Activity
From 08/23/2018 to 08/29/2018
08/29/2018
-
11:37 PM Bug #15039: Random.urandom and SecureRandom arc4random use
- naruse (Yui NARUSE) wrote:
> I agree with your fundamental concept: if there's more code, there's more bugs.
>
> I... -
08:42 PM Bug #15039: Random.urandom and SecureRandom arc4random use
- I agree with your fundamental concept: if there's more code, there's more bugs.
I don't fully agree with "userspac... -
- shyouhei (Shyouhei Urabe) wrote:
> Freaky (Thomas Hurst) wrote:
> > The other is clarifying the intent of `Random.u... -
- jeremyevans0 (Jeremy Evans) wrote:
> Freaky (Thomas Hurst) wrote:
> > Would rand() stop being a userspace PRNG if i... -
08:41 AM Bug #15039: Random.urandom and SecureRandom arc4random use
- Freaky (Thomas Hurst) wrote:
> > Or is it not about the code quality in general? Maybe the problem here is that hav... -
05:55 AM Bug #15039: Random.urandom and SecureRandom arc4random use
- Freaky (Thomas Hurst) wrote:
> Would rand() stop being a userspace PRNG if it automatically called srand() every 200... -
- shyouhei (Shyouhei Urabe) wrote:
> Frankly, I'm not sure how less trustworthy FreeBSD libc is, compared to its kerne... -
- jeremyevans0 (Jeremy Evans) wrote:
> Freaky (Thomas Hurst) wrote:
> > jeremyevans0 (Jeremy Evans) wrote:
> >
> >... -
- Frankly, I'm not sure how less trustworthy FreeBSD libc is, compared to its kernel. They are developed by the same p...
-
- Freaky (Thomas Hurst) wrote:
> jeremyevans0 (Jeremy Evans) wrote:
>
> > On OpenBSD, arc4random is not really a us... -
- jeremyevans0 (Jeremy Evans) wrote:
> On OpenBSD, arc4random is not really a userspace PRNG in the sense that it is... -
09:05 PM Bug #15043: MinGW intermittent test failure - TestIO#test_select_leak
- @normalperson Eric,
Thanks for looking at it and also the patch. It passed test-all, but as mentioned, it's inter... -
07:52 PM Bug #15043: MinGW intermittent test failure - TestIO#test_select_leak
- [email protected] wrote:
> Bug #15043: MinGW intermittent test failure - TestIO#test_select_leak
> https://bugs.... -
- Applied in changeset trunk|r64590.
----------
test/ruby/test_io.rb (test_select_leak): use handle_interrupt
Interru... -
- @normalperson Eric,
Thought I'd ping you on this, as the test is very new. The above test is not failing on Travi... -
06:40 PM Bug #15044: ENV encoding not UTF-8 by default
- If I put this into a .rb file:
puts 'secret'.encoding
ENV['PASS'] = 'secret'
puts ENV['PASS'].encodi... -
02:44 PM Bug #15044 (Closed): ENV encoding not UTF-8 by default
- ~~~
$ irb
2.5.1 :001 > 'secret'.encoding
=> #<Encoding:UTF-8>
2.5.1 :002 > ENV['PASS'] = 'secret'; ENV['PASS'].e... -
04:16 PM Bug #15045 (Closed): Test suite fails/hangs with stronger cryptography enforcement
- Testing in Fedora Rawhide with ```$ sudo update-crypto-policies --set NEXT``` [1] which sets a higher bar for cryptog...
-
02:12 PM Bug #15027: When Struct#each method is overriden Struct#select and Struct#to_a use wrong collections
- ```diff
diff --git a/struct.c b/struct.c
index 7de46980aa..e4c875b5be 100644
--- a/struct.c
+++ b/struct.c
@@ -8... -
11:22 AM Feature #14975: String#append without changing receiver's encoding
- ioquatix (Samuel Williams) wrote:
> ```ruby
> u.force_encoding('BINARY') do
> b << u
> end
> ```
>
> @duers... -
11:08 AM Feature #14975: String#append without changing receiver's encoding
- ```ruby
b = 'a'.force_encoding('BINARY')
u = "\u00ff".force_encoding('UTF-8') # aside: force_encoding here is a n... -
- As mentioned, the general idea of Ruby m17n is that strings that only contain ASCII bytes (8th bits are all 0) are tr...
