposal.
-
- # Why ?
In many codebases, especially Rails apps, these two methods are the source of quite a lot of object allocations.
`Module#name` is often accessed for various introspection features, autoloading etc.
`Symbol#to_s` is acces... -
02:31 PM Revision efda52ba (git): Use cgit instead of svn.
-
-
-
01:42 PM Bug #15809: GC.verify_compaction_references - intermittent SEGV's on multiple platforms
- Recursive finalizer definition raises a similar error on my environment.
```
$ ./miniruby -ve 'def foo(c = 10); ObjectSpace.define_finalizer(Object.new) { foo(c - 1) } if c > 0; end; 10.times do foo; GC.verify_compaction_references e... -
12:58 PM Revision a47f598d (git): Reduce ONIG_NREGION from 10 to 4: power of 2 and testing revealed most pattern matches are less than or equal to 4 results
- Closes: https://github.com/ruby/ruby/pull/2135
-
12:30 PM Bug #15835 (Feedback): Path traversal symlink - WEBrick
- On Apache with `FollowSymLinks` enabled, it can traverse out of DocumentRoot.
hxxps://httpd.apache.org/docs/2.4/en/urlmapping.html
Therefore it's not a problem. -
09:33 AM Bug #15835 (Closed): Path traversal symlink - WEBrick
- **Summary:**
A path traversal issue was observed in WEBrick ( WEBrick/1.4.2 (Ruby/2.6.3/2019-04-16)) via symlink. WEBrick serves static page for the current directory once enabled, however using symlink attacker could view data outside ... -
- Applied in changeset commit:git|7d805e67f3275aef066d77aa9c32bef715c362ed.
----------
Avoid triggering autoload in Module#const_defined?(String)
[Bug #15780] -
- I made a patch for this: https://github.com/ruby/ruby/pull/2172
-
- [Bug #15780]
-
- I made a patch for this feature: https://github.com/ruby/ruby/pull/2173
- 08:36 AM Revision 6786fe44 (git): * 2019-05-07
-
- Since it was recently mentioned, in particular in regards to names after kaigi 2019,
I think Threads::Feather as such is a fine name. More generally, though, I think ultimately
it would (or should) be important to keep the thread/fiber... -
05:10 AM Revision 4dc5d3c5 (git): add new debug_counters about is_pointer_to_heap().
- is_pointer_to_heap() is used for conservative marking. To analyze
this function's behavior, introduce some debug_counters. -
04:11 AM Bug #15834 (Closed): Mutating the result of SortedSet.to_a mutates the original set, potentially violating the set's guarantees
- ``` ruby
require 'set'
set = SortedSet[1, 2, 3]
set.to_a << -1
set.to_a << 3
set.each do |x|
puts x
end
puts
set.add -2
set.each do |x|
puts x
end
```
```
1
2
3
-1
3
-2
1
2
3
```
As can be seen, this ... -
03:02 AM Bug #15794 (Third Party's Issue): Can not start Puma with Rails after bundle install
05/06/2019
-
- I'm considering to support XDG guideline with rubygems, irb and etc.
-
- References to dot-files under the HOME in standard libraries are:
* `~/.irbrc` by irb
* `~/.options/` by optparse
* `~/.rdoc` by rdoc
* `~/.gem/` and `~/.gemrc` by RubyGems
* `~/.rb_shell` by shell
`OptionParser#load` now support... -
04:53 PM Bug #15809: GC.verify_compaction_references - intermittent SEGV's on multiple platforms
- Continuing to see intermittent SEGV's, today:
```
/ruby/test/ruby/test_gc_compact.rb:114: [BUG] ROOT finalizers points to MOVED: 0x0000000004f37f88 -> 0x0000000009736f50 [0 ] T_ARRAY [E ] len: 1 (embed)
ruby 2.7.0dev (2019-05-06... -
- jeremyevans0 (Jeremy Evans) wrote:
> I think the `ordinary parameter is defined` error makes sense if you consider the block starting at the `->` and not the `{`. Especially when you consider the older lambda/proc syntax uses the same ... -
02:03 PM Bug #15490: socket.rb - recurring segmentation faults
- Some background to how I have worked around this for now, which may be useful.
I use the parallel gem https://github.com/grosser/parallel, which can parallelise tasks using threads of processes. When switching from processes to threads... -
- This issue is still happening with the latest version of Ruby 2.6.3. Happy to provide more logs / run tests if I can help.
-
- I wrote some patches for shared-root array.
But I can't decide whether to commit them because they include the modification of public header `include/ruby/ruby.h`.
