path: root/ChangeLog

Tue Aug  2 21:42:40 2016  Chia-sheng Chen  <qitar888@gmail.com>

	* math.c (tanh): make faster by the extract form if three
	  hyperbolic functions are unavailable.  [Feature #12647]

Tue Aug  2 12:37:00 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ext/socket/option.c, ext/socket/rubysocket.h (inet_ntop): share
	  the fallback definition.  [ruby-core:76646] [Bug #12645]

Tue Aug  2 04:07:29 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* win32/win32.c (set_pioinfo_extra): use more reliable way to search
	  the position of pioinfo of VC14, and also support debug library of it.
	  patched by davispuh AT gmail.com
	  [ruby-core:76644] [Bug #12644]
	  this fixes also [Bug #12631]

Mon Aug  1 21:39:52 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ext/extmk.rb: [EXPERIMENTAL] build extension libraries in
	  extracted gems.

Mon Aug  1 16:07:18 2016  URABE Shyouhei  <shyouhei@ruby-lang.org>

	* include/ruby/ruby.h (struct RStruct): no longer.

	* internal.h (struct RStruct): moved here.

	* struct.c (rb_struct_ptr): a compensation function for the lack
	  of RSTRUCT_PTR.  But now that we have RSTRUCT_GET/SET, that must
	  not be used anyway.  I mark this deprecated.  Dont use it.

Mon Aug  1 14:50:06 2016  Jeremy Evans <code@jeremyevans.net>

	* object.c (rb_obj_clone2): Allow Object#clone to take freeze:
	  false keyword argument to not freeze the clone.
	  [ruby-core:75017][Feature #12300]

	* test/ruby/test_object.rb (TestObject): test for it.

Mon Aug  1 12:16:19 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* ext/json/*, test/json/json_parser_test.rb: Update json-2.0.2.

Sun Jul 31 16:17:23 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ext/win32/resolv/resolv.c (get_dns_server_list): [Win32] get DNS
	  servers only for connected network devices by GetNetworkParams
	  API.  [Bug #12604]

Sat Jul 30 12:13:26 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* string.c (String#downcase), NEWS: Mentioned that case mapping for all
	  of ISO-8859-1~16 is now supported. [ci skip]

Sat Jul 30 12:00:01 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/iso_8859_2.c, test/ruby/enc/test_case_comprehensive.rb:
	  Implement non-ASCII case conversion for ISO-8859-2, by Yushiro Ishii.

Fri Jul 29 20:57:12 2016  chuanshuo  <lilijreey@126.com>

	* *.c: rename rb_funcall2 to rb_funcallv, except for extensions
	  which are/will be/may be gems.  [Fix GH-1406]

Fri Jul 29 10:51:34 2016  Koichi Sasada  <ko1@atdot.net>

	* proc.c (env_write): remove unused function.

Fri Jul 29 10:49:52 2016  Koichi Sasada  <ko1@atdot.net>

	* vm_core.h (VM_LOCAL_P): should return an integer value.
	  reported at
	  http://d.hatena.ne.jp/nagachika/20160728/ruby_trunk_changes_55764_55770

Fri Jul 29 04:23:08 2016  Koichi Sasada  <ko1@atdot.net>

	* vm_core.h (VM_ENV_LOCAL_P): return truthy (0 or not) value.

	* vm.c (rb_vm_make_proc_lambda): use VM_ENV_ESCAPED_P() macro.

Fri Jul 29 03:49:04 2016  Koichi Sasada  <ko1@atdot.net>

	* vm.c, internal.h: remove RubyVM::Env class and all of env objects
	  are imemo objects (imemo_env).

	* NEWS: describe this change. I believe nobody touch these objects
	  because there are no method defined.

	* vm_core.h: remove the following definitions.
	  * rb_cEnv decl.
	  * GetEnvPtr() because Env is no longer T_DATA object.

	* vm_core.h (rb_env_t): fix layout for imemo values.

	* vm_core.h (vm_assert_env): added.

	* vm_core.h (vm_env_new): added.

Thu Jul 28 19:53:21 2016  Koichi Sasada  <ko1@atdot.net>

	* vm_core.h: revisit the structure of frame, block and env.
	  [Bug #12628]

	  This patch introduce many changes.

	  * Introduce concept of "Block Handler (BH)" to represent
	    passed blocks.

	  * move rb_control_frame_t::flag to ep[0] (as a special local
	    variable). This flags represents not only frame type, but also
	    env flags such as escaped.

	  * rename `rb_block_t` to `struct rb_block`.

	  * Make Proc, Binding and RubyVM::Env objects wb-protected.

	  Check [Bug #12628] for more details.

Thu Jul 28 15:05:12 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* include/ruby/ruby.h (ruby_fl_type): use __extension__ to get rid
	  of pedantic warning against RUBY_FL_USER19.
	  https://github.com/skylightio/skylight-ruby/issues/64

	* include/ruby/ruby.h (rb_mul_size_overflow): ditto for use of
	  int128.

Wed Jul 27 10:32:59 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/windows_1253.c: Remove dead code found by Coverity Scan.

Tue Jul 26 22:43:36 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* gc.c (run_finalizer): make saved running finalizer state
	  volatile to ensure not to be clobbered by longjmp.

Tue Jul 26 19:26:00 2016  Koichi Sasada  <ko1@atdot.net>

	* vm_insnhelper.c: introduce rb_vm_pop_frame() and use it
	  instead of setting rb_thread_t::cfp directly.

	* vm_insnhelper.c (vm_pop_frame): return the result of
	  finish frame or not.

Tue Jul 26 19:06:39 2016  Koichi Sasada  <ko1@atdot.net>

	* gc.c (rb_raw_obj_info): support to show Proc obj.

Tue Jul 26 18:55:55 2016  Koichi Sasada  <ko1@atdot.net>

	* gc.c (gc_mark): add `inline' explicitly.
	  I expected to inline this function implicitly at the loop
	  (ex: marking T_ARRAY objects) but sometimes it remains as
	  normal call.

Tue Jul 26 16:33:16 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/windows_1257.c, test/ruby/enc/test_case_comprehensive.rb:
	  Implement non-ASCII case conversion for Windows-1257, by Sho Koike.

Tue Jul 26 16:19:41 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/windows_1250.c, test/ruby/enc/test_case_comprehensive.rb:
	  Implement non-ASCII case conversion for Windows-1250, by Sho Koike.

	* ChangeLog: Fixed order of previous two entries.

Tue Jul 26 15:54:17 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/windows_1253.c, test/ruby/enc/test_case_comprehensive.rb:
	  Implement non-ASCII case conversion for Windows-1253, by Takumi Koyama.

Tue Jul 26 15:30:37 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/windows_1251.c, test/ruby/enc/test_case_comprehensive.rb:
	  Implement non-ASCII case conversion for Windows-1251, by Shunsuke Sato.

Tue Jul 26 13:04:59 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* test/ruby/enc/test_case_comprehensive.rb: Add explicit skip test for
	  availability of Unicode data files.

Mon Jul 25 21:33:13 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* range.c (check_step_domain): check step argument domain by <=>
	  method, instead of < and >.

Mon Jul 25 21:11:32 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>

	* doc/maintainers.rdoc: fix filenames.

Mon Jul 25 16:59:00 2016  Koichi Sasada  <ko1@atdot.net>

	* debug.c (ruby_debug_printf): use rb_raw_obj_info()
	  instead of rb_inspect() because it is more robust way
	  to see object internal.

Sun Jul 24 16:33:13 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* regenc.h/c, include/ruby/oniguruma.h, enc/ascii.c, big5.c, cp949.c,
	  emacs_mule.c, euc_jp.c, euc_kr.c, euc_tw.c, gb18030.c, gbk.c,
	  iso_8859_1|2|3|4|5|6|7|8|9|10|11|13|14|15|16.c, koi8_r.c, koi8_u.c,
	  shift_jis.c, unicode.c, us_ascii.c, utf_16|32be|le.c, utf_8.c,
	  windows_1250|51|52|53|54|57.c, windows_31j.c, unicode.c:
	  Remove conditional compilation macro ONIG_CASE_MAPPING. [Feature #12386].

Sun Jul 24 12:53:42 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* doc/maintainers.rdoc: xmlrpc is bundled gem from Ruby 2.4.

Sun Jul 24 12:07:39 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* doc/maintainers.rdoc: Update OpenSSL maintainer.

Sat Jul 23 22:43:41 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* internal.h (Check_Type): inline check for the object type.

Sat Jul 23 04:06:04 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* include/ruby/ruby.h (RTEST, NIL_P): use RUBY prefixed name in
	  macros.

Sat Jul 23 01:41:29 2016  Eric Wong  <e@80x24.org>

	* lib/webrick/httpservlet/cgihandler.rb (do_GET): delete HTTP_PROXY
	* test/webrick/test_cgi.rb (test_cgi_env): new test
	* test/webrick/webrick.cgi (do_GET): new endpoint to dump env
	  [ruby-core:76511] [Bug #12610]

Fri Jul 22 19:55:20 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* vm.c (vm_set_main_stack): remove unnecessary check.  toplevel
	  binding must be initialized.  [Bug #12611] (N1)

	* win32/win32.c (w32_symlink): fix return type.  [Bug #12611] (N3)

	* string.c (rb_str_split_m): simplify the condition.
	  [Bug #12611](N4)

Fri Jul 22 17:13:37 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* string.c (String#dump): Change escaping of non-ASCII characters in
	  UTF-8 to use upper-case four-digit hexadecimal escapes without braces
	  where possible [Feature #12419].

	* test/ruby/test_string.rb (test_dump): Add tests for above.

Fri Jul 22 10:35:35 2016  Kouhei Sutou  <kou@cozmixng.org>

	* lib/rexml/attribute.rb (REXML::Attribute#to_string): Fix wrong
	  entry reference name of double quote.
	  [Bug #12609][ruby-core:76509]
	  Patch by Joseph Marrero. Thanks!!!

Fri Jul 22 10:32:13 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* template/unicode_norm_gen.tmpl: Remove
	  UnicodeNormalize::UNICODE_VERSION at origin [Feature #12546].

Fri Jul 22 09:23:51 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* LEGAL: Added entries for files under the USD license.
	  [Bug #12598][ruby-core:76428][ci skip]

Fri Jul 22 09:19:57 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* LEGAL: Added entry for `lib/rdoc/generator/template/darkfish/css/fonts.css`
	  [Misc #12550][ruby-core:76255][ci skip]

Fri Jul 22 06:28:32 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* gc.c (run_finalizer): push and exec tag just once, instead of
	  protecting for each finalizer.

	* gc.c (gc_start_internal, rb_gc_start): set finalizing flag
	  whenever calling deferred finalizers not to recurse.

Thu Jul 21 22:26:40 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* missing/strl{cat,cpy}.c: Update latest upstream files.
	  [Misc #12205][ruby-core:74487]
	* LEGAL: Update license for missing/strl{cat,cpy}.c.

Thu Jul 21 21:53:30 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* LEGAL: added file list with Public domain license.
	  [ruby-core:76254][Bug #12549]

Wed Jul 20 17:44:07 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* enumerator.c (lazy_uniq): new method Enumerator::Lazy#uniq.
	  [Feature #11090]

	* enum.c (enum_uniq): new method Enumerable#uniq.
	  [Feature #11090]

Wed Jul 20 17:35:23 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* hash.c (rb_hash_add_new_element): add new element or do nothing
	  if it is contained already.

	* array.c (ary_add_hash, ary_add_hash_by): use
	  rb_hash_add_new_element.

Tue Jul 19 18:21:17 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* lib/unicode_normalize/tables.rb: Remove
	  UnicodeNormalize::UNICODE_VERSION (#12546).

Tue Jul 19 15:38:59 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* variable.c (rb_local_constants_i): exclude private constants
	  when excluding inherited constants too.  [Bug #12345]

Sun Jul 17 23:42:00 2016  Kenta Murata  <mrkn@mrkn.jp>

	* numeric.c (num_finite_p, num_infinite_p): Add Numeric#finite? and
	  Numeric#infinite?  [Feature #12039] [ruby-core:73618]

	* complex.c (rb_complex_finite_p): Add Complex#finite?

	* complex.c (rb_complex_infinite_p): Add Complex#infinite?

	* test/ruby/test_bignum.rb: Add test for Integer#finite? and
	  Integer#infinite?

	* test/ruby/test_fixnum.rb: ditto.

	* test/ruby/test_rational.rb: Add test for Rational#finite? and
	  Rational#infinite?

	* test/ruby/test_complex.rb: Add test for Complex#finite? and
	  Complex#infinite?

Sun Jul 17 20:59:24 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* common.mk, enc/depend (casefold.h, name2ctype.h): move to
	  unicode data directory per version.

Sat Jul 16 06:26:00 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* common.mk, enc/Makefile.in: moved timestamp files for
	  directories under the specific directory, to get rid of match
	  with files under the source directory.

Fri Jul 15 22:05:13 2016  Naohisa Goto  <ngotogenome@gmail.com>

	* string.c (str_buf_cat): Fix potential integer overflow of capa.
	  In addition, termlen is used instead of +1.

Fri Jul 15 21:30:38 2016  Naohisa Goto  <ngotogenome@gmail.com>

	* string.c (str_buf_cat): Fix capa size for embed string.
	  Fix bug in r55547. [Bug #12536]

Fri Jul 15 18:13:15 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* gems/bundled_gems: update latest gems.

Fri Jul 15 17:08:57 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* util.c (ruby_strtod): do not underflow only by preceeding zeros,
	  which may be canceled out by the exponent.
	  http://twitter.com/kazuho/status/753829998767714305

Fri Jul 15 09:53:48 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* enc/unicode/case-folding.rb, tool/enc-unicode.rb: check if
	  Unicode versions are consistent with each other.

Fri Jul 15 08:25:15 2016  Jeremy Evans <code@jeremyevans.net>

	* string.c (STR_BUF_MIN_SIZE): reduce from 128 to 127
	  [ruby-core:76371] [Feature #12025]
	* string.c (rb_str_buf_new): adjust for above reduction

Thu Jul 14 17:26:00 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* Makefile.in (enc/unicode/name2ctype.h): remove stale recipe,
	  which did not support Unicode age properties.

	* common.mk (enc/unicode/name2ctype.h): update by --header option
	  of tool/enc-unicode.rb.  enc/unicode/name2ctype.kwd file has not
	  been used.

	* common.mk (enc/unicode/name2ctype.kwd): rule to create from
	  Unicode data files, used only when the target does not exist.

Thu Jul 14 13:10:54 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ext/json/lib/json/ext: remove stale directory.  bundled
	  extension libraries are placed under the directory for each
	  architectures, but not mixed with plain text script libraries.

Thu Jul 14 12:48:47 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* ext/json/**/*.rb: merge original files from upstream repository.
	  It only fixes styles of frozen string literal.

Wed Jul 13 22:23:03 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* test/json/json_common_interface_test.rb: use assert_raise instead of
	  assert_raises.

Wed Jul 13 22:14:23 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* test/lib/test/unit.rb: added test files with `_test` suffix for json
	  upstream.
	* test/json: merge original test files from json upstream.

Wed Jul 13 18:09:42 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/iso_8859_9.c, test/ruby/enc/test_case_comprehensive.rb:
	  Implement non-ASCII case conversion for ISO-8859-9, by Kazuki Iijima.

	* enc/iso_8859_9.c: Exclude dotless i/I with dot from case-insensitive
	  matching because they are not a case pair.

	* test/ruby/enc/test_iso_8859.rb: Make test coverage for ISO-8859-9
	  a bit more complete.

Wed Jul 13 17:21:24 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/windows_1252.c, test/ruby/enc/test_case_comprehensive.rb:
	  Implement non-ASCII case conversion for Windows-1252, by Serina Tai.

	* test/ruby/enc/test_case_comprehensive.rb: Fix order of encodings.

Wed Jul 13 16:19:14 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/iso_8859_7.c, test/ruby/enc/test_case_comprehensive.rb:
	  Implement non-ASCII case conversion for ISO-8859-7, by Kosuke Kurihara.

	* test/ruby/enc/test_case_comprehensive.rb: Fix order of encodings.

Wed Jul 13 16:08:08 2016  Koichi Sasada  <ko1@atdot.net>

	* gc.c (gc_mark_roots): should mark the VM object itself to mark
	  singleton class of the VM object.
	  Before this patch, we only set mark bit for the VM object and
	  invoke mark function separately.
	  [Bug #12583]

	* test/ruby/test_gc.rb: add a test.

Wed Jul 13 15:59:59 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* math.c (_USE_MATH_DEFINES): it must be set before including internal.h
	  because internal.h includes ruby.h, ruby.h includes win32.h, and
	  win32.h includes system's math.h.
	  this change is to get rid of a compiler warning (redefinition of
	  a macro) introduced at r55641.

Wed Jul 13 15:19:03 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/iso_8859_1.c, enc/iso_8859_4.c: Avoid setting modification flag if
	  there is no modification.

Wed Jul 13 14:40:04 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/iso_8859_5.c, test/ruby/enc/test_case_comprehensive.rb:
	  Implement non-ASCII case conversion for ISO-8859-5, by Masaru Onodera.

	* test/ruby/enc/test_case_comprehensive.rb: Fix order of encodings.

Wed Jul 13 14:28:33 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* math.c (rb_math_sqrt): r55646 must use f_signbit.

Wed Jul 13 14:22:50 2016  Koichi Sasada  <ko1@atdot.net>

	* iseq.c (Init_ISeq): undef ISeq.translate and ISeq.load_iseq
	  to prevent calling super classes' methods.

	  Without this patch, you can write workaround like:

	    class << RubyVM::InstructionSequence
	      def translate; end
	      undef translate
	    end

	* test/ruby/test_iseq.rb: add a test.

Wed Jul 13 14:16:03 2016  Koichi Sasada  <ko1@atdot.net>

	* vm_method.c (method_entry_get_without_cache): check
	  undefined method even if ruby_running is FALSE.

	  We haven't call "undef"ed methods before ruby_running.
	  So that this issue does not make troubles.

Wed Jul 13 14:15:22 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/windows_1254.c: Adjust variable/macro names.

Wed Jul 13 13:19:12 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/iso_8859_9.c, enc/windows_1254.c: Split Windows-1254 from
	  ISO-8859-9 to be able to implement different case conversions.

Wed Jul 13 13:08:30 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/iso_8859_7.c, enc/windows_1253.c: Split Windows-1253 from
	  ISO-8859-7 to be able to implement different case conversions.

Wed Jul 13 10:50:12 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/iso_8859_13.c, test/ruby/enc/test_case_comprehensive.rb:
	  Implement non-ASCII case conversion for ISO-8859-13, by Kanon Shindo.

Wed Jul 13 10:31:39 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/iso_8859_13.c, enc/windows_1257.c: Split Windows-1257 from
	  ISO-8859-13 to be able to implement different case conversions.

Wed Jul 13 09:02:30 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/iso_8859_3.c, test/ruby/enc/test_case_comprehensive.rb:
	  Implement non-ASCII case conversion for ISO-8859-3, by Takuya Miyamoto.

	* test/ruby/enc/test_case_comprehensive.rb: Extend special treatment
	  for Turkic.

	* enc/iso_8859_3.c: Exclude dotless i/I with dot from case-insensitive
	  matching because they are not a case pair.

Wed Jul 13 08:40:21 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* test/ruby/enc/test_iso_8859.rb: Exclude dotless i/I with dot from
	  case-insensitive matching because they are not a case pair.

Tue Jul 12 23:13:43 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* math.c (rb_math_sqrt): [EXPERIMENTAL] move Complex sqrt support
	  from mathn.rb.

Tue Jul 12 01:25:09 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* configure.in, lib/mkmf.rb, win32/Makefile.sub (CSRCFLAG): make the
	  compiler option replacable in Makefile.

	* win32/Makefile.sub (OUTFLAG, COUTFLAG): ditto.

	* win32/Makeile.sub, win32/setup.mak (CC): should not append `-nologo`
	  option forcely.

Mon Jul 11 18:05:40 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/iso_8859_1.c: Moved test for lowercase characters without
	  uppercase equivalent.

Mon Jul 11 17:49:25 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/iso_8859_4.c, enc/iso_8859_10.c, enc/iso_8859_14.c,
	  enc/iso_8859_15.c, enc/iso_8859_16.c: Replace case-by-case code with
	  lookup in ENC_ISO_8859_xx_TO_LOWER_CASE table.

Mon Jul 11 16:00:56 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ext/stringio/stringio.c (strio_each, strio_readlines): convert
	  arguments just once before reading, instead of conversions for
	  each lines, as r55603.

Sun Jul 10 19:53:41 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/iso_8859_10.c, test/ruby/enc/test_case_comprehensive.rb:
	  Implement non-ASCII case conversion for ISO-8859-10, by Toya Hosokawa.

Sun Jul 10 19:33:47 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* test/ruby/enc/test_case_comprehensive.rb: Changed testing logic to
	  catch unintended modifications of characters that do not have a case
	  equivalent in the respective encoding.
	* enc/iso_8859_1.c, enc/iso_8859_15.c: Fixed unintended modifications of
	  micro sign and y with diaeresis.

Sun Jul 10 17:05:36 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/iso_8859_4.c, test/ruby/enc/test_case_comprehensive.rb:
	  Implement non-ASCII case conversion for ISO-8859-4, by Kotaro Yoshida.

Sun Jul 10 16:17:47 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* test/ruby/enc/test_case_comprehensive.rb: Fixed a comment
	  (message belongs to last commit). [ci skip]

Sun Jul 10 14:27:25 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* util.c (ruby_dtoa): [EXPERIMENTAL] adjust the case that the
	  Float value is close to the exact but unrepresentable middle
	  value of two values in the given precision, as r55604.

Sun Jul 10 08:57:20 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* thread.c: Fixed implicit conversion error with Apple clang-800.0.31

Sat Jul  9 12:43:09 2016  Shugo Maeda  <shugo@ruby-lang.org>

	* lib/getoptlong.rb: use false instead of FALSE.

Fri Jul  8 21:49:28 2016  Naohisa Goto  <ngotogenome@gmail.com>

	* thread.c (rb_wait_for_single_fd): Clean up fds.revents every time
	  before calling ppoll(2). [Bug #12575] [ruby-dev:49725]

Fri Jul  8 14:16:48 2016  Shugo Maeda  <shugo@ruby-lang.org>

	* vm_args.c (vm_caller_setup_arg_block): call rb_sym_to_proc()
	  directly to reduce method dispatch overhead.

Fri Jul  8 08:43:31 2016  Shugo Maeda  <shugo@ruby-lang.org>

	* io.c (rb_io_s_read): add description of pipes to the documentation
	  of IO.read.

Fri Jul  8 03:54:22 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* vm_args.c (setup_parameters_complex): don't raise ArgumentError
	  if an array is given for instance_exec with optional argument.
	  [ruby-core:76300] [Bug #12568]
	  https://github.com/rails/rails/pull/25699

Fri Jul  8 00:47:36 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* vm_eval.c (yield_under): change prototype to get argc/argv.

	* vm_eval.c (specific_eval): change for above.

	* vm_eval.c (rb_obj_instance_exec): avoid object allocation.

	* vm_eval.c (rb_mod_module_exec): ditto.

Thu Jul  7 20:08:37 2016  Shugo Maeda  <shugo@ruby-lang.org>

	* vm_args.c (vm_caller_setup_arg_block): disable symbol block
	  argument optimization when tail call optimization is enabled,
	  in order to avoid SEGV.  [ruby-core:76288] [Bug #12565]

Thu Jul  7 16:37:53 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* numeric.c (flo_round): [EXPERIMENTAL] adjust the case that the
	  receiver is close to the exact but unrepresentable middle value
	  of two values in the given precision.
	  http://d.hatena.ne.jp/hnw/20160702

Thu Jul  7 16:31:07 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* io.c (rb_io_s_foreach, rb_io_s_readlines): convert arguments
	  just once before reading, instead of conversions for each lines.

Wed Jul  6 19:54:17 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/iso_8859_14.c, test/ruby/enc/test_case_comprehensive.rb:
	  Implement non-ASCII case conversion for ISO-8859-14, by Yutaro Tada.

Wed Jul  6 19:24:48 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/iso_8859_1.c, enc/iso_8859_15.c, enc/iso_8859_16.c:
	  Align indenting to onigmo convention.

Wed Jul  6 18:59:13 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/iso_8859_15.c, test/ruby/enc/test_case_comprehensive.rb:
	  Implement non-ASCII case conversion for ISO-8859-15, by Maho Harada.

Wed Jul  6 18:34:21 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/iso_8859_16.c, test/ruby/enc/test_case_comprehensive.rb:
	  Implement non-ASCII case conversion for ISO-8859-16, by Satoshi Kayama.

Wed Jul  6 14:44:56 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* lib/mkmf.rb (create_makefile): store $headers in LOCAL_HDRS for
	  depend files.

	* ext/digest/digest_conf.rb (digest_conf): add implementation
	  specific headers to $header.

	* ext/digest/{md5,rmd160,sha1,sha2}/depend: add LOCAL_HDRS to the
	  dependencies.

Wed Jul  6 08:59:35 2016  Shugo Maeda  <shugo@ruby-lang.org>

	* lib/net/http/generic_rquest.rb (write_header): A Request-Line must
	  not contain CR or LF.

Wed Jul  6 07:11:27 2016  Shugo Maeda  <shugo@ruby-lang.org>

	* lib/net/ftp.rb (putline): raise an ArgumentError when
	  CR or LF is included in a line.

Tue Jul  5 20:49:30 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* ext/json/*, test/json/*: Update json-2.0.1.
	  Changes of 2.0.0: https://github.com/flori/json/blob/f679ebd0c69a94e3e70a897ac9a229f5779c2ee1/CHANGES.md#2015-09-11-200
	  Changes of 2.0.1: https://github.com/flori/json/blob/f679ebd0c69a94e3e70a897ac9a229f5779c2ee1/CHANGES.md#2016-07-01-201
	  [Feature #12542][ruby-dev:49706][fix GH-1395]

Tue Jul  5 19:39:49 2016  Naohisa Goto  <ngotogenome@gmail.com>

	* string.c (rb_str_change_terminator_length): New function to change
	  termlen and resize heap for the terminator. This is split from
	  rb_str_fill_terminator (str_fill_term) because filling terminator
	  and changing terminator length are different things. [Bug #12536]

	* internal.h: declaration for rb_str_change_terminator_length.

	* string.c (str_fill_term): Simplify only to zero-fill the terminator.
	  For non-shared strings, it assumes that (capa + termlen) bytes of
	  heap is allocated. This partially reverts r55557.

	* encoding.c (rb_enc_associate_index): rb_str_change_terminator_length
	  is used, and it should be called whenever the termlen is changed.

	* string.c (str_capacity): New static function to return capacity
	  of a string with the given termlen, because the termlen may
	  sometimes be different from TERM_LEN(str) especially during
	  changing termlen or filling terminator with specific termlen.

	* string.c (rb_str_capacity): Use str_capacity.

Tue Jul  5 11:07:14 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* pack.c (pack_pack): use union instead of bare variable to ease
	  optimizations and avoid assigning x87 floating point number.
	  [ruby-core:74496] [Bug #12209]

	* pack.c (pack_unpack): ditto.

Mon Jul  4 13:56:34 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* process.c: define sig_t if not exist.
	  at least Solaris 10 and 11 doesn't have sig_t.

Mon Jul  4 13:08:48 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* random.c (random_ulong_limited): avoid left shift count >= width of
	  type on 32bit environment.

Sun Jul  3 18:51:42 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* test/ruby/enc/test_case_comprehensive.rb, test_regex_casefold.rb,
	  test/test_unicode_normalize.rb: Replace UNICODE_VERSION from
	  UnicodeNormalize with RbConfig::CONFIG['UNICODE_VERSION'] from
	  feature 12460.

Sun Jul  3 06:04:09 2016  Eric Wong  <e@80x24.org>

	* process.c (disable_child_handler_fork_child): simplify
	  [ruby-core:75781] [Misc #12439]

Sun Jul  3 05:25:46 2016  Eric Wong  <e@80x24.org>

	* tool/asm_parse.rb: add description
	* tool/change_maker.rb: ditto
	* tool/downloader.rb: ditto
	* tool/eval.rb: ditto
	* tool/expand-config.rb: ditto
	* tool/extlibs.rb: ditto
	* tool/fake.rb: ditto
	* tool/file2lastrev.rb: ditto
	* tool/gem-unpack.rb: ditto
	* tool/gen_dummy_probes.rb: ditto
	* tool/gen_ruby_tapset.rb: ditto
	* tool/generic_erb.rb: ditto
	* tool/id2token.rb: ditto
	* tool/ifchange: ditto
	* tool/insns2vm.rb: ditto
	* tool/instruction.rb: ditto
	* tool/jisx0208.rb: ditto
	* tool/merger.rb: ditto
	* tool/mkrunnable.rb: ditto
	* tool/node_name.rb: ditto
	* tool/parse.rb: ditto
	* tool/rbinstall.rb: ditto
	* tool/rbuninstall.rb: ditto
	* tool/rmdirs: ditto
	* tool/runruby.rb: ditto
	* tool/strip-rdoc.rb: ditto
	* tool/vcs.rb: ditto
	* tool/vtlh.rb: ditto
	* tool/ytab.sed: ditto
	* tool/enc-unicode.rb: fix typo
	* tool/mk_call_iseq_optimized.rb: ditto
	* tool/update-deps: ditto
	  [ruby-core:76215] [Bug #12539]
	  by Noah Gibbs <the.codefolio.guy@gmail.com>

Sat Jul  2 18:04:24 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* lib/uri/mailto.rb (initialize): RFC3986_Parser#split sets opaque
	  only if the URI has path-rootless, not path-empty.
	  [ruby-core:76055] [Bug #12498]
	  patched by Chris Heisterkamp <cheister@squareup.com>

Sat Jul  2 04:26:14 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* regcomp.c (noname_disable_map): don't optimize out group 0
	  Ruby's Regexp doesn't allow normal numbered groups if the regexp
	  has named groups. On such case it optimizes out related NT_ENCLOSE.
	  But even on the case it can use \g<0>.
	  This fix not to remove NT_ENCLOSE whose regnum is 0.
	  [ruby-core:75828] [Bug #12454]

Sat Jul  2 03:09:27 2016  Naohisa Goto  <ngotogenome@gmail.com>

	* string.c: Partially reverts r55547 and r55555.
	  ChangeLog about the reverted changes are also deleted in this file.
	  [Bug #12536] [ruby-dev:49699] [ruby-dev:49702]

Sat Jul  2 02:22:22 2016  Naohisa Goto  <ngotogenome@gmail.com>

	* string.c (str_fill_term): When termlen increases, re-allocation
	  of memory for termlen should always be needed.
	  In this fix, if possible, decrease capa instead of realloc.
	  [Bug #12536] [ruby-dev:49699]

Fri Jul  1 20:20:20 2016  Naohisa Goto  <ngotogenome@gmail.com>

	* string.c: Specify termlen as far as possible.
	  Additional fix for [Bug #12536] [ruby-dev:49699].

	* string.c (str_new_static): Specify termlen from the given encoding
	  when creating a new String object is needed.

	* string.c (rb_tainted_str_new_with_enc): New function to create a
	  tainted String object with the given encoding. This means that
	  the termlen is correctly specified. Currently static function.
	  The function name might be renamed to rb_tainted_enc_str_new
	  or rb_enc_tainted_str_new.

	* string.c (rb_external_str_new_with_enc): Use encoding by using the
	  above rb_tainted_str_new_with_enc().

Fri Jul  1 19:38:57 2016  Naohisa Goto  <ngotogenome@gmail.com>

	* test/fiddle/test_pointer.rb (test_to_str, test_to_s, test_aref_aset):
	  Attempt to use independent strings for destructive tests that
	  directly modify values on memory by using Fiddle::Pointer.
	  [Bug #12537] [ruby-dev:49700]

Fri Jul  1 18:20:00 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* .gdbinit (rb_ps_thread): show the detail of cfunc in ruby level
	  backtrace.

Fri Jul  1 13:26:39 2016  Naohisa Goto  <ngotogenome@gmail.com>

	* string.c (rb_str_subseq, str_substr): When RSTRING_EMBED_LEN_MAX
	  is used, TERM_LEN(str) should be considered with it because
	  embedded strings are also processed by TERM_FILL.
	  Additional fix for [Bug #12536] [ruby-dev:49699].

Fri Jul  1 12:11:01 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* .gdbinit (rb_count_objects): added gdb version of count_objects().

Fri Jul  1 04:32:52 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* .gdbinit (rb_ps_thread): show ruby level backtrace.
	  Usually you can call `rb_ps` to show ruby level backtraces
	  for all living threads.
	  Note that it can call with core file like `gcore <pid>`
	  and `gdb ruby core.<pid>`.

Thu Jun 30 19:15:13 2016  Naohisa Goto  <ngotogenome@gmail.com>

	* string.c: Fix memory corruptions when using UTF-16/32 strings.
	  [Bug #12536] [ruby-dev:49699]

	* string.c (rb_str_new_with_class): Use TERM_LEN of the "obj".

	* string.c (rb_str_plus, rb_str_justify): Use str_new0 which is aware
	  of termlen.

	* string.c (str_shared_replace): Copy +termlen bytes instead of +1.

	* string.c (rb_str_times): termlen should not be included in capa.

	* string.c (RESIZE_CAPA_TERM): When using RSTRING_EMBED_LEN_MAX,
	  termlen should be counted with it because embedded strings are
	  also processed by TERM_FILL.

	* string.c (rb_str_capacity, str_shared_replace, str_buf_cat): ditto.

	* string.c (rb_str_drop_bytes, rb_str_setbyte, str_byte_substr): ditto.

Wed Jun 29 22:24:37 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* ext/psych/lib/psych_jars.rb: removed needless file required to JRuby.

Wed Jun 29 22:21:38 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl_ocsp.c: The "reuse" behavior of d2i_ functions does
	  not work well with OpenSSL 1.0.0t. So avoid it.

Wed Jun 29 15:18:28 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* insns.def (opt_succ): optimize like r55515. (but this argument is
	  constant)

Wed Jun 29 12:41:58 2016  Shugo Maeda  <shugo@ruby-lang.org>

	* test/ruby/test_refinement.rb: skip
	  test_prepend_after_refine_wb_miss on ARM or MIPS.
	  [ruby-core:76031] [Bug #12491]

Wed Jun 29 08:45:53 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* proc.c (passed_block): convert passed block symbol to proc.
	  based on the patch by Daisuke Sato in [ruby-dev:49695].
	  [Bug #12531]

Wed Jun 29 03:34:41 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* bignum.c (rb_big2ulong): the old logic seems to try to avoid
	  calculating `-(long)(num-1)-1` if `num` is not LONG_MIN. (Note that
	  `-LONG_MIN` may be larger than LONG_MAX) But C compilers can
	  optimize it into single NEG instruction.
	  Therefore those two conditions can be single if-body.

	* bignum.c (rb_big2long): ditto.

	* bignum.c (rb_big2ull): ditto.

	* bignum.c (rb_big2ll): ditto.

