Activity
From 11/24/2016 to 11/30/2016
11/30/2016
-
05:29 PM Feature #12831 (Closed): /\X/ (extended grapheme cluster) can't pass unicode.org's GraphemeBreakTest
- Applied in changeset r56949.
----------
Regexp supports Unicoe 9.0.0's \X
* meta character \X matches Unicode 9.0.0 characters with some workarounds
for UTR #51 Unicode Emoji, Version 4.0 emoji zwj sequences.
[Feature #12831] [ruby... - 05:29 PM Revision 48beedd5 (git): * 2016-12-01
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56950 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * meta character \X matches Unicode 9.0.0 characters with some workarounds
for UTR #51 Unicode Emoji, Version 4.0 emoji zwj sequences.
[Feature #12831] [ruby-core:77586]
The term "character" can have many meanings bytes, codepoints,... -
03:58 PM Feature #12906: do/end blocks work with ensure/rescue/else
- Checking that I submitted this correctly, I usually do Github, so might have gotten this wrong.
-
03:53 PM Bug #12994 (Feedback): Embedded ruby vs2015 build crashes
- I don't know what is "an ruby embedded engine".
But I can say that you have to call `rb_w32_sysinit` before calling `ruby_init` on Windows. -
03:38 PM Bug #12994 (Closed): Embedded ruby vs2015 build crashes
- When an embedded ruby engine in a vs2015 application is initialised using ruby_init the function set_pioinfo_extra in win32.c is not called.
This results in a crash when ruby_init is called and it is trying to derefence the __pioinfo po... -
03:18 PM Bug #12993: SEGV in OpenSSL::TestSSL#test_close_after_socket_close
- Actually, I thought the fix was applied in the package according to the OpenSSL changelog:
```
* Tue Nov 22 2016 Tomáš Mráz <tmraz@redhat.com> 1.1.0c-2
- revert SSL_read() behavior change
- patch from upstream (#1394677)
- EC curv... -
02:46 PM Bug #12993 (Closed): SEGV in OpenSSL::TestSSL#test_close_after_socket_close
- Imported at r56946.
-
- Not really a segfault but rb_sys_fail() is calling rb_bug() because errno is 0. OpenSSL 1.1.0c introduced a breaking change in SSL_read()[1], and it is already reverted in the OpenSSL-1_1_0-stable branch.
I have patched ruby/openssl[2... -
- r56693 used to build for me and now it fails, so this is apparently some change in Fedora Rawhide.
-
- Testing the latest version of Ruby from trunk on Fedora Rawhide, the OpenSSL::TestSSL#test_close_after_socket_close segfaults (see the attached log). Not sure what might be possible the reason, if it is some change in Fedora in Ruby.
... -
02:54 PM Revision e680bfb1 (git): ossl.c: cast
- * ext/openssl/ossl.c (ossl_pem_passwd_cb): cast to int. it's safe
because len does not exceed int max_len.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56948 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
02:46 PM Feature #12921: Retrieve user and password for proxy from env
- Shyouhei Urabe wrote:
> Yuri Samoilenko wrote:
> ...
Yuri Samoilenko said than it's operation system problem, if any users can read you env. Not ruby.
Anyway, http-proxy it's standart way set proxy and with auth too.
-
02:42 PM Feature #12921: Retrieve user and password for proxy from env
- Yuri Samoilenko wrote:
> What do you mean when say "insecure"? Storing login and password in filesystem and then read it and pass to http request manually is secure? Insecure is availability to pass login/password in plain form like "ht... -
08:08 AM Feature #12921: Retrieve user and password for proxy from env
- Shyouhei Urabe wrote:
> You have to understand that environment variables are in fact insecure. It is a very bad idea to store passwords there. Surprising or not, insecure is insecure. We shall not introduce this.
What do you mean... -
- * parse.y (parser_tokadd_utf8): refine error message at bad char
in unicode escape, "invalid" instead of "unterminated".
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56947 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Import Ruby/OpenSSL 2.0.0. The full commit history since 2.0.0 beta.2
(imported at r56098) can be found at:
https://github.com/ruby/openssl/compare/v2.0.0.beta.2...v2.0.0
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56946 b2dd03... -
02:24 PM Revision 0a5abaf7 (git): Fixed error with `make install`
- rbinstall.rb couldn't load "psych/versions".
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56945 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- It fixed segmentation fault related with GC.
https://github.com/ruby/psych/pull/296
and removed deprecated code of Ruby 1.8.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56944 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * parse.y (parser_tokadd_codepoint): extract from
parser_tokadd_utf8.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56943 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * parse.y (parser_tokadd_utf8): make open/close braces constant.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56942 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Thx for implementing this. ```--with-compress-debug-sections=no``` seems to disable the compression of debug sections.