To ruby core team (I'm not assuming anyone in particular):
Can I mod... -
08:10 AM Misc #15782: DevelopersMeeting20190522Japan
- greggzst (Grzegorz Jakubiak) wrote:
> * [Feature #15323] [PATCH] Proposal: Add Enumerable#filter_map
> ...
Updated
-
-
- What is the difference towards e. g. `hash.delete()`?
People may ask about this difference. The '!'
too since you can use e. g. `Hash #delete` as-is.
Also without '!', I am not sure if extract is a good
name per se, but these are... - 02:08 AM Revision 970a25b1 (git): * 2019-05-06
-
- This reverts commit 5776ae347540ac19c40d146a3566a806cd176bf1.
Mistaken `max` as `min`.
05/05/2019
-
11:36 PM Bug #15789: Parse error when numbered parameter is used in a lambda that is a default value of other optarg
- nobu (Nobuyoshi Nakada) wrote:
> > ```ruby
> ...
I think the `ordinary parameter is defined` error makes sense if you consider the block starting at the `->` and not the `{`. Especially when you consider the older lambda/proc syntax u... -
- jeremyevans0 (Jeremy Evans) wrote:
> Attached is a one-line patch that fixes this issue, hopefully without causing additional issues:
Thank you, committed the patch.
> ...
This should be the "outside block" error, I think. -
- jeremyevans0 (Jeremy Evans) wrote:
> Attached is a one-line patch that fixes this issue, hopefully without causing additional issues:
Here's a patch that includes tests for this fix. -
- Attached is a one-line patch that fixes this issue, hopefully without causing additional issues:
```ruby
m(a = ->{@1}); a end
m.call(1)
# => 1
m2 = ->(a = ->{@1}) {a}
m2.call.call(2)
# => 2
m3 = ->(a: ->{@1}) {a}
m3.call.c... -
08:52 PM Bug #15832 (Closed): IPAddr.new accepts invalid mask IP address
- IPAddr.new accepts IP address with multiple mask and/or zero padded one.
```
$ cat ipaddr.rb
require 'ipaddr'
addrs = %w[192.0.2.0/24/16
192.0.2.0/00024
192.0.2.0/24/00016
192.0.2.0/00024/16
... -
08:52 PM Feature #15831 (Rejected): Add `Array#extract`, `Hash#extract`, and `ENV.extract`
- ## Add `Array#extract`
The method removes and returns the elements for which the block returns a true value.
If no block is given, an Enumerator is returned instead.
```ruby
numbers = [0, 1, 2, 3, 4, 5, 6, 7, 8, 9]
odd_numbers =... -
08:51 PM Feature #15829: Object#then_if(condition){}
- Related to https://bugs.ruby-lang.org/issues/15557.
-
08:30 PM Feature #15829: Object#then_if(condition){}
- A more complex example:
``` ruby
scope :blacklisted_at, -> (seller=nil, flag=true) {
then_if(! flag.nil?) {
joins(:disk_seller_maps)
.then_if(seller) {|q|
q.where(disk_seller_map:{seller_id:seller})
}
... -
- A (somehow contrived) example:
``` ruby
[1,2,3,4,5].map{|n| n.then_if(n.even?){'even'}}
=> [1, "even", 3, "even", 5]
```
-
- ``` ruby
class Object
def then_if(condition, &block)
if condition
self.then(&block)
else
self
end
end
end
```
-
- I'd like to propose some sugar to Object#then
There should be `Object#then_if(condition, &block)`
The block should only be applied when the condition is true,
otherwise the object should be returned without applying the block.
... -
- ``` ruby
require 'set'
class SortedSetWithoutLowest < SortedSet
def initialize contents=[]
puts "initialize, #{contents}"
super
lowest = to_a[0]
puts "lowest from #{contents} is #{lowest}, removing"
delete... -
07:27 PM Feature #15240: Set operations check for is_a?(Set), rather than allowing duck typing
- Any news on this one? Can we get the ball rolling? :)
I really don't mind doing the work, but would like to get a bit of feedback on the correct direction to make sure that the contribution does get accepted. -
- Not related to the issue at hand, but perhaps it would be good if travis could
also run/use tests similar as what Greg is reporting on windows; then it could
be simpler to get to test ruby on haiku as a first-class citizen in general.
... -
03:51 PM Bug #15827 (Rejected): Haiku: SSL tests failing
- Every SSL-aware testcase fails like this:
```
Generating RDoc documentation
No newer files.
Files: 0
Classes: 0 (0 undocumented)
Modules: 0 (0 undocumented)
Constants: 0 (0 undocumented)
Attributes: ... -
- We need something along theese lines, but integrated as default in ruby:
```
rubver=`ruby --version | cut -d" " -f2 | cut -d. -f3 --complement` #had no better idea :(
export RUBY_VER="$rubver.0"
export GEM_PATH="/boot/sys... -
- According to the guidelines, no program should create any file directly in $HOME
For config files we should use the ~/config/settings/ruby folder for .rubyrc (<- this file also should have no dot at the beginning!)