Tue Jun 28 22:55:00 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* lib/pstore.rb (PStore::CHECKSUM_ALGO): extract the algorithm for
	  checksum, instead of qualified names for each times.

Tue Jun 28 22:29:36 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* bootstraptest/runner.rb: do not use safe navigation operator.
	  this runner may run on older ruby.  partially revert r53110
	  (GH-1142 patched by @mlarraz).

Tue Jun 28 22:09:09 2017  Akio Tajima <artonx@yahoo.co.jp>

	* lib/fileutils.rb: rescue Errno:EACCES for chown.
	  [Bug #12520]

Tue Jun 28 18:38:09 2016  Naohisa Goto  <ngotogenome@gmail.com>

	* ext/digest/md5/md5ossl.h: Remove excess semicolons.
	  Suppress warning on Solaris with Oracle Solaris Studio 12.
	  [ruby-dev:49692] [Bug #12524]

	* ext/digest/md5/md5cc.h: ditto.
	* ext/digest/sha1/sha1cc.h: ditto.
	* ext/digest/sha1/sha1ossl.h: ditto.
	* ext/digest/sha2/sha2cc.h: ditto.
	* ext/digest/sha2/sha2ossl.h: ditto.
	* ext/openssl/ossl_pkey_rsa.c: ditto.

Tue Jun 28 15:56:48 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* test/ruby/enc/test_case_comprehensive.rb: noting to test if
	  Unicode data files are available.
	  [ruby-core:76160] [Bug #12433]

	* test/test_unicode_normalize.rb: ditto.

Tue Jun 28 15:20:18 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* lib/net/http.rb (Net::HTTP#proxy_uri): cache the case no proxy
	  is used.

Tue Jun 28 09:56:29 2016  Stefan Schussler  <mail@stefanschuessler.de>

	* object.c (rb_mod_eqq): [DOC] Fix typo in RDoc.  [Fix GH-1393]

Tue Jun 28 02:41:32 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* insns.def (opt_plus): use `- 1` instead of `& (~1)` to allow
	  compilers to use x86 LEA instruction (3 operand).
	  Even if 3 operand LEA's latency is 3 cycle after SandyBridge,
	  it reduces code size and can be faster because of super scalar.

	* insns.def (opt_plus): calculate and use rb_int2big.
	  On positive Fixnum overflow, `recv - 1 + obj` doesn't carry
	  because recv's msb and obj's msb are 0, and resulted msb is 1.
	  Therefore simply rshift and cast as signed long works fine.
	  On negative Fixnum overflow, it will carry because both arguments'
	  msb are 1, and resulted msb is also 1.
	  In this case it needs to restore carried sign bit after rshift.

Mon Jun 27 16:58:32 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* lib/fileutils.rb (FileUtils#install): accecpt symbolic mode, as
	  well as chmod.

	* lib/fileutils.rb (FileUtils#install): add owner and group
	  options.

Mon Jun 27 08:56:55 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* compile.c (ADD_TRACE): ignore trace instruction on non-positive
	  line.

	* parse.y (coverage): get rid of ArgumentError when the starting
	  line number is not positive.  [ruby-core:76141] [Bug #12517]

Sun Jun 26 10:20:25 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ext/win32/lib/Win32API.rb (Win32API#initialize): Cygwin
	  2.5.2-1 (perhaps) seems to no longer append ".dll" suffix
	  implicitly.

	* ext/win32/lib/win32/resolv.rb (Win32::Resolv): ditto.  Fix the
	  error reported by yamataka AT u08.itscom.net in
	  [ruby-list:50339], and pointed out and patched by cerberus AT
	  m3.kcn.ne.jp in [ruby-list:50341].

Sat Jun 25 10:07:52 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* test/openssl/test_ocsp.rb: Ignore errors caused by bugs that exist in
	  LibreSSL >= 2.3.1.

Sat Jun 25 02:33:33 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* vm_method.c (vm_respond_to): try method_missing if respond_to?
	  is undefined, as if it is the default definition.
	  [ruby-core:75377] [Bug #12353]

Fri Jun 24 17:04:21 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* ext/psych/*, test/psych/*: Update psych 2.1.0
	  This version fixed [Bug #11988][ruby-core:72850]

Fri Jun 24 13:12:41 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* lib/rubygems.rb, lib/rubygems/*, test/rubygems/*:
	  Update rubygems 2.6.5 and 2.6.6.
	  Release note of 2.6.5: https://github.com/rubygems/rubygems/commit/656f5d94dc888d78d0d00f3598a4fa37391aac80
	  Release note of 2.6.6: https://github.com/rubygems/rubygems/commit/ccb9c3300c063f5b5656669972d24a10ef8afbf5

Fri Jun 24 09:17:15 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* common.mk (lib/unicode_normalize/tables.rb): should not depend
	  on Unicode data files unless ALWAYS_UPDATE_UNICODE=yes, to get
	  rid of downloading Unicode data unnecessary.  [ruby-dev:49681]

	* common.mk (enc/unicode/casefold.h): update Unicode files in a
	  sub-make, not to let the header depend on the files always.

	* enc/unicode/case-folding.rb: if gperf is not usable, assume the
	  existing file is OK.

Tue Jun 21 19:44:54 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* test/ruby/enc/test_regex_casefold.rb: Add Windows-1251, KOI8-R, and
	  KOI8-U to encodings; definitely removed EUC-JP.

Tue Jun 21 19:32:23 2016  Mark St.Godard  <markstgodard@gmail.com>

	* lib/webrick/httprequest.rb (setup_forwarded_info): Use the first
	  value in X-Forwarded-Proto, if header contains multiple comma
	  separated values.  Some middlewares may add these values to the
	  list, not replacing.  [Fix GH-1386]

Tue Jun 21 17:17:42 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* test/ruby/test_io.rb: Skip test_readpartial_with_not_empty_buffer,
	  test_read_buffer_error, test_read_unlocktmp_ensure,
	  test_readpartial_unlocktmp_ensure, and
	  test_sysread_unlocktmp_ensure on cygwin,
	  because these tests repeatedly hang. This makes test_io.rb
	  complete in finite time on cygwin.

	* ChangeLog: Fix test_in.rb -> test_io.rb (two instances).

Tue Jun 21 16:38:14 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* string.c (rb_str_casemap): do not put code with side effects
	  inside RSTRING_PTR() macro which evaluates the argument multiple
	  times.

Tue Jun 21 16:13:45 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* string.c (rb_str_casemap): fix memory leak.

Tue Jun 21 16:12:21 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* string.c (rb_str_casemap): int is too small for string size.

Tue Jun 21 15:42:22 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* test/ruby/test_io.rb: Skip test_read_buffer_error on cygwin,
	  because this test repeatedly hangs.

Tue Jun 21 15:35:14 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* LEGAL: Added filenames and copyrights for some files in
	  enc/trans/JIS.

Tue Jun 21 00:56:47 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* win32/win32.c (get_special_folder): fix calling convention of
	  SHGetPathFromIDListEx, which should be WINAPI.  pointed out by
	  @arton at http://twitter.com/arton/status/744884064277016576

Tue Jun 21 00:22:02 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* configure.in, include/ruby/defines.h (RUBY_USE_SETJMPEX):
	  include setjmpex.h only when using setjmpex() for RUBY_SETJMP.
	  the header of mingw32 overrides setjmp() by setjmpex().

Mon Jun 20 18:39:16 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* test/ruby/test_io.rb: Skip test_open_fifo_does_not_block_other_threads
	  on cygwin. Fifos seem to work okay in cygwin, but this test repeatedly
	  hangs.

Mon Jun 20 13:35:06 2016  Shugo Maeda  <shugo@ruby-lang.org>

	* vm.c (invoke_bmethod, invoke_block_from_c_0): revert r52104
	  partially to avoid "self has wrong type to call super in this
	  context" errors.
	  [ruby-core:72724] [Bug #11954]

Mon Jun 20 12:53:38 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* test/rubygems/test_gem_remote_fetcher.rb: OpenSSL::PKey::DH#priv_key=
	  is not defined when ext/openssl is built with OpenSSL 1.1.0.
	  https://github.com/rubygems/rubygems/pull/1648
	  [ruby-core:75225] [Feature #12324]

Sun Jun 19 21:25:43 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl_ocsp.c: Add OCSP::SingleResponse that represents an
	  OCSP SingleResponse structure. Also add two new methods #responses
	  and #find_response to OCSP::BasicResponse. A BasicResponse has one or
	  more SingleResponse. We have OCSP::BasicResponse#status that returns
	  them as an array of arrays, each containing the content of a
	  SingleResponse, but this is not useful. When validating an OCSP
	  response, we need to look into the each SingleResponse and check their
	  validity but it is not simple. For example, when validating for a
	  certificate 'cert', the code would be like:

	    # certid_target is an OpenSSL::OCSP::CertificateId for cert
	    basic = res.basic
	    result = basic.status.any? do |ary|
	      ary[0].cmp(certid_target) &&
	        ary[4] <= Time.now && (!ary[5] || Time.now <= ary[5])
	    end

	  Adding OCSP::SingleResponse at the same time allows exposing
	  OCSP_check_validity(). With this, the code above can be rewritten as:

	    basic = res.basic
	    single = basic.find_response(certid_target)
	    result = single.check_validity

	* test/openssl/test_ocsp.rb: Test this.

Sun Jun 19 18:40:19 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl_ocsp.c (ossl_ocspbres_add_status): Allow specifying
	  the times (thisUpdate, nextUpdate and revocationTime) with Time
	  objects. Currently they accepts only relative seconds from the current
	  time. This is inconvenience, especially for revocationTime. When
	  Integer is passed, they are still treated as relative times. Since the
	  type check is currently done with rb_Integer(), this is a slightly
	  incompatible change. Hope no one passes a relative time as String or
	  Time object...
	  Also, allow passing nil as nextUpdate. It is optional.

	* ext/openssl/ruby_missing.h: Define RB_INTEGER_TYPE_P() if not defined.
	  openssl gem will be released before Ruby 2.4.0.

Sun Jun 19 18:39:38 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl_ocsp.c: Implement OCSP::{CertificateId,Request,
	  BasicResponse,Response}#initialize_copy.
	  [ruby-core:75504] [Bug #12381]

	* test/openssl/test_ocsp.rb: Test them.

Sun Jun 19 18:29:50 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl_pkey_dh.c, ext/openssl/ossl_pkey_dsa.c,
	  ext/openssl/ossl_pkey_ec.c, ext/openssl/ossl_pkey_rsa.c: Implement
	  initialize_copy method for OpenSSL::PKey::*.
	  [ruby-core:75504] [Bug #12381]

	* test/openssl/test_pkey_dh.rb, test/openssl/test_pkey_dsa.rb,
	  test/openssl/test_pkey_ec.rb, test/openssl/test_pkey_rsa.rb: Test they
	  actually copy the OpenSSL objects, and modifications to cloned object
	  don't affect the original object.

Sun Jun 19 16:55:16 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* test/ruby/test_dir_m17n.rb: Skip tests with non-UTF-8 encodings
	  on cygwin. Cygwin can use the Unicode PUA (private use area) to store
	  bytes from non-UTF-8 filenames (see
	  https://cygwin.com/cygwin-ug-net/using-specialnames.html#pathnames-specialchars),
	  but we are not supporting this. [Bug #12443]

Sun Jun 19 15:01:18 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* localeinit.c: Fix filesystem encoding for cygwin to UTF-8 (see
	  https://cygwin.com/cygwin-ug-net/using-specialnames.html#pathnames-unusual)

Sun Jun 19 14:31:07 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl_pkey.h, ext/openssl/ossl_pkey_dh.c,
	  ext/openssl/ossl_pkey_dsa.c, ext/openssl/ossl_pkey_rsa.c: A few days
	  ago, OpenSSL changed {DH,DSA,RSA}_get0_*() to take const BIGNUM **.
	  https://git.openssl.org/gitweb/?p=openssl.git;a=commit;h=fd809cfdbd6e32b6b67b68c59f6d55fbed7a9327
	  [ruby-core:75225] [Feature #12324]

Sun Jun 19 11:19:43 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* variable.c (rb_path_to_class): consider the string length
	  instead of a terminator.

	* variable.c (rb_path_to_class): search the constant at once
	  instead of checking if defined and then getting it.

Sat Jun 18 14:01:40 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* test/rubygems/test_gem_installer.rb: Fixed broken test with extension
	  build. https://github.com/rubygems/rubygems/pull/1645

Sat Jun 18 13:59:54 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* lib/rubygems.rb, lib/rubygems/*, test/rubygems/*: Update rubygems
	  HEAD(2c6d256). It contains to update vendored Molinillo to 0.5.0.
	  https://github.com/rubygems/rubygems/pull/1638

Sat Jun 18 10:13:37 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* common.mk (build-ext), ext/extmk.rb: use variable EXTENCS
	  different than ENCOBJS, to get rid of circular dependency.
	  build libencs when linking encodings statically.
	  [ruby-core:75618] [Bug #12401]

Sat Jun 18 08:52:46 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ext/stringio/stringio.c (strio_getline): fix pointer index
	  overflow.  reported by Guido Vranken <guido AT guidovranken.nl>.

Thu Jun 16 16:35:35 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* class.c (Init_class_hierarchy): prevent rb_cObject which is the
	  class tree root, from GC.  [ruby-dev:49666] [Bug #12492]

Thu Jun 16 12:17:52 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* string.c (tr_trans): adjust buffer size by processed and rest
	  lengths, instead of doubling repeatedly.

Thu Jun 16 11:15:25 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* string.c (tr_trans): consider terminator length and fix heap
	  overflow.  reported by Guido Vranken <guido AT guidovranken.nl>.

Thu Jun 16 00:02:32 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl_ocsp.c (ossl_ocspreq_verify, ossl_ocspbres_verify):
	  Use ossl_clear_error() so that they don't print warnings to stderr and
	  leak errors in the OpenSSL error queue. Also, check the return value
	  of OCSP_*_verify() correctly. They can return -1 on verification
	  failure.

Wed Jun 15 19:52:23 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl_ocsp.c (ossl_ocspreq_sign, ossl_ocspbres_sign): Allow
	  specifying hash algorithm used in signing. They are hard coded to use
	  SHA-1.
	  Based on a patch provided by Tim Shirley <tidoublemy@gmail.com>.
	  [ruby-core:70915] [Feature #11552] [GH ruby/openssl#28]

	* test/openssl/test_ocsp.rb: Test sign-verify works.

Wed Jun 15 01:46:16 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* numeric.c: [DOC] fix rdoc directive, and an example of negative
	  value.  [ruby-core:76025] [Bug #12487]

Wed Jun 15 01:44:42 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* tool/mkconfig.rb: provide Unicode Version information as
	  RbConfig::CONFIG['UNICODE_VERSION'].
	  [ruby-core:75845] [Feature #12460]

Wed Jun 15 00:01:18 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>

	* test/ruby/enc/test_case_comprehensive.rb
	  (TestComprehensiveCaseFold::read_data): use \A and \z instead of
	  ^ and $ in regexp.

Tue Jun 14 23:43:00 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* include/ruby/backward.h (rb_cFixnum, rb_cBignum): remove the
	  backward compatibility macros, to fail incompatible extension
	  libraries early.  [Bug #12427]

Tue Jun 14 22:22:07 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* strftime.c (rb_strftime_with_timespec): limit the result string
	  size by the format length, to get rid of unlimited memory use.

Tue Jun 14 22:11:11 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl_ocsp.c (ossl_ocspcid_get_issuer_name_hash,
	  ossl_ocspcid_get_issuer_key_hash, ossl_ocspcid_get_hash_algorithm):
	  Add accessor methods OCSP::CertificateId#issuer_name_hash,
	  #issuer_key_hash, #hash_algorithm.
	  Based on a patch provided by Paul Kehrer <paul.l.kehrer@gmail.com>.
	  [ruby-core:48062] [Feature #7181]

	* test/openssl/test_ocsp.rb: Test these new methods.

Tue Jun 14 22:07:25 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ext/date/date_strftime.c (date_strftime_with_tmx): reject too
	  large precision to get rid of buffer overflow.
	  reported by Guido Vranken <guido AT guidovranken.nl>.

Tue Jun 14 21:40:42 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl_ocsp.c (ossl_ocspbres_to_der, ossl_ocspcid_to_der):
	  Implement #to_der methods for OCSP::BasicResponse and
	  OCSP::CertificateId.

	  (ossl_ocspreq_initialize, ossl_ocspres_initialize): Use GetOCSP*()
	  instead of raw DATA_PTR().

	  (ossl_ocspbres_initialize, ossl_ocspcid_initialize): Allow
	  initializing from DER string.

	  (Init_ossl_ocsp): Define new #to_der methods.

	* test/openssl/test_ocsp.rb: Test these changes. Also add missing tests
	  for OCSP::{Response,Request}#to_der.

Tue Jun 14 21:35:00 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/openssl_missing.h (DH_set0_pqg, RSA_set0_key):
	  DH_set0_pqg() allows 'q' to be NULL. Fix a typo in RSA_set0_key().
	  Fixes r55285.  [ruby-core:75225] [Feature #12324]

Mon Jun 14 10:19:00 2016  Kenta Murata  <mrkn@mrkn.jp>

	* NEWS: describe Integer#digits.

Mon Jun 13 21:09:40 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* thread.c (debug_deadlock_check): show thread lock dependency and
	  backtrace [Feature #8214] [ruby-dev:47217]

	* thread.c (thread_status_name): show "sleep_forever" instead of
	  "sleep" if called from inspect.

Mon Jun 13 20:50:07 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* parse.y (reg_named_capture_assign_iter): remove named capture
	  conflict warnings.  it is just annoying rather than useful.
	  [ruby-core:75416] [Bug #12359]

Mon Jun 13 20:04:00 2016  Kenta Murata  <mrkn@mrkn.jp>

	* numeric.c (rb_int_digits, rb_fix_digits, rb_int_digits_bigbase):
	  Add Integer#digits to extract columns in place-value notation
	  [Feature #12447] [ruby-core:75799]

	* test/ruby/test_integer.rb: Add tests for the above change.

	* test/ruby/test_bignum.rb: ditto.

Mon Jun 13 20:34:53 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* include/ruby/ruby.h (RUBY_INTEGER_UNIFICATION): macro to tell if
	  Integer is integrated.  [ruby-core:75718][Bug #12427]

	* include/ruby/backward.h, internal.h (rb_cFixnum, rb_cBignum):
	  fallback to rb_cInteger.

	* bignum.c, numeric.c, ext/json/generator/generator.{c,h}: use the
	  macro.

Mon Jun 13 16:58:53 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* test/ruby/enc/test_case_comprehensive.rb: Add tests for full Unicode
	  swapcase.

Sun Jun 12 14:48:00 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* test/ruby/enc/test_case_comprehensive.rb: Add tests for ASCII-only
	  swapcase; store calculated values in hashes.

Sun Jun 12 14:05:45 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl_cipher.c (ossl_cipher_get_auth_tag,
	  ossl_cipher_set_auth_tag): Check if the cipher flags retrieved by
	  EVP_CIPHER_CTX_flags() includes EVP_CIPH_FLAG_AEAD_CIPHER to see if
	  the cipher supports AEAD. AES-GCM was the only supported in OpenSSL
	  1.0.1.

	  (Init_ossl_cipher): Fix doc; OpenSSL::Cipher::AES.new(128, :GCM) can't
	  work.

	* ext/openssl/openssl_missing.h: Define EVP_CTRL_AEAD_{GET,SET}_TAG if
	  missing. They are added in OpenSSL 1.1.0, and have the same value as
	  EVP_CTRL_GCM_{GET,SET}_TAG and EVP_CTRL_CCM_{GET,SET}_TAG.

Sun Jun 12 13:47:42 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* test/openssl/test_engine.rb (test_openssl_engine_builtin,
	  test_openssl_engine_by_id_string): Skip test if 'openssl' engine is
	  already loaded. And test the number increased by Engine.load{_by_id,},
	  not the total count of loaded engines. Previously, we called
	  OpenSSL::Engine.cleanup every time running a test case, but we no
	  longer can do it.
	  [ruby-core:75225] [Feature #12324]

Sun Jun 12 09:24:34 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* file.c (append_fspath): normalize directory name to be appended
	  on OS X.  [ruby-core:75957] [Ruby trunk Bug#12483]
	  https://github.com/rails/rails/issues/25303#issuecomment-224834804

Sat Jun 11 23:07:32 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* lib/forwardable.rb (_delegator_method): get rid of a warning which
	  causes test failures introduced at r55376.

Sat Jun 11 18:37:58 2016  Marcus Stollsteimer  <sto.mar@web.de>

	* ext/json/lib/*.rb: Removed some comments. Because these are unnecessary
	  class description. [ci skip][Bug #12255][ruby-core:74835]

Sat Jun 11 15:19:38 2016  takiy33  <takiy33@users.noreply.github.com>

	* lib/net/smtp.rb: [DOC] Remove dead link on documentation for
	  Japanese of SMTP.  [Fix GH-1380]

Sat Jun 11 15:02:45 2016  Grant Hutchins  <github@nertzy.com>

	* string.c (rb_str_oct): [DOC] fix typo, hornored -> honored.
	  [Fix GH-1379]

Sat Jun 11 14:04:19 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ext/objspace/objspace_dump.c: generate valid JSON for dump_all.

Sat Jun 11 13:52:33 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* lib/forwardable.rb (_delegator_method): remove __send__ call if
	  possible, so that more optimizations will be enabled.

Sat Jun 11 11:24:36 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* enc/iso_8859.h (SHARP_s): name frequently used codepoint.

Sat Jun 11 09:58:45 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/iso_8859_1.c: Revert to older version of code.

Sat Jun 11 09:46:17 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/iso_8859_1.c: Implement non-ASCII case mapping.

	* test/ruby/enc/test_case_comprehensive.rb: Tests for above.

	* string.c: Add iso-8859-1 to supported encodings.

Sat Jun 11 09:31:28 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* lib/forwardable.rb (_delegator_method): leave the backtrace
	  untouched during accessor.  forwardable.rb does not appear in
	  the backtrace during delegated method because of tail-call
	  optimization.

Sat Jun 11 01:38:31 2016  Naohisa Goto  <ngotogenome@gmail.com>

	* include/ruby/defines.h (GCC_VERSION_SINCE): Fix logic error by
	  adding parentheses. Fix failures of TestMkmf::TestConvertible
	  with GCC 3.4.3 on Solaris 10. [Bug #12479] [ruby-dev:49660]

Fri Jun 10 21:54:24 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* lib/forwardable.rb (Forwardable._delegator_method): extract
	  method generator and deal with non-module objects.
	  [ruby-dev:49656] [Bug #12478]

Fri Jun 10 17:35:11 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* string.c: Special-case :ascii option in rb_str_capitalize_bang and
	  rb_str_swapcase_bang.

Fri Jun 10 17:12:24 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* string.c: Special-case :ascii option in rb_str_upcase_bang (retry).

Fri Jun 10 14:48:36 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* hash.c (get_env_cstr): ensure NUL-terminated.
	  [ruby-dev:49655] [Bug #12475]

	* string.c (rb_str_fill_terminator): return the pointer to the
	  NUL-terminated content.

Thu Jun  9 21:42:00 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl_asn1.c (asn1integer_to_num): Use
	  ASN1_ENUMERATED_to_BN() to convert an ASN1_ENUMERATED to a BN.
	  Starting from OpenSSL 1.1.0, ASN1_INTEGER_to_BN() rejects
	  non-ASN1_INTEGER objects. The format of INTEGER and ENUMERATED are
	  almost identical so they behaved in the same way in OpenSSL <= 1.0.2.
	  [ruby-core:75225] [Feature #12324]

	* test/openssl/test_asn1.rb (test_decode_enumerated): Test that it
	  works.

Thu Jun  9 21:10:04 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>

	* tool/ifchange: fix timestamp error when target without
	  directory.

Thu Jun  9 19:46:22 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl_ssl.c: Add define guards for OPENSSL_NO_EC.
	  SSL_CTX_set_ecdh_auto() is defined even when ECDH is disabled in
	  OpenSSL's configuration. This fixes r55214.

	* test/openssl/test_pair.rb (test_ecdh_curves): Skip if the OpenSSL does
	  not support ECDH.

	* test/openssl/utils.rb (start_server): Ignore error in
	  SSLContext#ecdh_curves=.

Thu Jun  9 18:12:42 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* Makefile.in (un-runnable): fail with proper error message.
	  [ruby-core:75905] [Bug #12472]

Thu Jun  9 15:32:17 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* common.mk (RBCONFIG): use ifchange tool to see if the content is
	  changed and update the timestamp file.

	* tool/mkconfig.rb: remove ifchange features.

	* tool/ifchange: make target directory if it does not exist with
	  its parent directories.

	* win32/ifchange.bat: drop support for command.com used on old
	  systems.

Thu Jun  9 15:03:35 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/extconf.rb: Check for CRYPTO_malloc() and SSL_new().
	  OpenSSL_add_all_digests() and SSL_library_init() are deprecated and
	  converted to macros in OpenSSL 1.1.0.
	  [ruby-core:75225] [Feature #12324]

Wed Jun  8 23:09:51 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>

	* string.c (rb_str_ascii_casemap): fix compile error.

Wed Jun  8 22:22:24 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* string.c: Revert previous commit (possibility of endless loop).

Wed Jun  8 21:57:41 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* string.c: Special-case :ascii option in rb_str_upcase_bang.

Wed Jun  8 21:28:36 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* string.c: New static function rb_str_ascii_casemap; special-casing
	  :ascii option in rb_str_upcase_bang and rb_str_downcase_bang.

	* regenc.c: Fix a bug (wrong use of unnecessary slack at end of string).

	* regenc.h -> include/ruby/oniguruma.h: Move declaration of
	  onigenc_ascii_only_case_map so that it is visible in string.c.

Wed Jun  8 20:33:44 2016  Naohisa Goto  <ngotogenome@gmail.com>

	* include/ruby/intern.h: Remove excess semicolons in PUREFUNC().
	  Fix failure of TestMkmf::TestConvertible on Solaris with
	  Oracle Solaris Studio 12. [ruby-dev:49651] [Bug #12470]
	* internal.h: ditto.

Wed Jun  8 16:03:09 2016  Shugo Maeda  <shugo@ruby-lang.org>

	* lib/net/smtp.rb (getok, get_response): raise an ArgumentError when
	  CR or LF is included in a line, because they are not allowed in
	  RFC5321.  Thanks, Jeremy Daer.

Tue Jun  7 21:27:25 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* test/rubygems/*_{cert,cert_32}.pem: Regenerate test certificates for
	  OpenSSL 1.1.0. This is already in upstream.
	  https://github.com/rubygems/rubygems/commit/9be5c53939440a61c4bba73cfffbeb5cfadf72be
	  [ruby-core:75225] [Feature #12324]

Tue Jun  7 21:27:17 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* test/open-uri/test_ssl.rb: Regenerate test certificates. The test CA
	  certificate was incorrectly generated. A CA certificate must have the
	  basic constraints extension with cA bit set to TRUE. OpenSSL <= 1.0.2
	  allowed the error when the certificate is in the trusted store but
	  OpenSSL 1.1.0 no longer does.
	  [ruby-core:75225] [Feature #12324]

Tue Jun  7 21:20:38 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* test/openssl/test_x509name.rb: Don't register OID for 'emailAddress'
	  and 'serialNumber'. A recent change in OpenSSL made OBJ_create()
	  reject an already existing OID. They were needed to run tests with
	  OpenSSL 0.9.6 which is now unsupported.
	  https://git.openssl.org/gitweb/?p=openssl.git;a=commit;h=52832e470f5fe8c222249ae5b539aeb3c74cdb25
	  [ruby-core:75225] [Feature #12324]

	* test/openssl/test_ssl_session.rb (test_server_session): Duplicate
	  SSL::Session before re-adding to the session store. OpenSSL 1.1.0
	  starts rejecting SSL_SESSION once removed by SSL_CTX_remove_session().
	  https://git.openssl.org/gitweb/?p=openssl.git;a=commit;h=7c2d4fee2547650102cd16d23f8125b76112ae75

	* test/openssl/test_pkey_ec.rb (setup): Remove X25519 from @keys. X25519
	  is new in OpenSSL 1.1.0 but this is for key agreement and not for
	  signing.

	* test/openssl/test_pair.rb, test/openssl/test_ssl.rb,
	  test/openssl/utils.rb: Set security level to 0 when using aNULL cipher
	  suites.

	* test/openssl/utils.rb: Use 1024 bits DSA key for client certificates.

	* test/openssl/test_engine.rb: Run each test in separate process.
	  We can no longer cleanup engines explicitly as ENGINE_cleanup() was
	  removed.
	  https://git.openssl.org/gitweb/?p=openssl.git;a=commit;h=6d4fb1d59e61aacefa25edc4fe5acfe1ac93f743

	* ext/openssl/ossl_engine.c (ossl_engine_s_cleanup): Add a note to the
	  RDoc for Engine.cleanup.

	* ext/openssl/lib/openssl/digest.rb: Don't define constants for DSS,
	  DSS1 and SHA(-0) when using with OpenSSL 1.1.0. They are removed.

	* test/openssl/test_digest.rb, test/openssl/test_pkey_dsa.rb,
	  test/openssl/test_pkey_dsa.rb, test/openssl/test_ssl.rb,
	  test/openssl/test_x509cert.rb, test/openssl/test_x509req.rb: Don't
	  test unsupported hash functions.

Tue Jun  7 17:49:52 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* test/ruby/enc/test_case_comprehensive: Change test for encodings
	  without any non-ASCII case conversions from ASCII-only test
	  to full test.

Tue Jun  7 17:18:39 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* string.c (rb_str_upcase_bang, rb_str_capitalize_bang,
	  rb_str_swapcase_bang): Switch to use primitive.

Tue Jun  7 16:44:16 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* string.c (rb_str_downcase_bang): Switch to use primitive except if
	  conversion can be done ASCII-only.

Tue Jun  7 16:13:36 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* test/ruby/enc/test_case_comprehensive: Add regression tests for
	  current ASCII-only case conversion.

Tue Jun  7 15:28:38 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* test/ruby/enc/test_case_comprehensive: Fix method name
	  (generate_casefold_tests -> generate_case_mapping_tests).

Tue Jun  7 15:05:13 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* regenc.h/c: Rename onigenc_not_support_case_map to
	  onigenc_ascii_only_case_map.

	* regenc.h: Add definition of onigenc_single_byte_ascii_only_case_map.

	* enc/iso_8859_X.c, windows_125X.c, ascii.c, us-ascii.c, koi8_x.c:
	  Replace onigenc_not_support_case_map by
	  onigenc_single_byte_ascii_only_case_map.

	* enc/big5.c, cp949.c, emacs_mule.c, euc_X.c, gbX.c, shift_jis.c,
	  windows_31j.c: Replace onigenc_not_support_case_map by
	  onigenc_ascii_only_case_map.

Tue Jun  7 14:57:09 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/extconf.rb: Check for SSL_CTX_set_min_proto_version()
	  macro added in OpenSSL 1.1.0. Version-specific methods, such as
	  TLSv1_method(), are deprecated in OpenSSL 1.1.0. We need to use
	  version-flexible methods (TLS_*method() or SSLv23_*method()) and
	  disable other protocol versions as necessary.
	  [ruby-core:75225] [Feature #12324]

	* ext/openssl/ossl_ssl.c: Use SSL_CTX_set_{min,max}_proto_version() to
	  fix the protocol version.

Tue Jun  7 12:55:34 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* regenc.c (onigenc_not_support_case_map): Move to end of file;
	  (onigenc_single_byte_ascii_only_case_map): Add new function.

Tue Jun  7 09:26:37 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* regenc.c (onigenc_not_support_case_map): Rewrite to work correctly
	  in ASCII range.

Mon Jun  6 23:00:00 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>

	* appveyor.yml: Update libressl version to 2.3.5.

Mon Jun  6 18:37:57 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* vm_insnhelper.c (vm_throw_start): check if the iseq is symbol
	  proc, class definition should not be a symbol proc.
	  [ruby-core:75856] [Bug #12462]

Mon Jun  6 18:36:34 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* string.c: Added UTF-16BE/LE and UTF-32BE/LE to supported encodings
	  for Unicode case mapping.

	* test/ruby/enc/test_case_comprehensive.rb: Tests for above
	  functionality; fixed an encoding issue in assertion error message.

Mon Jun  6 17:29:35 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* test/ruby/enc/test_case_comprehensive.rb: Speed up testing for small
	  encodings by preselecting codepoints.

Mon Jun  6 17:10:50 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl_cipher.c (ossl_cipher_free): Use EVP_CIPHER_CTX_free()
	  to free EVP_CIPHER_CTX allocated by EVP_CIPHER_CTX_new().
	  [ruby-core:75225] [Feature #12324]

Mon Jun  6 13:37:08 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* string.c Change rb_str_casemap to use encoding primitive
	  case_map instead of directly calling onigenc_unicode_case_map.

Mon Jun  6 13:16:46 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* test/ruby/enc/test_case_mapping.rb:
	  Remove :lithuanian guard for Unicode case mapping.

Mon Jun  6 10:39:56 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/openssl_missing.h: Include ruby/config.h. r55285 added
	  some inline functions but VC does not recognize 'inline' keyword.

Mon Jun  6 09:25:34 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* thread.c (thread_start_func_2): report raised exception if
	  report_on_exception flag is set.  [Feature #6647]

Mon Jun  6 01:36:24 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/extconf.rb: Check existence of SSL_is_server(). This
	  function was introduced in OpenSSL 1.0.2.
	  [ruby-core:75225] [Feature #12324]

	* ext/openssl/openssl_missing.h: Implement SSL_is_server() if missing.

	* ext/openssl/ossl_ssl.c (ssl_info_cb): Use SSL_is_server() to see if
	  the SSL is server. The state machine in OpenSSL was rewritten and
	  SSL_get_state() no longer returns SSL_ST_ACCEPT.

	  (ossl_ssl_cipher_to_ary, ossl_sslctx_session_get_cb): Add some
	  `const`s to suppress warning.

Mon Jun  6 01:18:10 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl_asn1.c (decode_bool): Do the same thing as
	  d2i_ASN1_BOOLEAN() does by ourselves. This function is removed in
	  OpenSSL 1.1.0.
	  [ruby-core:75225] [Feature #12324]

Mon Jun  6 00:34:16 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/extconf.rb: Check existence of accessor functions that
	  don't exist in OpenSSL 0.9.8. OpenSSL 1.1.0 made most of its
	  structures opaque and requires use of these accessor functions.
	  [ruby-core:75225] [Feature #12324]

	* ext/openssl/openssl_missing.[ch]: Implement them if missing.

	* ext/openssl/ossl*.c: Use these accessor functions.

	* test/openssl/test_hmac.rb: Add missing test for HMAC#reset.