-
08:36 AM Bug #12990: unicode_case_mapping_tests
- A short additional note: If cases like these increase with future versions of Unicode, I plan to change the special treatment of U+03B9/U+03BC/U+A64B to something more data-based.
-
- Applied in changeset r56941.
----------
fix uppercasing for U+A64B, CYRILLIC SMALL LETTER MONOGRAPH UK
* enc/unicode.c: Add U+A64B to the special cases 03B9 and 03BC
at the end of onigenc_unicode_case_map (Bug #12990).
* enc/unicode... -
- Just for the record, I think this is where the problem was introduced:
https://bugs.ruby-lang.org/projects/ruby-trunk/repository/revisions/55478/diff/test/ruby/enc/test_case_comprehensive.rb
That explains why almost all characters are ... -
- * enc/unicode.c: Add U+A64B to the special cases 03B9 and 03BC
at the end of onigenc_unicode_case_map (Bug #12990).
* enc/unicode/case-folding.rb: Add U+A64B to the special cases
03B9 and 03BC. Add a comment pointing to enc/unicode.... -
- Shyouhei Urabe wrote:
> The new default is called the "banker's round" for reasons. Money related people should prefer this.
We don't know to what extent such an argument applies around the world.
> ...
Definitely. But while we... -
- Jerome Cornet wrote:
> But does this mean that Float rounds in a different way as BigDecimal by default then ?
> ...
Correct. It is true that we fixed one inconsistency to introduce another. This might perhaps be a problem. Not sure... -
02:43 AM Bug #12958: Breaking change in how `#round` works
- > Am I making myself understood? I am distinguishing Floats and BigDecimals. We did not change how a BigDecimal rounds. Money related calculations should use that class instead. Not Float. Using Float for money is in fact toxic; for...
-
- Jerome Cornet wrote:
> Unfortunately there are many instances where money math requires floating point calculation, like taxes for example (discounts are also another source of float-related money math)
I would like to disagree.
I... -
- Unfortunately there are many instances where money math requires floating point calculation, like taxes for example (discounts are also another source of float-related money math)
Most jurisdictions that charge sales taxes specify rou... -
- Jerome Cornet wrote:
> Shyouhei Urabe wrote:
> ...
The new default is called the "banker's round" for reasons. Money related people should prefer this. And to say frankly, using Floats for money-related calculation is the source of w... -
- * parse.y (parser_yylex): do not warn parentheses of lambda
argument list, because there is no name.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56940 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
01:22 AM Revision f7d130f4 (git): Supress warnings.
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56939 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- You could close this issue by including `[Bug #12988]` in the commit message.
11/29/2016
-
08:59 PM Feature #12979: Avoid exception for #dup on Integer (and similar cases)
- Benoit Daloze wrote:
> `1.clone(freeze: false)` does not do what you say.
> ...
Oh wow. That's entirely my fault. I totally confused Ruby's dup/cloning with another language (and/or my imagination, haha).
Fortunately my point still... -
08:39 PM Feature #12979: Avoid exception for #dup on Integer (and similar cases)
- Mike Vastola wrote:
> Nobuyoshi Nakada wrote:
> ...
`1.clone(freeze: false)` does not do what you say.
The keyword :freeze has default value true for #clone, which means if the original object is frozen so will be the clone.
See #123... -
- Nobuyoshi Nakada wrote:
> Should `Kernel#clone` be same?
The original consensus was that the fix should be for both **`#dup`** and **`#clone`** to fail silently on immediate objects. The person who created this issue rendered **`#clo... -
02:52 PM Feature #12979: Avoid exception for #dup on Integer (and similar cases)
- Integer (Fixnum) can dup now.
But Integer (Bignum) can't dup.
~~~
% ruby -ve '1.dup'
ruby 2.4.0dev (2016-11-29 trunk 56928) [x86_64-linux]
% ruby -ve '(1<<64).dup'
ruby 2.4.0dev (2016-11-29 trunk 56928) [x86_64-linux]
-e:1:in `d... -
02:00 PM Feature #12979: Avoid exception for #dup on Integer (and similar cases)
- I vote for ArgumentError, because `clone(freeze: false)` is invalid operation for immediate (thus unfreezable) objects.
Matz.
-
- I remember `clone` has `freeze` option.
What should `1.clone(freeze: false)` do, ignore the option, or raise an `ArgumentError`? -
05:29 PM Bug #12988 (Closed): Calling `inspect` sometimes causes a segv
- Should be fixed in r56938
-
- Confirmed the overrun, although it didn't segfault.
Please commit the fix. -
12:31 AM Bug #12988: Calling `inspect` sometimes causes a segv
- tenderlove@ruby-lang.org wrote:
> * Backport: 2.1: UNKNOWN, 2.2: UNKNOWN, 2.3: REQUIRED
I think a backport will be required for all supported versions.
I'm surprised this remained undiscovered for so many years.