For --user-install... -
- nobu (Nobuyoshi Nakada) wrote:
> Sorry, I have wrongly got this as a rubygems specific error by just reading the subject.
> ...
Sorry, but because the missing knowledge i cannot give you a correct and helpful answer, but i would gladly... -
- Sorry, I have wrongly got this as a rubygems specific error by just reading the subject.
Opening in non-blocking mode is not to block the whole process by loading from a fifo, then reset it to get rid of problems on the `DATA` stream.
... -
- Here is my patchset against the current master. Please let me know if anything else required.
Patch tested working on Haiku. -
- This is required for the tests.
Haiku comes currently in 2 different flavour: 32 and 64 bit.
64 bit have only one toolchain, so it is easy, but 32 bit comes with 2 different (GCC2 primary and GCC7 secondary). Ruby built with GCC7 on ... -
- I don't want to comment so much on the issue as such, but I would like to point
out that it may be a better idea to start with a new idea not on highest
complexity. I am not sure if everyone fully understood pattern matching yet so
in... -
12:00 AM Feature #15824 (Open): respond_to pattern for pattern match
- rubyに追加されそうなパターンマッチの機能について...
Interger, Array など constantを書くことで、オブジェクトのクラスにマッチングさせる機能があるのを拝見しました。
duck typingを確認するようなパターンマッチがかけるとrubyらしいのではないかと思ってこちらに書いてみることにしました。
respond_to pattern
```ruby
class Runner
def run
end
d... -
- ko1 (Koichi Sasada) wrote:
> it is local problem on it?
I guess, no. It is not a local problem.
It will cause by `RARRAY_ASET` with shared array.
> ...
Not only `rb_ary_replace()` has the issue but also the array.c functions that... -
- As outlined in the proposal, the following diff fixes the WB miss local to `process_options`
```
diff --git a/ruby.c b/ruby.c
index bf94b8ee13..b7648934d6 100644
--- a/ruby.c
+++ b/ruby.c
@@ -1726,6 +1726,7 @@ process_options(int... -
- I didn't understand completely, but it is local problem on it? or all of `rb_ary_replace()` calls have the same issue?
-
07:52 AM Revision 594a033f (git): Improve description of Ruby in README
- Use improved description as suggested by Olivier Lacan (@olivierlacan),
see https://github.com/ruby/www.ruby-lang.org/pull/1888. -
-
07:32 AM Revision bb4ac7a6 (git): Fix use of numbered parameter inside proc that is default value of optarg
- This allows cases such as:
```ruby
m ->(a = ->{@1}) {a}
m.call.call(1)
m2 ->(a: ->{@1}) {a}
m2.call.call(2)
```
Previously, this would cause a syntax error.
[Bug#15789] -
- This allows cases such as:
```ruby
m ->(a = ->{@1}) {a}
m.call.call(1)
m2 ->(a: ->{@1}) {a}
m2.call.call(2)
```
Previously, this would cause a syntax error.
[Bug#15789] -
- Applied in changeset commit:git|b8f3be295b694964e88960c0228459b8aadd114a.
----------
Fix a case where numbered parameters should not be allowed
Because `proc{|| @1}` is a syntax error, the following should
also be syntax errors:
```ru... -
- Probably I might think the latter case could be allowed (and forgot to separate these cases), but it doesn't seem worth complicating.
-
- Because `proc{|| @1}` is a syntax error, the following should
also be syntax errors:
```ruby
proc { |
| @1}
proc { |; a| @1 }
```
The attached patch makes both of those cases syntax errors. -
- Because `proc{|| @1}` is a syntax error, the following should
also be syntax errors:
```ruby
proc { |
| @1}
```
```ruby
proc { |; a| @1 }
```
This fixes both cases.
[Bug #15825] -
- Applied in changeset commit:git|374c8f4ebab1a740990330c732b9de965c5e8d10.
----------
Fixed about ARGF.lineno
[Bug #15823] -
- Thank you, that document is wrong.
-
- [Bug #15823]
- 02:13 AM Revision 84e71e9f (git): * 2019-05-05
- 02:13 AM Revision f1b52d3a (git): * expand tabs.