-
- gotoken (Kentaro Goto) wrote:
> For example, `"ss"` and `"st"` are mapped `"ß"` (`"\u00DF"`) and `"st"` (`"\uFB06"`... -
- Eric Wong wrote:
> It will be difficult/unsafe to use auto-fiber/Thread::Coro
> [Feature #13618] without this
I... -
- Koichi Sasada wrote:
> Bug #595: Fiber ignores ensure clause
> http://redmine.ruby-lang.org/issues/595
What's t... -
- [email protected] wrote:
> (I'm not sure why fork removes other fibers, but)
fib = Fiber.new { puts " HI" }
fork ... -
- Applied in changeset trunk|r64589.
----------
cont.c: set th->root_fiber to current fiber at fork
Otherwise, th->ro... -
07:20 AM Bug #15041: [PATCH] cont.c: set th->root_fiber to current fiber at fork
- (I'm not sure why fork removes other fibers, but)
I think it is okay. let's try.
-
- cont.c: set th->root_fiber to current fiber at fork
Otherwise, th->root_fiber can point to an invalid Fiber,
beca...
08/28/2018
-
- On OpenBSD, arc4random is not really a userspace PRNG in the sense that it is reseeded on a regular basis using new e...
-
- Random.urandom defaults to arc4random() on a lot of platforms, including FreeBSD.
On all currently released versio... -
- I just noticed that the Appveyor MinGW build is not showing the installed bundled gems in the console output, althoug...
-
08:22 AM Feature #14097: Add union and difference to Array
- [@Student](https://bugs.ruby-lang.org/users/6774) as I just answered in the [PR](https://github.com/ruby/ruby/pull/17...
-
08:19 AM Feature #15036 (Closed): after upgrade when run apt upgrade metasploit
- ...
-
- Applied in changeset trunk|r64583.
----------
rest parameter optimization [Feature #15010]
* vm_args.c: rb_ary_dup(... -
- It seems fine.
I'll commit it. -
- Eregon (Benoit Daloze) wrote:
> There are certainly cases where a C extension wants to check the argument is an Inte... -
03:58 AM Bug #15035: Windows make install failures
- @MSP-Greg
Thanks for your patches. I will merge a patch of `tool/rbinstall.rb` after confirming with my windows en... -
- @hsbt
Thanks for applying the patch. I should have been clearer, as the patch fixes two issues, both involving Ru... -
- Applied in changeset trunk|r64582.
----------
Fixed test failures in mswin environment at r64555.
[ruby-core:8869... -
- @hsbt
Windows has been failing on install, seems due to a combination of RubyGems changes.
ruby-loco is passing... -
- On 2018/08/27 18:17, Eric Wong wrote:
>>> A joke; at least as far as we know it is a joke :>
>>
>> I'm not sure w... -
- I hope this is relevant.
I found an interesting article here: http://www.oilshell.org/blog/2016/12/11.html
It d... -
12:12 AM Bug #15034 (Feedback): Crash in Opensuse Software management
- Thank you for your report.
This is a issue tracker for ruby development, and I'm not at all familiar with Yast2 pa...
08/27/2018
-
09:37 PM Bug #15034 (Closed): Crash in Opensuse Software management
- Hi,
the problem appears when starting the Software Management component of Open Suse's Yast2 package manager.
Th... -
09:19 PM Bug #15033 (Closed): Encoding fallback uses wrong character when multiple conversions are required
- When converting a string from one encoding to another that involves multiple conversions, the proc passed to `encode`...
