Activity
From 09/25/2019 to 10/01/2019
10/01/2019
-
10:38 PM Revision 51456215 (git): bump up teeny version to 2.6.6.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_6@67822 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
09:53 PM Bug #16195 (Closed): Invalid certificate on https://ruby-lang.org/
- https://ruby-lang.org/ is serving a Let’s Encrypt certificate for lists.ruby-lang.org, which is invalid for ruby-lang.org.
-
07:31 PM Bug #16136: String corruption in 2.6.4
- Thanks so much! <3
-
01:50 PM Bug #16136: String corruption in 2.6.4
- FYI: Ruby 2.6.5 has been released, which includes the fix for this issue.
https://www.ruby-lang.org/en/news/2019/10/01/ruby-2-6-5-released/ -
04:33 PM Bug #16194 (Feedback): Can not compile the recent ruby 2.6.5 release due to some openssl-related issue (see issue description)
- Could you check the followings?
openssl.so refers `EC_GROUP_new_curve_GF2m` defined in libcrypto.so.
```
$ ldd .ext/x86_64-linux/openssl.so
linux-vdso.so.1 (0x00007ffc594d3000)
libruby.so.2.7 => not found
libssl.so.1.1 => /... -
02:57 PM Bug #16194 (Closed): Can not compile the recent ruby 2.6.5 release due to some openssl-related issue (see issue description)
- I am trying to compile the latest 2.6.5 release, just released recently:
https://ftp.ruby-lang.org/pub/ruby/2.6/ruby-2.6.5.tar.xz
I had a problem, which will be mentioned soon; but I first made sure that I do not
have this pro... -
03:03 PM Bug #16181: return from a proc in a module/class body returns out of script. Should be LJE.
- I added a comment about removing the pre 2.7 spec as it is unintended behavior.
-
- I can not comment/answer on the issue and questions; I think this is for
matz and the core team to decide either way, whatever the way.
I did, however had, want to add that:
json in {a:, b:, c:}
is quite difficult to read (... -
01:19 PM Revision 15606963 (git): expose assert_raise and assert_join_threads
-
12:26 PM Revision 55de0282 (git): Remove draft-release.yml [ci skip]
- I thought default branch's workflow runs on any tags,
but it does not run for stable branches without draft-release.yml.
So I abandoned, and use workflow in ruby/actions instead. - 11:06 AM Revision 1a45b044 (git): lib/shell/command-processor.rb (Shell#[]): prevent unknown command
- `FileTest.send(command, ...)` allows to call not only FileTest-related
methods but also any method that belongs to Kernel, Object, etc.
patched by <mame@ruby-lang.org>
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_4@67820... - 11:05 AM Revision 8eff476b (git): merge revision(s) 3ce238b5f9795581eb84114dcfbdf4aa086bfecc
- WEBrick: prevent response splitting and header injection
This is a follow up to d9d4a28.
The commit prevented CRLR, but did not address an isolated CR or an
isolated LF.
Co-Authored-By: NARUSE, Yui <naruse@airemix.jp>
git-svn-id: svn... - 11:05 AM Revision 38d2d0db (git): merge revision(s) 36e057e26ef2104bc2349799d6c52d22bb1c7d03
- Loop with String#scan without creating substrings
Create the substrings necessary parts only, instead of cutting the
rest of the buffer. Also removed a useless, probable typo, regexp.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branc... - 11:04 AM Revision 88387876 (git): merge revision(s) a0a2640b398cffd351f87d3f6243103add66575b
- Fix for wrong fnmatch patttern
* dir.c (file_s_fnmatch): ensure that pattern does not contain a
NUL character. https://hackerone.com/reports/449617
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_4@67817 b2dd03c8-39d4-4... - 11:02 AM Revision 7dce4316 (git): Forgotten to update :)
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_5@67815 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 11:01 AM Revision 3af01ae1 (git): lib/shell/command-processor.rb (Shell#[]): prevent unknown command
- `FileTest.send(command, ...)` allows to call not only FileTest-related
methods but also any method that belongs to Kernel, Object, etc.
patched by <mame@ruby-lang.org>
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_5@67814... - 11:01 AM Revision f98b3023 (git): merge revision(s) 3ce238b5f9795581eb84114dcfbdf4aa086bfecc
- WEBrick: prevent response splitting and header injection
This is a follow up to d9d4a28.
The commit prevented CRLR, but did not address an isolated CR or an
isolated LF.
Co-Authored-By: NARUSE, Yui <naruse@airemix.jp>
git-svn-id: svn... - 10:59 AM Revision 05cdcdc6 (git): merge revision(s) 36e057e26ef2104bc2349799d6c52d22bb1c7d03
- Loop with String#scan without creating substrings
Create the substrings necessary parts only, instead of cutting the
rest of the buffer. Also removed a useless, probable typo, regexp.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branc... -
- `FileTest.send(command, ...)` allows to call not only FileTest-related
methods but also any method that belongs to Kernel, Object, etc.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_6@67810 b2dd03c8-39d4-4d8f-98ff-823fe69b... - 10:58 AM Revision 02ea1fdf (git): merge revision(s) a0a2640b398cffd351f87d3f6243103add66575b
- Fix for wrong fnmatch patttern
* dir.c (file_s_fnmatch): ensure that pattern does not contain a
NUL character. https://hackerone.com/reports/449617
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_5@67809 b2dd03c8-39d4-4... -
- WEBrick: prevent response splitting and header injection
This is a follow up to d9d4a28f1cdd05a0e8dabb36d747d40bbcc30f16.
The commit prevented CRLR, but did not address an isolated CR or an
isolated LF.