Mon Jun  6 00:00:13 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/openssl_missing.[ch]: Implement EVP_PKEY_get0_*() and
	  {RSA,DSA,EC_KEY,DH}_get0_*() functions.
	  OpenSSL 1.1.0 makes EVP_PKEY/RSA/DSA/DH opaque. We used to provide
	  setter methods for each parameter of each PKey type, for example
	  PKey::RSA#e=, but this is no longer possible because the new API
	  RSA_set0_key() requires the 'n' at the same time. This commit adds
	  deprecation warning to them and adds PKey::*#set_* methods as direct
	  wrapper for those new APIs. For example, 'rsa.e = 3' now needs to be
	  rewritten as 'rsa.set_key(rsa.n, 3, rsa.d)'.
	  [ruby-core:75225] [Feature #12324]

	* ext/openssl/ossl_pkey*.[ch]: Use the new accessor functions. Implement
	  RSA#set_{key,factors,crt_params}, DSA#set_{key,pqg}, DH#set_{key,pqg}.
	  Emit a warning with rb_warning() when old setter methods are used.

	* test/drb/ut_array_drbssl.rb, test/drb/ut_drb_drbssl.rb,
	  test/rubygems/test_gem_remote_fetcher.rb: Don't set a priv_key for DH
	  object that are used in tmp_dh_callback. Generating a new key pair
	  every time should be fine - actually the private exponent is ignored
	  in OpenSSL >= 1.0.2f/1.0.1r even if we explicitly set.
	  https://www.openssl.org/news/secadv/20160128.txt

Sun Jun  5 22:06:00 2016  Kenta Murata  <mrkn@mrkn.jp>

	* configure.in: Fix the timing to detect the appropriate C++ compiler
	  in OS X.

Sun Jun  5 21:42:24 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/extconf.rb: Check absence of CRYPTO_lock() to see if the
	  OpenSSL has the new threading API. In OpenSSL <= 1.0.2, an application
	  had to set locking callbacks to use OpenSSL in a multi-threaded
	  environment. OpenSSL 1.1.0 now finds pthreads or Windows threads so we
	  don't need to do something special.
	  [ruby-core:75225] [Feature #12324]

	  Also check existence of *_up_ref(). Some structures in OpenSSL have
	  a reference counter. We used to increment it with CRYPTO_add() which
	  is a part of the old API.

	* ext/openssl/openssl_missing.h: Implement *_up_ref() if missing.

	* ext/openssl/ossl.c: Don't set locking callbacks if unneeded.

	* ext/openssl/ossl_pkey.c, ext/openssl/ossl_ssl.c,
	  ext/openssl/ossl_x509cert.c, ext/openssl/ossl_x509crl.c,
	  ext/openssl/ossl_x509store.c: Use *_up_ref() instead of CRYPTO_add().

Sun Jun  5 21:38:13 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/extconf.rb: Check if RAND_pseudo_bytes() is usable. It is
	  marked as deprecated in OpenSSL 1.1.0.
	  [ruby-core:75225] [Feature #12324]

	* ext/openssl/ossl_rand.c: Disable Random.pseudo_bytes if
	  RAND_pseudo_bytes() is unavailable.

	* test/openssl/test_random.rb: Don't test Random.pseudo_bytes if not
	  defined.

Sun Jun  5 19:06:40 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* NEWS: Add news about Unicode-wide case mapping for
	  String/Symbol#upcase/downcase/swapcase/capitalize(!).

Sun Jun  5 15:24:33 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* test/ruby/enc/test_case_comprehensive.rb:
	  Remove :lithuanian guard for Unicode case mapping.

Sun Jun  5 14:46:34 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* string.c: Remove :lithuanian guard for Unicode case mapping.

Sat Jun  4 10:54:52 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* missing/crypt.h (struct crypt_data): remove unnecessary member
	  "initialized".

	* missing/crypt.c (des_setkey_r): nothing to be initialized in
	  crypt_data.

	* configure.in (struct crypt_data): check for "initialized" in
	  struct crypt_data, which may be only in glibc, and isn't on AIX
	  at least.

Sat Jun  4 10:38:39 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* win32/win32.c: unify MAX_PATH, _MAX_PATH, and MAXPATHLEN to
	  PATH_MAX, except for MAX_PATH in get_special_folder for an API
	  limit.

Fri Jun  3 21:27:22 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ruby.c (process_options): rb_str_conv_enc() never set encoding
	  of the source string, but returns the string itself if the
	  conversion failed.  then the instance variable does not need to
	  be set again.

Fri Jun  3 18:04:37 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* ruby.c (process_options): need to duplicate path before passing it to
	  rb_str_conv_enc() because the function might call rb_enc_associate()
	  internally.  this fixes test failures on Windows introduced at r55260.

Fri Jun  3 17:44:25 2016  Reiner Herrmann  <reiner@reiner-h.de>

	* lib/mkmf.rb (create_makefile): sort lists of source and object
	  files in generated Makefile, unless given by extconf.rb.
	  [Fix GH-1367]

Thu Jun  2 21:18:10 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* win32/win32.c (get_special_folder): use SHGetPathFromIDListEx if
	  available instead of old SHGetPathFromIDListW, to check the
	  buffer size.

Thu Jun  2 17:05:19 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ruby.c (ruby_init_loadpath_safe): remove MAXPATHLEN restriction
	  for Windows 10.

Thu Jun  2 16:51:35 2016  Koichi ITO  <koic.ito@gmail.com>

	* misc/ruby-mode.el (ruby-here-doc-beg-re),
	  (ruby-here-doc-beg-match, ruby-parse-partial): Support for
	  `squiggly heredoc' syntax in ruby-mode.  [Fix GH-1372]

Thu Jun  2 10:24:48 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* string.c: Raise ArgumentError when invalid string is detected in
	  case mapping methods.

	* enc/unicode.c: Check for invalid string and signal with negative
	  length value.

	* test/ruby/enc/test_case_mapping.rb: Add tests for above.

	* test/ruby/test_m17n_comb.rb: Add a message to clarify test failure.

Wed Jun  1 21:41:05 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/extconf.rb: Check existence of ASN1_TIME_adj(). The old
	  ASN1_TIME_set() is not Year 2038 ready on sizeof(time_t) == 4
	  environment. This function was added in OpenSSL 1.0.0.
	  [ruby-core:45552] [Bug #6571]

	* ext/openssl/ossl_asn1.c (ossl_time_split): Added. Split the argument
	  (Time) into the number of days elapsed since the epoch and the
	  remainder seconds to conform to ASN1_TIME_adj().
	  (obj_to_asn1utime, obj_to_asn1gtime): Use ossl_time_split() and
	  ASN1_*TIME_adj().

	* ext/openssl/ossl_asn1.h: Add the function prototype for
	  ossl_time_split().

	* ext/openssl/ossl_x509.[ch]: Add ossl_x509_time_adjust(). Similarly to
	  obj_to_asn1*time(), use X509_time_adj_ex() instead of X509_time_adj().

	* ext/openssl/ossl_x509cert.c, ext/openssl/ossl_x509crl.c,
	  ext/openssl/ossl_x509revoked.c: Use ossl_x509_time_adjust().

Wed Jun  1 15:58:20 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* configure.in: revert r55237.  replace crypt, not crypt_r, and
	  check if crypt is broken more.

	* missing/crypt.c: move crypt_r.c

	* string.c (rb_str_crypt): use crypt_r if provided by the system.

Wed Jun  1 14:07:53 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* missing/crypt_r.c (a64toi): initialize statically and fix out of
	  bounds access when salt is not 7bit clean.

Wed Jun  1 11:34:59 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* win32/Makefile.sub (MISSING): fixed build error introduced at r55237.

Wed Jun  1 09:48:06 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* string.c (rb_str_crypt): use reentrant crypt_r.

Wed Jun  1 09:37:26 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* missing/crypt.c (des_setkey): void function never returns any value.

Wed Jun  1 09:16:22 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* crypt.h: separate header file from missing/crypt.c.

	* missing/crypt.c (crypt_r, setkey_r, encrypt_r): add reentrant
	  versions.

Wed Jun  1 02:25:38 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* missing/crypt.c: fix size macros to use configured values
	  for platforms long is larger than 32bit.
	  [ruby-core:75792] [Bug #12446]

Tue May 31 17:28:46 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* lib/optparse.rb (OptionParser::Completion.candidate): get rid of
	  nil as key names.  [ruby-core:75773] [Bug #12438]

	* lib/optparse.rb (OptionParser#make_switch): char class option
	  cannot be NoArgument, default to RequiredArgument.

Tue May 31 00:30:11 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* ext/socket/raddrinfo.c (host_str, port_str): Use StringValueCStr
	  instead of (Safe)StringValue, to detect NUL byte in the string.

Mon May 30 22:02:01 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl_asn1.c (time_to_time_t): Use NUM2TIMET() instead of
	  NUM2LONG(). time_t may be larger than long.
	  [ruby-core:45552] [Bug #6571]

Mon May 30 21:15:37 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* string.c: Document current behavior for other case mapping methods
	  on String. [ci skip]

Mon May 30 20:00:25 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* string.c: Document current situation for String#downcase. [ci skip]

Mon May 30 18:29:28 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl_ssl.c (ossl_sslctx_s_alloc): Enable the automatic
	  curve selection for ECDH by calling SSL_CTX_set_ecdh_auto(). With
	  this a TLS server automatically selects a curve which both the client
	  and the server support to use in ECDH. This changes the default
	  behavior but users can still disable ECDH by excluding 'ECDH' cipher
	  suites from the cipher list (with SSLContext#ciphers=). This commit
	  also deprecate #tmp_ecdh_callback=. It was added in Ruby 2.3.0. It
	  wraps SSL_CTX_set_tmp_ecdh_callback() which will be removed in OpenSSL
	  1.1.0. Its callback receives two values 'is_export' and 'keylength'
	  but both are completely useless for determining a curve to use in
	  ECDH. The automatic curve selection was introduced to replace this.

	  (ossl_sslctx_setup): Deprecate SSLContext#tmp_ecdh_callback=. Emit a
	  warning if this is in use.

	  (ossl_sslctx_set_ecdh_curves): Add SSLContext#ecdh_curves=. Wrap
	  SSL_CTX_set1_curves_list(). If it is not available, this falls back
	  to SSL_CTX_set_tmp_ecdh().

	  (Init_ossl_ssl): Define SSLContext#ecdh_curves=.

	* ext/openssl/extconf.rb: Check the existence of EC_curve_nist2nid(),
	  SSL_CTX_set1_curves_list(), SSL_CTX_set_ecdh_auto() and
	  SSL_CTX_set_tmp_ecdh_callback().

	* ext/openssl/openssl_missing.[ch]: Implement EC_curve_nist2nid() if
	  missing.

	* test/openssl/test_pair.rb (test_ecdh_callback): Use
	  EnvUtil.suppress_warning to suppress deprecated warning.

	  (test_ecdh_curves): Test that SSLContext#ecdh_curves= works.

	* test/openssl/utils.rb (start_server): Use SSLContext#ecdh_curves=.

Mon May 30 16:28:53 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ext/socket/raddrinfo.c (host_str, port_str): use RSTRING_LEN
	  instead of strlen, since RSTRING_PTR StringValueCStr may not be
	  NUL-terminated when SHARABLE_MIDDLE_SUBSTRING=1.  reported by
	  @tmtms, http://twitter.com/tmtms/status/736910516229005312

Mon May 30 16:20:26 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* string.c (str_fill_term): return new pointer reallocated by
	  filling terminator.

Mon May 30 14:54:58 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ext/stringio/stringio.c (enc_subseq): share the return value and
	  the buffer as possible.

Mon May 30 14:50:25 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* string.c (str_substr, rb_str_aref): refactor not to create
	  unnecessary empty string.

	* string.c (str_byte_substr, str_byte_aref): ditto.

Mon May 30 00:09:37 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* ext/-test-/auto_ext.rb: fixed a heedless bug introduced at r55198.
	  this change will make RubyCI green.

Sun May 29 22:58:19 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* regexec.c (ONIGENC_IS_MBC_ASCII_WORD): redefine optimized one.
	  WORD of Ruby's ascii compatible encoding is always [a-zA-Z0-9_].

Sun May 29 22:44:19 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* regexec.c (match_at): make compilers optimize harder.

Sun May 29 12:08:42 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ext/-test-/auto_ext.rb (auto_ext): utility method to create
	  extension libraries for tests.

Sat May 28 20:40:57 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* misc/ruby-additional.el (ruby-insert-heredoc-code-block): insert
	  here document code block for assert_separately mainly.

Sat May 28 20:34:19 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* test/test_unicode_normalize.rb: Add test to check for availability of
	  Unicode data file; refactoring; fix an error with tests for destructive
	  method (unicode_normalize!).

Sat May 28 19:08:36 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* test/ruby/enc/test_case_comprehensive.rb: Add error messages to tests
	  for data file availability; refactoring.

Sat May 28 14:00:10 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/lib/openssl/ssl.rb (SSLSocket): Move the implementation of
	  SSLSocket#initialize to C. Initialize the SSL (OpenSSL object) in it.
	  Currently this is delayed until ossl_ssl_setup(), which is called from
	  SSLSocket#accept or #connect. Say we call SSLSocket#hostname= with an
	  illegal value. We expect an exception to be raised in #hostname= but
	  actually we get it in the later SSLSocket#connect. Because the SSL is
	  not ready at #hostname=, the actual call of SSL_set_tlsext_host_name()
	  is also delayed.
	  This also fixes: [ruby-dev:49376] [Bug #11724]

	* ext/openssl/ossl_ssl.c (ossl_ssl_initialize): Added. Almost the same
	  as the Ruby version but this instantiate the SSL object at the same
	  time.

	  (ossl_ssl_setup): Adjust to the changes. Just set the underlying IO to
	  the SSL.

	  (ssl_started): Added. Make use of SSL_get_fd(). This returns -1 if not
	  yet set by SSL_set_fd().

	  (ossl_ssl_data_get_struct): Removed. Now GetSSL() checks that the SSL
	  exists.

	  (ossl_ssl_set_session): Don't call ossl_ssl_setup() here as now the
	  SSL is already instantiated in #initialize.

	  (ossl_ssl_shutdown, ossl_start_ssl, ossl_ssl_read_internal,
	   ossl_ssl_write_internal, ossl_ssl_stop, ossl_ssl_get_cert,
	   ossl_ssl_get_peer_cert, ossl_ssl_get_peer_cert_chain,
	   ossl_ssl_get_version, ossl_ssl_get_cipher, ossl_ssl_get_state,
	   ossl_ssl_pending, ossl_ssl_session_reused,
	   ossl_ssl_get_verify_result, ossl_ssl_get_client_ca_list,
	   ossl_ssl_npn_protocol, ossl_ssl_alpn_protocol, ossl_ssl_tmp_key): Use
	  GetSSL() instead of ossl_ssl_data_get_struct(). Use ssl_started().

	  (Init_ossl_ssl): Add method declarations of SSLSocket#{initialize,
	  hostname=}.

	* ext/openssl/ossl_ssl.h (GetSSL): Check that the SSL is not NULL. It
	  should not be NULL because we now set it in #initialize.

	* ext/openssl/ossl_ssl_session.c (ossl_ssl_session_initialize): No need
	  to check if the SSL is NULL.

Sat May 28 10:47:40 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* gems/bundled_gems: Update latest releases, power_assert-0.3.0,
	  test-unit 3.1.9, minitest 5.9.0, did_you_mean 1.0.1

Sat May 28 10:45:40 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* addr2line.c: drop support for ATARI ST platform. It was
	  discontinued more than two decades ago. [fix GH-1350] Patch by
	  @cremno
	* include/ruby/ruby.h: ditto.
	* io.c: ditto.

Sat May 28 10:39:47 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* test/ruby/enc/test_case_comprehensive.rb: Converted exception for
	  unavailable Unicode data files to failed assertion.

Sat May 28 10:26:18 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* lib/cgi/util.rb: added missing quote.
	  [fix GH-1363][ci skip] Patch by @dwaller

Fri May 27 17:38:49 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* variable.c (rb_local_constants_i): exclude hidden constants.
	  [ruby-core:75575] [Bug #12389]

Fri May 27 17:09:44 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* transcode.c (str_transcode0): scrub in the given encoding when
	  the source encoding is given, not in the encoding of the
	  receiver.  [ruby-core:75732] [Bug #12431]

Fri May 27 15:07:32 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* include/ruby/ruby.h (rb_scan_args): remove nul padding which
	  caused syntax error if fmt is not a string literal.

	* include/ruby/ruby.h (rb_scan_args_verify): suppress array-bounds
	  warnings by old clang.

	* include/ruby/ruby.h (rb_scan_args0): make extractor macros
	  inline functions, which do not validate the format and are
	  unnecessary to be expanded.

Fri May 27 01:00:36 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* symbol.c (is_identchar): use ISDIGIT instead of rb_enc_isalnum.
	  Though rb_enc_isalnum is encoding aware function, its argument here
	  is *m, which is a single byte. Therefore ISDIGIT is faster.

	* symbol.c (is_special_global_name): ditto.

	* symbol.c (rb_enc_symname_type): ditto.

Fri May 27 00:39:40 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* include/ruby/ruby.h (rb_scan_args): add nul padding here to
	  apply to all references.

	* include/ruby/ruby.h (rb_scan_args_verify): move length mismatch
	  check outside conditional operators.

Thu May 26 14:21:10 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl_pkey_dh.c (ossl_dh_compute_key): Check that the DH
	  has 'p' (the prime) before calling DH_size(). We can create a DH with
	  no parameter but DH_size() does not check and dereferences NULL.
	  [ruby-core:75720] [Bug #12428]

	* ext/openssl/ossl_pkey_dsa.c (ossl_dsa_sign): Ditto. DSA_size() does
	  not check dsa->q.

	* ext/openssl/ossl_pkey_rsa.c (ossl_rsa_public_encrypt,
	  ossl_rsa_public_decrypt, ossl_rsa_private_encrypt,
	  ossl_rsa_private_decrypt): Ditto. RSA_size() does not check rsa->n.

Thu May 26 14:13:52 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* include/ruby/ruby.h (rb_scan_args_count): verify length with
	  counting variables together.

Thu May 26 09:45:41 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* test/ruby/enc/test_case_comprehensive.rb: Add set of comprehensive
	  (across most Unicode characters; later across most character encodings)
	  tests for case mapping.

Thu May 26 05:00:13 2016  Benoit Daloze  <eregontp@gmail.com>

	* class.c (rb_define_class): Fix documentation.

Wed May 25 20:50:12 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* re.c (unescape_nonascii): scan hex up to only 3 characters.
	  [Bug #12420] [Bug #12423]

Wed May 25 19:07:19 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode.c: Handle DOTLESS_i by hand because it isn't involved in folding.

Wed May 25 18:30:53 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* regparse.c (fetch_token_in_cc): raise error if given octal escaped
	  character is too big. [Bug #12420] [Bug #12423]

Wed May 25 17:45:15 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl, test/openssl: Drop OpenSSL < 0.9.8 support.

Wed May 25 17:43:30 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/openssl_missing.h, ext/openssl/ossl.h: Remove
	  unnecessary 'extern "C"' blocks. We don't use C++ and these headers
	  are local to ext/openssl, so there is no need to enclose with it.

Wed May 25 17:42:58 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/extconf.rb: Remove check of OPENSSL_FIPS macro. This is
	  unneeded because we can check the macro directly in source code,
	  just as we already do for OPENSSL_NO_* macros.

	* ext/openssl/ossl.c: Replace occurrences of HAVE_OPENSSL_FIPS with
	  OPENSSL_FIPS.

Wed May 25 17:13:35 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* class.c (rb_scan_args): merge code for n_trail.

Wed May 25 17:11:34 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* include/ruby/ruby.h (rb_scan_args_validate): move failed
	  condition to the terminal.  [ruby-core:75714] [Bug #12426]

Wed May 25 13:13:37 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* regcomp.c: remove condition for debug output because prelude
	  doesn't use regexp now.

Wed May 25 13:10:30 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* regcomp.c (compile_length_tree): return error code immediately
	  if compile_length_tree raised error [Bug #12418]

Wed May 25 08:01:39 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode.c: Fix flag error for switch from titlecase to lowercase.

	* test/ruby/enc/test_case_mapping.rb: Tests for above error.

Wed May 25 01:13:55 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl_pkey_ec.c (ec_key_new_from_group): Create a new
	  EC_KEY on given EC group. Extracted from ossl_ec_key_initialize().
	  (ossl_ec_key_s_generate): Added. Create a new EC instance and
	  generate a random private and public key.
	  (ossl_ec_key_initialize): Use ec_key_new_from_group().
	  (Init_ossl_ec): Define the new method EC.generate. This change is
	  for consistency with other PKey types.  [ruby-core:45541] [Bug #6567]

	* test/openssl/test_pkey_ec.rb: Test that EC.generate works.

Wed May 25 00:37:16 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl_pkey_ec.c (ossl_ec_key_generate_key): Fix up RDoc.
	  (Init_ossl_ec): Rename EC#generate_key to EC#generate_key!. Make the
	  old name an alias of #generate_key!. This change is for consistency
	  with other PKey types.  [ruby-core:45541] [Bug #6567]

	* test/openssl/test_pkey_ec.rb: Use EC#generate_key! instead of
	  EC#generate_key.

Wed May 25 00:23:05 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* include/ruby/ruby.h (rb_scan_args_set): check the arity after
	  adjusting argc for an option hash, for optimization in simpler
	  cases.

Wed May 25 00:21:52 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* configure.in (XCFLAGS): merge flags only for ruby itself from
	  ruby_cflags.

Tue May 24 22:04:15 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl_cipher.c (ossl_cipher_set_key, ossl_cipher_set_iv):
	  Reject too long values as well as too short ones. Currently they
	  just truncate the input but this would hide bugs and lead to
	  unexpected encryption/decryption results.

	* test/openssl/test_cipher.rb: Test that Cipher#key= and #iv= reject
	  Strings with invalid length.

Tue May 24 21:32:21 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl_x509ext.c (ossl_x509ext_set_value): Use
	  ASN1_OCTET_STRING_set() instead of M_ASN1_OCTET_STRING_set(). Macros
	  prefixed by "M_" are discouraged to be used from outside OpenSSL
	  library[1].
	  (ossl_x509ext_get_value): Likewise, use ASN1_STRING_print() instead
	  of M_ASN1_OCTET_STRING_print().
	  [1] https://git.openssl.org/gitweb/?p=openssl.git;a=blob;f=CHANGES;h=bf61913d7b01212b4d8b2f3c13d71d645914f67c;hb=b6079a7835f61daa9fb2cbf9addfa86049523933#l878

	* ext/openssl/ossl.h: Include openssl/asn1.h instead of
	  openssl/asn1_mac.h. It just includes openssl/asn1.h and defines some
	  additional "M_" macros.

Tue May 24 18:52:11 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* include/ruby/ruby.h (rb_scan_args_verify): verify the format to
	  scan if no invalid chars and variable argument length matching,
	  at the compile time if possible.

Tue May 24 17:18:46 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* include/ruby/defines.h (ERRORFUNC, WARNINGFUNC): add fallback
	  definitions.

Tue May 24 16:37:43 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* configure.in (ruby_cflags): separate from optflags [Bug #12409]
	  -fexcess-precision=standard and -fp-model precise are set to this now.

	* configure.in (cflags): use ruby_cflags.

Tue May 24 16:20:53 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* configure.in (ERRORFUNC, WARNINGFUNC): __error__ and __warning__
	  attributes take a parenthesized string literal.

Tue May 24 12:35:56 2016  URABE Shyouhei  <shyouhei@ruby-lang.org>

	* common.mk: sort lines, and add missing dependencies suggested
	  by tool/update-deps

Mon May 23 21:33:36 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* include/ruby/missing.h (isfinite): move from numeric.c.

Mon May 23 21:09:06 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl.c (Init_openssl): Avoid reference to unset global
	  variable. ossl_raise() may be called before dOSSL is set. Since
	  global variables default to 0 and the default value of dOSSL set in
	  Init_openssl() is also Qfalse, there is no real issue but confusing.
	  Patch by Bertram Scharpf <software@bertram-scharpf.de>
	  [ruby-core:58264] [Bug #9101]

Mon May 23 20:32:16 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl_asn1.c, ext/openssl/ossl_bn.c,
	  ext/openssl/ossl_cipher.c, ext/openssl/ossl_digest.c
	  ext/openssl/ossl_engine.c, ext/openssl/ossl_ns_spki.c
	  ext/openssl/ossl_pkcs12.c, ext/openssl/ossl_pkcs7.c
	  ext/openssl/ossl_pkey.c, ext/openssl/ossl_pkey_ec.c
	  ext/openssl/ossl_rand.c, ext/openssl/ossl_ssl.c
	  ext/openssl/ossl_x509attr.c, ext/openssl/ossl_x509cert.c
	  ext/openssl/ossl_x509ext.c, ext/openssl/ossl_x509store.c: Use
	  StringValueCStr() where NUL-terminated string is expected.

Mon May 23 20:20:12 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* ext/bigdecimal/bigdecimal.c (isfinite): get rid of a warning on
	  cygwin. [Bug #12417][ruby-core:75691]

Mon May 23 19:41:27 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl_rand.c (ossl_rand_egd, ossl_rand_egd_bytes):
	  RAND_egd{_bytes,}() return -1 on failure, not 0.
	  Patch by cremno phobia <cremno@mail.ru>
	  [ruby-core:63795] [Bug #10053]
	  (ossl_pseudo_bytes): Similar, RAND_pseudo_bytes() may return 0 or
	  -1 on failure.

Mon May 23 15:52:07 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* ext/bigdecimal/bigdecimal.c (isfinite): isfinite does not always
	  exist.  fixed build error on Windows introduced at r55123.

Mon May 23 13:19:41 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* include/ruby/ruby.h (rb_scan_args0): make compile error if the
	  format is wrong or does not match with the variable argument
	  length if possible.

Mon May 23 12:47:09 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* include/ruby/ruby.h (rb_scan_args0): raise fatal error if
	  variable argument length does not match, it is a bug in the code
	  which uses rb_scan_args, not a runtime error.

Mon May 23 12:30:29 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ext/bigdecimal/bigdecimal.c (GetVpValueWithPrec): consider
	  non-finite float values not to raise FloatDomainError.
	  [ruby-core:75682] [Bug #12414]

Mon May 23 12:21:18 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* array.c (rb_ary_fill): suppress warnings: 'item' may be used
	  uninitialized in this function

Mon May 23 07:41:49 2016  Eric Wong  <e@80x24.org>

	* dir.c (dir_close): update RDoc for 2.3 #close change
	  [ruby-core:75679] [Bug #12413]

Sun May 22 20:01:21 2016  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>

	* lib/drb/timeridconv.rb: use finalizer trick instead of thread.

	* test/drb/ut_timerholder.rb: ditto.

Sun May 22 17:25:18 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* test/ruby/enc/test_case_options.rb: adjust test class name
	  to match file name

Sun May 22 17:24:07 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* rename test/ruby/enc/test_casing_options.rb to test_case_options.rb
	  for consistency

Sun May 22 17:06:55 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* transcode.c (enc_arg, str_transcode_enc_args, econv_args):
	  remove volatile, and add GC guards in callers.
	  [ruby-core:75664] [Bug #12411]

Sun May 22 16:27:00 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ext/-test-/integer/core_ext.c: move testutil/integer.c.

	* test/lib/-test-/integer.rb: extract implementation details from
	  test/unit/assertions.rb.  [Bug #12408]

Sun May 22 14:57:43 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* include/ruby/oniguruma.h: Extend OnigEncodingTypeDefine to define a
	  new encoding primitive 'case_map' for case mapping

	* enc/utf-8.c, utf_16be/le.c, utf_32be/le.c:
	  add onigenc_unicode_case_map as case_map primitive

	* enc/ascii.c, big5.c, cp949.c, emacs_mule.c, euc_jp/kr/tw.c, gb18030.c,
	  gbk.c, iso_8859_1/2/3/4/5/6/7/8/9/10/11/13/14/15/16.c, koi8_r/u.c,
	  shift_jis.c, us_ascii.c, windows_1250/1251/1252.c:
	  add onigenc_not_support_case_map as case_map primitive

Sun May 22 14:45:45 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* regenc.h/c: Define new function onigenc_not_support_case_map

Sun May 22 12:14:06 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* include/ruby/ruby.h (rb_scan_args): use original rb_scan_args
	  when fmt is dynamic.

Sun May 22 11:41:12 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* class.c (rb_scan_args): moved to bottom of the file to make the
	  effect of `#undef rb_scan_args` the minimum.

	* include/ruby/ruby.h (rb_scan_args): overwrite only if GCC and
	  optimized. Visual C++ 14 or later can compile it but make it
	  conservative.

Sat May 21 22:45:50 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* include/ruby/ruby.h (rb_scan_args): don't use ALWAYS_INLINE with
	  `inline`.  if gcc needs this duplication, do in ALWAYS_INLINE macro.

Sat May 21 21:11:56 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* include/ruby/ruby.h (rb_scan_args): use __VA_ARGS__ instead of
	  va_arg to allow compilers optimize more aggressive.
	  https://gustedt.wordpress.com/2011/07/10/avoid-writing-va_arg-functions/
	  rb_scan_args is now expected to be statically resolved.

Sun May 22 02:41:52 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* ext/zlib/zlib.c: remove hacky macro introduced at r30437.

	* ext/zlib/zlib.c (gzfile_make_header): cast as long (instead of int).

	* ext/zlib/zlib.c (gzfile_make_footer): ditto.

Sat May 21 21:07:18 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* configure.in (ALWAYS_INLINE): force compilers the function inlined.

Sat May 21 16:16:03 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl_ssl.c (ossl_ssl_stop): Don't free the SSL struct
	  here. Since some methods such as SSLSocket#connect releases GVL,
	  there is a chance of use after free if we free the SSL from another
	  thread. SSLSocket#stop was documented as "prepares it for another
	  connection" so this is a slightly incompatible change. However when
	  this sentence was added (r30090, Add toplevel documentation for
	  OpenSSL, 2010-12-06), it didn't actually. The current behavior is
	  from r40304 (Correct shutdown behavior w.r.t GC., 2013-04-15).
	  [ruby-core:74978] [Bug #12292]

	* ext/openssl/lib/openssl/ssl.rb (sysclose): Update doc.

	* test/openssl/test_ssl.rb: Test this.

Sat May 21 14:41:14 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl.c: [DOC] Fix SSL client example. The variable name
	  was wrong. Patch by Andreas Tiefenthaler <at@an-ti.eu> (@pxlpnk).
	  [GH ruby/openssl#32]

Sat May 21 14:25:38 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl_pkey_ec.c: rename PKey::EC#private_key? and
	  #public_key? to #private? and #public? for consistency with other
	  PKey types. Old names remain as alias.  [ruby-core:45541] [Bug #6567]

	* test/openssl/test_pkey_ec.rb (test_check_key): check private? and
	  public? works correctly.

Sat May 21 12:40:36 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/lib/openssl/buffering.rb (read_nonblock, readpartial):
	  Remove impossible EOFError raise. Patch by Zach Anker
	  <zanker@squareup.com>.  [GH ruby/openssl#23]

Sat May 21 11:18:42 2016  Evgeni Golov  <evgeni@golov.de>

	* lib/ipaddr.rb: [DOC] fix documentation of IN6MASK to mention
	  IPv6.  [Fix GH-1349]

Sat May 21 11:12:53 2016  Dan Martinez  <dfm@razorwind.org>

	* io.c (Init_IO): [DOC] define dummy ARGF instead of ARGF.class to
	  re-enable the generation of ARGF documentation.  [Fix GH-1358]

Sat May 21 11:07:29 2016  0x01f7  <souk.0x01f7@gmail.com>

	* doc/syntax/methods.rdoc (Method Names): add proper closing tag.
	  [Fix GH-1356]

Sat May 21 09:26:28 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* configure.in (RUBY_CHECK_SETJMP): fix missing macro definition
	  for the configured result.  fix up r55021.

Sat May 21 00:36:32 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* process.c (rb_execarg_commandline): build command line string
	  from argument vector in rb_execarg.
	  [ruby-core:75611] [Bug #12398]

Fri May 20 23:25:42 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl.c (ossl_pem_passwd_value): Added. Convert the
	  argument to String with StringValue() and validate the length is in
	  4..PEM_BUFSIZE. PEM_BUFSIZE is a macro defined in OpenSSL headers.
	  (ossl_pem_passwd_cb): When reading/writing encrypted PEM format, we
	  used to pass the password to PEM_def_callback() directly but it was
	  problematic. It is not NUL character safe. And surprisingly, it
	  silently truncates the password to 1024 bytes.  [GH ruby/openssl#51]

	* ext/openssl/ossl.h: Add function prototype declaration of newly
	  added ossl_pem_passwd_value().

	* ext/openssl/ossl_pkey.c (ossl_pkey_new_from_data): Use
	  ossl_pem_passwd_value() to validate the password String.

	* ext/openssl/ossl_pkey_dsa.c (ossl_dsa_initialize, ossl_dsa_export):
	  ditto.

	* ext/openssl/ossl_pkey_ec.c (ossl_ec_key_initialize,
	  ossl_ec_key_to_string): ditto.

	* ext/openssl/ossl_pkey_rsa.c (ossl_rsa_initialize, ossl_rsa_export):
	  ditto.

	* test/openssl/test_pkey_{dsa,ec,rsa}.rb: test this.

Fri May 20 23:45:53 2016  Naohisa Goto  <ngotogenome@gmail.com>

	* id_table.c (list_id_table_init): When unaligned word access is
	  prohibited and sizeof(VALUE) is 8 (64-bit machines),
	  capa should always be even number for 8-byte word alignment
	  of the values of a table. This code assumes that sizeof(ID) is 4,
	  sizeof(VALUE) is 8, and xmalloc() returns 8-byte aligned memory.
	  This fixes bus error on 64-bit SPARC Solaris 10.
	  [Bug #12406][ruby-dev:49631]

Fri May 20 22:30:09 2016  Naohisa Goto  <ngotogenome@gmail.com>

	* symbol.h (rb_id2sym): Use HAVE_BUILTIN___BUILTIN_CONSTANT_P

Fri May 20 22:19:00 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>

	* ext/date/date_core.c: [DOC] fix typos.
	  [fix GH-1360] patched by @soundasleep

Fri May 20 21:26:58 2016  Naohisa Goto  <ngotogenome@gmail.com>

	* include/ruby/defines.h (RB_GNUC_EXTENSION, RB_GNUC_EXTENSION_BLOCK):
	  macros for skipping __extension__ on non-GCC compilers.
	* eval_error.c (warn_print): use RB_GNUC_EXTENSION_BLOCK instead of
	  __extension__ because __extension__ is a GNU extension.
	  Fix compile error on Solaris 10 with Oracle Solaris Studio 12.x.
	  [Bug #12397] [ruby-dev:49629].
	* internal.h (rb_fstring_cstr, rb_fstring_enc_cstr): ditto
	* include/ruby/encoding.h (rb_enc_str_new, rb_enc_str_new_cstr): ditto
	* include/ruby/intern.h (rb_str_new, rb_str_new_cstr,
	  rb_usascii_str_new, rb_utf8_str_new, rb_tainted_str_new_cstr,
	  rb_usascii_str_new_cstr, rb_utf8_str_new_cstr,
	  rb_external_str_new_cstr, rb_locale_str_new_cstr,
	  rb_str_buf_new_cstr, rb_str_cat_cstr, rb_exc_new_cstr): ditto

Fri May 20 21:17:13 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ext/win32ole/win32ole.c (fole_missing): make substring or dup to
	  share the content if possible.