05/04/2019
-
04:54 PM Bug #15823 (Closed): ARGF.lineno behavior and documentation disagree
- The documentation for `ARGF.each` at http://ruby-doc.org/core-2.6.3/ARGF.html#method-i-each
describes `ARGF.lineno` as if it behaved like `ARGF.file.lineno`, when it behaves for me like `$.`
Here's the sample code from the documentat... -
- * parse.y (case_args): moved "duplicated when clause" warning from
compile phase, so that `ruby -wc` shows them. - 02:32 PM Revision 848edb03 (git): ignore test_RangeError
-
01:50 PM Feature #15822 (Closed): Add Hash#except
- According to `Hash#slice`, I would like to add `Hash#except`. Same for `ENV.except`.
-
-
-
-
10:29 AM Bug #15711 (Closed): Remove use of _id2ref from DRb - Applied in changeset commit:git|8980b53a48b1f55e09c5223008225e6bfa765405.
----------
add DRb::WeakIdConv (Bug #15711) - 10:29 AM Revision 7d30cd47 (git): * remove trailing spaces.
- 10:28 AM Revision 8980b53a (git): add DRb::WeakIdConv (Bug #15711)
-
- ## Problem
Ruby interpreter may cause error "WB miss (O->Y)" on some conditions that are `RGENGC_CHECK_MODE=5` and `RUBY_DEBUG=gc_stress`
## How to reproduce
1. build ruby with high RGENGC_CHECK_MODE
* `make ruby optflags="-... -
05/03/2019
-
- [Misc #15775][ruby-core:92332]
-
06:41 PM Bug #15775: Update broken URL in Float documentation
- @hsbt: Just wanted to ping you about this small update of the Float documentation (maybe you can reassign it to someone who could review it?). Thanks!
-
- Ok, i consulted with PulkoMandy, one of the main Haiku developer, he meant it is not possible to unset "O_NONBLOCK or O_NDELAY because we don't allow it" on packaged file-system. This is why this check fails.
We make it sure the file op... - 03:00 PM Revision a3cbce78 (git): * 2019-05-04
- 02:59 PM Revision 04fd98d5 (git): * expand tabs.
-
-
-
- The fix for the potential memory leak (https://github.com/ruby/ruby/commit/5e23b1138f16af0defb184d7deeffadfd2ce3c04) should be backported into stable branches.
-
- > IHMO, it's too difficult to guess proper command.
Agree - File::NULL works in e. g. irb too and should work for ri. I found rdoc/ri
a bit cumbersome to use API-wise; in one project, that displays rdoc documentation
for methods (a ... -
04:58 AM Bug #15819: cannot read document of File::NULL with "ri File::NULL"
- When I type `ri File::NULL`, got
> Nothing known about File::NULL
> ...
but
* `ruby -e 'p File::NULL'` prints `"/dev/null"` succesfully
* Documents are already written in #7365
* I could read it with `ri File::File::Constants`
... -
-
05/02/2019
- 09:26 PM Revision 58cd4b7b (git): * 2019-05-03
- 09:24 PM Revision 7d02bab0 (git): Nil cannot and should not convert to a string
- 09:16 PM Revision 6610b77e (git): merge revision(s) 66324: [Backport #15385]
- Modify insn list only when compiling
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_5@67708 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- As far as I know, ruby-doc.org is a third-party project. The footer of the site says:
> Ruby-doc.org is hosted and maintained by James Britt and the Neurogami secret laboratory.
> ...
So, please contact on that email address.
(I'... -
- I don't have a particular pro/con opinion on the suggestion itself, but I do agree with the problem
that we often end up finding old(er) documentation by default - most certainly when doing a search
through Google but perhaps also by u... -
05:03 PM Feature #15818 (Rejected): Please hide ruby-doc.org pages for EOL releases from the Google index
- It's a very common experience that if I google a Ruby method or class, the top Google link to me points to the docs for Ruby 2.2 or 2.1 or even 1.9 or 1.8. As a recent example, googling [ruby unicode_normalize] finds me https://ruby-doc....
-
04:56 PM Feature #1089: Stable sorting for sort and sort_by
- One use case I have for a stable sort is a large CSV file which is tracked in a git repository. I have a script which loads the file, adds rows, and then sorts all rows by a particular column before writing them back to the file. I nee...
-
- > "map + select" is much more frequent, but it is not introduced yet
I think it would also be nice if `filter_map` was added. However, a specific justification for adding `tally_by` is to avoid an extra array allocation. `filter_map... - 01:44 PM Revision 5c87bb3b (git): * expand tabs.
-
-
10:43 AM Bug #15794: Can not start Puma with Rails after bundle install
- I resolved my issue, it due to some versions of my gems. I updated my gem to the latest version and it works now.
Thank you -
- Title: Warnings for undef_method and remove_method on initialize()
Right now ruby warns you if you remove or undefine initialize, at the least if you
run in verbose mode e. g. the -w flag.
Example for such a warning:
gtk_co... - 05:47 AM Revision d6efb386 (git): * 2019-05-02
-