... -
- Loop with String#scan without creating substrings
Create the substrings necessary parts only, instead of cutting the
rest of the buffer. Also removed a useless, probable typo, regexp.
git-svn-id: svn+ssh://ci.ruby-lang... -
- Fix for wrong fnmatch patttern
* dir.c (file_s_fnmatch): ensure that pattern does not contain a
NUL character. https://hackerone.com/reports/449617
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_6@67806 ... -
- https://github.com/ruby/ruby/pull/2514
-
- burdettelamar@yahoo.com (Burdette Lamar) wrote:
> Thanks, @alanwu. I'm refreshing diff.txt with more fulsome documentation, along with enhanced testing.
Thank you for the patch.
```diff
+ * Deletes the environment variable for +... -
- This is a follow up to d9d4a28f1cdd05a0e8dabb36d747d40bbcc30f16.
The commit prevented CRLR, but did not address an isolated CR or an
isolated LF.
Co-Authored-By: NARUSE, Yui <naruse@airemix.jp> -
- Create the substrings necessary parts only, instead of cutting the
rest of the buffer. Also removed a useless, probable typo, regexp. -
- * dir.c (file_s_fnmatch): ensure that pattern does not contain a
NUL character. https://hackerone.com/reports/449617 -
- @jonathanhefner Thank you for your comment.
> * `<pattern> <~ <expr>`
The difficult part is that `<pattern>` is not distinguishable from `<expr>` for a parser. For example, `[1, 2, x, y]` is valid not only as `<pattern>` but also ... -
- I think this is a bug, for the same reason as [Bug #16173].
https://github.com/ruby/ruby/pull/2512
09/30/2019
-
-
11:47 PM Bug #16191 (Feedback): Crash when setting conditional to false
- I believe this bug has been fixed and backported (I think it is solved by commit:0628973435abeccf676f3c2563579b3dbab06976). Can you try 2.5.6 and see if the problem still occurs?
-
09:28 PM Bug #16191 (Closed): Crash when setting conditional to false
- changed a conditional from "if variable.present?" to "if false" and loading the class containing the method containing that line now crashes ruby
-
- I can not say whether it is a bug or not, but I believe that either way,
it would be good to mention, in the documentation, which behaviour is to
be expected e. g. for the example you gave:
ENV.update('foo' => 'value_from_hash')... -
10:12 PM Bug #16192 (Closed): ENV.update hash values ignored when block given
- Doc:
```
update(hash) → Hash
update(hash) { |name, old_value, new_value| block } → Hash
```
Adds the contents of hash to the environment variables. If no block is specified entries with duplicate keys are overwritten, otherwise ... -
- I've added a pull request that makes `return` in `proc` in `class`/`module` a `LocalJumpError`: https://github.com/ruby/ruby/pull/2511
-
- Ok cool. Let's hope we get some more consensus on this. I don't like deviating from MRI in behavior so I hope to see more come to the same conclusion.
-
- enebo (Thomas Enebo) wrote:
> Jeremy, since you are only person who has looked would you say semantically this should be some error vs silently only executing part of a file?
I believe if `return` directly inside `class`/`module` is... -
- I agree this is impossible to be done by parser or at iseq generation time. It must be done when return is invoked.
JRuby will return LocalJumpError for this for 9.2.9.0. It also had from 9.2.6.0 and earlier.
As a runtime LJE we... -
- I don't have much experience dealing with Mac OS X crash file analysis, but the crash file attached doesn't mention Ruby at all, which seems odd. Are you sure that is the crash file related to the Rails project crash? Can you also prov...
-
06:49 PM Bug #16190 (Closed): Strange Crash
- Just started a new Rails 6 Project with Ruby 2.6.3 - added a secrets file and configured a database.yml and got this crash.
- 04:14 PM Revision 01e75b6b (git): * 2019-10-01 [ci skip]
-
04:14 PM Revision 671ca212 (git): remove `unused var` warning
-
02:18 PM Bug #16135: Segmentation fault in Ruby 2.6.1
- We have a Ruby swig interface through which we are calling Kysy dll (written in C++) .I have upgraded swig interface from Ruby 2.2.6 to Ruby 2.6.1 .The same testcases are fine in Ruby 2.2.6 .I even tried with Ruby 2.5.6 versions and it w...
-
02:06 PM Revision a8c436d9 (git): Issue keyword flag warning even with no arguments
- If the keyword flag is set, there should be at least one argument,
if there isn't, that is a sign the keyword flag was passed when it
should not have been. -
- Currently, there is not a way to create a sized enumerator in C
with a different set of arguments than provided by Ruby, and
correctly handle keyword arguments. This function allows that.
The need for this is fairly uncommon, but it oc... -
- by explicitly removing the old definition.
-
- "in void context" by assigning the result to a dummy variable.
-
- This reverts commit c3b84f2de83a27acc638f99743bfa2c44bac621c.
Backtrace shows it is before running tests and debug check was
nonsense. -
- callable_method_entry_p is for rb_callable_method_entry_t.
-
-
-
- There are random failures:
> lib/rubygems/core_ext/kernel_require.rb:61:in `require':
> ...
http://ci.rvm.jp/results/trunk-jemalloc@silicon-docker/2275159
To check this failure, I added a small check code. -
- chkbuild (CI process) shows methods list before
running tests and sometimes it can fails. This
commit a code part to emulate this method listing
feature. -
-
- method_def_location() expects that rb_method_attr_t::location
is 0 (Qfalse) or not. -
- Some tests failed because https://github.com/ruby/ruby/blob/649a64ae29318501472f74798a12eb60871ab990/lib/rubygems/specification.rb#L106-L107 set TODAY as 2019-09-29, and tests run on 2019-09-30.
https://rubyci.org/logs/rubyci.s3.amazo... -
- This adds rb_funcall_passing_block_kw, rb_funcallv_public_kw,
and rb_yield_splat_kw. This functions are necessary to easily
handle cases where rb_funcall_passing_block, rb_funcallv_public,
and rb_yield_splat are currently used and a key... -
01:26 AM Revision fba8627d (git): delete unnecessary branch
- At last, not only myself but also your compiler are fully confident
that the method entries pointed from call caches are immutable. We
don't have to worry about silent updates. Just delete the branch
that is now always false.