Fri May 20 19:48:48 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* internal.h (NEW_PARTIAL_MEMO_FOR): shrink buffer array not to
	  mark non-VALUE fields.  fix check_rvalue_consistency abort with
	  RGENGC_CHECK_MODE=2.

	* internal.h (NEW_CMP_OPT_MEMO): exclude struct cmp_opt_data from
	  the valid array range.

	* enum.c (slicewhen_i): exclude inverted too.

Thu May 19 21:21:57 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>

	* re.c (rb_reg_match_m_p): [DOC] fix return value in rdoc.

	* test/ruby/test_regexp.rb (TestRegexp#test_match_p): add some
	  tests from document.

Thu May 19 13:22:44 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl.c (Init_openssl): register an ex_data index for
	  X509_STORE and X509_STORE_CTX respectively. Since they don't share
	  the ex_data index registry, we can't use the same index.
	  (ossl_verify_cb): use the the correct index.

	* ext/openssl/ossl_ssl.c (ossl_ssl_verify_callback): ditto.

	* ext/openssl/ossl_x509store.c (ossl_x509store_set_vfy_cb): ditto.
	  (ossl_x509stctx_verify): ditto.

	* ext/openssl/ossl.h (void ossl_clear_error): add extern declarations
	  of ossl_store_{ctx_,}ex_verify_cb_idx.

	* ext/openssl/openssl_missing.c: remove X509_STORE_set_ex_data and
	  X509_STORE_get_ex_data.

	* ext/openssl/openssl_missing.h: implement X509_STORE_get_ex_data,
	  X509_STORE_set_ex_data and X509_STORE_get_ex_new_index as macros.

Thu May 19 13:11:35 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl_x509attr.c (ossl_x509attr_set_value): check that the
	  argument is an OpenSSL::ASN1::Data before converting to ASN1_TYPE.
	  This fixes SEGV on OpenSSL::X509::Attribute#value=(non-asn1-value).

	* test/openssl/test_x509attr.rb: add tests for OpenSSL::X509::Attribute.

Thu May 19 12:10:10 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* re.c (rb_reg_match_m_p): fix match against empty string.
	  rb_str_offset returns the end when the position exceeds the
	  length.  fix the range parameter of onig_search.
	  [ruby-core:75604] [Bug #12394]

Thu May 19 11:37:36 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* re.c (rb_reg_match_m_p): should return false if no match, as the
	  document says.  [Feature #8110]

Thu May 19 00:17:01 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* re.c (reg_names_iter): specify capacify

Wed May 18 21:29:59 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* thread.c (recursive_list_access): a object id may be a Bignum.  so,
	  the list must be a objhash, instead of a identhash.
	  this fixes many test errors on mswin64 CI.

Wed May 18 19:33:54 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* re.c (rb_reg_match_m_p): Introduce Regexp#match?, which returns
	  bool and doesn't save backref.

Wed May 18 16:52:03 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl_pkcs12.c (ossl_pkcs12_initialize): pop errors
	  leaked by PKCS12_parse(). This is a bug in OpenSSL, which exists
	  in the versions before the version 1.0.0t, 1.0.1p, 1.0.2d.

Wed May 18 16:04:54 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* tool/downloader.rb (Downloader::RubyGems.download): verify gems
	  only if RubyGems is 2.4 or later.  old RubyGems fails to verify
	  almost all of bundled gems.

Wed May 18 14:52:38 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* string.c (rb_str_modify_expand): check integer overflow.
	  [ruby-core:75592] [Bug #12390]

Wed May 18 13:11:44 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* re.c (match_ary_subseq): get subseq of match array without creating
	  temporary array.

	* re.c (match_ary_aref): get element(s) of match array without creating
	  temporary array.

	* re.c (match_aref): Use match_ary_subseq with handling irregulars.

	* re.c (match_values_at): Use match_ary_aref.

Wed May 18 13:03:07 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl_x509cert.c (ossl_x509_verify): X509_verify()
	  family may put errors on 0 return (0 means verification failure).
	  Clear OpenSSL error queue before return to Ruby. Since the queue is
	  thread global, remaining errors in the queue can cause an unexpected
	  error in the next OpenSSL operation.  [ruby-core:48284] [Bug #7215]

	* ext/openssl/ossl_x509crl.c (ossl_x509crl_verify): ditto.

	* ext/openssl/ossl_x509req.c (ossl_x509req_verify): ditto.

	* ext/openssl/ossl_x509store.c (ossl_x509stctx_verify): ditto.

	* ext/openssl/ossl_pkey_dh.c (dh_generate): clear the OpenSSL error
	  queue before re-raising exception.

	* ext/openssl/ossl_pkey_dsa.c (dsa_generate): ditto.

	* ext/openssl/ossl_pkey_rsa.c (rsa_generate): ditto.

	* ext/openssl/ossl_ssl.c (ossl_start_ssl): ditto.

	* test/openssl: check that OpenSSL.errors is empty every time after
	  running a test case.

Wed May 18 12:07:42 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl.c (ossl_clear_error): Extracted from
	  ossl_make_error(). This prints errors in the OpenSSL error queue if
	  OpenSSL.debug is true, and clears the queue.
	  (ossl_make_error): use ossl_clear_error().

	* ext/openssl/ossl.h: add prototype declaration of ossl_make_error().
	  (OSSL_BIO_reset) use ossl_clear_error() to clear the queue. Clearing
	  silently makes debugging difficult.

	* ext/openssl/ossl_engine.c (ossl_engine_s_by_id): ditto.

	* ext/openssl/ossl_ns_spki.c (ossl_spki_initialize): ditto.

	* ext/openssl/ossl_pkcs7.c (ossl_pkcs7_verify): ditto.

	* ext/openssl/ossl_pkey_dsa.c (ossl_dsa_initialize): ditto.

	* ext/openssl/ossl_pkey_ec.c (ossl_ec_key_initialize): ditto.
	  (ossl_ec_group_initialize): ditto.

	* ext/openssl/ossl_ssl.c (ossl_ssl_shutdown): ditto.

Wed May 18 11:53:49 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl_pkey_ec.c (ossl_ec_point_mul): Validate the
	  arguments before passing to EC_POINT(s)_mul(). Add description of this
	  method.  [ruby-core:65152] [Bug #10268]

	* test/openssl/test_pkey_ec.rb (test_ec_point_mul): Test that
	  OpenSSL::PKey::EC::Point#mul works.

Wed May 18 11:19:59 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/ossl_bn.c (try_convert_to_bnptr): Extracted from
	  GetBNPtr(). This doesn't raise exception but returns NULL on error.
	  (GetBNPtr): Raise TypeError if conversion fails.
	  (ossl_bn_eq): Implement BN#==.
	  (ossl_bn_eql): #eql? should not raise TypeError even if the argument
	  is not compatible with BN.
	  (ossl_bn_hash): Implement BN#hash.

	* ext/openssl/ossl_bn.c (Init_ossl_bn): Define #== and #hash.

	* test/openssl/test_bn.rb: Test BN#eql?, #== and #hash

Wed May 18 10:17:41 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* include/ruby/ruby.h (RB_INTEGER_TYPE_P): new macro and
	  underlying inline function to check if the object is an
	  Integer (Fixnum or Bignum).

Wed May 18 09:52:00 2016  Kenta Murata  <mrkn@mrkn.jp>

	* enum.c (enum_sum, hash_sum, hash_sum_i, enum_sum_i, sum_iter):
	  Optimize for hashes when each method isn't redefined.

Wed May 18 09:14:00 2016  Kenta Murata  <mrkn@mrkn.jp>

	* enum.c (enum_sum, int_range_sum): Extract int_range_sum from
	  enum_sum.

Wed May 18 03:16:06 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* re.c (match_values_at): fix regression at r55036.
	  MatchData#values_at accepts Range.

Wed May 18 02:02:58 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* re.c (match_aref): remove useless condition and call rb_fix2int.
	  rb_reg_nth_match handles negative index.

Wed May 18 01:57:43 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* re.c (match_values_at): MatchData#values_at supports named captures
	  [Feature #9179]

	* re.c (namev_to_backref_number): separated.

Wed May 18 00:05:00 2016  Kenta Murata  <mrkn@mrkn.jp>

	* enum.c (enum_sum): Optimize for a range from int to int.

	* test/ruby/test_enum.rb (test_range_sum): Move from test_range.rb,
	  and add assertions for some conditions.

	* test/ruby/test_enum.rb (test_hash_sum): Move from test_hash.rb.

	* test/ruby/test_hash.rb, test/ruby/test_range.rb: Remove test_sum.

Tue May 17 23:08:00 2016  Kenta Murata  <mrkn@mrkn.jp>

	* enum.c (enum_sum): [DOC] Write documentation.

Tue May 17 22:53:00 2016  Kenta Murata  <mrkn@mrkn.jp>

	* enum.c (enum_sum): Implement Enumerable#sum.

	* test/ruby/test_enum.rb (test_sum): Test sum for Enumerable.

	* test/ruby/test_hash.rb (test_sum): Test sum for Hash.

	* test/ruby/test_range.rb (test_sum): Test sum for Range.

Tue May 17 22:11:41 2016  Tanaka Akira  <akr@fsij.org>

	* object.c, numeric.c, enum.c, ext/-test-/bignum/mul.c,
	  lib/rexml/quickpath.rb, lib/rexml/text.rb, lib/rexml/xpath_parser.rb,
	  lib/rubygems/specification.rb, lib/uri/generic.rb,
	  bootstraptest/test_eval.rb, basictest/test.rb,
	  test/-ext-/bignum/test_big2str.rb, test/-ext-/bignum/test_div.rb,
	  test/-ext-/bignum/test_mul.rb, test/-ext-/bignum/test_str2big.rb,
	  test/csv/test_data_converters.rb, test/date/test_date.rb,
	  test/json/test_json_generate.rb, test/minitest/test_minitest_mock.rb,
	  test/openssl/test_cipher.rb, test/rexml/test_jaxen.rb,
	  test/ruby/test_array.rb, test/ruby/test_basicinstructions.rb,
	  test/ruby/test_bignum.rb, test/ruby/test_case.rb,
	  test/ruby/test_class.rb, test/ruby/test_complex.rb,
	  test/ruby/test_enum.rb, test/ruby/test_eval.rb,
	  test/ruby/test_iseq.rb, test/ruby/test_literal.rb,
	  test/ruby/test_math.rb, test/ruby/test_module.rb,
	  test/ruby/test_numeric.rb, test/ruby/test_range.rb,
	  test/ruby/test_rational.rb, test/ruby/test_refinement.rb,
	  test/ruby/test_rubyvm.rb, test/ruby/test_struct.rb,
	  test/ruby/test_variable.rb, test/rubygems/test_gem_specification.rb,
	  test/thread/test_queue.rb: Use Integer instead of Fixnum and Bignum.

Tue May 17 15:26:10 2016  Tanaka Akira  <akr@fsij.org>

	* [Feature #12005] Unify Fixnum and Bignum into Integer

	* include/ruby/ruby.h (rb_class_of): Return rb_cInteger for fixnums.

	* insns.def (INTEGER_REDEFINED_OP_FLAG): Unified from
	  FIXNUM_REDEFINED_OP_FLAG and BIGNUM_REDEFINED_OP_FLAG.

	* vm_core.h: Ditto.

	* vm_insnhelper.c (opt_eq_func): Use INTEGER_REDEFINED_OP_FLAG instead
	  of FIXNUM_REDEFINED_OP_FLAG.

	* vm.c (vm_redefinition_check_flag): Use rb_cInteger instead of
	  rb_cFixnum and rb_cBignum.
	  (C): Use Integer instead of Fixnum and Bignum.

	* numeric.c (fix_succ): Removed.
	  (Init_Numeric): Define Fixnum as Integer.

	* bignum.c (bignew): Use rb_cInteger instead of rb_cBignum.
	  (rb_int_coerce): replaced from rb_big_coerce and return fixnums
	  as-is.
	  (Init_Bignum): Define Bignum as Integer.
	  Don't define ===.

	* error.c (builtin_class_name): Return "Integer" for fixnums.

	* sprintf.c (ruby__sfvextra): Use rb_cInteger instead of rb_cFixnum.

	* ext/-test-/testutil: New directory to test.
	  Currently it provides utilities for fixnum and bignum.

	* ext/json/generator/generator.c: Define mInteger_to_json.

	* lib/mathn.rb (Fixnum#/): Redefinition removed.

Tue May 17 11:58:58 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* configure.in (RUBY_CHECK_BUILTIN_SETJMP): declare t as NORETURN
	  to suppress warnings by -Wsuggest-attribute=noreturn.
	  [ruby-core:75510] [Bug #12383]

Tue May 17 10:40:18 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* configure.in (RUBY_CHECK_SETJMP): needs the header and proper
	  arguments for builtin setjmp functions.

Mon May 16 20:00:30 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode.h: Additional uses of ONIG_CASE_MAPPING compilation switch

Mon May 16 19:46:33 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* include/ruby/oniguruma.h: Introducing ONIG_CASE_MAPPING compilation
	  switch

	* include/ruby/oniguruma.h, enc/unicode.h: Using ONIG_CASE_MAPPING
	  compilation switch

Mon May 16 19:29:31 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* gems/bundled_gems: Update xmlrpc-0.1.1. xmlrpc-0.1.0 didn't allow
	  to install on 2.4.0dev.

Mon May 16 13:28:59 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* configure.in (FUNC_STDCALL, FUNC_CDECL, FUNC_FASTCALL): set
	  macro names explicitly to the old names, which are accidentally
	  changed at r54985, for backward compatibilities.
	  fiddle also depends on these names to fallback to ANSI names.
	  [ruby-core:75494] [Bug #12377]

Mon May 16 11:39:02 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* lib/xmlrpc.rb, lib/xmlrpc/*, test/xmlrpc: XMLRPC is bundled gem
	  on Ruby 2.4. It is extracted to https://github.com/ruby/xmlrpc
	  [Feature #12160][ruby-core:74239]
	* gems/bundled_gems: ditto.

Mon May 16 06:06:21 2016  Eric Wong  <e@80x24.org>

	* proc.c: fix RDoc of Proc#===/call/yield/[]
	  [Bug #12332]

Sun May 15 20:55:31 2016  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>

	* lib/drb/timeridconv.rb: don't use keeper thread. [Bug #12342]

	* test/drb/ut_timerholder.rb: ditto.

Sun May 15 16:15:25 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* array.c (rb_ary_entry): extract rb_ary_elt to organize if-conditions
	  and check whether is embedded at once.

Sun May 15 10:57:26 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* vm_insnhelper.c (vm_get_ev_const): warn deprecated constant even
	  in the class context.  [ruby-core:75505] [Bug #12382]

Sun May 15 03:13:01 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* iseq.h (struct iseq_compile_data): use struct rb_id_table
	  instead of st_table.

	* iseq.c (prepare_iseq_build): don't allocate ivar_cache_table
	  until it has at least one element.

	* iseq.c (compile_data_free): free ivar_cache_table only if it
	  is allocated.

	* compile.c (get_ivar_ic_value): allocate if the table is not
	  allocated yet.

Sat May 14 09:04:34 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* lib/mkmf.rb (pkg_config): use xsystem consistently to set up
	  library path environment variable as well as latter pkg-config
	  calls.  [ruby-dev:49619] [Bug #12379]

Sat May 14 00:16:54 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* random.c (make_seed_value): append leading-zero-guard and get
	  rid of making a local copy of the seed.

Fri May 13 08:46:42 2016  cremno  <cremno@mail.ru>

	* NEWS: drop FreeBSD < 4 support.  [Fix GH-1339]
	  The most recent version affected by this is 3.5 and was released
	  in 2000.
	  https://www.freebsd.org/releases/3.5R/announce.html
	  https://en.wikipedia.org/wiki/History_of_FreeBSD#Version_history

Fri May 13 03:12:09 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* include/ruby/defines.h (GCC_VERSION_SINCE): moved from internal.h.

Fri May 13 03:11:20 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* configure.in (__builtin_constant_p): check.

	* internal.h: Use HAVE_BUILTIN___BUILTIN_CONSTANT_P

Fri May 13 03:10:39 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* configure.in: use alternative keyword
	  to avoid macros conflicts with them.

Thu May 12 01:54:08 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* lib/mkmf.rb (try_func): get rid of conflict of declarations of
	  main().  checking local symbol reference does not make sense.

Thu May 12 00:18:19 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* win32/Makefile.sub (HAVE_QSORT_S): use qsort_s only for Visual Studio
	  2012 or later, because VS2010 seems to causes a SEGV in
	  test/ruby/test_enum.rb.

Wed May 11 23:59:47 2016  Masaya Tarui  <tarui@ruby-lang.org>

	* vm_insnhelper.c (vm_getivar): describe fast-path explicit
	  (compiler friendly). [Bug #12274].

Wed May 11 21:30:07 2016  Masaya Tarui  <tarui@ruby-lang.org>

	* compile.c (iseq_compile_each): share InlineCache during same
	  instance variable accesses. Reducing memory consumption,
	  rising cache hit rate and rising branch prediction hit rate
	  are expected. A part of [Bug #12274].

	* iseq.h (struct iseq_compile_data): introduce instance
	  variable IC table for sharing.

	* iseq.c (prepare_iseq_build, compile_data_free):
	  construct/destruct above table.

Wed May 11 17:18:53 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* util.c (ruby_qsort): use qsort_s if available, for Microsoft
	  Visual Studio 2005 (msvcr80.dll) and mingw.

Wed May 11 10:33:26 2016  Marcus Stollsteimer  <sto.mar@web.de>

	* COPYING: Remove trailing-whitespaces.
	  [ci skip][fix GH-1348]

Tue May 10 21:05:45 2016  Benoit Daloze  <eregontp@gmail.com>

	* insns.def (defineclass): Also raise an error when redeclaring the
	  superclass of a class as Object and it has another superclass.
	  [Bug #12367] [ruby-core:75446]

	* test/ruby/test_class.rb: test for above.

Tue May 10 14:57:09 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* random.c (obj_random_bytes): base on bytes method instead of
	  rand method, not to call toplevel rand method.

Tue May 10 13:07:28 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* configure.in (-fexcess-precision=standard): before r54895 -std=c99
	  is specified and it implied -fexcess-precision=standard.
	  Now with -std=gnu99, it should be explicitly specified.
	  https://gcc.gnu.org/onlinedocs/gcc/Optimize-Options.html

Mon May  9 10:51:59 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* thread.c (rb_thread_atfork_internal): move th to an argument.

	* thread.c (rb_thread_atfork): do not repeat GET_THREAD().

Mon May  9 10:46:36 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* thread.c (rb_thread_atfork, rb_thread_atfork_before_exec): do
	  nothing unless working fork is available.

	* thread_sync.c (rb_mutex_abandon_all): define only if working
	  fork is available.

	* thread_sync.c (rb_mutex_abandon_keeping_mutexes): ditto.

	* thread_sync.c (rb_mutex_abandon_locking_mutex): ditto.

	* thread_win32.c (gvl_init): never used.

Mon May  9 07:18:06 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* include/ruby/{defines,ruby}.h: need to define function attributes
	  alternatives in defines.h instead of ruby.h, because they are used
	  in oniguruma.h and the header used without including ruby.h at
	  encoding library sources.

Mon May  9 06:30:12 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* include/ruby/ruby.h (CONSTFUNC, PUREFUNC): fixed build errors on
	  non-gcc build environments introduced at r54952.

Mon May  9 02:51:51 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* gc.c (rb_gc_unprotect_logging): throw rb_memerror when it cannot
	  allocate memory. This is pointed out by Facebook's Infer.

	* gc.c (gc_prof_setup_new_record): ditto.

	* regparse.c (parse_regexp): ditto.

	* util.c (MALLOC): use xmalloc and xfree like above.

Mon May  9 02:39:16 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* configure.in: check function attribute const and pure,
	  and define CONSTFUNC and PUREFUNC if available.
	  Note that I don't add those options as default because
	  it still shows many false-positive (it seems not to consider
	  longjmp).

	* vm_eval.c (stack_check): get rb_thread_t* as an argument
	  to avoid duplicate call of GET_THREAD().

Sun May  8 21:01:14 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* ext/openssl/extconf.rb: assume it doesn't have SSLv2 related
	  functions when OPENSSL_NO_SSL2 is defined.
	  Usually openssl's header and the library (libssl) have the same
	  set of functions, but on some environment the library has functions
	  whose headers doesn't declare. (openssl/opensslconf.h and libssl.so
	  aren't be synchronized)
	  To detect such case explicitly check feature macro and remove
	  related functions.

Sun May  8 18:51:33 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* file.c (rb_home_dir_of): return the default home path if the
	  user name is the current user name, on platforms where struct
	  pwd is not supported.  a temporary measure against
	  [Bug #12226].

Sun May  8 08:51:38 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* configure.in: add -Wsuggest-attribute=format and suppress warnings.

Sun May  8 08:31:03 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* configure.in: add -Wsuggest-attribute=noreturn and suppress warnings.

Sun May  8 08:19:16 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* configure.in: add -Werror=implicit-int to avoid missing type of
	  function declaration.

Sat May  7 22:22:37 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* lib/webrick/ssl.rb: Accept string value for SSLCertName. It is used
	  to invoke ssl server with command line.
	  [fix GH-1329] Patch by @kerlin
	* test/webrick/test_ssl_server.rb: Added test for GH-1329

Sat May  7 21:55:12 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* test/webrick/test_ssl_server.rb: Added basic test for `webrick/ssl`

Sat May  7 16:22:13 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* random.c (int_pair_to_real_inclusive): optimize to multiply
	  without Bignum.

Sat May  7 07:58:02 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* process.c (rb_exec_getargs): honor the expected argument types
	  over the conversion method.  the basic language functionality
	  should be robust.  [ruby-core:75388] [Bug #12355]

Fri May  6 08:16:26 2016  David Silva  <david.silva@digital.cabinet-office.gov.uk>

	* enum.c (enum_find): [DOC] add more examples to the documentation
	  of Enumerable#detect, to show that it equals to Enumerable#find.
	  [Fix GH-1340]

Thu May  5 18:08:31 2016  Yuichiro Kaneko  <yui-knk@ruby-lang.org>

	* test/ruby/test_complexrational.rb: Remove duplicated raise.

Thu May  5 14:41:05 2016  Yuichiro Kaneko  <yui-knk@ruby-lang.org>

	* string.c (rb_str_sub): Fix a special match variable name.
	  [ci skip]

Thu May  5 12:22:17 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* vm_eval.c (rb_eval_cmd, rb_catch_obj): use TH_JUMP_TAG with the
	  same rb_thread_t used for TH_PUSH_TAG, instead of JUMP_TAG with
	  the current thread global variable.

Thu May  5 10:49:33 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* random.c (fill_random_bytes_syscall): use arc4random_buf if
	  available.

Wed May  4 23:13:58 2016  Yuichiro Kaneko  <yui-knk@ruby-lang.org>

	* numeric.c (fix_plus): Remove rb_nucomp_add prototype
	  declaration.

	* numeric.c (fix_mul): Remove rb_nucomp_mul prototype
	  declaration.

	* internal.h (rb_nucomp_add, rb_nucomp_mul): add prototype
	  declarations.

Wed May  4 18:38:00 2016  Kazuki Tsujimoto  <kazuki@callcc.net>

	* lib/net/http/header.rb (Net::HTTPHeader#{each_header,each_name,
	  each_capitalized_name,each_value,each_capitalized}): Return
	  sized enumerators.

	* test/net/http/test_httpheader.rb: add test for above.

Wed May  4 17:53:15 2016  Kazuki Tsujimoto  <kazuki@callcc.net>

	* lib/set.rb (Set#{delete_if,keep_if,collect!,reject!,select!,classify,divide},
	  SortedSet#{delete_if,keep_if}): Return sized enumerators.

	* test/test_set.rb: add test for above.

Tue May  3 23:25:48 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>

	* numeric.c: [DOC] Update result of 123456789 ** -2.
	  [ruby-dev:49606] [Bug #12339]

Tue May  3 23:13:16 2016  Yuichiro Kaneko  <yui-knk@ruby-lang.org>

	* internal.h (RCOMPLEX_SET_IMAG): undef RCOMPLEX_SET_IMAG
	  instead of duplicated undef RCOMPLEX_SET_REAL.

Tue May  3 22:55:07 2016  Yuichiro Kaneko  <yui-knk@ruby-lang.org>

	* complex.c (rb_complex_set_imag): Fix to properly set imag
	  of complex.

Tue May  3 22:19:55 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* configure.in (warnflags): use -std=gnu99 instead of
	  -std=iso9899:1999. [Feature #12336]

Tue May  3 22:10:09 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* string.c (count_utf8_lead_bytes_with_word): Use __builtin_popcount
	  only if it can use SSE 4.2 POPCNT whose latency is 3 cycle.

	* internal.h (rb_popcount64): use __builtin_popcountll because now
	  it is in fast path.

Tue May  3 14:19:18 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* parse.y (new_if_gen): set newline flag to NODE_IF to trace all
	  if/elsif statements.  [ruby-core:67720] [Bug #10763]

Tue May  3 05:35:54 2016  Eric Wong  <e@80x24.org>

	* process.c (disable_child_handler_fork_child):
	  initialize handler for SIGPIPE for !POSIX_SIGNAL

Mon May  2 23:03:42 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* win32/win32.c, include/ruby/win32.h (rb_w32_utruncate): implements new
	  truncate alternative which accepts UTF-8 path.

	* file.c (truncate): use above function.
	  [Bug #12340]

Mon May  2 20:59:21 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* re.c (str_coderange): to avoid function call when the string already
	  has coderange information.

	* re.c (rb_reg_prepare_enc): add shortcut path when the regexp has
	  the same encoding of given string.

	* re.c (rb_reg_prepare_re): avoid duplicated allocation of
	  onig_errmsg_buffer.


Mon May  2 12:34:52 2016  Tanaka Akira  <akr@fsij.org>

	* test/ruby/test_refinement.rb (test_inspect): Use Integer instead of
	  Fixnum.

Mon May  2 06:58:38 2016  Tanaka Akira  <akr@fsij.org>

	* complex.c: Don't refer rb_cFixnum and rb_cBignum.
	  (k_fixnum_p): Use FIXNUM_P.
	  (k_bignum_p): Use RB_TYPE_P.

Mon May  2 01:27:59 2016  Tanaka Akira  <akr@fsij.org>

	* test/ruby/test_numeric.rb (test_step): Use Integer::FIXNUM_MAX.

Mon May  2 01:15:01 2016  Tanaka Akira  <akr@fsij.org>

	* test/ruby/test_module.rb (test_name): Use Integer instead of Fixnum.

Mon May  2 01:00:04 2016  Tanaka Akira  <akr@fsij.org>

	* test/lib/test/unit/assertions.rb (assert_fixnum): Defined.
	  (assert_bignum): Defined.

	* test/ruby/test_bignum.rb: Use assert_bignum.

	* test/ruby/test_integer_comb.rb: Use assert_fixnum and assert_bignum.

	* test/ruby/test_optimization.rb: Ditto.

Mon May  2 00:41:53 2016  Tanaka Akira  <akr@fsij.org>

	* vm_trace.c (recalc_remove_ruby_vm_event_flags): Add a cast to
	  avoid signed integer overflow.

Mon May  2 00:06:04 2016  Tanaka Akira  <akr@fsij.org>

	* test/lib/envutil.rb: Define Integer::{FIXNUM_MIN,FIXNUM_MAX}.

	* test/ruby/test_bignum.rb: Use Integer::{FIXNUM_MIN,FIXNUM_MAX}.

	* test/ruby/test_bignum.rb: Ditto.

	* test/ruby/test_integer_comb.rb: Ditto.

	* test/ruby/test_marshal.rb: Ditto.

	* test/ruby/test_optimization.rb: Ditto.

Sun May  1 23:59:59 2016  Kenta Murata  <mrkn@mrkn.jp>

	* array.c (rb_ary_sum): fix for mathn

	* test/ruby/test_array.rb (test_sum): ditto.

Sun May  1 23:51:54 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* test/lib/test/unit.rb (Options#non_options): fixed wrong regexp.
	  if both positives and negatives were specified, positives had to
	  be specified from the beginning.

Sun May  1 21:00:07 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* win32/win32.c: drop Win2K support.

Sun May  1 20:39:47 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* cont.c, hash.c, random.c, win32/win32.c: cleanup some Win9x/ME/NT4
	  support leftovers.
	  [fix GH-1328] patched by @cremno

Sun May  1 07:30:44 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* string.c (search_nonascii): use nlz on big endian environments.

	* internal.h (nlz_intptr): defined.

Sun May  1 00:03:30 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* configure.in (__builtin_ctz): check.

	* configure.in (__builtin_ctzll): check.

	* internal.h (rb_popcount32): defined for ntz_int32.
	  it can use __builtin_popcount but this function is not used on
	  GCC environment because it uses __builtin_ctz.
	  When another function uses this, using __builtin_popcount
	  should be re-considered.

	* internal.h (rb_popcount64): ditto.

	* internal.h (ntz_int32): defined for ntz_intptr.

	* internal.h (ntz_int64): defined for ntz_intptr.

	* internal.h (ntz_intptr): defined as ntz for uintptr_t.

	* string.c (search_nonascii): unroll and use ntz.

Sat Apr 30 21:54:13 2016  Tanaka Akira  <akr@fsij.org>

	* numeric.c (Init_Numeric): Gather Fixnum method definitions.

Sat Apr 30 21:28:14 2016  Tanaka Akira  <akr@fsij.org>

	* numeric.c (rb_int_div): Define Integer#/.

	* bignum.c (rb_big_div): Don't define Bignum#/.

	* lib/mathn.rb (Integer#/): Replace Integer#/ instead of Bignum#/.

Sat Apr 30 21:11:08 2016  Tanaka Akira  <akr@fsij.org>

	* numeric.c (rb_int_plus): Define Integer#+.

	* bignum.c (rb_big_plus): Don't define Bignum#+.

Sat Apr 30 21:01:20 2016  Tanaka Akira  <akr@fsij.org>

	* numeric.c (rb_int_minus): Define Integer#-.

	* bignum.c (rb_big_minus): Don't define Bignum#-.

Sat Apr 30 20:53:33 2016  Tanaka Akira  <akr@fsij.org>

	* numeric.c (rb_int_mul): Define Integer#*.

	* bignum.c (rb_big_mul): Don't define Bignum#*.

Sat Apr 30 20:30:44 2016  Tanaka Akira  <akr@fsij.org>

	* numeric.c (rb_int_modulo): Define Integer#%.

	* bignum.c (rb_big_modulo): Don't define Bignum#%.

Sat Apr 30 20:17:08 2016  Tanaka Akira  <akr@fsij.org>

	* numeric.c (int_equal): Define Integer#==.

	* bignum.c (rb_big_eq): Don't define Bignum#==.

Sat Apr 30 19:41:15 2016  Tanaka Akira  <akr@fsij.org>

	* numeric.c (int_gt): Define Integer#>.

	* bignum.c (rb_big_gt): Don't define Bignum#>.
	  Renamed from big_gt.

	* internal.h (rb_big_gt): Declared.

Sat Apr 30 19:24:40 2016  Tanaka Akira  <akr@fsij.org>

	* numeric.c (int_ge): Define Integer#>=.

	* bignum.c (rb_big_ge): Don't define Bignum#>=.
	  Renamed from big_ge.

	* internal.h (rb_big_ge): Declared.

Sat Apr 30 19:20:40 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* doc/standard_library.rdoc: Remove obsoleted classes and modules.

Sat Apr 30 19:09:23 2016  Tanaka Akira  <akr@fsij.org>

	* numeric.c (int_lt): Define Integer#<.

	* bignum.c (rb_big_lt): Don't define Bignum#<.
	  Renamed from big_lt.

	* internal.h (rb_big_lt): Declared.

Sat Apr 30 18:44:05 2016  Tanaka Akira  <akr@fsij.org>

	* numeric.c (int_le): Define Integer#<=.

	* bignum.c (rb_big_le): Don't define Bignum#<=.
	  Renamed from big_le.

	* internal.h (rb_big_le): Declared.

Sat Apr 30 18:11:44 2016  Tanaka Akira  <akr@fsij.org>

	* bignum.c (Init_Bignum): Define Integer::GMP_VERSION.

Sat Apr 30 16:58:18 2016  Tanaka Akira  <akr@fsij.org>

	* numeric.c (int_remainder): Define Integer#remainder.

	* bignum.c (rb_big_remainder): Don't define Bignum#remainder.

	* internal.h (rb_big_remainder): Declared.

Sat Apr 30 15:29:24 2016  Tanaka Akira  <akr@fsij.org>

	* numeric.c (rb_int_uminus): {Fixnum,Bignum}#-@ is unified into
	  Integer.

	* bignum.c (rb_big_uminus): Don't define Bignum#-@.

Sat Apr 30 14:42:20 2016  Tanaka Akira  <akr@fsij.org>

	* numeric.c (rb_int_idiv): {Fixnum,Bignum}#div is unified into
	  Integer.

	* bignum.c (rb_big_idiv): Don't define Bignum#div.

Sat Apr 30 14:25:55 2016  Tanaka Akira  <akr@fsij.org>

	* numeric.c (rb_int_modulo): {Fixnum,Bignum}#modulo is unified into
	  Integer.

	* bignum.c (rb_big_modulo): Don't define Bignum#modulo.

Sat Apr 30 14:04:30 2016  Tanaka Akira  <akr@fsij.org>

	* numeric.c (int_divmod): {Fixnum,Bignum}#divmod is unified into
	  Integer.

	* bignum.c (rb_big_divmod): Don't define Bignum#divmod.

Sat Apr 30 13:20:00 2016  Tanaka Akira  <akr@fsij.org>

	* numeric.c (int_fdiv): {Fixnum,Bignum}#fdiv is unified into
	  Integer.

	* bignum.c (rb_big_fdiv): Don't define Bignum#fdiv.

Sat Apr 30 12:25:43 2016  Tanaka Akira  <akr@fsij.org>

	* numeric.c (rb_int_pow): {Fixnum,Bignum}#** is unified into
	  Integer.

	* bignum.c (rb_big_pow): Don't define Bignum#**.

Sat Apr 30 12:28:59 2016  Tanaka Akira  <akr@fsij.org>

	* bignum.c (rb_big_comp): Renamed from rb_big_neg.

	* numeric.c (fix_comp): Renamed from fix_rev.

Sat Apr 30 12:25:43 2016  Tanaka Akira  <akr@fsij.org>

	* numeric.c (int_comp): {Fixnum,Bignum}#~ is unified into
	  Integer.

	* bignum.c (rb_big_neg): Don't define Bignum#~.

	* internal.h (rb_big_neg): Declared.

Sat Apr 30 12:07:42 2016  Tanaka Akira  <akr@fsij.org>

	* numeric.c (int_and): {Fixnum,Bignum}#& is unified into
	  Integer.

	* bignum.c (rb_big_and): Don't define Bignum#&.

Sat Apr 30 11:56:15 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ext/thread: removed dummy extension library.  thread_sync.c
	  provides "thread.rb" already.