> I've attached ... -
- If you have code like this:
```ruby
class A
def initialize
@a = nil
@b = nil
@c = nil
@d = nil
@e = nil
end
end
x = A.new
y = x.clone
100.times { |z| x.instance_variable_set(:"@foo#{z}", nil) }
puts y.inspect
``... -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56937 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- Follow net/http and open-uri. Don't rely on the constants/methods from
OpenSSL::TestUtils.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56936 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Follow net/http. Define own test DH parameters and use.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56935 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Make test code independent of test/openssl/utils.rb. The development of
openssl library has moved to a separate repository and
OpenSSL::TestUtils may be modified at any time.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56934 b2dd0... -
- * object.c (special_object_p): no longer raise a TypeError for
Integer and Float, and return itself instead. [Feature#12979]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56933 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- [ci skip]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56932 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 03:23 PM Revision 40c9f1f3 (git): get rid of test failures on CI introduced at r56927
- * lib/matrix.rb: now ruby warns ambiguous parentheses after a space in method definitions.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56931 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Shyouhei Urabe wrote:
> As far as I can see the Rails breakage only happens in views, where 30secs is shown "less than a minute" instead of "1 minute". That sounds very trivial to me. And yet, did not break silently.
Actually, any... - 03:14 PM Revision 18658d7d (git): * 2016-11-30
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56930 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * object.c (rb_obj_clone2): no longer raise a TypeError for
special constants, and return itself instead. however, if
freeze option is false, raise an ArgumentError. [Feature#12979]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trun... -
02:27 PM Feature #12992 (Open): ArgumentError if hostname is missing
- Tue Nov 29 15:07:31 2016 Mohamed Osama <mohamed.o.alnagdy@gmail.com>
* lib/net/http/generic_request.rb: Throw ArgumentError if hostname is
not provided in URI -
- Applied in changeset r56928.
----------
thread.c: fix doc of abort_on_exception [ci skip]
* thread.c (rb_thread_s_abort_exc, rb_thread_s_abort_exc_set):
[DOC] the raised exception will be re-raised in the main thread,
and then foll... -
12:09 PM Bug #12991 (Closed): Documentation of Thread.abort_on_exception is wrong: it raises error again in main thread
- Currently, ruby doc says that:
```
When set to true, all threads will abort (the process will exit(0)) if an exception is raised in any thread.
```
But actually, the error raised in thread (with abort_on_exception = true) will ... -
- * thread.c (rb_thread_s_abort_exc, rb_thread_s_abort_exc_set):
[DOC] the raised exception will be re-raised in the main thread,
and then follows the ordinary exception sequence, exit status is
not 0. [ruby-core:78415] [Bug #12991]... -
- * parse.y (parser_yylex): warn ambiguous parentheses after a space
in method definitions.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56927 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Fix documentation of String#casecmp? (examples didn't have the '?').
Add an example with non-ASCII characters. Clarify that casecmp,
unlike casecmp?, only does case-insensitivity on A-Z/a-z.
[ci skip]
git-svn-id: svn+ssh://ci.ruby-lang.... -
- Ryan Ringler wrote:
> I see the changeset implements the new method by calling `String#downcase` on both strings and compares the results for equality.
Yes, this is as proposed at https://bugs.ruby-lang.org/issues/12786#note-1.
... -
03:42 AM Feature #12786: String#casecmp?
- Thank you for incorporating this!
I see the changeset implements the new method by calling `String#downcase` on both strings and compares the results for equality. While this does allow unicode support in advance of `String#casecmp`,... -
- The culprit is line 4555 in enc/unicode/9.0.0/casefold.h:
```
{0xa64b, {2, {0x1c88, 0xa64a}}},
```
With comments, this looks as follows:
```
{0xa64b /* ꙋ */, {2, {0x1c88, 0xa64a /* ᲈ, Ꙋ */}}},
```
Correct would be
... -
- Nobuyoshi Nakada wrote:
> I found that tests generated by `TestComprehensiveCaseFold.unicode_case_mapping_tests` compares `target` with same `target`, which should be always true usually.
Nice catch, thanks! This is indeed a serious ... -
- Applied in changeset r56924.
----------
Fix erroneous test of target against target
* test/ruby/enc/test_case_comprehensive.rb: fix test condition,
add a temporary check for U+A64B, the only character where the tests
currently fail... -
- Failed results:
```
$ ./ruby test/runner.rb test/ruby/enc/test_case_comprehensive.rb
Run options:
# Running tests:
[258/422] TestComprehensiveCaseFold#test_UTF-16BE_capitalize = 0.30 s
1) Failure:
TestComprehensiveCaseFold... -
- I found that tests generated by `TestComprehensiveCaseFold.unicode_case_mapping_tests` compares `target` with same `target`, which should be always true usually.