Calculat... -
- Now that we have eliminated most destructive operations over the
rb_method_entry_t / rb_callable_method_entry_t, let's make them
mostly immutabe and mark them const.
One exception is rb_export_method(), which destructively modifies
visi... -
- Tired of rb_method_entry_create(..., rb_method_definition_create(
..., &(rb_method_foo_t) {...})) maneuver. Provide a function that
does the thing to reduce copy&paste. -
- Make things more immutable.
-
- The deleted function was to destructively overwrite existing method
entries, which is now considered to be a bad idea. Delete it, and
assign a newly created method entry instead. -
- Instead of destructively write fields of method entries, create a
new entry and let it overwrite its owner. -
- Before this changeset rb_method_definition_create only allocated a
memory region and we had to destructively initialize it later.
That is not a good design so we change the API to return a complete
struct instead. -
- No longer used.
-
- Did you know that C functions can return structs? That has been
true since the beginning, but not that very useful until C99. Now
that we can write compound literals, this feature is much easier
to use. By allowing struct-returning fu... -
- Most (if not all) of the fields of rb_method_definition_t are never
meant to be modified once after they are stored. Marking them const
makes it possible for compilers to warn on unintended modifications. -
- This is a pure refactoring to reduce copy & paste. Also the new
function is made visible from other parts of the interpreter, to
be used later. -
- Not used from anywhere.
-
- Because rb_method_definition_t tracks its own complemented_count,
we no longer have to check it in rb_method_entry_t side.
09/29/2019
-
- VM_NO_KEYWORDS was introduced first in vm_core.h, but it is best
to only use a single definition for this. -
- In general RB_PASS_CALLED_KEYWORDS should only be set if we are
sure the arguments passed come directly from Ruby. For direct calls
to these C functions, we should not assume that keywords are passed.
Add static *_internal versions of t... -
10:16 PM Revision 59060aad (git): Run test-unit tests in CI
-
- * See https://github.com/test-unit/test-unit/issues/165
-
- I'll get to the point because I think we're starting to discuss unrelated things in this issue.
headius (Charles Nutter) wrote:
> I'd argue it's actually NEVER safe to modify the result of calling `to_s` [...]
Yes, and it has neve... -
07:29 PM Feature #16150: Add a way to request a frozen string from to_s
- > - `-@` could be `dedup`, or `deduplicate`.
`dedup` is an interesting choice because it implies the opposite of `dup`.
Although you might run into naming collisions if / when you extend this functionality to other types. For ex... -
12:29 PM Feature #16150: Add a way to request a frozen string from to_s
- > However it's still not safe to modify it since it would modify the original string too.
IMHO that's two different use cases. Either:
- You want to be sure not to mutate the string, then you use `.dup`
- You don't care about... -
- Eregon (Benoit Daloze) wrote:
> In #14183, keyword arguments became further separated from positional arguments.
> ...
Internally, there are not really performance implications for the choice to treat keyword arguments as a last positi... -
- In #14183, keyword arguments became further separated from positional arguments.
Contrary to the original design though, keyword and positional arguments are not fully separated for methods not accepting keyword arguments.
Example: `... -
- > We want to ask English speaker about this proposal. Do you feel natural on this syntax?
(I am a native English speaker.) The syntax feels confusing to me. When I read `X in Y`, I expect `Y` to be a collection of things, and `X` to... -
-
- Some warnings are because the @o.rb_scan_args call doesn't
include keyword arguments, but the first argument is passed to
rb_scan_args may have a last hash treated as keywords. Those
should be handled using rb_scan_args_kw on Ruby 2.7.
... -
- Current implementation is since r38225 (fe6b2e20e9f17ed2c2900aa72994e075ffdc7124).
I think this was discussed in the mailing list (probably ruby-dev), but I can't find it now.
And I think which should be uninterruptible is only printin... -
- As `Kernel#p` is equivalent to `Thread.handle_interrupt(Object => :never) {puts args.map(&:inspect)}`, all interrupts never be invoked.
-
-
-
-
- * numeric.c (num_step): pass the extracted argument from keyword
argument, not the last argument itself which should have been
warned already. - 04:02 PM Revision f39314a9 (git): * 2019-09-30 [ci skip]
-
-
-
-
-
-
- * Writing shell scripts in a Makefile is very error-prone.
* TEST_BUNDLED_GEMS_ALLOW_FAILURES seemed to not work before. -
- * See https://github.com/test-unit/test-unit/issues/165
-
- * Otherwise, it takes a very long time to notice those tests broke.
-
- https://github.com/ruby/stringio/commit/f0e5027279
-
- https://github.com/ruby/stringio/commit/5892663e32
-
- https://github.com/ruby/stringio/commit/a37ab7c419
-
- https://github.com/ruby/stringio/commit/e8065153b8
-
- https://github.com/ruby/stringio/commit/4dfd997e0a
-
- https://github.com/ruby/zlib/commit/00ead8cb2c
-
- https://github.com/ruby/zlib/commit/8f43b264cd
-
-
- [ruby-list:50826]
- 04:17 AM Revision 289cad50 (git): * 2019-09-29 [ci skip]
-
04:15 AM Revision 6e1b72ae (git): Upgrade to the latest did_you_mean 1.3.1
-
02:26 AM Feature #16147: List Comprehensions in Ruby
- ## ruby_list_comprehension gem documentation
https://github.com/SammoMichael/Ruby_List_Comprehension
## ruby_list_comprehension gem page
https://rubygems.org/gems/ruby_list_comprehension
Here is a gem to give a better idea what I mea...
09/28/2019
-
-
- RDoc cannot know if bare words are method name or not.