Sat Apr 30 11:53:48 2016  Tanaka Akira  <akr@fsij.org>

	* numeric.c (int_or): {Fixnum,Bignum}#| is unified into
	  Integer.

	* bignum.c (rb_big_or): Don't define Bignum#|.

Sat Apr 30 11:18:47 2016  Yuichiro Kaneko  <yui-knk@ruby-lang.org>

	* vm_trace.c: Fix typos. [ci skip]

Sat Apr 30 10:09:04 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ext/pty/pty.c (establishShell): honor USER environment variable
	  and login name over uid, one uid can be shared by some login
	  names.

Fri Apr 29 22:40:28 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* doc/maintainers.rdoc (ext/io/nonblock): still maintained, as
	  well as ext/io/wait, which is the origin.

Fri Apr 29 21:18:12 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* doc/maintainers.rdoc (ext/win32): of course, it's still maintained.

Fri Apr 29 21:03:10 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* gems/bundled_gems: Update latest gems, test-unit-3.1.8 and rake-11.1.2.

Fri Apr 29 20:43:02 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* doc/maintainers.rdoc: Update latest maintainers list on Ruby 2.4

Fri Apr 29 19:52:45 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* doc/maintainers.rdoc: Removed deprecated entries. These are already deleted.

Fri Apr 29 19:48:45 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* doc/maintainers.rdoc: Removed Ruby 1.8 entries. It's not necessary to
	  Ruby 2.4 or later.

Thu Apr 28 17:03:17 2016  Nicholas Maccharoli  <nmaccharoli@gmail.com>

	* hash.c (rb_hash_update_{block,func}_callback): dry up hash
	  update callback code.  [Fix GH-1338]

Thu Apr 28 16:52:05 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* re.c (rb_reg_prepare_enc): use rb_enc_asciicompat(enc) instead of
	  rb_enc_str_asciicompat_p(str) to avoid useless rb_enc_get(str) call.

Thu Apr 28 16:33:41 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* lib/optparse.rb: [DOC] fix example code.  base on the code by
	  Semyon Gaivoronskiy in [ruby-core:75224].  [Bug #12323]

Thu Apr 28 09:33:03 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* lib/rubygems.rb, lib/rubygems/*, test/rubygems/*: Update rubygems-2.6.4.
	  Please see entries of 2.6.4 on
	  https://github.com/rubygems/rubygems/blob/master/History.txt

Thu Apr 28 04:49:07 2016  Rei Odaira  <Rei.Odaira@gmail.com>

	* configure.in (rb_cv_lgamma_r_pm0): check if lgamma_r(+0.0)
	  returns positive infinity, in addition to lgamma_r(-0.0).
	  AIX returns an incorrect result of negative infinity.

	* math.c (ruby_lgamma_r): handle +0.0, in addition to -0.0.

Thu Apr 28 01:11:14 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* time.c: define _DEFAULT_SOURCE because glibc 2.20 deprecates
	  _BSD_SOURCE.
	  https://sourceware.org/glibc/wiki/Release/2.20

Thu Apr 28 00:27:55 2016  Tanaka Akira  <akr@fsij.org>

	* numeric.c (int_xor): {Fixnum,Bignum}#^ is unified into
	  Integer.

	* bignum.c (rb_big_xor): Don't define Bignum#^.

Wed Apr 27 20:53:59 2016  Tanaka Akira  <akr@fsij.org>

	* numeric.c (int_aref): {Fixnum,Bignum}#[] is unified into
	  Integer.

	* bignum.c (rb_big_aref): Don't define Bignum#<<.

	* internal.h (rb_big_aref): Declared.

Wed Apr 27 16:10:35 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* tool/instruction.rb: fix to follow current implementation.

Wed Apr 27 15:47:54 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ext/stringio/stringio.c (strio_s_new): warn if a block is given,
	  as well as IO.new.

Wed Apr 27 14:29:47 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* error.c (ruby_only_for_internal_use): raise fatal error when
	  deprecated function only for internal use is called, not just a
	  warning.

Tue Apr 26 23:42:30 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* tool/redmine-backporter.rb (rel): should not raise exceptions even if
	  the user input is wrong.  only reports the error and continue process.

Tue Apr 26 23:35:23 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ruby.c (process_options): convert -e script to the encoding
	  given by a command line option on Windows.  assume it is the
	  expected encoding.  [ruby-dev:49461] [Bug #11900]

Tue Apr 26 21:11:02 2016  Tanaka Akira  <akr@fsij.org>

	* numeric.c (rb_int_lshift): {Fixnum,Bignum}#<< is unified into
	  Integer.

	* bignum.c (rb_big_lshift): Don't define Bignum#<<.

Tue Apr 26 20:59:40 2016  Tanaka Akira  <akr@fsij.org>

	* numeric.c (rb_int_rshift): {Fixnum,Bignum}#>> is unified into
	  Integer.

	* bignum.c (rb_big_rshift): Don't define Bignum#>>.

Tue Apr 26 20:46:16 2016  Tanaka Akira  <akr@fsij.org>

	* numeric.c (int_size): {Fixnum,Bignum}#size is unified into Integer.

	* bignum.c (rb_big_size_m): Don't define Bignum#size.

	* internal.h (rb_big_size_m): Declared.

Tue Apr 26 20:09:08 2016  Tanaka Akira  <akr@fsij.org>

	* numeric.c (rb_int_bit_length): {Fixnum,Bignum}#bit_length is
	  unified into Integer.

	* bignum.c (rb_big_bit_length): Don't define Bignum#bit_length.

	* internal.h (rb_big_bit_length): Declared.

Tue Apr 26 19:56:16 2016  Tanaka Akira  <akr@fsij.org>

	* numeric.c (int_abs): Integer#{abs,magnitude} moved from
	  Fixnum and Bignum.

	* bignum.c (rb_big_abs): Don't define Bignum#{abs,magnitude}.

	* internal.h (rb_big_abs): Declared.

Mon Apr 25 14:39:11 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ext/rbconfig/sizeof/extconf.rb: just check the existence of each
	  types, to reduce configuration time, especially cross-compiling.

	* template/sizes.c.tmpl: calculate sizes of checked types at
	  compilation time.

Mon Apr 25 11:27:27 2016  Marcus Stollsteimer  <sto.mar@web.de>

	* doc/extension.rdoc: Improvements to english grammars.
	  [Bug #12246][ruby-core:74792][ci skip]

Mon Apr 25 11:17:50 2016  Marcus Stollsteimer  <sto.mar@web.de>

	* encoding.c: Fix return value of `Encoding::ISO8859_1.name`
	  [Bug #12313][ruby-core:75147][ci skip]
	* ext/bigdecimal/bigdecimal.c: Fix code sample of `BigDecimal.new`

Sun Apr 24 23:29:16 2016  Rei Odaira  <Rei.Odaira@gmail.com>

	* configure.in: add missing -lm for AIX.

Sun Apr 24 18:33:58 2016  Kazuki Tsujimoto  <kazuki@callcc.net>

	* vm_insnhelper.c (INLINE): disable r54738 if __NO_INLINE__ is defined.
	  It caused "undefined reference to `vm_getivar'".

Sun Apr 24 09:32:12 2016  Yuichiro Kaneko  <yui-knk@ruby-lang.org>

	* test/ruby/test_array.rb: Add test cases for Array#sum with
	  non-numeric objects.

Sun Apr 24 04:21:27 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* vm_insnhelper.c (INLINE): define as `inline` when it is optimized.
	  define as `static inline` when it is not optimized to keep
	  the symbol generated.

	* vm_insnhelper.c (vm_getivar): use `INLINE` to force inline
	  so that a compiler inlines it into vm_getinstancevariable
	  and optimizes out is_attr and related branches.

	* vm_insnhelper.c (vm_getivar): use `inline` to recommend inline.
	  Without this vm1_ivar_set is degraded.

	  benchmark results:
	  minimum results in each 5 measurements.
	  Execution time (sec)
	  name    ruby 2.4.0dev (2016-04-23 trunk 54727) [x86_64-linux]   ruby 2.4.0dev (2016-04-23 trunk 54733) [x86_64-linux]       built-ruby
	  loop_whileloop 0.641                                                   0.642 0.646
	  vm1_ivar* 1.002                                                   0.999 0.831
	  vm1_ivar_set* 0.369                                                   1.106 0.362

	  Speedup ratio: compare with the result of `ruby 2.4.0dev (2016-04-23
	  trunk 54727) [x86_64-linux]' (greater is better)
	  name    ruby 2.4.0dev (2016-04-23 trunk 54733) [x86_64-linux]
	  built-ruby
	  loop_whileloop
	  0.998        0.991
	  vm1_ivar*
	  1.003        1.205
	  vm1_ivar_set*
	  0.334        1.018

Sat Apr 23 18:01:21 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* vm_insnhelper.c (vm_getivar): specify inline instead of static inline.
	  vm_getivar is called by vm_call_ivar and vm_getinstancevariable.
	  At least with GCC 4.8 and 5.3 on Linux, they are inlining it into
	  vm_call_ivar but not vm_getinstancevariable.
	  By `inline`, they correctly inline it and gains performance.

	  Speedup ratio: compare with the result of `ruby 2.4.0dev (2016-04-23
	  trunk 54727) [x86_64-linux]' (greater is better)
	  name    built-ruby
	  loop_whileloop       1.001
	  vm1_ivar*            1.189
	  vm1_ivar_set*        1.024

	  Note the `inline`'s meaning is different between old GCC
	  and C99. Old GCC's inline means C99's extern inline.
	  https://gcc.gnu.org/onlinedocs/gcc/Inline.html
	  Since Ruby specify -std=iso9899:1999, it works like C99.

Sat Apr 23 16:11:39 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* include/ruby/ruby.h (rb_mul_size_overflow): use UNLIKELY
	  by user side to improve generality.

Sat Apr 23 16:10:02 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* include/ruby/ruby.h (RB_LIKELY): use prefix in ruby.h.

	* intern.h (LIKELY): define with RB_LIKELY.

Sat Apr 23 13:27:25 2016  Yuichiro Kaneko  <yui-knk@ruby-lang.org>

	* NEWS: Add descriptions for Time#to_time updates.
	  [Bug #12271]

Sat Apr 23 13:21:24 2016  Yuichiro Kaneko  <yui-knk@ruby-lang.org>

	* NEWS: Add descriptions for DateTime#to_time updates.
	  [Bug #12189]

Sat Apr 23 11:21:27 2016  Marcus Stollsteimer  <sto.mar@web.de>

	* ext/date/date_core.c (Init_date_core): [DOC] Convert DateTime
	  documentation to RDoc from Markdown.
	  [ruby-core:75136] [Bug #12311]

Sat Apr 23 09:03:35 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ruby.c: cygwin does not use w32_cmdvector, command line can be
	  other than UTF-8.  [ruby-dev:49519] [Bug #12184]

Sat Apr 23 01:00:03 2016  Rei Odaira  <Rei.Odaira@gmail.com>

	* configure.in: don't use the system-provided round(3) on AIX.
	  In AIX, round(0.49999999999999994) returns 1.0.
	  Use round() in numeric.c instead.

Fri Apr 22 21:00:44 2016  Tanaka Akira  <akr@fsij.org>

	* test/ruby/test_time_tz.rb: Tests depends on Europe/Moscow removed
	  to avoid test failures due to the tzdata change.
	  https://github.com/eggert/tz/commit/8ee11a301cf173afb0c76e0315b9f9ec8ebb9d95
	  Found by naruse.

Fri Apr 22 20:18:40 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* include/ruby/ruby.h (rb_mul_size_overflow): added to handle
	  mul overflow efficiently.

	* include/ruby/ruby.h (rb_alloc_tmp_buffer2): use rb_mul_size_overflow
	  and avoid division where it can define DSIZE_T.

	* gc.c (xmalloc2_size): moved from ruby.h and use rb_mul_size_overflow.

Fri Apr 22 20:34:04 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* time.c (time_asctime): [DOC] add ctime example, not only
	  asctime.  [ruby-core:75126] [Bug #12310]

Fri Apr 22 18:44:32 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* variable.c: use uint32_t instead of long to avoid confusion about
	  the type of ivtbl->numiv.

Fri Apr 22 15:09:27 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* eval_jump.c (exec_end_procs_chain): restore previous error info
	  for each end procs.  [ruby-core:75038] [Bug #12302]

Fri Apr 22 15:04:56 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* tool/redmine-backporter.rb: the fullpath of merger.rb is too long to
	  copy&paste on Windows.  show shorter name instead on the platform.
	  I'm sure that the user of this command on Windows is only me.

Fri Apr 22 14:52:04 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* tool/merger.rb: remove temporary file.

Fri Apr 22 11:27:03 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* lib/net/http.rb: Improve documentation for SSL requests via GET method.
	  [fix GH-1325][ci skip] Patch by @jsyeo

Fri Apr 22 10:51:13 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* lib/webrick/ssl.rb: Support to add SSLCiphers option.
	  [fix GH-1321] Patch by @rhadoo

Fri Apr 22 10:43:19 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* file.c, win32/file.c: Removed obsoleted safe level checks.
	  [fix GH-1327] Patch by @cremno

Fri Apr 22 10:01:48 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* benchmark/bm_so_meteor_contest.rb: fix a typo.
	  [fix GH-1330][ci skip] Patch by @sachin21

Fri Apr 22 04:57:01 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* gc.c (rb_alloc_tmp_buffer_with_count): added like xmalloc2 to
	  avoid duplicated check of size.

	* gc.c (ruby_xmalloc2): added to keep separate layers.

	* include/ruby/ruby.h (rb_alloc_tmp_buffer2): added to check
	  the size more statically.

Fri Apr 22 04:54:40 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* include/ruby/ruby.h (LIKELY): moved from internal.h.

	* include/ruby/ruby.h (UNLIKELY): ditto.

Thu Apr 21 01:44:19 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* gc.c (objspace_malloc_prepare): remove size check because it is
	  used by objspace_xmalloc and objspace_xcalloc.
	  objspace_xmalloc introduces its own check in this commit.
	  objspace_xcalloc checks with xmalloc2_size (ruby_xmalloc2_size).

	* gc.c (objspace_xmalloc0): common xmalloc function.

	* gc.c (objspace_xmalloc): introduce its own size check.

	* gc.c (objspace_xmalloc2): separated from ruby_xmalloc2 to clarify
	  the layer who has the responsibility to check the size.

	* gc.c (objspace_xrealloc): remove duplicated size check.

	* gc.c (ruby_xmalloc2): use objspace_xmalloc2.

	* include/ruby/ruby.h (ruby_xmalloc2_size): follow the size limit
	  as SSIZE_MAX. Note that ISO C says size_t is unsigned integer.

Thu Apr 21 12:14:04 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* configure.in: check if succeeded in creating config.h.

	* tool/ifchange: ignore failures when TEST_COLORS unmatched. just
	  use the default value if expected name is not contained in it.
	  [ruby-core:75046] [Bug #12303]

Wed Apr 20 17:33:31 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ext/cgi/escape/escape.c (cgiesc_unescape): define unescape
	  method instead of _unescape, and should pass the optional
	  argument to the super method.

	* lib/cgi/util.rb (CGI::Util#_unescape): remove intermediate
	  method.

Wed Apr 20 15:52:28 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* error.c (syntax_error_initialize): move the default message,
	  "compile error", from parse.y.  the default parameter should
	  belong to the class definition.

	* parse.y (yycompile0): use the default parameter.

Wed Apr 20 10:25:53 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* compile.c (append_compile_error): use rb_syntax_error_append.

	* error.c (rb_syntax_error_append): append messages into a
	  SyntaxError exception instance.

	* parse.y (yycompile0): make new SyntaxError instance in main
	  mode, otherwise error_buffer should be a SyntaxError if error
	  has occurred.

Tue Apr 19 17:42:47 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* error.c (err_vcatf): rename, and separate appending message from
	  creating a string buffer.

	* error.c (rb_syntax_error_append): merge rb_error_vsprintf and
	  rb_compile_err_append.

	* parse.y (parser_compile_error): use rb_syntax_error_append.

Tue Apr 19 13:46:19 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* compile.c (append_compile_error, compile_bug): pass iseq and get
	  error info and file from it, not by the thread error info.

	* error.c (rb_report_bug_valist): take va_list instead of variadic
	  arguments, and just report the bug but not abort.

Tue Apr 19 13:18:12 2016  Naotoshi Seo  <sonots@gmail.com>

	* lib/time.rb: revert r54167 because it would break
	  backward compatibilities, and it is documented that
	  Time.parse does not take into account time zone
	  abbreations other than ones described in RFC 822

Tue Apr 19 13:12:03 2016  Naotoshi Seo  <sonots@gmail.com>

	* ChangeLog: Fix dates of previous commits

Tue Apr 19 12:45:03 2016  Naotoshi Seo  <sonots@gmail.com>

	* ChangeLog: Add descriptions for logger updates
	* NEWS: Add descriptions for logger updates

Tue Apr 19 12:45:02 2016  Naotoshi Seo  <sonots@gmail.com>

	* lib/logger.rb: Add shift_period_suffix option

Tue Apr 19 12:45:01 2016  Naotoshi Seo  <sonots@gmail.com>

	* lib/logger.rb: Allow specifying logger parameters in constructor
	  such as level, progname, datetime_format, formatter.

Mon Apr 18 16:07:01 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* compile.c (iseq_peephole_optimize): should not replace the
	  current target INSN, not to follow the replaced dangling link in
	  the caller.  [ruby-core:74993] [Bug #11816]

Mon Apr 18 12:56:31 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* numeric.c (flo_truncate): add an optional parameter, digits, as
	  well as Float#round.  [Feature #12245]

	* numeric.c (int_truncate): add an optional parameter, digits, as
	  well as Integer#round.  [Feature #12245]

Sun Apr 17 04:18:56 2016  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>

	* tool/redmine-backporter.rb: revisions are strings.

Sat Apr 16 14:26:49 2016  Yuichiro Kaneko  <yui-knk@ruby-lang.org>

	* ext/date/date_core.c : remove not used f_getlocal macro.
	  After r54553 f_getlocal macro is not used.

Sat Apr 16 14:15:24 2016  Yuichiro Kaneko  <yui-knk@ruby-lang.org>

	* ext/date/date_core.c : remove not used f_utc6 macro.
	  After r54169 f_utc6 macro is not used.

Sat Apr 16 10:00:11 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* struct.c (struct_make_members_list, rb_struct_s_def): member
	  names should be unique. [ruby-core:74971] [Bug #12291]

	* struct.c (struct_make_members_list): extract making member name
	  list from char* va_list, with creating symbols without
	  intermediate IDs.

Sat Apr 16 01:33:27 2016  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>

	* tool/redmine-backporter.rb: sort revisions.

Sat Apr 16 01:16:02 2016  Tanaka Akira  <akr@fsij.org>

	* array.c (rb_ary_sum): Don't yield same element twice.
	  Found by nagachika.

Sat Apr 16 01:03:32 2016  Tanaka Akira  <akr@fsij.org>

	* array.c (rb_ary_sum): Fix SEGV by [1/2r, 1].sum.

Fri Apr 15 23:52:00 2016  Kenta Murata  <mrkn@mrkn.jp>

	* rational.c (rb_rational_plus): rename from rb_rational_add
	  to be aligned with rb_fix_plus.

	* array.c (rb_ary_sum): ditto.

	* internal.h: ditto.

Fri Apr 15 23:42:00 2016  Kenta Murata  <mrkn@mrkn.jp>

	* rational.c (rb_rational_add): rename from nurat_add.

	* array.c (rb_ary_sum): use rb_rational_add directly.

	* test/ruby/test_array.rb (test_sum): add assertions for an array of
	  Rational values.

Fri Apr 15 22:31:00 2016  Kenta Murata  <mrkn@mrkn.jp>

	* array.c (rb_ary_sum): apply the precision compensated algorithm
	  for an array in which Rational and Float values are mixed.

	* test/ruby/test_array.rb (test_sum): add assertions for the above
	  change.

Fri Apr 15 22:30:01 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* thread.c (rb_thread_setname): defer setting native thread name
	  set in initialize until the native thread is created.
	  [ruby-core:74963] [Bug #12290]

Fri Apr 15 20:27:16 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* lib/irb/ext/save-history.rb: Fix NoMethodError when method is not defined.

Fri Apr 15 15:38:58 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* common.mk (benchmark): order options for built-ruby and compare-ruby.

Fri Apr 15 14:14:00 2016  Kenta Murata  <mrkn@mrkn.jp>

	* test/ruby/test_array.rb (test_sum): add assertions for Rational and
	  Complex numbers.

Fri Apr 15 10:07:11 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ext/io/console/console.c (console_key_pressed_p): raise the same
	  exception, "unknown virtual key code", for names with nul chars.
	  though console_win32_vk() considers the length and can deal with
	  nul chars, rb_sprintf() raised at PRIsVALUE previously, so quote
	  it if it is unprintable.

Fri Apr 15 09:02:58 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ext/io/console/console.c (rb_sym2str): fallback definition for
	  older ruby.  [ruby-core:74953] [Bug #12284]

Thu Apr 14 21:46:36 2016  Tanaka Akira  <akr@fsij.org>

	* array.c (rb_ary_sum): Support the optional argument, init, and
	  block.

Thu Apr 14 19:02:41 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* lib/irb/ext/save-history.rb: suppress warning: method redefined;
	  discarding old save_history=.

Thu Apr 14 14:58:14 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ext/tk/tkutil/tkutil.c (tk_hash_kv): the third argument can be
	  nil not only an Array.  reported by @windwiny at
	  https://github.com/ruby/ruby/commit/cdaa94e#commitcomment-17096618

Thu Apr 14 14:28:55 2016  cremno phobia  <cremno@mail.ru>

	* cont.c (fiber_initialize_machine_stack_context): fix wrong
	  _MSC_VER check, should be decimal but not hexadecimal.
	  [ruby-core:74936] [Bug #12279]

Wed Apr 13 22:51:38 2016  Tanaka Akira  <akr@fsij.org>

	* array.c (rb_ary_sum): Array#sum is implemented.
	  Kahan's compensated summation algorithm for precise sum of float
	  numbers is moved from ary_inject_op in enum.c.

	* enum.c (ary_inject_op): Don't specialize for float numbers.

	  [ruby-core:74569] [Feature #12217] proposed by mrkn.

Wed Apr 13 15:56:35 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* numeric.c (flo_ceil): add an optional parameter, digits, as
	  well as Float#round.  [Feature #12245]

	* numeric.c (flo_floor): add an optional parameter, digits, as
	  well as Float#round.  [Feature #12245]

	* numeric.c (int_ceil): add an optional parameter, digits, as
	  well as Integer#round.  [Feature #12245]

	* numeric.c (int_floor): add an optional parameter, digits, as
	  well as Integer#round.  [Feature #12245]

Wed Apr 13 14:47:47 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* file.c (Init_File): add alias File.empty? to File.zero?.
	  [Feature #9969]

Wed Apr 13 14:36:24 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* parse.y (assign_in_cond): allow multiple assignment in
	  conditional expression.  [Feature #10617]

Wed Apr 13 14:11:59 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* bignum.c (rb_big_size): add wrapper function of BIGSIZE and
	  rename the method function with _m suffix.

	* numeric.c (int_round_zero_p): extracted from rb_int_round.
	  optimize for Bignum, and convert VALUE returned by Numeric#size
	  to long.

Wed Apr 13 12:00:08 2016  Koichi Sasada  <ko1@atdot.net>

	* test/ruby/test_basicinstructions.rb: add a test to check access
	  instance variables on special const objects.

	  All of such objects are frozen, so that we can not set instance
	  variables for them. But we can read instance variables and return
	  default value (nil).

Tue Apr 12 20:40:35 2016  Kaneko Yuichiro  <spiketeika@gmail.com>

	* ext/date/date_core.c (time_to_time): should preserve timezone
	  info.  [ruby-core:74889] [Bug #12271]

Tue Apr 12 11:51:18 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* compile.c (new_label_body): initialize bit fields, since
	  compile_data_alloc does not clear the memory.  [Bug #12082]

Mon Apr 11 20:18:43 2016  Koichi Sasada  <ko1@atdot.net>

	* vm_backtrace.c (frame2klass): filter only for imemo_ment.
	  T_IMEMO/imemo_iseq can be passed here.

Mon Apr 11 17:43:04 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* compile.c (iseq_optimize): disable tail call optimization in
	  rescued, rescue, and ensure blocks.
	  [ruby-core:73871] [Bug #12082]

Mon Apr 11 06:54:39 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* file.c (apply2files): apply to a VALUE vector instead of a
	  temporary array.

Sun Apr 10 20:54:16 2016  Joe Swatosh  <joe.swatosh@gmail.com>

	* ext/win32/lib/win32/registry.rb (DeleteValue, DeleteKey): fix
	  API names.  [ruby-core:74863] [Bug #12264]

Sun Apr 10 17:47:42 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* file.c (rb_realpath_internal): no argument conversions since
	  this internal function does not need to_path and encoding
	  conversions, not to be affected by the default internal
	  encoding.

Sat Apr  9 10:03:12 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* load.c (rb_f_load): raise with the original path name before
	  encoding conversion.

Sat Apr  9 02:05:10 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* bignum.c (rb_cstr_parse_inum): [EXPERIMENTAL] new function to
	  parse integer in C-string with length.  the name and the
	  arguments may be changed in the future.

	* bignum.c (rb_str_to_inum): preserve encoding of the argument in
	  error messages, and no longer needs to copy non-terminated
	  strings.

	* bignum.c (rb_str2big_{poweroftwo,normal,karatsuba,gmp}): ditto.

Thu Apr  7 19:04:03 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* doc/regexp.rdoc (comments): [DOC] terminators cannot appear in
	  comments.  [ruby-core:74838] [Bug #12256]

Thu Apr  7 11:24:14 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ext/tk/tkutil/tkutil.c (cbsubst_initialize): fix out-of-bound
	  access when no arguments given.  `p Tk::Event.new` crashed.

Fri Apr  1 01:26:00 2016  Benoit Daloze  <eregontp@gmail.com>

	* ext/coverage/coverage.c: Fully reset coverage to not persist global state.
	  It was returning old file coverages as empty arrays to the user.
	  [ruby-core:74596] [Bug #12220]

	* ext/coverage/coverage.c (rb_coverages): remove unused static state.

	* thread.c: Moved and renamed coverage_clear_result_i to reset_coverage_i.

	* test/coverage/test_coverage.rb: improve precision of tests.

Wed Apr  6 22:41:31 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* configure.in (rb_cv_lgamma_r_m0): fix the condition for
	  lgamma_r(-0.0).  [Bug #12249]

Wed Apr  6 17:38:42 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* tool/downloader.rb (RubyGems.download): follow the change of the
	  rubygems ssl_certs directory tree introduced by previous commit.

Wed Apr  6 15:00:27 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* lib/rubygems.rb, lib/rubygems/*, test/rubygems/*: Update rubygems-2.6.3.
	  Please see entries of 2.6.3 on
	  https://github.com/rubygems/rubygems/blob/master/History.txt

Wed Apr  6 14:13:28 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* configure.in (rb_cv_lgamma_r_m0): check if lgamma_r(-0.0)
	  returns negative infinity.  [Bug #12249]

	* math.c (ruby_lgamma_r): define by the configured result.

Wed Apr  6 10:56:15 2016  Anton Davydov  <antondavydov.o@gmail.com>

	* lib/logger.rb (Logger#level=): remove unnecessary local
	  variable.

	* lib/logger.rb (Logger#initialize, Logger#reopen): [DOC] mention
	  the default values.  cherrypicked from [GH-1319].

Wed Apr  6 10:17:53 2016  cremno phobia  <cremno@mail.ru>

	* math.c (ruby_lgamma_r): missing/lgamma_r.c is used on Windows,
	  since msvcrt does not provide it.

	* missing/lgamma_r.c (lgamma_r): fix lgamma(-0.0).
	  [ruby-core:74823] [Bug #12249]

Wed Apr  6 01:22:55 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* math.c (ruby_lgamma_r): mswin's lgamma_r also seems to be wrong.
	  cf. [Bug #12249]

Wed Apr  6 00:53:31 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* math.c (ruby_lgamma_r): fix lgamma(-0.0) on mingw and OSX.

	* math.c (ruby_tgamma): fix tgamma(-0.0) on mingw.
	  [ruby-core:74817] [Bug #12249]

Tue Apr  5 14:50:28 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* ext/nkf/nkf-utf8/nkf.c (mime_putc): fix typo.
	  [Bug #12202] [ruby-core:74802]

Tue Apr  5 00:06:44 2016  Aeris  <aeris@imirhil.fr>

	* ext/openssl/ossl_ssl.c (ossl_ssl_tmp_key): Access to ephemeral
	  TLS session key in case of forward secrecy cipher.  Only
	  available since OpenSSL 1.0.2.  [Fix GH-1318]

	* ext/openssl/extconf.rb: Check for SSL_get_server_tmp_key.

Mon Apr  4 23:37:05 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* vm_core.h (rb_vm_struct): make at_exit a single linked list but
	  not RArray, not to mark the registered functions by the write
	  barrier.  based on the patches by Evan Phoenix.
	  [ruby-core:73908] [Bug #12095]

Mon Apr  4 17:43:45 2016  Koichi Sasada  <ko1@atdot.net>

	* gc.c: change default value of
	  RUBY_GC_HEAP_FREE_SLOTS_MIN_RATIO 0.3 -> 0.2
	  RUBY_GC_HEAP_FREE_SLOTS_MAX_RATIO 0.8 -> 0.65

	  These values are same as Ruby 2.0.0.

	  This change cause GC counts.
	  However, generational GC reduced each (minor) GC time and
	  increase memory locality. So that not so big impact on my
	  benchmarking results.
	  (surprisingly, this fix speed up programs on some cases)

	  You can change these values by environment variables
	  if you feel wrong.

Mon Apr  4 17:36:52 2016  Koichi Sasada  <ko1@atdot.net>

	* gc.c (get_envparam_double): take an upper_bound.

	  And also take an accept_zero flag which allow to accept zero
	  even if lower_bound is set.

	* gc.c (ruby_gc_set_params): fix parameters.

	  RUBY_GC_HEAP_FREE_SLOTS_MAX_RATIO set 0.9 as *lower_bound*, so that
	  it should be upper_bound.
	  Set RUBY_GC_HEAP_FREE_SLOTS_MIN_RATIO as lower bound.

	  Also set lower/upper bound of RUBY_GC_HEAP_FREE_SLOTS_GOAL_RATIO to
	  RUBY_GC_HEAP_FREE_SLOTS_MIN/MAX_RATIO.

Mon Apr  4 16:41:32 2016  Koichi Sasada  <ko1@atdot.net>

	* vm.c (Init_VM): should pass tokens.

Sun Apr  3 09:34:29 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* numeric.c (flodivmod): round division if it is a finite number
	  and module is required.

	* numeric.c (dbl2ival): do not round here.

	* numeric.c (flo_ceil): use dbl2ival.

	* numeric.c (flo_round): round explicitly.

Sat Apr  2 15:24:18 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* include/ruby/intern.h (rb_check_arity): returns argc.

Fri Apr  1 20:58:33 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode/case-folding.rb, casefold.h: Data generation to implement
	  swapcase functionality for titlecase characters. Swapcase isn't defined
	  by Unicode, because the purpose/usage of swapcase is unclear anyway.
	  The implementation follows a proposal from Nobu, swapping the case of
	  each component of a titlecase character individually.
	  This means that the titlecase characters have to be decomposed.

	* enc/unicode.c: Code using the above data.

	* test/ruby/enc/test_case_mapping.rb: Tests for the above.

Fri Apr  1 14:55:28 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* configure.in (AC_CONFIG_FILES): $srcdir/.git can be a file pointing
	  the real git_dir, such as when the git working tree is a "linked
	  working tree" (a working tree created by git-worktree). So use
	  git-rev-parse --git-dir to check if $srcdir is the top-level of a git
	  repository, not just checking if the $srcdir/.git directory does exist
	  or not.  [ruby-core:74759] [Bug #12239]

	* tool/change_maker.rb: use tool/vcs.rb to detect VCS. This used to have
	  its own VCS detection code, while we have tool/vcs.rb.

	* tool/vcs.rb (detect): remove code duplication

Fri Apr  1 04:50:44 2016  Eric Wong  <e@80x24.org>

	* ext/openssl/ossl_ssl.c (ossl_sslctx_s_alloc):
	  enable SSL_MODE_ACCEPT_MOVING_WRITE_BUFFER by default
	  [Bug #12126]

Fri Apr  1 01:13:55 2016  Benoit Daloze  <eregontp@gmail.com>

	* thread.c (update_coverage): Do not track coverage in loaded files
	  after Coverage.result. Avoids out-of-bounds access. [Bug #12237]

	* ext/coverage/coverage.c (coverage_clear_result_i): document.

Thu Mar 31 19:16:16 2016  Koichi Sasada  <ko1@atdot.net>

	* gc.c: need to set initial value of GC_HEAP_FREE_SLOTS_GOAL_RATIO.

Thu Mar 31 17:50:27 2016  Koichi Sasada  <ko1@atdot.net>

	* gc.c: change additional allocation policy.

	  Introduce new environment variable
	  RUBY_GC_HEAP_FREE_SLOTS_GOAL_RATIO (goal_ratio) to calculate the
	  ratio of additional memory.

	  Before this change, we add pages with the following formula
	  (when free_slots < total_pages * RUBY_GC_HEAP_FREE_SLOTS_MIN_RATIO):
	    next_pages = total_pages * RUBY_GC_HEAP_GROWTH_FACTOR

	  This addition can allocate too much.

	  With this change, we increase pages to satisfy the following formula:
	    next_free_slots = next_total_slots * goal_ratio
	  where
	    next_free_slots = free_slots + adding_slots
	    next_total_slots = total_slots + adding_slots.

	  If you want to prepare many free slots, increase this ratio.

	  If this variable is 0, then simply multiply
	  RUBY_GC_HEAP_GROWTH_FACTOR.

	* gc.c (get_envparam_double): enable to accept 0.

Thu Mar 31 17:48:25 2016  Koichi Sasada  <ko1@atdot.net>

	* gc.c (gc_marks_finish): fix syntax error.

Thu Mar 31 16:49:36 2016  Koichi Sasada  <ko1@atdot.net>

	* gc.c: simplify allocate/free detecting logic at the end of marking.

	  Before this change, heap_pages_min_slots are calculated at the
	  beginning sweeping phase. And this value is used at the end of
	  *next* marking phase.

	  To simplify it, we use this value at the end of this marking phase.
	  It means that we don't need to store this value as global state.

	  Also heap_pages_max_slots is calculated at the begging of sweeping
	  phase and used at the end of sweeping phase.
	  To simplify this logic, we introduced new global value
	  heap_pages_freeable_pages it means extra pages count we can free.
	  gc_sweep_step() checks this value and moves empty pages to tomb_heap
	  not more than this value.

	  Because of this fix, heap_pages_swept_slots is no longer needed.