I suspect this is unintentional, tried comparison `target` with `result`,
... -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56925 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * test/ruby/enc/test_case_comprehensive.rb: fix test condition,
add a temporary check for U+A64B, the only character where the tests
currently fail. (Bug #12990)
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56924 b2dd03c8-39d4-... -
- Errno::ECONNRESET was raised by test_tls_post_connection_check and
test_tls_unknown_ca on mswin64 CI.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56923 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Encoding argument should be prior to `binmode` option.
```ruby
open(IO::NULL, "r", binmode: true){|f| p [f.binmode?, f.external_encoding]} #=> [true, #<Encoding:ASCII-8BIT>]
open(IO::NULL, "r:utf-8", binmode: true){|f| p [f.binmode?... -
- * test/ruby/test_keyword.rb (test_block_required_keyword): more
assertions for positional and require keyword parameters.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56922 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * test/ruby/test_keyword.rb (test_block_required_keyword):
assert_raise_with_message returns the raised exception object.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56921 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * string.c (rb_str_casemap): use xmalloc simply instead of
ALLOC_N.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56920 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- The TLS handshake timeout can be specified independently, because the TLS
handshake doesn't start just after the underlying connection is established,
but after the AUTH command is completed. It's also useful for testing
purposes. Howe... -
- Otherwise, @sock.read in Net::FTP#close hungs until read_timeout exceeded.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56918 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 02:21 AM Revision d8a0f14b (git): * 2016-11-29
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56917 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- Otherwise, file descriptor leaks may occur in Net::FTP.new.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56916 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
11/28/2016
-
- Hi,
> Is there any problem with the following code?
> ...
2 problems
1. It's slower than it could be (as Naruse demonstrates)
2. It's very noisy when `-d` is enabled.
In Psych, I am trying to avoid noise from `-d`. That means... -
- Here is a sample program:
~~~ruby
Encoding.default_external = Encoding::UTF_8
Encoding.default_internal = Encoding::UTF_8
reader, writer = IO.pipe(binmode: true)
reader.binmode? # => true
reader.external_encoding # =... -
- `rb_obj_inspect` calls `rb_ivar_count ` to find the number of instance variables on an object. `rb_ivar_count` uses `tbl->num_entries` on the instance variable index table to determine how far in to the instance variable array it should...
-
05:31 PM Bug #12984: `rescue *[]` should be equivalent to `rescue` as `method_call(*[])` is equivalent to `method_call`
- bug hit wrote:
> Nobuyoshi Nakada wrote:
> ...
The difference between rescue and super is that there is such a thing as an explicit empty `super()` that passes nothing, but there is no corresponding explicit empty `rescue()` that rescu... -
- Nobuyoshi Nakada wrote:
> It's similar to:
> ...
I guess there's some similarity. But super has a very explicit definition. Only a naked super is auto-forwarding, any attempt to pass args turns it into manual super. So `super(*[])` i... -
- It's similar to:
```ruby
super(*[])
``` -
- * string.c (mapping_buffer): get rid of zero-length array member,
which is not a part of C90.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56915 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
12:47 PM Bug #12987: Win32 API.rb can not call function without argument.
- Thank you very much.
I just learned that only nil was clearly stating that there are no arguments in the previous document.
I used a single 'V' to specify no arguments, so it was up to this idea. -
- For workaround, you can use '' instead of '0'.
-
- In Win32API.rb, it is not possible to call a function without arguments.
I could avoid it for the time being if I did the following (against trunk).
~~~
diff --git a/ext/win32/lib/Win32API.rb b/ext/win32/lib/Win32API.rb
index d03ec... - 10:17 AM Revision f2a1f32f (git): hide exceptions in server thread
- * test/net/ftp/test_ftp.rb (test_tls_connect_timeout): sleeping to give a chance of complete of accept is a timing issue at all. therefore, simply ignore the exception.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56914 b2dd03c8-... -
- * string.c (rb_str_casecmp_p): [DOC] move forward declaration of
rb_str_downcase to enable rdoc.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56913 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Applied in changeset r56912.
----------
implement String/Symbol#casecmp? including Unicode case folding
* string.c: Implement String#casecmp? and Symbol#casecmp? by using
String#downcase :fold for Unicode case folding. This does not ... -
-
- * string.c: Implement String#casecmp? and Symbol#casecmp? by using
String#downcase :fold for Unicode case folding. This does not include
options such as :turkic, because these currently cannot be combined
with the :fold option. Thi... -
04:04 AM Feature #12986 (Open): HTTP/request basic_auth context should be switched
- HTTP basic auth should be persisted over the HTTP object and not the request object. The consequence is having to authenticate on every request.
-
- I don't really care if this gets low priority. I'd just like to have a quick discussion and get it approved.
-
- Need to implement an RFC-compliant cookie handling method. Specifically, set-cookies need to be handled properly; but we really should put in an overall proper handling.