-
04:43 AM Feature #16150: Add a way to request a frozen string from to_s
- Hi,
In message "Re: [ruby-core:95142] [Ruby master Feature#16150] Add a way to request a frozen string from to_s"
on Sat, 28 Sep 2019 04:33:32 +0000 (UTC), headius@headius.com writes:
>Issue #16150 has been updated by headius (Ch... -
04:33 AM Feature #16150: Add a way to request a frozen string from to_s
- > Not exactly, -@ and +@ makes this much simpler
I do like the unary operators, but they also have some precedence oddities:
```
> ...
=> -3
>> (-"foo").size
=> 3
```
And it doesn't work at all if you're chaining method call... -
- ruby_2_6 r67805 merged revision(s) 8b3774be3dd9f472bddd99e84e3c9fe2ff99d7ac.
-
- Fix memory leak
* string.c (str_make_independent_expand): free independent buffer.
[Bug# 15935]
Co-Authored-By: luke-gru (Luke Gruber) <luke.gru@gmail.com>
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches... -
03:46 AM Feature #15973: Let Kernel#lambda always return a lambda
- I have updated my fix for [Bug #15620] to also issue a warning in cases it
returns the argument it receives without modification.
Since both 2.5.x and 2.6.x has [Bug #15620], there may be code in the wild that
depend on the unintend...
09/27/2019
-
- matz (Yukihiro Matsumoto) wrote:
> I think the following code behavior is wrong:
> ...
Here's a pull request for that: https://github.com/ruby/ruby/pull/2502
Note that it breaks some tests/specs. I believe the reason methods with k... -
- It looked into this behavior, which started in Ruby 2.4 (when top level return started being allowed).
`return` not being allowed directly in class/module body is a parse error, because the parser knows that it cannot be valid. Howev... -
10:49 PM Bug #16121: Stop making a redundant hash copy in Hash#dup
- ko1 (Koichi Sasada) wrote:
> I found that
> ...
I tried doing that and it caused test failures because Hash#replace stop rehashing keys in ruby 2.6. I've opened a bug for that: https://bugs.ruby-lang.org/issues/16187
I've opened ht... -
- Here is a script that shows the problem
```ruby
mutable_key = [1]
h = { mutable_key => 'a' }
mutable_key[0] = 2
p(h == {}.replace(h))
```
which outputs `true` in ruby 2.5 and lower versions and `false` in ruby 2.6 and later v... -
10:34 PM Bug #16186 (Closed): Calling Net::HTTP from within an #inspect implementation hang when invoked from p
- Using this script:
``` ruby
require 'net/http'
class Victim
def inspect
puts "before"
Net::HTTP.get('example.com', '/index.html') # this hangs. I haven't tried other network libraries
puts "After"
return "success"
end
... -
- This is due to calling rb_mod_module_eval directly instead of using
rb_funcall_passing_block.
The problem with calling directly is it does not create a new VM
frame, so rb_mod_module_eval was called with no arguments, but with
the keywo... -
- If a method accepts no keywords and was called with a keyword, an
ArgumentError was not always issued previously. Force methods that
accept no keywords to go through setup_parameters_complex so that
an ArgumentError is raised if keyword... -
- This may be too verbose, if so, maybe it should be moved lower in
the document, or to a separate document. -
- Applied in changeset commit:git|a1fda16b238f24cf55814ecc18f716cbfff8dd91.
----------
Optimize Array#flatten and flatten! for already flattened arrays (#2495)
* Optimize Array#flatten and flatten! for already flattened arrays
* Add benc... -
- It looks like I made a mistake in my benchmarking of non-flattened arrays, since flatten! from the first iteration would cause further iterations to actually test with a flat array. It looks like the performance for arrays starting with...
- 04:24 PM Revision 707c4885 (git): * 2019-09-28 [ci skip]
- 04:24 PM Revision a1fda16b (git): Optimize Array#flatten and flatten! for already flattened arrays (#2495)
- * Optimize Array#flatten and flatten! for already flattened arrays
* Add benchmark for Array#flatten and Array#flatten!
[Bug #16119] -
- Applied in changeset commit:git|869e2dd8c8efc1e7a043c9eee82d97c47befbcc7.
----------
Warn for URI.{,un}{escape,encode}, even if not in verbose mode
The verbose mode warning has been present for almost 10 years.
If we ever plan to remov... -
- Applied in changeset commit:git|a2c26fe1c6d49b54494b7c6301cc4c77596eec0c.
----------
Fix fallback in URI.encode_www_form_component to include #
Patch from Matthew Kerwin.
Fixes [Bug #14358] -
- Applied in changeset commit:git|43a16c98df392e726040f0331a3e09d00c53d513.
----------
Do not escape + in Shellwords.escape
+ is not a character that requires escaping in Bourne sh.
Fixes [Bug #14429] -
- Applied in changeset commit:git|2b6a9f3a1ffcdb00bf89798979d475c6d189d419.
----------
Ignore Errno::EPIPE when sending requests in net/http
An EPIPE when sending the request should be ignored. Even if you
cannot write more data, you ma... -
- The verbose mode warning has been present for almost 10 years.
If we ever plan to remove these methods, we should make the warning
a regular deprecation warning so that people are aware.
Implements [Feature #15961] -
- Patch from Matthew Kerwin.
Fixes [Bug #14358] -
- + is not a character that requires escaping in Bourne sh.
Fixes [Bug #14429] -
- An EPIPE when sending the request should be ignored. Even if you
cannot write more data, you may still be able to read the server's
response.
Fixes [Bug #14466] -
-
- @kanemoto @reiodaira Could you treat this?
-
11:58 AM Bug #16185 (Feedback): basictest failure on AIX 6.1 for 64bit build
- We are building ruby 2.6.3(both 32 and 64bit) on AIX 6.1.
The build options used for 64bit build are:-
export LDFLAGS="-L/opt/freeware/lib64 -L/opt/freeware/lib -Wl,-blibpath:/opt/freeware/lib64:/usr/lib:/lib"
and for 32bit build are:... -
- Applied in changeset commit:git|79d5332a2da0e2743a77480aebfca256a54a962e.