-
- > https://bugs.ruby-lang.org/issues/13618
Btw, many of my recent changes ([Misc #15014] (rb_hrtime_t)),
and proc... -
06:34 PM Feature #14912: Introduce pattern matching syntax
- "The patterns are run in sequence until the first one that matches."
This means O(n) time for matching. If we are ... -
04:07 PM Bug #14920: Backport r63857 to fix performance problem on Time class in MacOs systems
- Hi @usa.
63994 alone doesn't fix the issue. I just cloned ruby_2_4 with r64560 and run the same benchmark script ... -
01:59 PM Bug #14920: Backport r63857 to fix performance problem on Time class in MacOs systems
- ruby_2_4 r64560 merged revision(s) 63994.
-
- ruby_2_4 r64564 merged revision(s) 64092.
-
- ruby_2_4 r64563 merged revision(s) 64071.
-
- ruby_2_4 r64562 merged revision(s) 64014.
-
- ruby_2_4 r64561 merged revision(s) 64007,64019,64020.
-
- ruby_2_4 r64559 merged revision(s) 63696.
-
- ruby_2_4 r64558 merged revision(s) 62731,62735.
-
- [email protected] wrote:
> There are certainly cases where a C extension wants to check the argument is an Integer,... -
10:02 AM Feature #15030: Have T_INTEGER constant for checking Integer types in C API
- There are certainly cases where a C extension wants to check the argument is an `Integer`, and its range doesn't matt...
-
- +1 for Hanmac.
C extension writers should check `T_FIXNUM` or `T_BIGNUM` explicitly.
-
05:49 AM Feature #15030: Have T_INTEGER constant for checking Integer types in C API
- hm that is not that easy, while `Fixnum` and `Bignum` are deprecated on the ruby side,
on the C side they still exis... -
05:12 AM Feature #15030 (Rejected): Have T_INTEGER constant for checking Integer types in C API
- After Integer unification in Ruby 2.4, I think it makes sense to introduce a T_INTEGER
type that helps in directly ... -
- ko1 (Koichi Sasada) wrote:
> `rb_obj_is_range()` can be accepted like `rb_obj_is_proc` and so on.
> `T_xxx` shows d... -
- `rb_obj_is_range()` can be accepted like `rb_obj_is_proc` and so on.
`T_xxx` shows data layout of `RVALUE`.
-
- Range is an in-built type of Ruby and it would be better for C extension
writers if there existed a constant `T_RANG... -
- > Is there a use-case for using this new method with non-binary strings?
Yes, predictably appending strings withou... -
- I agree the current behavior for appending with binary strings is surprising.
Especially since the result of the ope... -
10:08 AM Feature #15010: Reduce allocation for rest parameters
- @ko1, I've added new field rest_dupped to args_info.
Updated patch https://bugs.ruby-lang.org/attachments/7344/Re... -
- ko1 (Koichi Sasada) wrote:
> # Idea (as my understanding)
>
> ~a rest parameter" is dup multiple times because of... -
- Sorry for late response.
# Idea (as my understanding)
~a rest parameter" is dup multiple times because of curre... -
- @ko1, It would be great if you could review https://bugs.ruby-lang.org/attachments/7343/Reduce-allocation-for-rest-pa...
-
- Koichi Sasada <[email protected]> wrote:
> On 2018/08/27 16:16, Eric Wong wrote:
> > Yep. Maybe some small systems ca... -
- On 2018/08/27 16:16, Eric Wong wrote:
> Yep. Maybe some small systems can benefit with
> usec/msec resolution and... -
- [email protected] wrote:
> Thanks!
>
> > + * Hi-res monotonic clock. It is currently nsec resolution, which has ove... -
- Thanks!
> + * Hi-res monotonic clock. It is currently nsec resolution, which has over 500 years of range (unsigne... -
06:31 AM Bug #13671: Regexp with lookbehind and case-insensitivity raises RegexpError only on strings with certain characters
- Thanks shyouhei for your pointing out.