-
- * insns.def (checkmatch): adjust type of the index variable, to
get rid of (potential) overflow.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56911 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 03:14 AM Revision 8e7e71d6 (git): Give a chance to complete server.accept
- * test/net/ftp/test_ftp.rb (test_tls_connect_timeout): get rid of a test error on Windows.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56910 b2dd03c8-39d4-4d8f-98ff-823fe69b080e - 02:02 AM Revision a530b970 (git): * 2016-11-28
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56909 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- * compile.c (iseq_peephole_optimize): remove splatarray following
always-array insn.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56908 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
11/27/2016
-
10:23 PM Feature #12973: [RFC] net/http: use require_relative to reduce syscalls
- I have no particular pro or con opinion, not that it would be relevant anyway - but I am watching this closely out of interest what others have to say. :)
-
- Splatting an empty array to a construct that takes a list is supposed to be equivalent to specifying no list
```ruby
def foo
end
foo *[] #works
```
So `rescue *[]` should be equivalent to `rescue`
```ruby
begin
raise '... -
03:00 PM Bug #12950: irb: 'input-method.rb:151: [BUG] Segmentation fault' / 'malloc(): smallbin double linked list corrupted'
- Umm, I guess the issue may be readine's and this can be reproduced without ruby.
```
#include <stdio.h>
#include <readline/readline.h>
int main() {
FILE *fp;
rl_outstream = NULL;
while(1) {
fp = fdopen... -
- Should `Kernel#clone` be same?
-
- * signal.c (check_stack_overflow): [EXPERIMENTAL] consider the
segfault a stack overflow when the fault address is between SP
and BP.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56907 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * object.c (rb_obj_dup): no longer raise a TypeError for special
constants, and return itself instead. [Feature#12979]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56906 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Seems a simple stack overflow, but it's strange that `sigsegv` handler hasn't been called.
Could you elaborate what you did? -
10:17 AM Bug #12983 (Closed): compatibility_version and current_version on Darwin are reversed
- Committed as r56905.
-
- Please commit it.
-
- The `compatibility_version` should have an API version and the `current_version` should have a program version of Ruby, but they have been reversed and the binary compatibility has never worked.
For example, the ruby 2.3.3 program/dyl... - 10:08 AM Revision 673a6ad9 (git): Reverse compatibility_version and current_version for Darwin
- The `compatibility_version` should have an API version and the
`current_version` should have a program version of Ruby, but they have
been reversed and the binary compatibility has never worked.
git-svn-id: svn+ssh://ci.ruby-lang.org/ru... - 09:40 AM Revision c46946ca (git): * 2016-11-27
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56904 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
09:40 AM Revision edaf46da (git): gems/bundled_gems: Update to power_assert 0.4.1
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56903 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- Nobuyoshi Nakada wrote:
> `String#casecmp` also should have those options, I think.
I agree that once String#casecmp supports full Unicode, it should have these options (and more, because for Unicode sorting, you need/can have a lot ... -
- Nobuyoshi Nakada wrote:
> `String#casecmp` also should have those options, I think.
I agree, and feel strongly that the logic to support those options should live there. Someone smarter than I should probably make the changes Martin... -
- `String#casecmp` also should have those options, I think.
-
- Ryan Ringler wrote:
> The attached patch implements `#casecmp?` by calling `#casecmp` and checking whether the return value is zero. I think this provides a clean separation of concerns (keeping all the comparison logic in `#casecmp`... -
04:12 AM Feature #10912: Add method(s) to IPAddr for determining whether an address is link local
- I also want a method to determine whether an address is link-local.
There is a proposal for a fix on https://github.com/ruby/ruby/pull/1494 .
11/26/2016
-
- Martin Dürst wrote:
> I gave an implementation in Ruby, but can somebody provide an equivalent patch in C?
The attached patch implements `#casecmp?` by calling `#casecmp` and checking whether the return value is zero. I think this p... -
- Yukihiro Matsumoto wrote:
> `casecmp?` accepted.
> ...
Thank you! -
08:45 PM Feature #12745: String#(g)sub(!) should pass a MatchData to the block, not a String
- What about `sub_md` as a name?
-
02:54 PM Feature #12982 (Feedback): ruby 2.3.1 got crash on macos
- ```
VM Regions Near 0x7fff5920fc68:
MALLOC_TINY 00007fd315800000-00007fd315a00000 [ 2048K] rw-/rwx SM=PRV
--> STACK GUARD 00007fff55a10000-00007fff59210000 [ 56.0M] ---/rwx SM=NUL stack guard for thread 0
... -
- ```ruby
if n.bittest?(0b10100000)
```
If I encountered that code without having the context of this case, I wouldn't know what what the equivalent behaviour would be:
```ruby
if n & 0b10100000 != 0 #=> Is at least one ... -
01:33 PM Bug #12981 (Closed): Date.parse raises an Argument error under a specific condition
- Date.parse cannot accept incomplete strings whose delimiter is a hyphen like: "2016-12", although it can accept if a delimiter is a slash like: "2016/12"
I'm not really sure if this is a bug or an expected behavior, but a maintainer o... -
- _ wanabe wrote:
> after r42402, Readline.input in IRB::ReadlineInputMethod#gets causes "Too many open files - dup (Errno::EMFILE)".