----------
Drop eliminated catch-entries
Drop catch table entries used in eliminated block, as well as
call_infos. [Bug #16184] -
10:05 AM Bug #16184: Entry persists in catch table even though its labels were removed, which may cause [BUG]
- > And having `position` field of labels initialized (as in my attached patch) makes crash totally impossible.
I think having this field properly initialized is a good idea anyway, so I created a pull request: https://github.com/ruby/r... -
- > So whether Ruby crashes or not depends on the contents of the memory where labels are allocated by compile_rescue function.
And having `position` field of labels initialized (as in my attached patch) makes crash totally impossible. -
- > Could you post a reproducer that reliably crashes Ruby?
I guess this is not really possible to crash Ruby reliably via this issue. Here is a reproducer that crashes on my MIPS32 LE machine one out of 20 times or so:
```ruby
puts... -
- Could you show the backtrace?
-
- Drop catch table entries used in eliminated block, as well as
call_infos. [Bug #16184] -
-
- ruby_2_6 r67804 merged revision(s) 93faa011d393bb4b5cf31a0cbb46922f0a5e7cdc.
-
- Tag string shared roots to fix use-after-free
The buffer deduplication codepath in rb_fstring can be used to free the buffer
of shared string roots, which leads to use-after-free.
Introudce a new flag to tag str... -
08:35 AM Feature #16183: Hash#with_default
- zverok (Victor Shepelev) wrote:
> Eregon (Benoit Daloze) wrote:
> ...
Well, yes, but at least in my view, `merge` is a two-sided (symmetric) operation. There are two hashes, and you merge them. It would be strange if one of them is cha... -
06:02 AM Feature #16183: Hash#with_default
- > That means a copy of the Hash is necessary on each call to `#with_default`.
Yes, the same way it is for, say, `merge`, and we still use it in a lot of cases even when source hash would be dropped -- for the sake of chainability:
... -
- > In order to safely get a frozen string from an object, you must ALWAYS do `some_str.to_s.dup.freeze`. In order to safely get a mutable string from an object, you must ALWAYS do `some_str.to_s.dup`.
Not exactly, `-@` and `+@` makes t... -
- Applied in changeset commit:git|eff15a269fdc37d2b09cf1dfe8c1b1bf6e377a32.
----------
[EXPERIMENTAL] Make NilClass#to_s, TrueClass#to_s and FalseClass#to_s return a frozen String
* Always the same frozen String for each of these val... -
-
-
- * Always the same frozen String for each of these values.
* Avoids extra allocations whenever calling these 3 methods.
* See [Feature #16150] -
- https://github.com/ruby/fileutils/commit/ba3bd6fdfd
-
- https://github.com/ruby/fileutils/commit/17a9de6d9b
-
- https://github.com/ruby/fileutils/commit/ba81f024cf
-
- Previously this would copy the symlink root as a symlink instead
of creating a new root directory. This modifies the source
to expand it using File.realpath before starting the copy.
Fixes Ruby Bug 12123
https://github.com/ruby/fileut... -
- Previously this would give an error such as:
TestFileUtils#test_cp_r_dev [c:/fileutils/test/fileutils/test_fileutils.rb:455]:
[RuntimeError] exception expected, not.
Class: <TypeError>
Message: <"no implicit conversion of nil into Strin... -
- If FileUtils is included into another object, and verbose mode is
used, a FrozenError is currently raised unless the object has the
@fileutils_output and @fileutils_label instance variables.
This fixes things so that it does not attempt... -
- https://github.com/ruby/fileutils/commit/20bb9ec789
-
- Previously, this was broken. Trying to copy a FIFO would raise a
NoMethodError if File.mkfifo was defined. Trying to copy a UNIX
socket would raise a RuntimeError as File.mknod is not something
Ruby defines.
Handle the FIFO issue usin... -
- https://github.com/ruby/fileutils/commit/b701353c53
-
- This fixes instance_exec and similar methods. It also fixes
Enumerator::Yielder#yield, rb_yield_block, and a couple of cases
with Proc#{<<,>>}.
This support requires the addition of rb_yield_values_kw, similar to
rb_yield_values2, for p... -
09/26/2019
-
- Oh I should point out another problem with the current situation...
```ruby
some_str.to_s.freeze
```
The above code is also unsafe right now, because since `to_s` returns self, we're now freezing the original string!
In order ... -
- > But granted that this now cause this somewhat weird situation where to_s might or might not return a frozen string. that being said it's already kinda the case
Yeah, this is precisely the problem. It is made even worse by the large ... -
- @headius I totally get your opinion on `String#to_s`, however on a very practical standpoint I agree with @eregon:
> Returning a frozen String for String#to_s would just add more allocations, since the current String#to_s just returns... -
- I'm glad to see some smaller parts of this are moving forward. A few updates from my end (JRuby):
* Frozen strings from Symbol etc:
We are on board with making Symbol, nil, true, false, and Module all return frozen strings, and hav... -
- Then store those objects in global variables, and `rb_gc_register_mark_object`.
-
- byroot (Jean Boussier) wrote:
> Over the 502 gems used in the repo, I found 5 impacted gems, all because of `Symbol#to_s`, and all in very minor ways:
A great report. For the record, I add another case I've heard:
* did_you_mean:... -
- @nobu Yes that's what I had in mind, but then from what I understand it means we have to lookup the fstring table every time rather than just return a existing pointer. So it's not really an optimisation, is it?
-
- Ok, so I just finished updating our application to be compatible with both `Symbol#to_s` and `Module#name` returning frozen strings.
Over the 502 gems used in the repo, I found 5 impacted gems, all because of `Symbol#to_s`, and all in... -
- byroot (Jean Boussier) wrote:
> For `nil/true/false` it is a bit trickier because the returned string are ASCII encoded, so I'm not quite sure how to statically create ASCII `fstrings` with the existing APIs.