	* gc.c (rb_objspace_t::heap_pages): restruct the objspace global
	  status.

	  remove the following fields
	    * swept_slots (and heap_pages_swept_slots)
	    * min_free_slots (and heap_pages_min_free_slots)
	    * max_free_slots (and heap_pages_max_free_slots)
	  And add the following filed.
	    * freeable_pages (and heap_pages_freeable_pages)

	* gc.c (heap_pages_free_unused_pages): unlink tomb heap pages
	  because tomb heap should have only freeable pages.

	* gc.c (heap_extend_pages): add parameters for future extension.

Thu Mar 31 16:43:02 2016  Koichi Sasada  <ko1@atdot.net>

	* gc.c: add GC parameters to configure the following values:
	  * RUBY_GC_HEAP_FREE_SLOTS_MIN_RATIO:
	    allocate additional pages when free slots is lower than
	    the value (total_slots * (this ratio)).
	  * RUBY_GC_HEAP_FREE_SLOTS_MAX_RATIO:
	    allow to free pages when free slots is greater than
	    the value (total_slots * (this ratio)).

	  Before this change, these values are hard coded.

	* gc.c (ruby_gc_params_t): ditto.

	* gc.c (ruby_gc_set_params): ditto.

Thu Mar 31 15:59:17 2016  Koichi Sasada  <ko1@atdot.net>

	* gc.c (gc_verify_heap_page): check the number of zombies.

	* gc.c (gc_verify_heap_pages): check also tomb heap.

Thu Mar 31 15:48:18 2016  Koichi Sasada  <ko1@atdot.net>

	* gc.c (gc_page_sweep): return free slots count.

	* gc.c (gc_sweep_step): use returned free slots count.

	* gc.c (gc_sweep_step): change variable name `next'
	  to `next_sweep_page'.

Thu Mar 31 11:33:49 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ext/date/date_core.c (d_lite_strftime, dt_lite_strftime): [DOC]
	  fix indent not to be a big sole verbatim.

Thu Mar 31 11:18:19 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ext/date/date_core.c (Init_date_core): [DOC] fix misplaced doc
	  of DateTime.  [ruby-core:74729] [Bug #12233]

Thu Mar 31 03:41:02 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* ext/nkf/nkf-utf8/nkf.c: Merge upstream 69f7e74dde.
	  fix indent.

Wed Mar 30 16:33:19 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* extension.rdoc, extension.ja.rdoc: [DOC] Fix some errors.
	  Renamed files, wrong method names or argument types; the example
	  GetDBM macro is now updated to the current version of the actual
	  code.  patch by Marcus Stollsteimer in [ruby-core:74690].
	  [Bug #12228]

Wed Mar 30 09:46:01 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* lib/open-uri.rb: Use `userinfo` for authenticated proxy.
	  [fix GH-1148] Patch by @SokichiFujita
	* test/open-uri/test_open-uri.rb: ditto.
	  [fix GH-1309] Patch by @jdamick

Wed Mar 30 01:56:06 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* ext/nkf/nkf-utf8/nkf.c: Merge upstream 4f3edf80a0.
	  patched by Anton Sivakov [Bug #12201] [Bug #12202]

Wed Mar 30 01:54:30 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* tool/redmine-backporter.rb: add given revision to current changesets
	  on associating the revision to the related ticket.

Wed Mar 30 01:53:17 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* tool/merger.rb: update revision.h before merge.

Tue Mar 29 19:33:54 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* addr2line.c: define toupper for its use. fix r54391.

Tue Mar 29 19:23:46 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* include/ruby/ruby.h (rb_isupper, rb_islower, rb_isalpha, rb_isdigit,
	  rb_isalnum, rb_isxdigit, rb_isblank, rb_isspace, rb_isblank,
	  rb_iscntrl, rb_isprint, rb_ispunct, rb_isgraph,
	  rb_tolower, rb_toupper): use inline function to avoid function call.

	* include/ruby/ruby.h (rb_isascii): use inline function to clarify
	  the logic.

Tue Mar 29 18:56:55 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* tool/redmine-backporter.rb (backport): show merger.rb's path.

	* tool/redmine-backporter.rb (show): show current issue again if no
	  ticket number is given.

	* tool/redmine-backporter.rb (rel): show error message if current
	  bugs.ruby-lang.org doesn't support the API.

Tue Mar 29 18:54:34 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* tool/merger.rb: support to backport header as backport identifier.
	  Now you can specify by 'merge revision(s) 49254: [Backport #10738]'.

Tue Mar 29 16:53:44 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode/case-folding.rb, casefold.h: Tweaked handling of 6
	  special cases in CaseUnfold_11_Table.

	* enc/unicode.c: Adjustments for above.

	* test/ruby/enc/test_case_mapping.rb: Tests for the above: Some tests in
	  test_titlecase activated; test_greek added. A test in test_cherokee fixed.

Tue Mar 29 13:31:00 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode.c: Cleaned up some comments.

Tue Mar 29 13:24:56 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode/case-folding.rb, casefold.h: Removing data for idempotent
	  titlecasing.

	* enc/unicode.c: Adjust code to data removal.

Tue Mar 29 12:45:18 2016  Laurent Arnoud  <laurent@spkdev.net>

	* lib/webrick/httpresponse.rb: Move error_body to method. It allow to
	  override the body more easily. [fix GH-1307]
	* test/webrick/test_httpresponse.rb: ditto.

Tue Mar 29 06:40:53 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* error.c (rb_compile_err_append): rb_thread_t::base_block is no
	  longer used.

	* iseq.c (rb_iseq_compile_with_option): ditto, no protection is
	  needed.

Tue Mar 29 06:39:22 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* parse.y (struct parser_params): move parse_in_eval flag from
	  rb_thread_t.

	* parse.y (rb_parser_set_context): set parsing context, not only
	  mild error flag.

	* iseq.c (rb_iseq_compile_with_option): the parser now refers no
	  thread local states to be restored.

	* vm_eval.c (eval_string_with_cref): ditto.

Mon Mar 28 21:24:02 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>

	* numeric.c (int_pos_p): fix typos.

Mon Mar 28 14:54:49 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode.c: Refactoring in preparation for data reduction for
	  titlecase.

Mon Mar 28 14:36:36 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode.c: Minor refactoring for I WITH DOT ABOVE.

Mon Mar 28 14:26:24 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode.c: Removed code now covered by data from table.

Mon Mar 28 11:49:21 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode.c: Adding comments. [ci skip]

Mon Mar 28 11:30:23 2016  Shinichi Maeshima  <netwillnet@gmail.com>

	* lib/rubygems.rb: Fix `Gem.find_spec_for_exe` picks oldest gem.
	  https://github.com/travis-ci/travis-ci/issues/5798
	  https://github.com/rubygems/rubygems/pull/1566
	* test/rubygems/test_gem.rb: ditto.

Mon Mar 28 11:26:31 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* lib/rubygems.rb, lib/rubygems/*, test/rubygems/*: Update rubygems-2.6.2.
	  Please see entries of 2.6.2 on
	  https://github.com/rubygems/rubygems/blob/master/History.txt

Mon Mar 28 11:02:31 2016  Hiroshi Shirosaki  <h.shirosaki@gmail.com>

	* lib/rubygems/test_case.rb: Fix test on Windows for inconsistent temp path.
	  https://github.com/rubygems/rubygems/pull/1554
	  [Bug #12193][ruby-core:74431]

Mon Mar 28 08:19:49 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* sprintf.c (rb_str_format): refactor floating point format of
	  Rational by using generic Integer functions.

	* sprintf.c (rb_str_format): fix buffer overflow, length must be
	  greater than precision.  reported by William Bowling <will AT
	  wbowling.info>.

Sun Mar 27 12:13:37 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* sprintf.c (rb_str_format): convert Rational to floating point
	  format by using generic Integer functions, not by methods which
	  can be overwritten.

Sat Mar 26 10:55:12 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* numeric.c (rb_int2str): conversion function to String for
	  generic Integer.

	* numeric.c (rb_int_round): rounding function for generic
	  Integers.

	* numeric.c (rb_int_{uminus,plus,minus,mul,idiv,modulo}): basic
	  arithmetic functions for generic Integers.

	* numeric.c (FIXNUM_{POSITIVE,NEGATIVE,ZERO}_P): predict macros
	  only for Fixnum.

Sat Mar 26 06:34:24 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* localeinit.c (rb_locale_charmap_index): fix prototype.
	  patched by Andreas Schwab [Bug #12218]

Fri Mar 25 16:40:48 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* test/ruby/enc/test_case_mapping.rb: Additional tests title case;
	  some not yet activated.

Fri Mar 25 13:38:11 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* ext/openssl/extconf.rb: check SSL_CTX_set_next_proto_select_cb
	  function rather than OPENSSL_NPN_NEGOTIATED macro. it exists
	  even if it is disabled by OpenSSL configuration.
	  [ruby-core:74384] [Bug #12182]

	* ext/openssl/ossl_ssl.c: update #ifdef(s) as above.

	* test/openssl/test_ssl.rb: skip NPN tests if NPN is disabled.

Fri Mar 25 11:08:37 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* lib/uri/http.rb (URI::HTTP#initialize): [DOC] fix example,
	  missing mandatory arguments.  [ruby-core:74540] [Bug #12215]

Fri Mar 25 01:50:58 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* thread_pthread.c (reserve_stack): fix reserving position where
	  the stack growing bottom to top. [Bug #12118]

Fri Mar 25 01:10:42 2016  Sebastian Schuberth  <sschuberth@gmail.com>

	* lib/mkmf.rb (find_executable0): On Windows, it is actually valid
	  to surround individual PATH directory entries with double
	  quotes. Remove these before joining the path as otherwise the
	  literal quotes would become part of the path, resulting in the
	  executable not to be found.  [Fix GH-1305]

Thu Mar 24 22:38:18 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* strftime.c (FMT, FMTV): remove recursive-assignments to get rid
	  of undefined behavior.  [ruby-core:74532] [Bug #12213]

Thu Mar 24 17:44:02 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* strftime.c (FMT_PADDING): extract format for padding.

	* strftime.c (FMT_PRECISION): extract precision formula.

	* strftime.c (FMTV): append formatted string to expand the result.

Thu Mar 24 14:20:21 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* strftime.c (STRFTIME): deal with case conversion flags for
	  recursive formats.

Thu Mar 24 12:43:26 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ext/date/date_core.c (dt_lite_iso8601): strftimev() always
	  returns a String, so append them directly.

	* ext/date/date_core.c (d_lite_jisx0301, iso8601_timediv),
	  (dt_lite_jisx0301): format by the format string in local buffer
	  to prevent intermediate strings from GC.

	* ext/date/date_core.c (mk_inspect_raw, mk_inspect): inspect by
	  "%+"PRIsVALUE, to prevent intermediate strings from GC.

Thu Mar 24 11:43:32 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* strftime.c (rb_strftime_with_timespec): remove unnecessary
	  check, as `s` equals to `endp` when recursed STRFTIME resized
	  the capacity same as the size.

Wed Mar 23 21:48:00 2016  Kenta Murata  <mrkn@mrkn.jp>

	* enum.c (ary_inject_op): put subtract operation out of if-clause.

Wed Mar 23 21:38:00 2016  Kenta Murata  <mrkn@mrkn.jp>

	* enum.c (ary_inject_op): Use Kahan's compensated summation algorithm
	  for summing up float values.

Wed Mar 23 20:56:59 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* strftime.c (rb_strftime_with_timespec): append formatted results
	  to the given string with expanding, and also deal with NUL chars.

	* strftime.c (rb_strftime, rb_strftime_timespec): return formatted
	  string, not the length put in the given buffer.

	* time.c (rb_strftime_alloc): no longer needs to retry with
	  reallocating buffers.

	* time.c (time_strftime): no longer needs to split by NUL chars.

Wed Mar 23 14:23:54 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* lib/rdoc/ri/driver.rb (interactive): rescue NotFoundError raised in
	  expand_name. (display_name rescues NotFoundError by itself,
	  the original logic looks buggy...)

Wed Mar 23 11:44:53 2016  cremno  <cremno@mail.ru>

	* marshal.c (r_long): cast to `signed char`, which is used
	  already, instead of SIGN_EXTEND_CHAR.

	* parse.y: SIGN_EXTEND_CHAR is no longer used.  [Fix GH-1302]

Wed Mar 23 11:38:47 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* cygwin/GNUmakefile.in (MSYS2_ARG_CONV_EXCL_PARAM):
	  * add missing parentheses and remove double quotes.
	  * rename to get rid of recursive references.
	  * as --excludes-dir option is for a path name, its argument
	    should be converted.
	  [ruby-dev:49526] [Bug #12199]

Wed Mar 23 10:39:38 2016  Koichi ITO  <koic.ito@gmail.com>

	* variable.c: Added documentation about order of `Module#constants`
	  [ci skip][Bug #12121][ruby-dev:49505][fix GH-1301]

Tue Mar 22 21:08:30 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* include/ruby/oniguruma.h: Additional flag for characters that are titlecase.

	* enc/unicode/case-folding.rb, casefold.h: Using above flag in data.

	* enc/unicode.c: Marking capitalized character as unmodified if it is
	  already titlecase.

	* test/ruby/enc/test_case_mapping.rb: Tests for above functionality.

Tue Mar 22 14:18:59 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* parse.y (lambda_body, parser_yylex): warn mismatched indentation
	  of lambda block.

Tue Mar 22 11:36:49 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* time.c (wmul): wrong condition.
	  fixed many test failures on 32bit and LLP64 platforms.

Tue Mar 22 10:31:34 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* time.c (wdiv, wmod): wdivmod0() assumes the 3rd and the 4th arguments
	  are valid pointers.
	  maybe checking them in wdivmod0() is better manner, but I guess that
	  passing real dummy pointers may be faster than checking and branching
	  in wdivmod0().
	  this commit fixes SEGV on 32bit and LLP64 platforms.

Tue Mar 22 10:24:04 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* time.c (divmodv): void function never returns any value.

Tue Mar 22 10:11:01 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* test/lib/test/unit.rb (Test::Unit::StatusLine#failed): print
	  failed messages only if replacing mode, otherwise defer them
	  until the end, to get rid of interleaving failures with progress
	  messages.  refix r54195.

Tue Mar 22 03:45:03 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* time.c (MUL_OVERFLOW_FIXWV_P): defined for FIXWV.

	* time.c (wmul): use MUL_OVERFLOW_FIXWV_P and only switch.

	* time.c (wmul): use mul which has Fixnum optimization.

	* time.c (rb_time_magnify): If WIDEVALUE_IS_WIDER, wmul() has the same
	  optimized logic, else mul() has also the similar logic for Fixnum.

	* time.c (rb_time_unmagnify): almost ditto.

Tue Mar 22 03:10:09 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* time.c (divmodv): add the case both arguments are Fixnum.

	* time.c (wquo): use quo which has Fixnum optimization.

	* time.c (wdivmod0): added for WIDEVALUE_IS_WIDER.

	* time.c (wdivmod): use wdivmod0 and divmodv.
	  divmodv has Fixnum optimization.

	* time.c (wdiv): use wdivmod0 and div to avoid the use of divmodv which
	  calls id_quo whose return value is array.

	* time.c (wmod): use wdivmod0 and mod to avoid the use of divmodv which
	  calls id_quo whose return value is array.

Mon Mar 21 22:32:50 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* internal.h (rb_fix_divmod_fix): like r54213, use FIX2NUM only if
	  x == FIXNUM_MIN && y == -1. This must be a rare case and it is
	  expected compiler to handle well.

Mon Mar 21 22:15:11 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* time.c (mod): Add Fixnum case.

	* time.c (quo): c can be Fixnum except a == FIXNUM_MIN && b == -1.
	  Such case can be optimized out because quo()'s argument is constant.

Mon Mar 21 22:09:24 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* internal.h (rb_fix_mul_fix): multiply converted values, not
	  object VALUEs.

Mon Mar 21 20:18:29 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* common.mk (TEST_EXCLUDES, EXCLUDE_TESTFRAMEWORK): use full spell
	  long option.

	* cygwin/GNUmakefile.in (MSYS2_ARG_CONV_EXCL): suppress path name
	  conversions by msys2.  [ruby-dev:49525] [Bug #12199]

Mon Mar 21 19:09:31 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* string.c (enc_succ_alnum_char): try to skip an invalid character
	  gap between GREEK CAPITAL RHO and SIGMA.
	  [ruby-core:74478] [Bug #12204]

Mon Mar 21 18:55:49 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* node.c (rb_gc_mark_node): NODE_MATCH2 can have nd_args, u3,
	  since r54100.

Sun Mar 20 21:17:13 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* internal.h (rb_int128t2big): declare only when HAVE_INT128_T.
	  fixed a compile error with VC++ introduced at r54203.

Sun Mar 20 20:10:14 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* internal.h (DLONG): defined if long is 32bit (and LONG_LONG is 64bit;
	  but LONG_LONG is always defined as 64bit), or there's int128_t.

	* internal.h (DL2NUM): defined if DLONG is defined.

	* internal.h (rb_fix_mul_fix): defined for `Fixnum * Fixnum`.

	* insns.def (opt_mul): use rb_fix_mul_fix().

	* numeric.c (fix_mul): ditto.

	* time.c (mul): ditto.

Sun Mar 20 18:53:49 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* numeric.c (fix_gt, fix_ge, fix_lt, fix_le): optimize comparisons
	  Fixnum against Bignum by rb_big_cmp in inversed order without
	  new Bignum instance.

Sun Mar 20 18:44:52 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* time.c (add): remove FIXABLE() which is in LONG2NUM().

	* time.c (sub): ditto.

	* time.c (mul): ditto.

Sun Mar 20 04:46:02 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* bignum.c (rb_big_cmp): reduce the code.

	* bignum.c (rb_big_eq): If normalized bignum is still bignum,
	  it must be larger than fixnum.

Sat Mar 20 00:58:00 2016  Kenta Murata  <mrkn@mrkn.jp>

	* include/ruby/intern.h (rb_big_odd_p, rb_big_even_p): move to
	  internal.h so that they are exported only for ruby itself.

	* internal.h (rb_big_odd_p, rb_big_even_p): ditto.

Sat Mar 19 21:56:23 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* test/lib/test/unit.rb (Test::Unit::StatusLine#failed): defer
	  failed messages until the end in verbose mode, to get rid of
	  interleaving failures with progress messages.

Sat Mar 19 21:53:35 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* numeric.c (fix_cmp): invert the result as the comparison is
	  inverted.

Sat Mar 19 18:32:00 2016  Kenta Murata  <mrkn@mrkn.jp>

	* numeric.c (int_to_f): raise NotImplementedError when a receiver
	  class is unknown.

	* test/-ext-/integer/test_my_integer.rb (test_my_integer_to_f): modify
	  a test for the above change.

Sat Mar 19 18:21:00 2016  Kenta Murata  <mrkn@mrkn.jp>

	* bignum.c (Bignum#<=>): remove it because they are unified with
	  Integer#<=>.

	* numeric.c (Integer#<=>, Fixnum#<=>): move <=> method from Fixnum to
	  Integer.

	* numeric.c (int_cmp): add this method for Integer#<=>.

	* test/-ext-/integer/test_my_integer.rb (test_my_integer_cmp): add a
	  test to examine Integer#<=> for unknown subclasses.

Sat Mar 19 14:46:18 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* iseq.c (rb_iseq_compile_with_option): make the parser in mild
	  error.

	* load.c (rb_load_internal0): ditto.

	* parse.y (yycompile0): return the error message within the error
	  to be raised.  [Feature #11951]

	* parse.y (parser_compile_error): accumulate error messages in the
	  error_buffer.

Sat Mar 19 03:57:13 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* time.c (LOCALTIME): organize #ifdefs.

	* time.c (GMTIME): define only ifndef HAVE_STRUCT_TM_TM_GMTOFF.

Sat Mar 19 03:53:31 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* configure.in (rb_cv_member_struct_tm_tm_gmtoff): For Linux (glibc)
	  define _BSD_SOURCE for time.h to define struct tm.tm_gmtoff.

	* time.c: define _BSD_SOURCE at the top.

Sat Mar 19 03:00:50 2016  Rei Odaira  <Rei.Odaira@gmail.com>

	* test/-ext-/time/test_new.rb (test_timespec_new): change a gmtoff
	  test to a better one that does not depend on whether the current
	  time is in summer time or not.

Fri Mar 19 00:00:00 2016  Kenta Murata  <mrkn@mrkn.jp>

	* bignum.c (rb_big_to_f, Bignum#to_f): removed them because they are
	  unified with int_to_f and Integer#to_f.

	* numeric.c (int_to_f): treat Bignum values directly.

Fri Mar 18 23:41:00 2016  Kenta Murata  <mrkn@mrkn.jp>

	* numeric.c (int_to_f, fix_to_f): rename fix_to_f to int_to_f, and add
	  treatment for subclasses which don't have definitions of to_f method.

	* numeric.c (Integer#to_f, Fixnum#to_f): move to_f method from Fixnum
	  to Integer.

	* ext/-test-/integer/my_integer.c: define helper class for testing
	  to_f method for a subclass of Integer.

	* ext/-test-/integer/extconf.rb: ditto.

	* ext/-test-/integer/init.c: ditto.

	* test/-ext-/integer/test_my_integer.rb: examine to_f method for a
	  subclass of Integer.

Fri Mar 18 22:32:00 2016  Kenta Murata  <mrkn@mrkn.jp>

	* include/ruby/intern.h (rb_big_hash): Move to internal.h.

	* internal.h: ditto.

Fri Mar 18 22:10:00 2016  Kenta Murata  <mrkn@mrkn.jp>

	* bignum.c (Bignum#eql?): remove its definition because it is unified
	  with Numeric#eql?.

	* numeric.c (num_eql): treat Bignum values directly.

Fri Mar 18 21:57:00 2016  Kenta Murata  <mrkn@mrkn.jp>

	* bignum.c (rb_big_to_s, Bignum#to_s): remove its definition because
	  it is unified with Integer#to_s.

	* numeric.c (int_to_s): treat Bignum values directly.

Fri Mar 18 21:30:00 2016  Kenta Murata  <mrkn@mrkn.jp>

	* numeric.c (int_to_s): Move from fix_to_s.

	* numeric.c (Integer#to_s): Move from Fixnum#to_s.

Fri Mar 18 16:22:24 2016  Victor Nawothnig  <Victor.Nawothnig@gmail.com>

	* parse.y (parse_numvar): NTH_REF must be less than a half of
	  INT_MAX, as it is left-shifted to be ORed with back-ref flag.
	  [ruby-core:74444] [Bug#12192] [Fix GH-1296]

Fri Mar 18 12:25:30 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* gc.c (tick): fix missing close parenthesis.  [Fix GH-1291]

Fri Mar 18 10:24:12 2016  Naotoshi Seo  <sonots@gmail.com>

	* ext/date/date_core.c (datetime_to_time): preserve timezone info
	  [Bug #12189] [Fix GH-1295]

Fri Mar 18 10:17:00 2016  Kenta Murata  <mrkn@mrkn.jp>

	* bignum.c (rb_big_hash): make it public function to be available in
	  other source files, and remove documentation comment for Bignum#hash.

	* bignum.c (Bignum#hash): remove its definition because it is unified
	  with Object#hash.

	* include/ruby/intern.h (rb_big_hash): add a prototype declaration.

	* hash.c (any_hash): treat Bignum values directly.

Fri Mar 18 02:35:12 2016  Naotoshi Seo  <sonots@gmail.com>

	* lib/time.rb (parse, strptime): Fix Time.parse/strptime does not
	  have compatibility with DateTime.parse/strptime in terms of parsing
	  timezone [Bug #12190] [Fix GH-1297]

Fri Mar 18 02:17:00 2016  Kenta Murata  <mrkn@mrkn.jp>

	* numeric.c (fix_zero_p, fix_even_p, fix_odd_p): remove needless
	  functions.

Fri Mar 18 02:15:00 2016  Kenta Murata  <mrkn@mrkn.jp>

	* numeric.c (int_even_p): treat Fixnum and Bignum values directly.

Fri Mar 18 02:07:00 2016  Kenta Murata  <mrkn@mrkn.jp>

	* bignum.c (Bignum#even?, Bignum#odd?): remove definitions
	  because they are unified with Integer#even? and Integer#odd?.

	* numeric.c (Fixnum#zero?, Fixnum#even?, Fixnum#odd?): remove
	  definitions because they are unified with Numeric#zero?,
	  Integer#even?, and Integer#odd?.

	* numeric.c (num_zero_p, int_odd_p): treat Fixnum and
	  Bignum values directly.

	* test/ruby/test_integer.rb (test_odd_p_even_p): remove meaningless
	  test case.

Fri Mar 18 01:51:00 2016  Kenta Murata  <mrkn@mrkn.jp>

	* bignum.c (rb_big_even_p, rb_big_odd_p): make them public functions
	  to be available in other source files.

	* include/ruby/intern.h (rb_big_even_p, rb_big_odd_p): add prototype
	  declarations.

Fri Mar 18 00:25:56 2016  Tanaka Akira  <akr@fsij.org>

	* enum.c (ary_inject_op): Implement the specialized code for sum of
	  float numbers.

Fri Mar 18 00:15:05 2016  Yusuke Endoh  <mame@ruby-lang.org>

	* numeric.c (num_step): use rb_equal for zero check. rb_num_coerce_cmp
	  created an object which caused extra overhead.

Thu Mar 17 22:21:34 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* include/ruby/ruby.h (RB_GC_GUARD_PTR): remove intermediate
	  macro, and expand for each RB_GC_GUARD.  [Fix GH-1293]

Thu Mar 17 22:08:33 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* compile.c (iseq_specialized_instruction): move specialization
	  for opt_newarray_max/min from translation phase.

Thu Mar 17 21:52:09 2016  Yusuke Endoh  <mame@ruby-lang.org>

	* array.c, enum.c: make rdoc format consistent.

Thu Mar 17 21:47:57 2016  Yusuke Endoh  <mame@ruby-lang.org>

	* NEWS: add Array#max, #min, and the optimization.  [Feature #12172]

Thu Mar 17 21:45:02 2016  Yusuke Endoh  <mame@ruby-lang.org>

	* compile.c (NODE_CALL): add optimization shortcut for Array#max/min.
	  Now `[x, y].max` is optimized so that a temporal array object is not
	  created in some condition.  [Feature #12172]

	* insns.def (opt_newarray_max, opt_newarray_min): added.

Thu Mar 17 21:35:52 2016  Yusuke Endoh  <mame@ruby-lang.org>

	* array.c (rb_ary_max, rb_ary_min): implement Array#max and min with
	  arguments.  replace super call with rb_nmin_run.

	* enum.c (nmin_run): exported (as rb_nmin_run).

	* internal.h: added a prototype for rb_nmin_run.

Thu Mar 17 21:24:52 2016  Yusuke Endoh  <mame@ruby-lang.org>

	* array.c (rb_ary_max, rb_ary_min): implement a block by itself instead
	  of delegating Enumerable#max/min.

Thu Mar 17 21:09:34 2016  Yusuke Endoh  <mame@ruby-lang.org>

	* array.c (rb_ary_max, rb_ary_min): Array#max and Array#min added.
	  [Feature #12172]

	* internal.h (OPTIMIZED_CMP): moved from enum.c so that array.c can
	  use it.

	* test/ruby/test_array.rb (test_max, test_min): tests for Array#max
	  and Array#min.

	* test/ruby/test_enum.rb (test_max, test_min): revised a bit to test
	  Enumerable#max and #min explicitly.

Thu Mar 17 21:02:42 2016  Yusuke Endoh  <mame@ruby-lang.org>

	* internal.c: struct cmp_opt_data added for refactoring out a data
	  structure for CMP_OPTIMIZABLE

	* array.c (struct ary_sort_data): use struct cmp_opt_data.

	* enum.c (struct min_t, max_t, min_max_t): use struct cmp_opt_data.

Thu Mar 17 20:55:21 2016  Tanaka Akira  <akr@fsij.org>

	* enum.c (ary_inject_op): Extracted from enum_inject.

Thu Mar 17 18:39:04 2016  Tanaka Akira  <akr@fsij.org>

	* enum.c (enum_inject): Implement the specialized code for sum of
	  integers including Bignums.

	* internal.h (rb_fix_plus): Declared to be usable from enum_inject.

	* numeric.c (rb_fix_plus): Defined.

Thu Mar 17 17:20:28 2016  Anton Davydov  <antondavydov.o@gmail.com>

	* thread_sync.c: [DOC] Update documentation for Queue class
	  description.  [Fix GH-1292]

Thu Mar 17 17:14:51 2016  Dinar Valeev  <dvaleev@suse.com>

	* gc.c (tick): Use __builtin_ppc_get_timebase for POWER arch.
	  [Fix GH-1291]

Thu Mar 17 11:51:48 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* lib/securerandom.rb (gen_random): to avoid blocking on Windows.
	  On Windows OpenSSL RAND_bytes (underlying implementation is
	  RAND_poll in crypto/rand/rand_win.c) may be blocked at
	  NetStatisticsGet.
	  https://wiki.openssl.org/index.php/Random_Numbers#Windows_Issues
	  Instead of this, use Random.raw_seed directly (whose implementation
	  CryptGenRandom is one of the source of
	  entropy of RAND_poll on Windows).
	  https://wiki.openssl.org/index.php/Random_Numbers
	  Note: CryptGenRandom function is PRNG and doesn't check its entropy,
	  so it won't block. [Bug #12139]
	  https://msdn.microsoft.com/ja-jp/library/windows/desktop/aa379942.aspx
	  https://tools.ietf.org/html/rfc4086#section-7.1.3
	  https://eprint.iacr.org/2007/419.pdf
	  http://www.cs.huji.ac.il/~dolev/pubs/thesis/msc-thesis-leo.pdf

Thu Mar 17 12:09:00 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode.c: Fixed two macro definitions.
	* test/ruby/enc/test_case_mapping.rb: Test cases that detected
	  the above bugs.

Thu Mar 17 11:36:27 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ext/socket/option.c (inspect_tcpi_msec): more accurate condition
	  for TCPI msec member inspection function.
	  [ruby-core:74388] [Bug #12185]

Thu Mar 17 08:13:43 2016  Rei Odaira  <Rei.Odaira@gmail.com>

	* test/-ext-/time/test_new.rb (test_timespec_new): Time#gmtoff values
	  are the same only when both or neither of the Time objects are in
	  summer time (daylight-saving time).

Thu Mar 17 07:17:36 2016  Eric Hodel  <drbrain@segment7.net>

	* marshal.c (r_object0):  raise ArgumentError when linking to undefined
	  object.

Thu Mar 17 00:45:00 2016  Kenta Murata  <mrkn@mrkn.jp>

	* test/ruby/test_bignum.rb: Make sure to use Bignum values in the tests.

Wed Mar 16 23:10:25 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* defs/keywords (alias, undef): symbol literals are allowed.

	* parse.y (parse_percent): should parse symbol literals for alias
	  and undef.  [ruby-dev:47681] [Bug #8851]

Wed Mar 16 21:39:39 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* enc/trans/JIS: update Unicode's notice. [Bug #11844]

Wed Mar 16 20:03:35 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* proc.c (proc_binding): proc from symbol can not make a binding.
	  [ruby-core:74100] [Bug #12137]

Wed Mar 16 18:42:45 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* test/ruby/enc/test_case_mapping.rb: Fixed and activated a test for Cherokee.

Wed Mar 16 17:58:56 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* test/ruby/enc/test_case_mapping.rb: Fixed a logical error.

Wed Mar 16 17:57:34 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* test/ruby/enc/test_case_mapping.rb: Adding tests for Cherokee.
	  One test not yet working.
	  (with Kimihito Matsui)

Wed Mar 16 15:44:05 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* test/ruby/enc/test_case_mapping.rb: Adding tests for actual Unicode
	  case mapping. Fixing some aliasing issues.
	  (with Kimihito Matsui)

Tue Mar 15 21:38:28 2016  Tanaka Akira  <akr@fsij.org>

	* enum.c (enum_inject): Consider redefinition of Fixnum#+.
	  [ruby-dev:49510] [Bug#12178] Reported by usa.

Tue Mar 15 20:32:57 2016  Tanaka Akira  <akr@fsij.org>

	* enum.c (enum_inject): Implement the specialized code for :+ operator
	  for Fixnums.

Tue Mar 15 20:21:01 2016  Tanaka Akira  <akr@fsij.org>

	* enum.c (enum_inject): Implement the specialized code for self is an
	  array and a symbol operator is given.

Tue Mar 15 16:29:51 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode.c: Eliminating common code.
	  (with Kimihito Matsui)

Tue Mar 15 16:17:09 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode.c: Expansion of some code repetition in preparation for
	  elimination of common code pieces.
	  (with Kimihito Matsui)

Tue Mar 15 13:49:23 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode.c: Additional macros and code to use mapping data in
	  CaseMappingSpecials array.
	  (with Kimihito Matsui)

Tue Mar 15 13:41:22 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* internal.h (rb_gc_mark_global_tbl): should be private,
	  but was accidentally exported.

Tue Mar 15 12:51:06 2016  Marcus Stollsteimer  <sto.mar@web.de>

	* doc/extension.ja.rdoc: Fix RDoc markup in doc/extension*.rdoc.
	  [ci skip][Bug #12143][ruby-core:74143]
	* doc/extension.rdoc: ditto.

Tue Mar 15 09:56:41 2016  Ian Lunderskov  <ian.lunderskov@gmail.com>

	* time.c: Minor typo in Time#dst? documentation.
	  [ci skip][fix GH-1290]

Tue Mar 15 04:36:41 2016  Charles Oliver Nutter  <headius@headius.com>

	* test/ruby/test_rubyoptions.rb (test_disable): add tests for
	  --disable-gems and --disable-did_you_mean.

Tue Mar 15 03:35:04 2016  Eric Hodel  <drbrain@segment7.net>

	* marshal.c (r_object0): Fix Marshal crash for corrupt extended object.

Tue Mar 15 01:22:27 2016  Charles Oliver Nutter  <headius@headius.com>

	* test/ruby/test_rubyoptions.rb: make version matching support
	  JRuby's version output.

Mon Mar 14 19:05:39 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* bignum.c (big2str_2bdigits): reduce div instruction.

Mon Mar 14 18:39:53 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* include/ruby/oniguruma.h, enc/unicode.c: Adjusting flag assignments
	  and macros to work with unified CaseMappingSpecials array.
	  (with Kimihito Matsui)

Mon Mar 14 16:53:37 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* compile.c (compile_named_capture_assign): optimize named capture
	  assignments, by replacing repeating global variable accesses
	  with `dup`, and by returning the matched result instead of
	  re-getting it from the MatchData.

	* parse.y (reg_named_capture_assign_gen): build just assignment
	  nodes for the optimization.

Mon Mar 14 16:02:59 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* file.c (ruby_is_fd_loadable): now return -1 if loadable but
	  may block.

	* ruby.c (open_load_file): wait to read by the result of
	  ruby_is_fd_loadable, without fstat.

Mon Mar 14 13:38:38 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* numeric.c (fix2str): improve r54092 like rb_int2big().