Sorry, it is not reproduced on real console. I have confirmed it on PTY.spawn only.
Ctrl-c doesn't ... -
- after r42402, Readline.input in IRB::ReadlineInputMethod#gets causes "Too many open files - dup (Errno::EMFILE)".
after r43439, the method causes SEGV.
Everything goes well without `set enable-keypad on` with any commits.
-
- Applied in changeset r56902.
----------
file.c: home directory from system
* file.c (rb_default_home_dir): resolve home directory from the
system database when HOME is not set. [Feature #12695] -
- * file.c (rb_default_home_dir): resolve home directory from the
system database when HOME is not set. [Feature #12695]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56902 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- * win32/win32.c (rb_w32_home_dir): move from win32/file.c to try
special folders.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56901 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56900 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 09:29 AM Revision 2e2fa3dd (git): * 2016-11-26
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56899 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- Net::FTP inherits ssl_socket_connect from Net::Protocol to implement timeout.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56898 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
07:47 AM Feature #12944: Change Kernel#warn to call Warning.warn
- Here's an updated patch that uses str_end_with_asciichar instead of just manually checking the last byte for \n.
-
- You can't compare and append the last byte simply in wide character encodings.
E.g., `str_end_with_asciichar` in io.c. -
- Here's an updated patch that uses rb_str_dup before calling rb_str_cat, so it doesn't modify a passed in string, and works with frozen strings.
-
06:16 AM Bug #12938: forwardable.rb is no longer compatible with any alternative implementation
- This is great, thanks nobu!
-
- matz, did you want to modify the Kernel#warn API? Currently, I don't think modifying the Kernel#warn API can be done in a backwards compatible way, as Kernel#warn accepts an arbitrary number of arguments and converts them all to strings...
11/25/2016
-
10:05 PM Bug #12970: == Equality of recursive sets fails
- Took me a while to understand what was going on.
The issue is that the Set would need to be rehashed. Indeed, when you add it to itself, the element you are adding is modified too.
This is, in a way, similar to:
```
a = []
s =... -
- It would help if you explained what is meant by "safe".
There is currently no valid usage of `chunk_while`, `slice_after`, `slice_before` or `slice_when` without a block and an argument, as they all raise an exception. So "extending" ... -
- We briefly looked at this issue in todays developer meeting but had no time to fully agree that the proposed extension is safe. We just did not reach a conclusion.
-
-
- We looked at this issue at todays developer meeting and no one (including matz who was there) was against it. I believe this is accepted.
-
- Herwin Quarantainenet wrote:
> I can't say the usage of `bittest?` is directly clear to me. Does it test if resulting integer is not equal to `0`? And would we have to use it this way?
> ...
Like
```ruby
if n.bittest?(0b10100000)
`... -
07:59 AM Feature #12753: Useful operator to check bit-flag is true or false
- I can't say the usage of `bittest?` is directly clear to me. Does it test if resulting integer is not equal to `0`? And would we have to use it this way?
```ruby
if (n & 0b10100000).bittest?
```
I think a name like `Integer#binar... -
- `bittest?` sounds reasonable. Accepted.
Matz.
-
- What about `bittest?`
-
- I understand the demand. But `and?` is an unacceptable name.
Any idea?
Matz.
-
- We looked at this issue today. It turned out that ruby do not fully implement what was needed originally now. Reopened the ticket to assign ko1.
-
-
- We (mainly mrkn and matz) discussed why Fixnum#round returns Integer today. The answer is because (unlike other languages, namely C's round(3)) in ruby we can safely store all possible return value that round can theoretically generate,...
-
- We looked at this issue at todays developer meeting. We referred issue #6539 and now we remember that each method (not the "Module meta programming" at once) should have separate considerations.
While Matz do not like `String.include... -
- We looked at this issue in todays developer meeting and had 2 feelings in common.
1. The autoload should not render NameError. It definitely is a bug that must be fixed. Ko1 is assigned.
2. Besides, we want to discourage people fr... -
-
01:31 PM Bug #12980: Time - Time to return a Rational
- I think of one philosophical reason and one practical reason.
A philosophical reason: many Time objects are inherently inexact. Since a Rational is considered exact, returning a Rational might convey the wrong message that the Time o... -
09:25 AM Bug #12980 (Feedback): Time - Time to return a Rational
-
- It is because ruby-dev:38446 by mame-san.
http://blade.nagaokaut.ac.jp/cgi-bin/scat.rb/ruby/ruby-dev/38446 -
08:26 AM Bug #12980: Time - Time to return a Rational
- Well, my assumption seems wrong.