```C
static VALUE
true... -
- > Do you want to do the PR for nil/true/false as well then?
I tried but I'm quite unsure how to do it, so I don't mind if you handle it.
On a side note I'm currently running our test suite with the following monkey patch:
```rub... -
- @byroot Great, I merged your PR.
Sorry, I should have asked first if you wanted to do it.
Do you want to do the PR for `nil`/`true`/`false` as well then? -
- Applied in changeset commit:git|9d0866c7d7b9cbe36a851744a37806e747e0e7a8.
----------
[EXPERIMENTAL] Make Module#name return a frozen String
* Always the same frozen String for a given Module or Class.
* Avoids extra allocations... -
- I got the `Module#name` one ready here: https://github.com/ruby/ruby/pull/2487
For `nil/true/false` it is a bit trickier because the returned string are ASCII encoded, so I'm not quite sure how to statically create ASCII `fstrings` wi... -
- Reopening, the issue was accidentally closed as I merged the Symbol#to_s change.
I will make a PR for nil/true/false.to_s and Module#name to return a frozen String. -
- Applied in changeset commit:git|6ffc045a817fbdf04a6945d3c260b55b0fa1fd1e.
----------
[EXPERIMENTAL] Make Symbol#to_s return a frozen String
* Always the same frozen String for a given Symbol.
* Avoids extra allocations whenever calling... -
byroot (Jean Boussier) wrote:
> I get you are worried about `String#to_s`, but what about others that have been mentioned here, namely:
> ...
Ok for the experiment.
Matz.-
10:41 PM Revision 293c6c8c (git): Add compaction support to `rb_ast_t`
- This commit adds compaction support to `rb_ast_t`.
-
- This requires adding rb_proc_call_kw to pass the keyword flag.
-
- We are calling this in a few other files, it is better to have it
in a header than adding prototypes to the other files. - 10:01 PM Revision 31339ef4 (git): Honor Syslog::Logger#level overrides
-
- I've been compiling with:
```
set -lx cflags '-std=c99 -Werror=pedantic -pedantic-errors'
```
But compilation would fail with the following:
```
cont.c:296:90: error: format specifies type 'void *' but the argument has type 'struct ... -
- zverok (Victor Shepelev) wrote:
> So, I propose side-effect-less method, that **is** acceptable for chaining, I believe.
That means a copy of the Hash is necessary on each call to #with_default.
I.e., it would be the same as:
```ru... -
- > Duplicate of #11761
I don't think it is a duplicate, though I haven't clarified an important behavior indeed:
```ruby
h = {a: 1, b: 2}
h2 = h.with_default(3)
h.default # => nil
h2.default # => 3
```
So, I propose side-effec... -
- sawa has it all covered. ;)
The explanation by matz is interesting.
"Use tap. Methods with side-effect should be handled with care. Making it chainable
has little benefit."
Personally I think other names, be these default_set, ... -
03:52 PM Feature #16183: Hash#with_default
- Duplicate of #11761
-
- Reasons: there is no way, currently, to *declaratively* define Hash with default value (for example, to store it in constant, or use in an expression). Which leads to code more or less like this:
```ruby
FONTS = {
title: 'Arial'... -
-
- This function is just a synonym for RB_OBJ_WRITTEN, so we can just
directly call that. -
- We don't use this array anymore so we can remove it
-
- We can mark everything via the instruction objects, so just execute the
write barrier instead of appending to the array -
- `freeze_string` essentially called iseq_add_mark_object_compile_time. I
need to know where all writes occur on the `rb_iseq_t`, so this commit
separates the function calls so we can add write barriers in the right
place. -
- Move the "add mark object" function to the location where we should be
calling RB_OBJ_WRITTEN. I'm going to add verification code next so we
can make sure the objects we're adding to the array are also reachable
from the mark function. -
- This commit scans the ISEQ arena for objects that can be marked and
marks them. This should make the mark array unnecessary. -
-
- We'll scan the secondary arena during GC mark. So, we should only
allocate "markable" instruction linked list nodes out of the secondary
arena. -
- Now we can allocate and free a secondary arena.
-
- This is so we can configure a new arena later
-
- headius (Charles Nutter) wrote:
> If we make the global default to freeze string literals, would `frozen-string-literal: false` for a given file make them mutable?
Yes, in fact shortly after `--enable-frozen-string-literal` was intro... -
- C API updates:
* adjusted indent.
* marked up ANYARGS as RDoc. -
- Eregon (Benoit Daloze) wrote:
> Dan0042 (Daniel DeLorme) wrote:
> ...
That's still incomplete, since it doesn't explain why `_1` doesn't decompose when used in lambdas. If it always decomposes, it'd be more self consistent.
Anyways, I... -
-
- Could you post a reproducer that reliably crashes Ruby? Interesting issue!
-
- Of course, while having some strange cache table entries would be pretty ok if they were not used (like it usually happens on x86_64). To make sure they are never used compiler should initialize the `position` field of the labels. I'm at...
-
- When `remove_unreachable_chunk` removes the code from within a rescue block, the catch table entry corresponding the block is not removed. Here is a simple reproducer (tested with ruby 2.5.5):
``` ruby
puts "BEGIN"
if false
b... -
- * got rid of inadvertent label lists.
* marked up resolve_feature_path method names.
* fixed indentation of UnboundMethod#bind_call and marked up as
RDoc. -
- Ensure that argument array size does not overflow as `int`, before
cast in thread_do_start after new thread created. - 03:02 PM Revision 3b9cdc59 (git): * 2019-09-27 [ci skip]
-
-
-
-
-
- It's unlikely anyone would actually hit these. The methods are
private, you only hit this code path if calling these methods
before performing the SSL connection, and there is already a
verbose warning issued. -
- This requires adding rb_proc_call_with_block_kw.
-
-
- The pull request was merged.