Mon Mar 14 10:02:23 2016  Eric Wong  <e@80x24.org>

	* ext/openssl/ossl_ssl.c (ossl_sslctx_setup): document as MT-unsafe
	  [ruby-core:73803] [Bug #12069]

Sun Mar 13 09:43:23 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* include/ruby/win32.h (O_SHARE_DELETE): change to fit Fixnum
	  limit.  [ruby-core:74285] [Bug #12171]

Sun Mar 13 09:15:45 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* numeric.c (rb_fix2str): fix edge case, accidentally generated
	  wrong Fixnum from LONG_MIN.

Sat Mar 12 09:50:27 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* vm_eval.c (rb_f_catch): [DOC] fix malformed RDoc syntax, "+...+"
	  cannot enclose non-identifier characters.
	  a patch by Sebastian S in [ruby-core:74278].  [Bug#12170]

Sat Mar 12 02:44:48 2016  Tanaka Akira  <akr@fsij.org>

	* test/lib/test/unit.rb: describe !/REGEXP/ in the help message.

Fri Mar 11 17:03:09 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* test/lib/test/unit.rb (Options#non_options): make regexp name
	  options prefixed with "!" negative filters.

	* common.mk (TEST_EXCLUDES): use negative filter to exclude memory
	  leak tests.  -x option excludes test files, not test methods.

Fri Mar 11 16:11:27 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode/case-folding.rb, casefold.h: Streamlining approach to
	  case mapping data not available from case folding by unifying all
	  three cases (special title, special upper, special lower).
	* enc/unicode.c: Adjust macro names for above (macros are currently inactive).
	  (with Kimihito Matsui)

Thu Mar 10 17:34:16 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* iseq.c (prepare_iseq_build): enable coverage by coverage_enabled
	  option, not by parse_in_eval flag in the thread context.

	* iseq.h (rb_compile_option_struct): add coverage_enabled flag.

	* parse.y (yycompile0): set coverage_enabled flag if coverage
	  array is made.

Thu Mar 10 15:19:54 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* node.c (dump_option): nd_compile_option is a hidden hash object,
	  cannot call inspect on it.

Thu Mar 10 09:49:54 2016  Rei Odaira  <Rei.Odaira@gmail.com>

	* test/socket/test_socket.rb (test_udp_recvmsg_truncation):
	  AIX does not set the MSG_TRUNC flag for a message partially read
	  by recvmsg(2) with the MSG_PEEK flag set.

Wed Mar  9 16:48:45 2016  Koichi Sasada  <ko1@atdot.net>

	* benchmark/driver.rb: fix my last commit (syntax error).

Wed Mar  9 16:41:44 2016  Koichi Sasada  <ko1@atdot.net>

	* benchmark/driver.rb: fix output messages.

	* benchmark/memory_wrapper.rb: use respond_to? because
	  member? does not work well.

Wed Mar  9 16:20:25 2016  Koichi Sasada  <ko1@atdot.net>

	* benchmark/driver.rb: support memory usage benchmark.
	  use `--measure-target=[target]'.
	  Now, we can use the following targets:
	    * real (default): real time which returns process time in sec.
	    * peak: peak memory usage (physical memory) in bytes.
	    * size: last memory usage (physical memory) in bytes.

	* benchmark/memory_wrapper.rb: ditto.

Wed Mar  9 15:04:22 2016  Koichi Sasada  <ko1@atdot.net>

	* benchmark/bm_vm3_gc_old_full.rb: add GC.start benchmark.

	* benchmark/bm_vm3_gc_old_immediate.rb: ditto.

	* benchmark/bm_vm3_gc_old_lazy.rb: ditto.

Wed Mar  9 14:59:11 2016  Koichi Sasada  <ko1@atdot.net>

	* benchmark/driver.rb: exit benchmarking if a benchmark process
	  receives signals.

Wed Mar  9 13:22:49 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* test/lib/memory_status.rb: make Memory::Status independent of
	  MiniTest::Skip.

	* test/lib/test/unit/assertions.rb (assert_no_memory_leak): skip
	  if Memory::Status is not available.

Wed Mar  9 09:19:55 2016  Rei Odaira  <Rei.Odaira@gmail.com>

	* test/io/wait/test_io_wait.rb (test_wait_readwrite_timeout):
	  select(2) in AIX returns "readable" for the write-side fd
	  of a pipe, so it is not possible to use a pipe to test
	  the read-write timeout of IO#wait on AIX.

Wed Mar  9 03:35:22 2016  Charles Oliver Nutter  <headius@headius.com>

	* test/ruby/test_require.rb (test_require_with_loaded_features_pop):
	  Only remove PATH so threads don't accidentally double-pop.

Wed Mar  9 00:29:46 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* vm_method.c (rb_alias): the original name should be properly
	  available method_added method, set the name before calling the
	  hook.

Wed Mar  9 00:07:03 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* lib/logger.rb (Logger::LogDevice#initialize): define using
	  keyword arguments.

Tue Mar  8 23:37:07 2016  Charles Oliver Nutter  <headius@headius.com>

	* test/ruby/test_array.rb: split out the test for no stack error
	  on large input for test_permutation, test_repeated_permutation,
	  and test_repeated_combination, and make them all timeout:30.

Tue Mar  8 17:20:21 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* intern.h (rb_divmod): assume compilers `/` and `%` comply C99
	  and reduce branching. If a compiler doesn't comply, add #ifdefs.

	* intern.h (rb_div): added for Ruby's behavior.

	* intern.h (rb_mod): added for Ruby's behavior.

	* insns.def (opt_div): use rb_div.

	* insns.def (opt_mod): use rb_mod.

	* numeric.c (fixdivmod): removed.

	* numeric.c (fix_divide): use rb_div.

	* numeric.c (fix_mod): use rb_mod.

	* numeric.c (fix_divmod): use rb_divmod.

Tue Mar  8 17:53:09 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* insns.def (opt_mod): show its method name on ZeroDivisionError.
	  [Bug #12158]

Tue Mar  8 17:33:38 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* win32/win32.c (rb_w32_write_console): now no need to check
	  ERROR_CALL_NOT_IMPLEMENTED because it is for old Win9X.

Tue Mar  8 16:54:29 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* win32/win32.c (rb_w32_write_console): stop the VT100 emulation if the
	  console supports it natively.

Tue Mar  8 08:13:01 2016  Rei Odaira  <Rei.Odaira@gmail.com>

	* test/net/imap/test_imap.rb (test_idle_timeout): Because of the
	  timeout specified in "imap.idle(0.2)", there is no guarantee that
	  the server thread has done all the work before the client thread
	  performs the assertions. It depends on the thread scheduling.
	  Add checks to avoid false positives (on AIX, particularly).

Tue Mar  8 00:42:22 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* ruby.c (warn_cr_in_shebang): meaningless check on DOSISH platforms.
	  fixed a test failure introduced at r53998.

Tue Mar  8 00:27:53 2016  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>

	* ext/tk/lib/tkextlib/tcllib/tablelist_tile.rb: fix method name typo.
	  [ruby-core:72513] [Bug #11893] The patch provided by Akira Matsuda.


Tue Mar  8 00:25:08 2016  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>

	* ext/tk/lib/tkextlib/tcllib/toolbar.rb: fix method name typo.
	  [ruby-core:72511] [Bug #11891] The patch provided by Akira Matsuda.


Tue Mar  8 00:21:58 2016  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>

	* ext/tk/lib/tkextlib/blt/tree.rb: fix method name typo.
	  [ruby-core:72510] [Bug #11890] The patch provided by Akira Matsuda.


Tue Mar  8 00:11:47 2016  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>

	* ext/tk/lib/tk/menubar.rb: fix a typo in font name. [ruby-core:72505]
	  [Bug #11886] The patch provided by Akira Matsuda.

	* ext/tk/sample/*.rb: ditto.

Mon Mar  7 13:32:58 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* class.c (rb_define_class, rb_define_class_id_under): raise
	  ArgumentError if super is 0, deprecated behavior which has been
	  warned long time.

Mon Mar  7 13:28:30 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* internal.h: move function declarations for class internals from
	  include/ruby/intern.h.

Mon Mar  7 10:58:07 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ext/win32ole/win32ole_event.c (rescue_callback): use
	  rb_write_error_str instead of rb_write_error, to respect
	  the encoding and prevent the message from GC.

	* internal.h (rb_write_error_str): export.

Mon Mar  7 01:38:41 2016  Rei Odaira  <Rei.Odaira@gmail.com>

	* test/ruby/test_process.rb (test_execopts_gid): Skip a test
	  that is known to fail on AIX. AIX allows setgid to
	  a supplementary group, but Ruby does not allow the "-e"
	  option when setgid'ed, so the test does not work as intended.

Sun Mar  6 22:43:41 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* io.c (rb_obj_display): [DOC] fix output of Array, as Array#to_s
	  is same as Array#inspect since 1.9.

Sat Mar  5 09:50:58 2016  Rei Odaira  <Rei.Odaira@gmail.com>

	* test/socket/test_addrinfo.rb (test_ipv6_address_predicates):
	  IN6_IS_ADDR_V4COMPAT and IN6_IS_ADDR_V4MAPPED are broken
	  on AIX, so skip related tests.

Sat Mar  5 09:17:54 2016  Rei Odaira  <Rei.Odaira@gmail.com>

	* test/rinda/test_rinda.rb (test_make_socket_ipv4_multicast):
	  The fifth argument to getsockopt(2) should be modified to
	  indicate the actual size of the value on return,
	  but not in AIX. This is a know bug. Skip related tests.
	* test/rinda/test_rinda.rb (test_ring_server_ipv4_multicast):
	  ditto.
	* test/rinda/test_rinda.rb (test_make_socket_unicast): ditto.
	* test/socket/test_basicsocket.rb (test_getsockopt): ditto.
	* test/socket/test_sockopt.rb (test_bool): ditto.

Sat Mar  5 07:36:27 2016  Rei Odaira  <Rei.Odaira@gmail.com>

	* test/-ext-/float/test_nextafter.rb: In AIX,
	  nextafter(+0.0,-0.0)=+0.0, and nextafter(-0.0,+0.0)=-0.0,
	  but they should return -0.0 and +0.0, respectively. This is
	  a known bug in nextafter(3) on AIX, so skip related tests.

Sat Mar  5 07:14:10 2016  Rei Odaira  <Rei.Odaira@gmail.com>

	* test/zlib/test_zlib.rb (test_adler32_combine, test_crc32_combine):
	  Skip two tests on AIX because zconf.h in zlib does not correctly
	  recognize _LARGE_FILES in AIX. The problem was already reported
	  to zlib, and skip these tests until it is fixed.

Sat Mar  5 03:07:40 2016  Rei Odaira  <Rei.Odaira@gmail.com>

	* thread_pthread.c (getstack): __pi_stacksize returned by
	  pthread_getthrds_np() is wrong on AIX. Use
	  __pi_stackend - __pi_stackaddr instead.

Fri Mar  4 19:19:42 2016  Koichi Sasada  <ko1@atdot.net>

	* gc.c: use 2 bits with unsigned int for rb_objspace::flags::mode
	  because it always returns 0 to 2 (non-negative value).

Fri Mar  4 18:42:08 2016  Koichi Sasada  <ko1@atdot.net>

	* gc.c: rename "enum gc_stat" to "enum gc_mode"
	  because there is a same name (no related) function gc_stat().

	  Also gc_stat_* are renamed to gc_mode_*,
	  gc_stat_transition() to gc_mode_transition(),
	  rb_objspace::flags::stat is renamed to rb_objspace::flags::mode.

	  Change rb_objspace::flags::mode from 2 bits to 3 bits because VC++
	  returns negative enum value with 2 bits.

	* gc.c (gc_mode): add a macro to access rb_objspace::flags::mode
	  with verification code (verification is enabled only on
	  RGENGC_CHECK_MODE > 0).

	* gc.c (gc_mode_set): same macro for setter.

Fri Mar  4 09:28:18 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* lib/rubygems.rb, lib/rubygems/*, test/rubygems/*: Update rubygems-2.6.1.
	  Please see entries of 2.6.0 and 2.6.1 on
	  https://github.com/rubygems/rubygems/blob/master/History.txt
	  [fix GH-1270] Patch by @segiddins

Thu Mar  3 14:09:00 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* lib/ostruct.rb (modifiable?, new_ostruct_member!, table!):
	  rename methods for internal use with suffixes and make private,
	  [ruby-core:71069] [Bug #11587]

Wed Mar  2 16:28:48 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* vm_eval.c (method_missing): call by found method entry and get
	  rid of searching the same method entry twice.

	* vm_eval.c (vm_call0_body): calling method_missing method is
	  method_missing().

Wed Mar  2 15:13:33 2016  herwinw  <herwin@quarantainenet.nl>

	* lib/xmlrpc.rb: Removed broken parser named XMLScanStreamParser.
	  It's not works with current Ruby version.
	  [fix GH-1271][ruby-core:59588][Bug #9369]
	* lib/xmlrpc/config.rb: ditto.
	* lib/xmlrpc/parser.rb: ditto.

Wed Mar  2 15:08:33 2016  herwinw  <herwin@quarantainenet.nl>

	* lib/xmlrpc.rb: Removed broken parser named XMLTreeParser.
	  Required gem of its parser didn't compile on newer Ruby versions.
	  [fix GH-1271][ruby-core:59590][Bug #9370]
	* lib/xmlrpc/config.rb: ditto.
	* lib/xmlrpc/parser.rb: ditto.

Tue Mar  1 11:25:48 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* lib/fileutils.rb: use keyword arguments instead of option
	  hashes.

Mon Feb 29 16:50:20 2016  hanachin  <hanachin@gmail.com>

	* array.c (rb_ary_push_m): [DOC] Remove trailing comma from
	  Array#push example, as other Array examples doesn't put trailing
	  comma.  [Fix GH-1279]

Mon Feb 29 16:31:01 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* common.mk, tool/mkconfig.rb: set cross_compiling option from
	  Makefile, but not from rbconfig.rb, which is just going to be
	  created by this command.

Sun Feb 28 23:13:49 2016  C.J. Collier  <cjcollier@linuxfoundation.org>

	* configure.in: Add summary to end of configure output.
	  [Fix GH-1275]

Sun Feb 28 20:23:36 2016  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>

	* lib/drb/drb.rb (error_print): Add verbose failure messages and
	  avoid infamous DRb::DRbConnError. [Feature #12101]

Sun Feb 28 13:40:46 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* error.c (nometh_err_initialize): add private_call? parameter.

	* error.c (nometh_err_private_call_p): add private_call? method,
	  to tell if the exception raised in private form FCALL or VCALL.
	  [Feature #12043]

	* vm_eval.c (make_no_method_exception): append private_call?
	  argument.

	* vm_insnhelper.c (ci_missing_reason): copy FCALL flag.

Sun Feb 28 10:19:47 2016  Ryan T. Hosford  <tad.hosford@gmail.com>

	* array.c (rb_ary_and): clarify that set intersection returns the
	  unique elements common to both arrays.

	* array.c (rb_ary_or): clarify that union preserves the order from
	  the given arrays.

Sat Feb 27 17:05:29 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode/case-folding.rb, casefold.h: Reducing size of TitleCase
	  table by eliminating duplicates.
	  (with Kimihito Matsui)

Fri Feb 26 14:40:48 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* numeric.c (num_step_scan_args): comparison String with Numeric
	  should raise TypeError. it is an invalid type, but not a
	  mismatch the number of arguments.  [ruby-core:62430] [Bug #9810]

Fri Feb 26 14:39:39 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* doc/extension.rdoc, doc/extension.ja.rdoc: add editor local
	  variables, with commenting out by :enddoc: directives which are
	  just ignored unless code object mode.  [Bug #12111]

Fri Feb 26 12:25:56 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* doc/extension.ja.rdoc: removed rendering error caused by editor specific
	  configuration on http://docs.ruby-lang.org/en/trunk/extension_rdoc.html .
	  [Bug #12111][ruby-core:73990]

Fri Feb 26 11:21:41 2016  herwinw  <herwin@quarantainenet.nl>

	* lib/xmlrpc.rb: Removed references to NQXML. It's obsoleted parser.
	  [fix GH-1245][ruby-core:59593][Feature #9371]
	* lib/xmlrpc/config.rb: ditto.
	* lib/xmlrpc/parser.rb: ditto.

Fri Feb 26 11:10:19 2016  Rick Salevsky  <rsalevsky@suse.com>

	* lib/tmpdir.rb: Unify to coding-style for method definition.
	  [fix GH-1252]

Fri Feb 26 11:02:04 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* README.md: update markdown syntax for anchor tag.
	  [fix GH-1265] Patch by @lukBarros

Fri Feb 26 10:52:29 2016  Alex Boyd  <alex@opengroove.org>

	* lib/irb.rb: avoid to needless truncation when using back_trace_limit option.
	  [fix GH-1205][ruby-core:72773][Bug #11969]

Fri Feb 26 08:11:58 2016  Aaron Patterson <tenderlove@ruby-lang.org>

	* gem_prelude.rb: Reduce system calls by activating the `did_you_mean`
	  gem before requiring the gem.  Activating the gem puts the gem on
	  the load path, where simply requiring the file will search every gem
	  that's installed until it can find a gem that contains the
	  `did_you_mean` file.

Thu Feb 25 19:04:13 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode/case-folding.rb: Adding possibility for debugging output
	  for TitleCase table in casefold.h.
	  (with Kimihito Matsui)

Wed Feb 24 22:31:13 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* include/ruby/oniguruma.h: Rearranging flag assignments and making
	  space for titlecase indices; adding additional macros to add or
	  extract titlecase index; adding comments for better documentation.
	* enc/unicode.c: Moving some macros to include/ruby/oniguruma.h;
	  activating use of titlecase indices.
	  (with Kimihito Matsui)

Wed Feb 24 21:03:04 2016  Tanaka Akira  <akr@fsij.org>

	* random.c (limited_rand): Add a specialized path for when the limit fits
	  in 32 bit.

Tue Feb 23 21:52:24 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode/case-folding.rb, casefold.h: Outputting actual titlecase
	  data (new table, with indices from other tables).
	* enc/unicode.c: Ignoring titlecase data indices for the moment.
	  (with Kimihito Matsui)

Tue Feb 23 15:21:14 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode/case-folding.rb, casefold.h: Reading casing data from
	  SpecialCasing.txt.
	  (with Kimihito Matsui)

Mon Feb 22 18:33:55 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode/case-folding.rb, casefold.h: Adding flag for title-case,
	  not yet operational.
	  (with Kimihito Matsui)

Mon Feb 22 18:17:03 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode/case-folding.rb, casefold.h: Fixed bug that avoided inclusion
	  of compatibility characters in upper-/lower-case mappings.
	  (with Kimihito Matsui)

Sun Feb 21 13:57:18 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* cgi/escape/escape.c: Optimize CGI.unescape performance by C ext
	  for ASCII-compatible encodings.  [Fix GH-1250]

Sun Feb 21 13:56:57 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* cgi/escape/escape.c: Optimize CGI.unescapeHTML performance by C
	  ext for ASCII-compatible encodings.  [Fix GH-1242]

Sat Feb 20 15:38:16 2016  Eric Wong  <e@80x24.org>

	* doc/extension.rdoc: update paths for defs/ directory

Sat Feb 20 14:44:15 2016  Lucas Buchala  <lucasbuchala@gmail.com>

	* vm_eval.c (rb_mod_module_eval): [DOC] Fix documentation
	  signature for Module#module_eval.  [Fix GH-1258]

Sat Feb 20 14:40:44 2016  Adam O'Connor  <northband@gmail.com>

	* README.md: a few grammatical changes to the main Ruby README.md.
	  [Fix GH-1259]

Sat Feb 20 13:04:22 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* dir.c (push_pattern, push_glob): deal with read paths as UTF-8
	  to stat later, on Windows as well as OS X.
	  [ruby-core:73868] [Bug #12081]

Sat Feb 20 01:53:33 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* object.c (rb_mod_const_get): make error message at uninterned
	  string consistent with symbols.  [ruby-dev:49498] [Bug #12089]

Fri Feb 19 23:37:52 2016  Masahiro Tomita  <tommy@tmtm.org>

	* lib/find.rb (Find#find): raise with the given path name if it
	  does not exist.  [ruby-dev:49497] [Bug #12087]

Fri Feb 19 12:44:57 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode.c: Activated use of case mapping data in CaseUnfold_11 array.
	  (with Kimihito Matsui)

Fri Feb 19 11:08:32 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ext/extmk.rb: add cygwin case, nothing excluded.
	  [ruby-core:73806] [Bug#12071]

Thu Feb 18 21:32:15 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>

	* man/irb.1: fix output in EXAMPLES.

Thu Feb 18 21:05:47 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* string.c (sym_match_m): delegate to String#match but not
	  String#=~.  [ruby-core:72864] [Bug #11991]

Thu Feb 18 14:15:38 2016  Shota Fukumori  <her@sorah.jp>

	* re.c: Add MatchData#named_captures
	  [Feature #11999] [ruby-core:72897]

	* test/ruby/test_regexp.rb(test_match_data_named_captures): Test for above.

	* NEWS: News about MatchData#named_captures.

Wed Feb 17 21:41:29 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* defs/id.def (predefined): add idLASTLINE and idBACKREF for $_
	  and $~ respectively.

	* parse.y: use idLASTLINE and idBACKREF instead of rb_intern.

Wed Feb 17 20:23:38 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* string.c (rb_str_init): fix segfault and memory leak, consider
	  wide char encoding terminator.

Wed Feb 17 12:14:59 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* string.c (rb_str_init): introduce String.new(capacity: size)
	  [Feature #12024]

Tue Feb 16 19:10:08 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode/case-folding.rb, casefold.h: Used only first element
	  (rather than all) of target in CaseUnfold_11 array.
	  (with Kimihito Matsui)

Tue Feb 16 18:24:38 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* numeric.c (compare_with_zero): fix variable name, rb_cmperr
	  requires VALUEs but not an ID.

Tue Feb 16 17:34:18 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* dir.c (rb_dir_s_empty_p): add Dir.empty? method, which tells the
	  argument is the name of an empty directory.  [Feature #10121]

Tue Feb 16 09:51:20 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* tool/rbinstall.rb (without_destdir): just strip a drive letter
	  which is prepended by with_destdir.
	  pointed out by @DavidEGrayson.
	  https://github.com/ruby/ruby/commit/0e5f9ae#commitcomment-16101763

Tue Feb 16 04:42:13 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* insns.def (opt_plus): simply use LONG2NUM() instead of wrongly
	  complex overflow case.

	* insns.def (opt_sub): ditto.

Tue Feb 16 02:49:41 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* tool/rbinstall.rb (without_destdir): compare with the destdir
	  after stripping a drive letter, on dosish platforms.
	  pointed out by @DavidEGrayson.
	  https://github.com/ruby/ruby/commit/d0cf23b#commitcomment-16100407

Mon Feb 15 15:44:09 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* parse.y (parse_ident): allow keyword arguments just after a
	  method where the same name local variable is defined.
	  [ruby-core:73816] [Bug#12073]

Mon Feb 15 14:43:28 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode/case-folding.rb: Added debugging option
	  (with Kimihito Matsui)

Sun Feb 14 17:31:50 2016  Lars Kanis  <lars@greiz-reinsdorf.de>

	* lib/mkmf.rb (with_{cpp,c,ld}flags): copy caller strings not to
	  be modified, in append_{cpp,c,ld}flags respectively.
	  [Fix GH-1246]

Sun Feb 14 16:18:57 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* eval.c (setup_exception): set the cause only if it is explicitly
	  given or not set yet.  [Bug #12068]

Sat Feb 13 21:44:58 2016  Tanaka Akira  <akr@fsij.org>

	* hash.c (rb_hash_invert): [DOC] more examples.

Sat Feb 13 17:30:49 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* lib/uri/generic.rb (URI::Generic#find_proxy): support CIDR in
	  no_proxy.  [ruby-core:73769] [Feature#12062]

Sat Feb 13 17:11:58 2016  Fabian Wiesel  <fabian.wiesel@sap.com>

	* lib/uri/generic.rb (find_proxy): exclude white-spaces and allow
	  for a leading dot in the domain name in no_proxy.
	  [ruby-core:54542] [Feature #8317]

Fri Feb 12 12:20:56 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* error.c (name_err_initialize, nometh_err_initialize): [DOC] fix
	  argument positions.  optional parameters except for the message
	  are placed at the last.

Fri Feb 12 11:49:49 2016  Anthony Dmitriyev  <antstorm@gmail.com>

	* net/ftp.rb: add NullSocket#closed? to fix closing not opened
	  connection.  [Fix GH-1232]

Fri Feb 12 11:17:38 2016  Bogdan  <bogdanvlviv@gmail.com>

	* re.c (rb_reg_initialize_m): [DOC] fix missing right bracket.
	  [Fix GH-1243]

Thu Feb 11 14:57:58 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* configure.in (RUBY_CHECK_SIZEOF, RUBY_DEFINT): fix for types
	  which are conditionally available depending on architectures
	  when universal binary, e.g., __int128.

Thu Feb 11 06:26:18 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* configure.in (RUBY_DEFINT): use Parameter Expansion.

Thu Feb 11 05:33:24 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* configure.in (int128_t): don't check HAVE_XXX (for example
	  HAVE___INT128) because RUBY_CHECK_SIZEOF() don't define it for
	  config.h and use of $ac_cv_sizeof___int128 alternates the check.
	  (and don't need to define because users shouldn't know that)

Wed Feb 10 12:03:41 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* configure.in (ARFLAGS): check if deterministic mode flag is
	  effective, which is on by default on Ubuntu.

Tue Feb  9 16:36:23 2016  Naotoshi Seo  <sonots@gmail.com>

	* lib/logger.rb: Remove block from Logger.add as it's not needed
	  patch provided by Daniel Lobato Garcia [fix GH-1240] [Bug #12054]

Tue Feb  9 14:32:23 2016  Zachary Scott  <zzak@ruby-lang.org>

	* ext/zlib/zlib.c: Document mtime header behavior with patch by @schneems
	  Fixes [GH-1129]: https://github.com/ruby/ruby/pull/1129

Tue Feb  9 13:52:49 2016  Zachary Scott  <zzak@ruby-lang.org>

	* re.c: Remove deprecated kcode argument from Regexp.new and compile
	  patch provided by Dylan Pulliam [Bug #11495]

Mon Feb  8 21:26:19 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode/case-folding.rb, enc/unicode/casefold.h: Flags for
	  upper/lower conversion added (titlecase and SpecialCasing still missing)
	  (with Kimihito Matsui)

Mon Feb  8 20:43:57 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* string.c, enc/unicode.c: Disassociating ONIGENC_CASE_FOLD flag from
	  ONIGENC_CASE_DOWNCASE.
	  (with Kimihito Matsui)

Mon Feb  8 13:00:17 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode.c: Shortened macros for enc/unicode/casefold.h to
	  single-letter; use flags in casefold.h for logic.

	* enc/unicode/case-folding.rb: Added flag for case folding.
	  Changed parameter passing.

	* enc/unicode/casefold.h: New flags added.
	  (with Kimihito Matsui)

Mon Feb  8 10:30:10 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ruby.c (feature_option): raise a runtime error if ambiguous
	  feature name is given, in the future.  [Bug #12050]

Mon Feb  8 09:43:57 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* common.mk: Removed enc/unicode/casefold.h from automatic build because
	  some CI systems don't have gperf. Creation of enc/unicode/casefold.h
	  is now possible with make unicode-up. This is intended as a temporary measure.

Sun Feb  7 22:10:08 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* common.mk: Added two more precondition files for enc/unicode/casefold.h

	* enc/unicode.c: Added shortening macros for enc/unicode/casefold.h

	* enc/unicode/case-folding.rb: Fixed file encoding for CaseFolding.txt
	  to ASCII-8BIT (should fix some ci errors). Clarified usage. Created
	  class MapItem. Partially implemented class CaseMapping.
	  (with Kimihito Matsui)

Sun Feb  7 14:12:32 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode/case-folding.rb: Fixing parameter passing.
	  (with Kimihito Matsui)

Sun Feb  7 11:44:03 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode/case-folding.rb: New classes CaseMapping/CaseMappingDummy
	  to pass as parameters; not yet implemented or used.
	  (with Kimihito Matsui)

Sun Feb  7 11:16:00 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* common.mk: using new option in recipe for enc/unicode/casefold.h

	* enc/unicode/case-folding.rb: Correctly specify argument to new option.
	  (with Kimihito Matsui)

Sun Feb  7 10:43:27 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	(this commit message applies to the previous commit)
	* common.mk: explicit recipe for enc/unicode/casefold.h

	* enc/unicode/case-folding.rb: Adding -m option to prepare for using
	  multiple data files.
	  (with Kimihito Matsui)

Sat Feb  6 22:30:57 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* lib/cgi/util.rb (escapeHTML, unescapeHTML): consider
	  ASCII-incompatible encodings.  [Fix GH-1239]

Sat Feb  6 15:18:28 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* test/ruby/enc/test_regex_casefold.rb: Added data-based testing for
	  String#downcase :fold.

	* enc/unicode.c: Fixed a range error (lowest non-ASCII character affected
	  by case operations is U+00B5, MICRO SIGN)

	* test/ruby/enc/test_case_mapping.rb: Explicit test for case folding of
	  MICRO SIGN to Greek mu.
	  (with Kimihito Matsui)

Sat Feb  6 14:51:23 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* test/ruby/enc/test_regex_casefold.rb: Tests for three case folding
	  primitives (mbc_case_fold, get_case_fold_codes_by_str,
	  apply_all_case_fold) in the various encodings. Currently only known
	  good encodings are tested to avoid test failures. For bug hunting,
	  start by adding more encodings with
	  generate_test_casefold encoding
	  (with Kimihito Matsui)

Sat Feb  6 14:37:16 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode.c, test/ruby/enc/test_case_mapping.rb: Implemented :fold
	  option for String#downcase by using case folding data from
	  regular expression engine, and added a few simple tests.
	  (with Kimihito Matsui)

Fri Feb  5 20:08:59 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* test/ruby/enc/test_case_mapping.rb: added tests for :ascii option.
	  (with Kimihito Matsui)

Fri Feb  5 12:22:20 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* insns.def (opt_mult): Use int128_t for overflow detection.

	* bignum.c (rb_uint128t2big): added for opt_mult.

	* bignum.c (rb_uint128t2big): added for rb_uint128t2big..

	* configure.in: define int128_t, uint128_t and related MACROs.
	  Initially introduced by r41379 but reverted by r50749.

Thu Feb  4 21:05:17 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode.c: Activated :ascii flag for ASCII-only case conversion
	  (with Kimihito Matsui)

Thu Feb  4 17:38:01 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* re.c (reg_set_source): make source string frozen without
	  copying.

	* re.c (rb_reg_initialize_m): refactor initialization with
	  encoding.

Thu Feb  4 15:35:29 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* string.c (rb_fstring_enc_new, rb_fstring_enc_cstr): functions to
	  make fstring with encoding.

Thu Feb  4 14:42:29 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* common.mk: Added Unicode data file SpecialCasing.txt to be additionally
	  downloaded (with Kimihito Matsui)

Thu Feb  4 12:39:08 2016  joker1007  <kakyoin.hierophant@gmail.com>

	* cgi/escape/escape.c: Optimize CGI.escape performance by C ext
	  for ASCII-compatible encodings.  [Fix GH-1238]

Thu Feb  4 11:53:56 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* common.mk: Introduce two variables (UNICODE_DATA_DIR and
	  UNICODE_SRC_DATA_DIR) to eliminate repetitions.

Wed Feb  3 12:13:20 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* string.c (str_new_frozen): if the given string is embeddedable
	  but not embedded, embed a new copied string. [Bug #11946]

Wed Feb  3 08:25:38 2016  boshan  <boshan@subsplash.com>

	* ext/openssl/ossl_pkey.c (Init_ossl_pkey): [DOC] Fix typo
	  "encrypted" to "decrypted".  [Fix GH-1235]

Wed Feb  3 08:21:32 2016  Seiei Miyagi  <hanachin@gmail.com>

	* ext/ripper/lib/ripper/lexer.rb (on_heredoc_dedent): Fix
	  Ripper.lex error in dedenting squiggly heredoc.  heredoc tree is
	  also an array of Elem in the outer tree.  [Fix GH-1234]

Wed Feb  3 02:33:39 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* re.c (rb_reg_prepare_enc): use already compiled US-ASCII regexp
	  if given string is ASCII only.
	  121.2s to 113.9s on my x86_64-freebsd10.2 Intel Core i5 661

Tue Feb  2 13:02:03 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* re.c: Introduce RREGEXP_PTR.
	  patch by dbussink.
	  partially merge https://github.com/ruby/ruby/pull/497

	* include/ruby/ruby.h: ditto.

	* gc.c: ditto.

	* ext/strscan/strscan.c: ditto.

	* parse.y: ditto.

	* string.c: ditto.

Tue Feb  2 09:08:27 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* lib/rubygems/specification.rb: `coding` is effective only first
	  line except shebang.

	* lib/rubygems/package.rb, lib/rubygems/package/*: ditto.

Mon Feb  1 21:41:58 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* lib/rubygems.rb, lib/rubygems/*, test/rubygems/*: Update rubygems-2.5.2.
	  It supports to enable frozen string literal and add `--norc` option for
	  disable to `.gemrc` configuration.
	  See 2.5.2 release notes for other fixes and enhancements.
	  https://github.com/rubygems/rubygems/blob/a8aa3bac723f045c52471c7b9328310a048561e0/History.txt#L3

Sun Jan 31 12:33:13 2016  Dan Kreiger  <dan@dankreiger.com>

	* test/drb/ut_large.rb (multiply, avg, median): add additional
	  math operations to DRbLarge.  [Fix GH-1086]

Sun Jan 31 12:19:15 2016  Kuniaki IGARASHI  <igaiga@gmail.com>

	* test/ruby/test_file_exhaustive.rb (test_lstat): Add lacking test
	  for File#lstat.  [Fix GH-1231]

Sun Jan 31 12:15:33 2016  Prayag Verma  <prayag.verma@gmail.com>

	* doc/standard_library.rdoc: fix typo  [Fix GH-1230]
	  Spelling mistakes -
	  outputing > outputting
	  publich > publish

Sat Jan 30 15:18:07 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* vm_eval.c (rb_check_funcall_with_hook): also should call the
	  given hook before returning Qundef when overridden respond_to?
	  method returned false.  [ruby-core:73556] [Bug #12030]

Fri Jan 29 17:40:07 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* win32/file.c (rb_readlink): drop garbage after the substitute
	  name, as rb_w32_read_reparse_point returns the expected buffer
	  size but "\??\" prefix is dropped from the result.

	* win32/win32.c (w32_readlink): ditto, including NUL-terminator.

Fri Jan 29 17:07:27 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* win32/win32.c (fileattr_to_unixmode, rb_w32_reparse_symlink_p): volume
	  mount point should be treated as directory, not symlink.
	  [ruby-core:72483] [Bug #11874]

	* win32/win32.c (rb_w32_read_reparse_point): check the reparse point is
	  a volume mount point or not.

	* win32/file.c (rb_readlink): follow above change (but this pass won't
	  be used).