0.5r.round also becomes 0, so this isn't really related to #12952
Still I'm wondering why Time - Time isn't Rational though. -
- Currently subtracting a Time from a Time returns a Float, which I guess causes #12952
Is there any reason this doesn't return a Rational? -
- `rb_str_cat` modifies the argument, it is unexpected to `Kernel#warn`, I think.
Also it doesn't work on a frozen string. -
- Accepted.
Matz.
-
- ping nobu.
-
- We looked at this issue at todays developer meeting but a "random subset with random size" still does not sound familiar to the attendees.
Can you show us in code how do you use such array? -
- We looked at this issue in todays developer meeting and assigned it to Yui. But no one there had implementation of this. It might need some time. Stay tuned!
-
-
- Of course it changes behaviour. A change always is. The problem is how vital this (rounding an inexact number to a yet more inexact number) is.
As far as I can see the Rails breakage only happens in views, where 30secs is shown "les... -
09:42 AM Bug #12958: Breaking change in how `#round` works
- Tests are not the only problem. I believe this change is dangerous because this code may change behavior in production without being easily noticed.
-
-
- LGTM, as long as we add a functionality to `warn` to include those info.
Matz.
-
- LGTM.
Matz.
-
- Below is PoC; it may have a path which raises an exception.
```diff
diff --git a/object.c b/object.c
index 05bef4d..5d63803 100644
--- a/object.c
+++ b/object.c
@@ -2750,17 +2750,60 @@ static VALUE
rb_f_integer(int argc, VALUE ... -
- Is there any problem with the following code?
```
Integer(str) rescue default_value
```
Matz. -
- shyouhei@ruby-lang.org wrote:
> P.S. we have /#{@mutex=Mutex.new}/o, so we already have such thing, to some extent at least.
Awesome, ROFL; you almost made me choke on my food!
Yes, //o is awesome, but this really takes the cake.... -
- I feel that Thread.exclusive is too big a primitive to merely initialize a mutex at a process startup. We could perhaps introduce pthread_once_t analogous more fine-grained light-weight control structure that does the job.
P.S. we ha... -
09:11 AM Feature #12612: Switch Range#=== to use cover? instead of include?
- I encountered the problem in production code, it was not strings, but specialized IPAddress class, to filter "if IP is in range" -- it was SUDDENLY 0.4 sec on each request to server spent just to `(IPAddress.new(from)..IPAddress.new(to))...
-
- I see no real-world use-case for `Range#===` with strings.
Besides that, using `cover?` behavior for string ranges would introduce incompatibility.
Matz. -
- I don't think it's worth adding which is easily done by `sub/gsub`.
Matz.
-
-
- I want to see a real-world use-case for the feature.
In addition, I don't think the name `merge` is the best for the functionality.
Matz.
-
- I'd like to keep compatibility rather than reducing confusion slightly.
In any way, modifying $LOAD_PATH on the fly is not encouraged in modern Ruby.
Matz.
-
- I gave an implementation in Ruby, but can somebody provide an equivalent patch in C?
-
- I gave an implementation in Ruby, but can somebody provide an equivalent patch in C?
-
- `casecmp?` accepted.
Matz.
-
- Well, I should say `yield_self` is not very readable in context (it raises question which is "self" in that context?):
~~~ruby
construct_url(*params)
.yield_self(&method(:open))
.read
.yield_self(&JSON.method(:parse))
.to... -
- `yield_self` is OK, but I don't think we are going to add `object.{|x| ... }`.
Matz.
-
- Do you want to treat nils specially? If so, the name `left_merge` does not indicate the intention.
Besides that, you don't explain why you need `a.left_merge(b)` when we can `b.merge(a)`.
If you have additional opinions, ideas or whate... -
- Reject because it is too similar to string literals.
Matz.
-
- As the developers' meeting today, proposed syntax extension `key::"value"`.
https://github.com/ruby/ruby/compare/trunk...nobu:feature/symbol-after-keyword -
- FIY both append and prepend are in ActiveSupport: https://github.com/rails/rails/blob/master/activesupport/lib/active_support/core_ext/array/prepend_and_append.rb
-
- Accepted.
Matz.
-
- I found that a rubygems test ,`TestGemSpecification#test_initialize_copy_broken` in test/rubygems/test_gem_specification.rb, depends on this exception.
And rubyspec fails as usual.
```diff
diff --git i/object.c w/object.c
index 05b... -
- If we don't care about dup-ability much, the proposal sounds reasonable.
And I agree this dup-ability matters less than confusion caused by exceptions.
Accepted.
Matz. -
- This is a proposal resulting from a discussion in Bug #11929. Because this is proposing a different solution from #11929, it has a new number.