-
-
-
-
01:13 PM Revision 41e3c204 (git): Reduce ISeq size of mjit_exec benchmark
- to avoid unwanted memory pressure
-
- Maybe 2.6.4 has same issue.
-
- Applied in changeset commit:git|93faa011d393bb4b5cf31a0cbb46922f0a5e7cdc.
----------
Tag string shared roots to fix use-after-free
The buffer deduplication codepath in rb_fstring can be used to free the buffer
of shared string roots, w... - 11:25 AM Revision 1070456b (git): * remove trailing spaces. [ci skip]
-
- * Always the same frozen String for a given Module or Class.
* Avoids extra allocations whenever calling Module#name.
* See [Feature #16150] -
-
-
-
- And test the fix we're adding.
https://github.com/rubygems/rubygems/commit/6f86637b98 -
- https://github.com/rubygems/rubygems/commit/cc255b774a
-
- https://github.com/rubygems/rubygems/commit/8cdb2fd66f
-
- Co-authored-by: Yusuke Endoh <mame@ruby-lang.org>
https://github.com/rubygems/rubygems/commit/92892bbc3a - 08:48 AM Revision 0b65a7a1 (git): [rubygems/rubygems] Fix test_gem_attr
- https://github.com/rubygems/rubygems/commit/7c5b66f9ef
-
- https://github.com/rubygems/rubygems/commit/506c5bce49
-
- https://github.com/rubygems/rubygems/commit/b3b5c2d379
-
08:48 AM Revision 73633683 (git): [rubygems/rubygems] add tests
- https://github.com/rubygems/rubygems/commit/8a7e27381c
-
- https://github.com/rubygems/rubygems/commit/1d18b12a26
-
- Previously it wouldn't play nice with the bundler version finder.
https://github.com/rubygems/rubygems/commit/d8bb81556d -
- https://github.com/rubygems/rubygems/commit/0832c079c8
-
- https://github.com/rubygems/rubygems/commit/7fd2460f76
-
- https://github.com/rubygems/rubygems/commit/d4ba75dfd2
-
- https://github.com/rubygems/rubygems/commit/7134c49179
-
- https://github.com/rubygems/rubygems/commit/6405a1e51a
- 08:48 AM Revision 5c872b29 (git): [rubygems/rubygems] Avoid adding OpenSSL::PKey::RSA instances
- https://github.com/rubygems/rubygems/commit/ba021fb4be
- 08:48 AM Revision d1c97ceb (git): [rubygems/rubygems] Update expectation in test_to_ruby_with_rsa_key
- https://github.com/rubygems/rubygems/commit/2e65f7d4ae
- 08:48 AM Revision 2e9c0787 (git): [rubygems/rubygems] Fix indentation in case statement
- https://github.com/rubygems/rubygems/commit/8ac0647659
- 08:48 AM Revision 8f2379b0 (git): [rubygems/rubygems] Make ruby_code method handle OpenSSL::PKey::RSA objects
- https://github.com/rubygems/rubygems/commit/b1d825ab3a
- 08:48 AM Revision 508afe2c (git): [rubygems/rubygems] Set SOURCE_DATE_EPOCH env var if not provided.
- Fixes #2290.
1. `Gem::Specification.date` returns SOURCE_DATE_EPOCH when defined,
2. this commit makes RubyGems set it _persistently_ when not provided.
This combination means that you can build a gem, check the build time,
and use tha... - 08:48 AM Revision 8436b271 (git): [rubygems/rubygems] Add a gem attr to the Gem::Package class.
- https://github.com/rubygems/rubygems/commit/5b81f364ae
-
- * Always the same frozen String for a given Symbol.
* Avoids extra allocations whenever calling Symbol#to_s.
* See [Feature #16150] -
- In #15865, a new syntax `<expr> in <pattern>` was introduced. By using this, we can write:
```
json = { foo: 1, bar: 2}
if json in { foo:, bar: }
p [foo, bar] #=> [1, 2]
end
```
However, we cannot write:
```
p(json in... -
- I wanted to dynamically generate benchmark cases to test various number
of methods. Thus I added a dedicated runner of benchmark-driver. -
- for all compilations and compaction.
Prior to this commit, the last-compiled code has not been used because
MJIT worker is stopped before setting the code, and compaction has also
been skipped.
But it was not intentional and `wait: tru... -
- Yes, it is same as that in irb.
I have attached a log of the output. -
-
- Since the introduction of STR_SHARED_ROOT, the word "shared"
has become very overloaded with respect to String's internal
states. Use a different name for STR_IS_SHARED_M and explain
its purpose. -
- The buffer deduplication codepath in rb_fstring can be used to free the buffer
of shared string roots, which leads to use-after-free.
Introudce a new flag to tag strings that at one point have been a shared root.
Check for it in rb_fstr... -
- Applied in changeset commit:git|3cee99808d629c0ec493955ce8ea019d1f8a637b.
----------
[EXPERIMENTAL] Expression with modifier `in`
[Feature #15865] -
- [Feature #15865]
-
- It works in Ruby 2.2.x and Ruby 2.5.x .
-
- GCC emits a lot of false positives for rb_scan_args because:
* `rb_scan_args(argc, argv, "*:", NULL, &opts);` makes `n_mand == 0`,
* `n_mand == argc + 1` implies `argc == -1`, and
* `memcpy(ptr, argv, sizeof(VALUE)*argc);` explodes
How... -
- https://github.com/ruby/io-console/commit/3d69c577a4
-
- https://github.com/ruby/io-console/commit/77ed8d5a06
09/25/2019
-
- > So how about making an experiment by the next preview(2) to see how big the incompatibility is?
Thank you, I will merge https://github.com/ruby/ruby/pull/2437 and mark the change as experimental.
> ...
I agree String#to_s should ... -
- @matz Great to hear.