I imagine another Rexexp option, say `//I`, which is almost the same as `/... -
- gotoken (Kentaro Goto) wrote:
> By the way, this expansion by `//i` option looks over kill for me.
> I wish case s... -
- Thanks znz. The workaround is helpful. And I understood what was happened.
https://github.com/k-takata/Onigmo/issues... -
03:46 AM Bug #13671: Regexp with lookbehind and case-insensitivity raises RegexpError only on strings with certain characters
- You can use `(?:s)` instead of `s` for workaround.
```
$ ruby -ve '/(?<=ast)/iu'
ruby 2.5.1p57 (2018-03-29 revis... -
- I encountered a non `ss` case. Is this a same problem?
```
% ruby -ve '"".match(/(?<=ast)/ui)'
ruby 2.6.0dev (2... -
- ioquatix (Samuel Williams) wrote:
> Doing a binary comparison would be acceptable.
I don't think so. As Akira al... -
- Doing a binary comparison would be acceptable.
-
- @nobu Isn't changing TZ enough that `#localtime` should do something?
-
- @marcandre I already added support for coverage of templates and it works pretty well. You need to be careful with te...
-
- Did you take a look at https://github.com/ioquatix/covered - I'd be interested in your feedback. Can we discuss furth...
-
- What about using trace points?
-
- * [Feature #14975] String#append without changing receiver's encoding (ioquatix)
* Can we review this PR? We need ...
08/26/2018
-
11:49 PM Misc #15007: Let all Init_xxx and extension APIs frequently called from init code paths be considered cold
- naruse (Yui NARUSE) wrote:
> methodmissing (Lourens Naudé) wrote:
> > ko1 (Koichi Sasada) wrote:
> > > I agree to ... -
- as a middle way, can't we just do the "is overwritten by user" check?
i think i have seen it on other classes like A... -
- Since matz asked for feedback, just a comment - I have no particular pro or con opinion
per se, mostly because I ver... -
- I'm against changing the current behavior. If you can override `each`, you can override other methods. If we wanted...
-
01:56 AM Bug #15027 (Feedback): When Struct#each method is overriden Struct#select and Struct#to_a use wrong collections
- The proposed behavior is more consistent but slower. I am not sure it's a good idea to hinder performance by supporti...
-
- Thank you for the answer. I'm assigning this ticket to matz.
-
- @mame
> However, I'm now getting confused. I thought you agreed that Array#union should be non-destructive, but in... -
08/25/2018
-
- methodmissing (Lourens Naudé) wrote:
> ko1 (Koichi Sasada) wrote:
> > I agree to introduce COLD attribute.
>
> I... -
05:29 PM Bug #15027 (Rejected): When Struct#each method is overriden Struct#select and Struct#to_a use wrong collections
- ### Bug
Here's the code snippet that should reproduce the problem:
~~~ ruby
class Foo < Struct.new(:bar)
de... -
- Ana, I watched the video of [your EuRuKo talk](https://www.youtube.com/watch?v=jUc8InwoA-E). I could understand the ...
-
- Applied in changeset trunk|r64533.
----------
thread.c: use rb_hrtime_t scalar for high-resolution time operations
... -
- I'd like somewhat to agree with the motivation. Indeed, I sometimes feel I want to insert separators between each pa...
-
- `Array#join` concatenates array elements recursively.
What do you expect as the index between different level elements? -
- Seems an already fixed bug.
08/24/2018
-
- For `%{a b c d}.join { |_,_,i| i.to_s } == "a0b1c2d"`, you could do:
~~~ ruby
*a, l = %w{a b c d}; a.each.with_ob... -
07:54 PM Feature #15024: Support block in Array#join
- Doesn't `#each_slice` create temporary array for each pair? Doesn't seem very efficient. But assuming that does not m...
-
- Your examples are both possible to implement using existing Array methods:
~~~ ruby
puts %w{a b c d}.each_slice(2... -
- I think it could be handy to have block support in Array#join.