#11929 shows that people are confused that e.g. 3.dup throws an exception (but Integer#dup ... -
- `unpack1` looks so ad-hoc but surely is attractive.
Matz.
-
- How about `Array#unpack1` ?
-
- Accepted.
Matz.
-
07:26 AM Feature #12871 (Assigned): Using the algorithm like math.fsum of Python for Array#sum
-
- The issue closed by #12979
Matz.
-
- 1. ability to be duped cannot be checked by `respond_to?`. It's simply wrong.
2. the "fail silently" proposal is attractive, but it should be a separate proposal (for the sake of bookkeeping).
Matz.
-
- Out of Shyouhei's 4 options,
1. not acceptable for compatibility's sake
2. not excited, may cause compatibility problem. besides that, `String#scan` cannot be enhanced by this option
3. looks nice but I have performance concern
4.... -
- I understand your proposal, but I don't see what it is for.
Do you have any no-artificial use-case for this feature?
If you come up with real-world use-case, please re-open the issue.
Matz.
-
- For the time being, use `math-tau` gem.
I will add it after (and only after) it became time-proven major, probably due to Python's tau.
Matz.
-
- Accepted.
Matz.
-
- * numeric.c (round_half_down, int_round_half_down): support
round-down mode.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56897 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
- Agreed.
Matz.
-
- Since ?up gives a string "up", it cannot be a replacement for :up symbol.
If someone come up with an use-case for this multi-character ? string literal, I will reinvestigate.
Matz.
11/24/2016
- 08:51 PM Revision b6d10b6c (git): * 2016-11-25
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56896 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
- 08:51 PM Revision f6f158c6 (git): use `rb_struct_ptr` for `RSTRUCT_PTR` macro
- * include/ruby/ruby.h (RSTRUCT_PTR): at r55788, `rb_struct_const_ptr` had been hidden and `rb_struct_ptr` had been implemented for backward compatiblity. but the definition of `RSTRUCT_PTR` was not modified to use it, probably by mistak...
-
- https://github.com/ruby/ruby/pull/1492
Alway add a space between a comma and the next element. These spaces were there sometimes, but not always. This keeps to code consistent.
-
- https://github.com/ruby/ruby/pull/1491
The same code is used for has_value? and value?, but according to http://blade.nagaokaut.ac.jp/cgi-bin/scat.rb/ruby/ruby-core/43765 has_value is deprecated. Use the non-deprecated syntax in the... -
02:09 PM Feature #12968: Allow default value via block for Integer(), Float() and Rational()
- Nobuyoshi Nakada wrote:
> Only "invalid value for Integer()" from a string?
I don't have a strong opinion on this. -
11:44 AM Feature #12921: Retrieve user and password for proxy from env
- > Why don't you use open-uri?
Because open-uri can't treat POST, HEAD, etc.
I think it is natural that net/http gets the same parameters from env for default unspecified parameters as open-uri does.
-
- Applied in changeset r56894.
----------
marshal.c: fix infinite recursion
* marshal.c (check_userdump_arg): marshal_dump should not return
an instance of the same class, otherwise it causes infinite
recursion. [ruby-core:78289] [B... -
- ```
zsh % lldb -- ./miniruby -v tmp.rb
(lldb) target create "./miniruby"
Current executable set to './miniruby' (x86_64).
(lldb) settings set -- target.run-args "-v" "tmp.rb"
(lldb) run
Process 91881 launched: './miniruby' (x86_64... -
- ```ruby
Hash.include Module.new {
def marshal_dump
dup
end
def marshal_load this
initialize_copy this
end
}
p Marshal.load(Marshal.dump(foo:"bar"))
```
This script dumps core. -
- * marshal.c (check_userdump_arg): marshal_dump should not return
an instance of the same class, otherwise it causes infinite
recursion. [ruby-core:78289] [Bug #12974]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56894 b2dd03c8... -
- * marshal.c (w_object): should not decrement negative limit which
means unlimited.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56893 b2dd03c8-39d4-4d8f-98ff-823fe69b080e -
10:19 AM Bug #12975 (Closed): Edit documentation for IO#print function
- Important changes made:
* Delete a duplicate sentence relating to a description of global variable $_
* Split the whole paragraph into three semantically related smaller ones.
To make it easier to digest the content.
* Emphasize name... -
- * test/fileutils/test_fileutils.rb (assert_output_lines): ensure
that write pipe is closed, not the reader thread to deadlock
when an exception raised in the given block.
[ruby-core:78053] [Bug #12910]
git-svn-id: svn+ssh://ci.rub... -
- git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56891 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
-
- Thx for fixing this. But is there chance to improve my first point? I.e. why the test suite gets stuck? It should just fail. The assert apparently failed, so the test case should fail ....
-
- Shugo Maeda wrote:
> ```ruby
> ...
It might be better to swap `dst_*` and `src_*`.