I get you are worried about `String#to_s`, but what about others that have been mentioned here, namely:
- `NilClass#to_s`
- `TrueClass/FalseClass#to_s`
- `Module#name`
-
07:51 PM Bug #16148: bugs.ruby-lang.org is not sending email notifications for watched issues
- test
-
- The hash-flag approach for ruby2_keywords has been merged at 3b302ea8c95d34d5ef072d7e3b326f28a611e479. That commit uses ruby2_keywords in delegate, fixing the keyword argument separation issues.
-
- This approach uses a flag bit on the final hash object in the regular splat,
as opposed to a previous approach that used a VM frame flag. The hash flag
approach is less invasive, and handles some cases that the VM frame flag
approach do... -
- matz decided to always treat `:` in `rb_scan_args` as `**opt`, so I have merged my `rb_scan_args-colon-always-keyword` branch at 80b5a0ff2a7709367178f29d4ebe1c54122b1c27.
-
- Cfuncs that use rb_scan_args with the : entry suffer similar keyword
argument separation issues that Ruby methods suffer if the cfuncs
accept optional or variable arguments.
This makes the following changes to : handling.
* Treats as *... -
- return is not allowed from class/module body. But if we insert a return into a block then we can invoke the block then it returns all the way out of the script. This has to be accidental behavior doesn't it? I believe the case below s...
-
- matz (Yukihiro Matsumoto) wrote:
> @eregon [ruby-core:95070] beats me. I am persuaded. I agree with:
> ...
Thank you for your decision.
I think this is going to save many bugs and make numbered parameters significantly simpler.
-
- Dan0042 (Daniel DeLorme) wrote:
> > How would we define the current semantics, without being very complex or confusing?
> ...
That's incomplete, it's much more tricky than that in the now previous semantics:
`_1` is the first numbered... -
12:39 PM Bug #16178: Numbered parameters: _1 should be the same as |x| and _0 should not exist
- > I think `_0` and `_1` are very confusing because people can consider it is sequential meaning. However, the meaning is different.
I agree, but matz chose _0 for the implicit parameter. I think it's a mistake to then use that naming ... -
- > Beautifully simple:
> ...
I think `_0` and `_1` are very confusing because people can consider it is sequential meaning. However, the meaning is different.
It is same as `$0` and `$1`, but they are completely different feature (pro... -
11:16 AM Bug #16178: Numbered parameters: _1 should be the same as |x| and _0 should not exist
- Just to make it more polarized, I like the change, it feels more natural.
-
- Dan0042 (Daniel DeLorme) wrote:
> How unfortunate that you managed to persuade matz :-(
Completely agreed. -
- > How would we define the current semantics, without being very complex or confusing?
Beautifully simple:
_0 is a single implicit parameter, as in x in { |x| }
_1 is the first numbered parameter, as in x in { |x,y,z,etc| }
_2 is th... -
- Applied in changeset commit:git|55e1e22b2d44a8a1eca0f6ed2b11dc0f564f7192.
----------
Changed numbered parameters semantics
* `_1` (and no other numbered parameters) to work as `|x|`.
* giving up `_0`.
[ruby-core:95074] [Bug #16178] -
- @eregon [ruby-core:95070] beats me. I am persuaded. I agree with:
* `_1` (and no other numbered parameters) to work as `|x|`.
* giving up `_0`.
Matz. - 03:27 PM Revision 4755e23d (git): * 2019-09-26 [ci skip]
-
-
- IO#ready? returns true or false only, since r50262(1baa57b0033).
-
- If you want random `Float` values up to `max`, use `Random#rand` or `Random.rand` instead of `Kernel#rand`.
-
11:34 AM Bug #16180: Random.rand(max) ignores / is broken when max ceiling value is a float
- your cases are documented there: https://ruby-doc.org/core-2.6.4/Kernel.html#method-i-rand
`When max.abs is greater than or equal to 1, rand returns a pseudo-random integer greater than or equal to 0 and less than max.to_i.abs.`
`N... -
11:07 AM Bug #16180 (Rejected): Random.rand(max) ignores / is broken when max ceiling value is a float
- ruby 2.6.3p62 (2019-04-16 revision 67580) [x86_64-darwin18]
ruby 2.7.0preview1 (2019-05-31 trunk c55db6aa271df4a689dc8eb0039c929bf6ed43ff) [x86_64-darwin18]
Calling `rand` with a float of > 1 always return 0.
Calling `rand` with a... -
- ARY_SHARED_ROOT_P(ary) is true, ARY_HEAP_CAPA(ary) should not
be called. -
- Introduce a new debug counter `obj_ary_extracapa` which counts
arrays which are `len < capa`. -
-
- Could you show what `ruby -e 'puts $:'` print, and is it same as in irb?
-
- We are building ruby 2.6.3(both 32 and 64bit) on AIX 6.1.
The build options used for 64bit build are:-
```
export LDFLAGS="-L/opt/freeware/lib64 -L/opt/freeware/lib -Wl,-blibpath:/opt/freeware/lib64:/usr/lib:/lib"
```
and for 32... -
- It occurs in KysyCore.dll and the extension library calling that DLL.
Have you asked the author? -
- Loading Kysy dll is required in order to run my test cases .
-
-
- I've expanded my pull request to deprecate taint/trust and related methods with verbose warnings, and make the methods no-ops. I believe this implements matz's plan for Ruby 2.7.
The changes involved removing tainting from all includ... -
- * `_1` (and no other numbered parameters) to work as `|x|`.
* giving up `_0`.
[ruby-core:95074] [Bug #16178] -
- I think the following code behavior is wrong:
```ruby
p instance_exec(["a"]){|*foo, **bar| foo } #=> ["a"]
```
It should return `[["a"]]`.
Matz.
-
-
- This reverts commit 5294ded681992ced2ecbd31c814b63265d2f591e.
-
-
- https://github.com/ruby/io-console/commit/a3ad851b6c
-
- https://github.com/ruby/io-console/commit/ab7653c543