For example
```ruby
> puts %w{a b c d}.join { |... -
08:03 PM Bug #15025 (Closed): Encountered a bug in interpreter or extension libraries
- /Users/suyesh/.rbenv/versions/2.6.0-preview2/lib/ruby/gems/2.6.0/gems/bootsnap-1.3.1/lib/bootsnap/compile_cache/iseq....
-
- Applied in changeset trunk|r64527.
----------
thread_pthread.c: use eventfd instead of pipe on Linux
Based on r6447... -
- `Time.at` supports `Rational` numbers too.
```ruby
puts Time.at(1486525793.995r).strftime("%s.%6N") #=> 148652579... -
- IEEE 754 floating point numbers are inexact, and cannot represent that value exactly.
```ruby
printf("%.20f\n", 1... -
03:41 PM Bug #15023 (Rejected): Time precision (in microsecond) inaccurate after performing strftime
- Please see the example below. I think we should expect the result to be "1486525793.995". Right?
OS version: MacOS... -
02:47 PM Bug #14997: Socket connect timeout exceeds the timeout value for
- If anyone is actually willing to confirm, that it is indeed an unwanted / unexpected behavior, I offer to fix it.
... -
- I think this is duplicate of https://bugs.ruby-lang.org/issues/14897 which contain the fix and the backport request t...
-
- Can reproduce this on 2.5, but not on trunk. Seems 2.5 needs a backport?
Not sure which is the exact revision though. -
09:44 AM Bug #15021: Segfault when compiling certain code on Ruby 2.5.1
- Sorry there's a small typo. The command to run the script is `ruby code.rb` not `ruby foo.rb`
-
## Steps to reproduce
save the following code to `code.rb`
```
code =<<CODE
if false
new(arg: $arg)
end
CO...-
- * [Feature #15022] Oneshot coverage
* I'd like to introduce a new kind of coverage to record whether each line is ... -
- I'd like to introduce a new feature to the coverage library, namely, "oneshot coverage".
## Synopsis
The follow... -
10:25 AM Bug #15020 (Closed): [PATCH] Add 'guards' for mswin, mingw, update test_readline.rb
- Applied in changeset trunk|r64526.
----------
ext/readline/extconf.rb: try using more readline APIs
on MinGW.
[Bug... -
- @k0kubun,
I forgot one patch that I have on `ext/readline/extconf.rb`, at [ruby-loco/patches/ext-readline-extconf.... -
- ~~~
Hello, I took a look at add-hash-support-for-transient-heap.patch.
It seems lines are copy & pasted from st.c... -
08/23/2018
-
- Hanmac (Hans Mackowiak) wrote:
> what happens when i would call such cold function outside of Init_xxx ?
I consid... -
- ko1 (Koichi Sasada) wrote:
> I agree to introduce COLD attribute.
I was wondering about Eric's suggestion about t... -
- what happens when i would call such cold function outside of Init_xxx ?
what functions in my own binding should be... -
- I agree to introduce COLD attribute.
----
> TLB (translation lookaside buffer) specific
Do PHP people specif... -
- When the mingw build was added, test_readline.rb failed. One test fails on mingw, several tests error due to the win...
-
- https://ruby-doc.org/stdlib-2.5.1/libdoc/net/http/rdoc/Net/HTTP.html claims that `At this time Net::HTTP does not sup...
-
- [email protected] wrote:
> Just a comment.
> API references (a list of provided functions with a comment) at the beg... -
- Just a comment.
API references (a list of provided functions with a comment) at the beginning of the hrtime.h will h... -
- ko1 (Koichi Sasada) wrote:
> sorry, which patch should I review?
Reduce-allocation-for-rest-parameters-v1.patch -
- sorry, which patch should I review?
-
-
- 見逃してました。
これ、実装見てたら、どーしょーもない気がするんで、制限のある仕様、ってことで駄目でしょうか。駄目かなあ。実装アイディア募集。 -
- Applied in changeset trunk|r64514.
----------
check trace flags at loading [Bug #14702]
* iseq.c (iseq_init_trace):...
Also available in: Atom