Fri Jan 29 16:17:07 2016  Lucas Buchala  <lucasbuchala@gmail.com>

	* enum.c (enum_take_while, enum_drop_while): rename block
	  parameter to obj, since they are generic objects.  [Fix GH-1226]

Fri Jan 29 14:15:26 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* lib/erb.rb (ERB::Compiler#detect_magic_comment): allow
	  frozen-string-literal in comment as well as encoding.

	* lib/erb.rb (ERB#def_method): insert def line just before the
	  first non-comment and non-empty line, not to leave duplicated
	  and stale magic comments.

Fri Jan 29 11:13:33 2016  Jeremy Evans  <code@jeremyevans.net>

	* lib/erb.rb (ERB#set_eoutvar): explicitly make mutable string as
	  a buffer to make ERB work with --enable-frozen-string-literal.
	  [ruby-core:73561] [Bug #12031]

Fri Jan 29 10:44:56 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* lib/net/http/header.rb: Warn nil variable on HTTP Header.
	  It caused to NoMethodError. [fix GH-952][fix GH-641] Patch by @teosz
	* test/net/http/test_httpheader.rb: Added test for nil HTTP Header.

Thu Jan 28 17:31:43 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ext/socket/socket.c (sock_gethostname): support unlimited size
	  hostname.

Wed Jan 27 21:03:45 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* test/-ext-/string/test_capacity.rb: Added missing library.

Wed Jan 27 18:53:40 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode.c: Fixed bit mask in macro OnigCodePointCount

Wed Jan 27 17:54:42 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode.c: Protect code point count by macro, in order to
	  be able to use the remaining bits for flags.
	  (with Kimihito Matsui)

Wed Jan 27 16:34:35 2016  boshan  <boshan@subsplash.com>

	* lib/tempfile.rb (Tempfile#initialize): [DOC] the first parameter
	  `basename` is optional and defaulted to an empty string since
	  [GH-523].  [Fix GH-1225]

Wed Jan 27 16:25:54 2016  Koichi ITO  <koic.ito@gmail.com>

	* array.c (rb_ary_dig): [DOC] fix the exception class to be raised
	  when intermediate object does not have dig method.  TypeError
	  will be raised now.  [Fix GH-1224]

Tue Jan 26 19:36:15 2016  Aggelos Avgerinos  <evaggelos.avgerinos@gmail.com>

	* array.c (permute0, rpermute0): [DOC] Substitute indexes ->
	  indices in documentation for consistency.  [Fix GH-1222]

Tue Jan 26 15:21:37 2016  Eric Wong  <e@80x24.org>

	* compile.c (caller_location): use rb_fstring_cstr for "<compiled>"
	  (it is converted to fstring anyways inside rb_iseq_new_with_opt)
	* iseq.c (iseqw_s_compile): ditto
	* iseq.c (rb_iseq_new_main): use rb_fstring_cstr for "<main>"
	* vm.c (Init_VM): ditto, share with with above
	* iseq.c (iseqw_s_compile_file): rb_fstring before rb_io_t->pathv
	  share "<main>" with above
	* vm.c (rb_binding_add_dynavars): fstring "<temp>" immediately

Tue Jan 26 15:14:01 2016  Kazuki Yamaguchi  <k@rhe.jp>

	* compile.c (iseq_peephole_optimize): don't apply tailcall
	  optimization to send/invokesuper instructions with blockiseq.
	  This is a follow-up to the changes in r51903; blockiseq is now
	  the third operand of send/invokesuper instructions.
	  [ruby-core:73413] [Bug #12018]

Tue Jan 26 14:26:46 2016  Eric Wong  <e@80x24.org>

	* signal.c (sig_list): use fstring for hash key
	* test/ruby/test_signal.rb (test_signal_list_dedupe_keys): added

Tue Jan 26 13:08:34 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* signal.c (rb_f_kill): should immediately deliver reserved
	  signals SIGILL and SIGFPE, not only SIGSEGV and SIGBUS.

Tue Jan 26 07:57:28 2016  Joseph Tibbertsma  <josephtibbertsma@gmail.com>

	* gc.c (RVALUE_PAGE_WB_UNPROTECTED): fix a typo of argument name.
	  [Fix GH-1221]

Mon Jan 25 17:26:54 2016  Eric Wong  <e@80x24.org>

	* ruby_assert.h (RUBY_ASSERT_WHEN): fix reference to macro name
	* vm_core.h: include ruby_assert.h before using
	  [ruby-core:73371]

Mon Jan 25 15:55:30 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* symbol.c (sym_check_asciionly): more informative error message
	  with the encoding name and the inspected content.
	  [ruby-core:73398] [Feature #12016]

Mon Jan 25 09:38:26 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* test/ruby/test_string.rb: added testcase for next!, succ and succ!
	  [fix GH-1213] Patch by @K0mAtoru

Mon Jan 25 09:32:25 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* lib/webrick/httpservlet/filehandler.rb: fix documentation for namespace.
	  [fix GH-1219][ci skip] Patch by @leafac

Sun Jan 24 19:34:23 2016  Eric Wong  <e@80x24.org>

	* vm_insnhelper.c (vm_check_if_namespace): tiny size reduction

Sun Jan 24 18:12:36 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* common.mk: Simplifying Unicode data file download logic to make
	  it more reliable (including additional fix not in r53633) [Bug #12007]

Sun Jan 24 16:54:11 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ext/io/wait/wait.c (io_wait_readwrite): [EXPERIMENTAL] allow to
	  wait for multiple modes, readable and writable, at once.  the
	  arguments may change in the future.  [Feature #12013]

Sat Jan 23 22:30:59 2016  K0mA  <mctj1218@gmail.com>

	* test/ruby/test_array.rb (test_keep_if): Add test for
	  Array#keep_if separate from Array#select!  [Fix GH-1218]

Sat Jan 23 20:54:26 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* common.mk: revert r53633. It broke rubyci and travis.
	  https://travis-ci.org/ruby/ruby/builds/104259623

Sat Jan 23 20:10:29 2016  Shugo Maeda  <shugo@ruby-lang.org>

	* range.c (range_eqq): revert r51585 because rb_call_super() is
	  called in range_include() and thus r51585 doesn't work when the
	  receiver Range object consists of non linear objects such as Date
	  objects.
	  [ruby-core:72908] [Bug #12003]

Sat Jan 23 18:37:37 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* ChangeLog: Fixing wrong time on previous commit, and adding
	  previous commit message to svn [ci skip]

Sat Jan 23 18:30:30 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* common.mk: Simplifying Unicode data file download logic to make
	  it more reliable [Bug #12007]

Sat Jan 23 16:29:42 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* tool/downloader.rb: Fixed a logical error, improved documentation

Sat Jan 23 11:42:43 2016  Peter Suschlik  <ps@neopoly.de>

	* README.md: Use SVG Travis badge over PNG for better quality and
	  device support.  [Fix GH-1214] [Fix GH-1216]

Sat Jan 23 11:29:16 2016  Pascal Betz  <pascal.betz@simplificator.com>

	* lib/csv.rb: Update documentation of CSV header converter for
	  r45498, [GH-575].  [Fix GH-1215]

Fri Jan 22 17:36:46 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* vm_core.h (VM_ASSERT): use RUBY_ASSERT instead of rb_bug.

	* error.c (rb_assert_failure): assertion with stack dump.

	* ruby_assert.h (RUBY_ASSERT): new header for the assertion.

Fri Jan 22 00:25:57 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* regparse.c (fetch_name_with_level): allow non word characters
	  at the first character.  [Feature #11949]

	* regparse.c (fetch_name): ditto.

Thu Jan 21 17:34:01 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* marshal.c (r_object0): honor Marshal.load post proc
	  value for TYPE_LINK.  by Hiroshi Nakamura <nahi@ruby-lang.org>
	  https://github.com/ruby/ruby/pull/1204 fix GH-1204

Thu Jan 21 16:37:50 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* Makefile.in (update-rubyspec): fix r53208 like r53451.

Wed Jan 20 20:58:25 2016  NAKAMURA Usaku  <usa@ruby-lang.org>

	* common.mk, Makefile.in: update-config_files is only for Unix
	  platforms.

Wed Jan 20 17:13:39 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* tool/extlibs.rb: add --cache option to change cache directory.

Tue Jan 19 17:03:40 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* common.mk: Added Unicode data file CaseFolding.txt to be additionally
	  downloaded (with Kimihito Matsui)

Tue Jan 19 10:09:58 2016  Sho Hashimoto  <sho-h@ruby-lang.org>

	* lib/shell.rb (Shell.debug_output_exclusive_unlock): remove
	  because Mutex#exclusive_unlock was already deleted. [fix GH-1185]

Tue Jan 19 09:38:27 2016  Nick Cox  <nick@nickcox.me>

	* vm_method.c: fix grammar in respond_to? warning.
	  [fix GH-1047]

Mon Jan 18 14:37:07 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* parse.y (parser_here_document): an escaped newline is not an
	  actual newline, and the rest part should not be dedented.
	  [ruby-core:72855] [Bug #11989]

Mon Jan 18 12:04:34 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* test/ruby/test_string.rb: Added extra testcase for test_rstrip_bang
	  and test_lstrip_bang. [fix GH-1178] Patch by @Matrixbirds

Mon Jan 18 11:47:27 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* string.c: fix a typo. [fix GH-1202][ci skip] Patch by @sunboshan

Sun Jan 17 21:15:30 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* configure.in: improve ICC (Intel C Compiler) support.

	* configure.in (CXX): The name of icc's c++ compiler is `icpc`.

	* configure.in (warnings): Add `-diag-disable=2259` to suppress
	  noisy warnings: "non-pointer conversion from "..." to "..." may
	  lose significant bits".

	* configure.in (optflags): Add `-fp-model precise` like -fno-fast-math.

	* lib/mkmf.rb: icc supports -Werror=division-by-zero
	  and -Werror=deprecated-declarations, but doesn't support
	  -Wdivision-by-zero and -Wdeprecated-declarations.

Sun Jan 17 20:40:10 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* string.c: Any kind of option is now taking the new code path for
	  upcase/downcase/capitalize/swapcase. :lithuanian can be used for
	  testing if no specific option is desired.
	* test/ruby/enc/test_case_mapping.rb: Adjusted to above.
	  (with Kimihito Matsui)

Sun Jan 17 20:10:10 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode.c: Fixed a logical error and some comments.
	* test/ruby/enc/test_case_mapping.rb: Made tests more general.
	  (with Kimihito Matsui)

Sun Jan 17 17:41:41 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode.c: Removed artificial expansion for Turkic,
	  added hand-coded support for Turkic, fixed logic for swapcase.
	* string.c: Made use of new case mapping code possible from upcase,
	  capitalize, and swapcase (with :lithuanian as a guard).
	* test/ruby/enc/test_case_mapping.rb: Adjusted for above.
	  (with Kimihito Matsui)

Sun Jan 17 15:30:57 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ext/socket/option.c (sockopt_bool): relax boolean size to be one
	  too not only sizeof(int).  Winsock getsockopt() returns a single
	  byte as a boolean socket option.  [ruby-core:72730] [Bug #11958]

Sun Jan 17 14:43:01 2016  Kuniaki IGARASHI  <igaiga@gmail.com>

	* test/ruby/test_env.rb: [Fix GH-1201]
	  * Extract test code for ENV#keep_if from ENV#select_bang
	  * Add a test case for ENV#select_bang,keep_if

Sun Jan 17 14:42:25 2016  Kuniaki IGARASHI  <igaiga@gmail.com>

	* test/ruby/test_env.rb: [Fix GH-1201]
	  * Extract test code for ENV#delete_if from ENV#reject_bang
	  * Add a test case for ENV#reject_bang,delete_if

Sun Jan 17 14:40:22 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ext/socket/option.c (check_size): extract a macro to check
	  binary data size, with a consistent message.

	* ext/socket/option.c (sockopt_byte): fix error message,
	  sizeof(int) differs from sizeof(unsigned char) in general.

Sat Jan 16 21:16:21 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* parse.y (xstring): reset heredoc indent after dedenting,
	  so that following string literal would not be dedented.
	  [ruby-core:72857] [Bug #11990]

Sat Jan 16 17:24:24 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* enc/unicode.c: Artificial mapping to test buffer expansion code.
	* string.c: Fixed buffer expansion logic.
	* test/ruby/enc/test_case_mapping.rb: Tests for above.
	  (with Kimihito Matsui)

Sat Jan 16 16:47:14 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* ext/openssl/lib/openssl/pkey.rb: Added 2048 bit DH parameter.
	* test/openssl/test_pkey_dh.rb: ditto.

Sat Jan 16 10:51:19 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* enc/unicode.c: fix implicit conversion error with clang. fixup r53548.
	* string.c: ditto.

Sat Jan 16 10:31:00 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* common.mk: test-sample was changed to test-basic.
	  [Feature #11982][ruby-core:72823]
	* basictest/runner.rb: ditto. rename from tool/rubytest.rb.
	* basictest/test.rb: ditto. rename from sample/test.rb.
	* defs/gmake.mk: picked from r53540
	* sample/test.rb: backward compatibility for chkbuild.

Sat Jan 16 10:23:23 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* string.c, enc/unicode.c: New code path as a preparation for Unicode-wide
	  case mapping. The code path is currently guarded by the :lithuanian
	  option to avoid accidental problems in daily use.
	* test/ruby/enc/test_case_mapping.rb: Test for above.
	* string.c: function 'check_case_options': fixed logical errors
	  (with Kimihito Matsui)

Fri Jan 15 20:20:20 2016  Naohisa Goto  <ngotogenome@gmail.com>

	* regint.h (PLATFORM_UNALIGNED_WORD_ACCESS): The value of
	  UNALIGNED_WORD_ACCESS should be used to determine whether
	  unaligned word access is allowed or not. After this commit,
	  ./configure CPPFLAGS="-DUNALIGNED_WORD_ACCESS=0" disables
	  unaligned word access even on platforms that support the feature.

Fri Jan 15 16:12:10 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* parse.y (string1): reset heredoc indent for each string literal
	  so that concatenated string would not be dedented.
	  [ruby-core:72857] [Bug #11990]

Thu Jan 14 20:01:00 2016  NARUSE, Yui  <naruse@ruby-lang.org>

	* lib/uri/generic.rb (URI::Generic#to_s): change encoding to
	  UTF-8 as Ruby 2.2/ by Koichi ITO <koic.ito@gmail.com>
	  https://github.com/ruby/ruby/pull/1188 fix GH-1188

Thu Jan 14 17:36:16 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* variable.c (rb_f_global_variables): add matched back references
	  only, as well as defined? operator.

Thu Jan 14 16:12:09 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* sprintf.c (rb_str_format): format exact number more exactly.

Thu Jan 14 15:08:43 2016  Tony Arcieri  <bascule@gmail.com>

	* Remove 512-bit DH group. It's affected by LogJam Attack.
	  https://weakdh.org/
	  [fix GH-1196][Bug #11968][ruby-core:72766]

Thu Jan 14 11:44:29 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* variable.c (rb_f_global_variables): add $1..$9 only if $~ is
	  set.  fix the condition removed at r14014.

Wed Jan 13 17:21:45 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* .travis.yml: removed commented-out code.

Wed Jan 13 17:14:54 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* .travis.yml: removed osx code. follow up with r53517

Wed Jan 13 16:56:19 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* iseq.c (rb_iseq_mark): mark parent iseq to prevent dynamically
	  generated iseq by eval from GC.  [ruby-core:72620] [Bug #11928]

Wed Jan 13 03:42:58 2016  Eric Wong  <e@80x24.org>

	* class.c (Init_class_hierarchy): resolve name for rb_cObject ASAP
	* object.c (rb_mod_const_set): move name resolution to rb_const_set
	* variable.c (rb_const_set): do class resolution here
	  [ruby-core:72807] [Bug #11977]

Wed Jan 13 00:37:12 2016  Satoshi Ohmori  <sachin21dev@gmail.com>

	* man/ruby.1: fix double word typo.  [Fix GH-1194]

Tue Jan 12 21:01:09 2016  Benoit Daloze  <eregontp@gmail.com>

	* common.mk: update URL and name for the Ruby spec suite.

Tue Jan 12 19:52:19 2016  sorah (Shota Fukumori)  <her@sorah.jp>

	* lib/forwardable.rb: Convert given accessors to String.

	  r53381 changed to accept only Symbol or String for accessors, but
	  there are several rubygems that pass classes (e.g. Array,
	  Hash, ...) as accessors. Prior r53381, it was accepted because Class#to_s
	  returns its class name. After r53381 given accessors are checked
	  with define_method, but it accepts only Symbol or String, otherwise
	  raises TypeError.

	      def_delegator Foo, :some_method

	  This change is to revert unexpected incompatibility. But this behavior
	  may change in the future.

Mon Jan 12 18:41:41 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* string.c: made a variable name more grammatically correct

Mon Jan 12 18:34:34 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* string.c: minor grammar fix [ci skip]

Mon Jan 12 16:09:09 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* test/ruby/enc/test_casing_options.rb: Tests for option
	  parsing/checking for upcase/downcase/capitalize/swapcase
	  (see r53503; with Kimihito Matsui)

Mon Jan 12 16:03:03 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* string.c: Added option parsing/checking for upcase/downcase/
	  capitalize/swapcase (with Kimihito Matsui)

Mon Jan 11 21:28:28 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* include/ruby/oniguruma.h: Added flags needed for upcase/downcase
	  Unicode addition (with Kimihito Matsui)

Mon Jan 11 09:50:24 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* configure.in: check if the API version number is consistent with
	  the program version number.

Sun Jan 10 20:57:25 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* compile.c (compile_massign_lhs): when index ends with splat,
	  append rhs value to it like POSTARG, since VM_CALL_ARGS_SPLAT
	  splats the last argument only.  [ruby-core:72777] [Bug #11970]

Sun Jan 10 15:45:10 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* include/ruby/missing.h (explicit_bzero_by_memset_s): remove
	  inline implementation by memset_s, which needs a macro before
	  including headers and can cause problems in extension libraries
	  by the order of the macro and headers.

Sun Jan 10 13:41:36 2016  Eric Wong  <e@80x24.org>

	* io.c (rb_deferr): remove long obsolete global

Sun Jan 10 09:14:42 2016  Eric Wong  <e@80x24.org>

	* ext/psych/lib/psych/visitors/yaml_tree.rb (visit_String):
	  eliminate chomp
	* lib/net/http.rb (connect): eliminate delete
	* lib/net/http/header.rb (basic_encode): ditto
	* lib/net/imap.rb (authenticate): eliminate gsub
	  (self.encode_utf7): shorten delete arg
	* lib/net/smtp.rb (base64_encode): eliminate gsub
	* lib/open-uri.rb (OpenURI.open_http): eliminate delete
	* lib/rss/rss.rb: ditto
	* lib/securerandom.rb (base64): ditto
	  (urlsafe_base64): eliminate delete!
	* lib/webrick/httpauth/digestauth.rb (split_param_value):
	  eliminate chop
	* lib/webrick/httpproxy.rb (do_CONNECT): eliminate delete
	  (setup_upstream_proxy_authentication): ditto
	  [ruby-core:72666] [Feature #11938]

Sat Jan  9 23:19:14 2016  Kuniaki IGARASHI  <igaiga@gmail.com>

	* test/ruby/test_hash.rb (test_try_convert): Add test for
	  Hash.try_convert.  [Fix GH-1190]

Sat Jan  9 23:15:25 2016  Jon Moss  <maclover7@users.noreply.github.com>

	* ext/openssl/ossl.c: Add missing variables to documentation
	  examples.  [Fix GH-1189]

Sat Jan  9 18:25:57 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* symbol.h (is_attrset_id): ASET is an attrset ID.  fix
	  unexpected safe call instead of an ordinary ASET.

Sat Jan  9 10:44:33 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* configure.in, win32/setup.mak: extract RUBY_PROGRAM_VERSION from
	  RUBY_VERSION in version.h instead of RUBY_API_VERSION numbers in
	  include/ruby/version.h, and cut it into version numbers.

Sat Jan  9 07:13:33 2016  Koichi Sasada  <ko1@atdot.net>

	* gc.c: rename PAGE_* to HEAP_PAGE_* because PAGE_SIZE is used
	  in Mac OS X.

	* test/ruby/test_gc.rb: catch up this fix.

Sat Jan  9 05:45:40 2016  Koichi Sasada  <ko1@atdot.net>

	* gc.c: PAGE_BITMAP_PLANES (the number of bitmap) is 4, not 3.

Sat Jan  9 05:42:57 2016  Koichi Sasada  <ko1@atdot.net>

	* gc.c: rename constant names HEAP_* to PAGE_*.

	  Keys of GC::INTERNAL_CONSTANTS are also renamed.

	* test/ruby/test_gc.rb: catch up this fix.

Fri Jan  8 22:30:06 2016  Akinori MUSHA  <knu@iDaemons.org>

	* doc/regexp.rdoc: [DOC] Elaborate on the \G anchor. [ci skip]

Fri Jan  8 19:49:27 2016  Koichi Sasada  <ko1@atdot.net>

	* gc.c: remove heap_page::body. Instead of this field,
	  heap_page::start field works well.

Fri Jan  8 19:31:52 2016  Koichi Sasada  <ko1@atdot.net>

	* gc.c: rename rb_heap_t::page_length to rb_heap_t::total_pages.

	  `page_length' is not clear (we may understand with length of
	  a page).

Fri Jan  8 17:07:14 2016  Koichi Sasada  <ko1@atdot.net>

	* gc.c: remove heap_page::heap. This field is only used to recognize
	  whether a page is in a tomb or not. Instead of this field,
	  heap_page::flags::in_tomb (1 bit field) is added.

	  Also type of heap_page::(total|free|final)_slots are changed from
	  int to short. 2B is enough for them.

Fri Jan  8 09:33:59 2016  Shugo Maeda  <shugo@ruby-lang.org>

	* iseq.c (rb_iseq_compile_with_option): move variable initialization
	  code to avoid maybe-uninitialized warnings by gcc 4.8.

Fri Jan  8 00:03:22 2016  Shugo Maeda  <shugo@ruby-lang.org>

	* enum.c (enum_min, enum_max): do the same optimization as r53454.

Thu Jan  7 22:32:21 2016  Kenta Murata  <mrkn@mrkn.jp>

	* ruby.h: undef HAVE_BUILTIN___BUILTIN_CHOOSE_EXPR_CONSTANT_P
	  and HAVE_BUILTIN___BUILTIN_TYPES_COMPATIBLE_P on C++.
	  [ruby-core:72736] [Bug #11962]

Thu Jan  7 22:02:21 2016  Shugo Maeda  <shugo@ruby-lang.org>

	* enum.c (enum_minmax): optimize object comparison in
	  Enumerable#minmax.

Thu Jan  7 14:49:12 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* thread.c (rb_thread_pending_interrupt_p): no pending interrupt
	  before initialization.

	* thread.c (thread_raise_m, rb_thread_kill): uninitialized thread
	  cannot interrupt.  [ruby-core:72732] [Bug #11959]

Thu Jan  7 11:34:14 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* include/ruby/backward.h (ruby_show_copyright_to_die): for source
	  code backward compatibility.

	* ruby.c (process_options): return Qtrue to exit the process
	  successfully.

	* version.c (ruby_show_copyright): no longer exit.

Wed Jan  6 17:22:53 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* lib/optparse.rb (OptionParser#order!): add `into` optional
	  keyword argument to store the results.  [Feature #11191]

Tue Jan  5 21:44:37 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* ChangeLog: fix wrong class name.

Tue Jan  5 21:43:50 2016  Kuniaki IGARASHI  <igaiga@gmail.com>

	* test/ruby/test_string.rb(test_chr): added test for String#chr
	  [fix GH-1179]

Tue Jan  5 21:32:26 2016  Kuniaki IGARASHI  <igaiga@gmail.com>

	* test/ruby/test_numeric.rb (test_nonzero_p): added test for Numeric#nonzero?
	  [fix GH-1187]

Tue Jan  5 11:47:23 2016  Damir Gaynetdinov  <damir.gaynetdinov@gmail.com>

	* doc/marshal.rdoc: Clarify object references example, that the
	  reference is same object.  [Fix GH-1156]

Tue Jan  5 05:06:51 2016  Eric Wong  <e@80x24.org>

	* ext/stringio/stringio.c (strio_binmode): implement to set encoding
	* test/stringio/test_stringio.rb (test_binmode): new test
	  [ruby-core:72699] [Bug #11945]

Mon Jan  4 15:44:37 2016  Sho Hashimoto  <sho-h@ruby-lang.org>

	* variable.c (rb_mod_deprecate_constant): [DOC] added
	  documentation for Module#deprecate_constant. [ci skip]

Mon Jan  4 15:36:38 2016  Sho Hashimoto  <sho-h@ruby-lang.org>

	* thread_sync.c: [DOC] remove SizedQueue#close argument.
	  [ci skip]

Mon Jan  4 10:14:24 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* test/coverage/test_coverage.rb: ignored test when enabled to coverage.
	  It lead to crash with `make test-all`.

Mon Jan  4 08:10:44 2016  Yuichiro Kaneko  <yui-knk@ruby-lang.org>

	* insns.def (opt_case_dispatch): Move a comment to the
	  appropriate position.
	  [ci skip]

Sun Jan  3 23:55:13 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* lib/rubygems/security.rb (DIGEST_ALGORITHM, KEY_ALGORITHM):
	  should check same name as the used constants.
	  [ruby-core:72674] [Bug #11940]

Sun Jan  3 19:22:01 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* aclocal.m4: add fallback file for non-aclocal environments.
	  [ruby-core:72683] [Bug #11942]

Sun Jan  3 13:56:49 2016  Yuichiro Kaneko  <yui-knk@ruby-lang.org>

	* eval_error.c (rb_print_undef): Use `rb_method_visibility_t`
	  instead of `int`.
	* eval_intern.h (rb_print_undef): ditto
	* proc.c (mnew_internal): ditto
	* vm_method.c (rb_export_method): ditto
	  [Misc #11649] [ruby-core:71311] [fix GH-1078]

Sun Jan  3 12:12:09 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* acinclude.m4: rename aclocal.m4, which should be generated by
	  aclocal.  [ruby-core:72675] [Bug #11941]

Sat Jan  2 21:07:55 2016  Eric Wong  <e@80x24.org>

	* thread_sync.c (queue_do_pop): avoid cast with Qfalse
	  (rb_szqueue_push): ditto, use queue_sleep wrapper

Sat Jan  2 16:16:14 2016  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>

	* lib/erb.rb: Allow ERB subclass to add token easily.
	  [Feature #11936]

	* test/erb/test_erb.rb: ditto.

Sat Jan  2 14:44:31 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* parse.y (regexp): set_yylval_num sets u1, should use nd_tag
	  instead of nd_state.  [ruby-core:72638] [Bug #11932]

Sat Jan  2 02:27:22 2016  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>

	* lib/ostruct.rb: Fix case of frozen object with initializer.
	  Bug revealed by RubySpec [ruby-core:72639]

Fri Jan  1 22:01:52 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>

	* NEWS: mention CSV's liberal_parsing option.

Fri Jan  1 19:38:23 2016  okkez  <okkez000@gmail.com>

	* doc/NEWS-2.3.0: fix double words typo.
	  [ci skip][fix GH-1183]

Fri Jan  1 15:28:56 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* compile.c (remove_unreachable_chunk): decrease count of
	  call_info in removed instructions.  fix up r53402.

Fri Jan  1 12:05:53 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* compile.c (remove_unreachable_chunk): remove unreferred label
	  to optimize away unreachable chunk.

Fri Jan  1 11:42:57 2016  James Edward Gray II  <james@graysoftinc.com>

	* lib/csv.rb (CSV): Add a liberal_parsing option.
	  Patch by Braden Anderson. [#11839]
	* test/csv/test_features.rb:  test liberal_parsing

Fri Jan  1 10:27:28 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* tool/mkconfig.rb (RbConfig): prefix SDKROOT to oldincludedir
	  not includedir, the latter is outside the ruby installation.
	  [ruby-core:72496] [Bug #11881]

Fri Jan  1 08:53:02 2016  Yuki Kurihara  <co000ri@gmail.com>

	* test/ruby/test_lazy_enumerator.rb (test_take_bad_arg): Add test
	  code in case of Enumerator::Lazy#take called with negative number.
	  [ruby-dev:49467] [Bug #11933]

Fri Jan  1 05:06:20 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* parse.y (parser_here_document): update indent for each line in
	  indented here document with single-quotes.
	  [ruby-core:72479] [Bug #11871]

Fri Jan  1 03:26:44 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* lib/ostruct.rb (freeze): define deferred accessors before
	  freezing to get rid of an error when just reading frozen
	  OpenStruct.

Thu Dec 31 14:36:45 2015  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>

	* lib/ostruct.rb: Fix new_ostruct_member to correctly avoid
	  redefinition [#11901]

Thu Dec 31 02:45:12 2015  NARUSE, Yui  <naruse@ruby-lang.org>

	* test/ruby/test_module.rb (test_classpath): r53376 may change
	  the order of m.constants.
	  `make TESTS='-v ruby/test_class.rb ruby/test_module.rb' test-all`
	  may fail after that.
	  http://rubyci.s3.amazonaws.com/tk2-243-31075/ruby-trunk/log/20151230T164202Z.log.html.gz

Thu Dec 31 02:20:00 2015  Benoit Daloze  <eregontp@gmail.com>

	* common.mk (help): Fix typo.

Wed Dec 30 20:53:09 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* lib/net/http/responses.rb: Added new response class for 451 status code.
	* lib/net/http.rb: documentation for HTTPUnavailableForLegalReasons

Wed Dec 30 20:45:45 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* lib/webrick/httpstatus.rb: Added HTTP 451 Status Code.
	  [fix GH-1167] Patch by @MuhammetDilmac
	  https://tools.ietf.org/html/draft-tbray-http-legally-restricted-status-00

Wed Dec 30 20:25:52 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* doc/syntax/calling_methods.rdoc: fix old operator for safe navigation
	  operator. [ci skip][fix GH-1182] Patch by @dougo

Wed Dec 30 16:43:23 2015  Kuniaki IGARASHI  <igaiga@gmail.com>

	* test/ruby/test_string.rb (test_ord): Add test for String#ord.
	  [Fix GH-1181]

Wed Dec 30 11:28:57 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* lib/forwardable.rb (def_instance_delegator): adjust backtrace of
	  method body by tail call optimization.  adjusting the delegated
	  target is still done by deleting backtrace.

	* lib/forwardable.rb (def_single_delegator): ditto.

Wed Dec 30 11:18:42 2015  Elliot Winkler  <elliot.winkler@gmail.com>

	* lib/forwardable.rb (def_instance_delegator) fix delegating to
	  'args' and 'block', clashing with local variables in generated
	  methods.  [ruby-core:72579] [Bug #11916]

	* lib/forwardable.rb (def_single_delegator): ditto.

Wed Dec 30 09:58:56 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* object.c (rb_class_inherited_p): search the corresponding
	  ancestor to prepended module from prepending class itself.
	  [ruby-core:72493] [Bug #11878]

Wed Dec 30 09:20:00 2015  Yuki Kurihara  <co000ri@gmail.com>

	* test/stringio/test_io.rb (test_flag): add assertion for error when
	  text and binary mode are mixed.
	  [ruby-dev:49465] [Feature #11921]

Wed Dec 30 08:43:59 2015  Yuki Kurihara  <co000ri@gmail.com>

	* test/stringio/test_stringio.rb (test_initialize): add test for
	  StringIO#initialize.  [ruby-core:72585] [Feature #11920]

Wed Dec 30 05:19:24 2015  Eric Wong  <e@80x24.org>

	* class.c (struct clone_const_arg): adjust for id_table
	  (clone_const): ditto
	  (clone_const_i): ditto
	  (rb_mod_init_copy): ditto
	  (rb_singleton_class_clone_and_attach): ditto
	  (rb_include_class_new): ditto
	  (include_modules_at): ditto
	* constant.h (rb_free_const_table): ditto
	* gc.c (free_const_entry_i): ditto
	  (rb_free_const_table): ditto
	  (obj_memsize_of): ditto
	  (mark_const_entry_i): ditto
	  (mark_const_tbl): ditto
	* internal.h (struct rb_classext_struct): ditto
	* object.c (rb_mod_const_set): resolve class name on assignment
	* variable.c (const_update): replace with const_tbl_update
	  (const_tbl_update): new function
	  (fc_i): adjust for id_table
	  (find_class_path): ditto
	  (autoload_const_set): st_update => const_tbl_update
	  (rb_const_remove): adjust for id_table
	  (sv_i): ditto
	  (rb_local_constants_i): ditto
	  (rb_local_constants): ditto
	  (rb_mod_const_at): ditto
	  (rb_mod_const_set): ditto
	  (rb_const_lookup): ditto
	  [ruby-core:72112] [Feature #11614]

Wed Dec 30 04:10:13 2015  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>

	* thread_pthread.c (rb_thread_create_timer_thread): destroy attr even
	  if pthread_create() failed.

Wed Dec 30 02:55:09 2015  Eric Wong  <e@80x24.org>

	* thread_pthread.c (setup_communication_pipe): delay setting owner
	  (rb_thread_create_timer_thread): until thread creation succeeds
	  [ruby-core:72590] [Bug #11922]

Tue Dec 29 19:12:46 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* ruby.c (proc_options): -W command line option should be able to
	  override -w in RUBYOPT environment variable.

Tue Dec 29 17:54:16 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* eval.c (ignored_block): warn if a block is given to `using`,
	  which is probably for `Module.new`.

Tue Dec 29 12:48:34 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* lib/ostruct.rb (OpenStruct): make respond_to? working on
	  just-allocated objects for workaround of Psych.
	  [ruby-core:72501] [Bug #11884]

Tue Dec 29 10:35:00 2015  Kenta Murata  <mrkn@mrkn.jp>

	* test/mkmf/test_have_func.rb (test_have_func):
	  Add assertion to examine the existence of HAVE_RUBY_INIT.

	* test/mkmf/test_have_func.rb (test_not_have_func):
	  Add assertion to examine the absence of HAVE_RUBY_INIT.

Tue Dec 29 06:50:42 2015  Eric Wong  <e@80x24.org>

	* thread_sync.c: static classes

Tue Dec 29 05:30:30 2015  Eric Wong  <e@80x24.org>

	* lib/resolv.rb (Resolv::IPv6.create): avoid modifying frozen
	* test/resolv/test_dns.rb (test_ipv6_create): test for above
	  [Bug #11910] [ruby-core:72559]

Mon Dec 28 14:55:57 2015  Kuniaki IGARASHI  <igaiga@gmail.com>

	* test/ruby/test_string.rb (TestString#test_rstrip_bang): Add test
	  for String#rstrip!.  [Fix GH-1176]

Mon Dec 28 09:18:53 2015  Kuniaki IGARASHI  <igaiga@gmail.com>

	* test/ruby/test_string.rb (TestString#test_lstrip_bang): Add test
	  for String#lstrip!.  [Fix GH-1176]

Sun Dec 27 23:32:26 2015  Masaki Suketa <masaki.suketa@nifty.ne.jp>

	* ext/win32ole/win32ole.c (ole_variant2val): refactoring.

Sun Dec 27 21:14:42 2015  NAKAMURA Usaku  <usa@ruby-lang.org>

	* process.c (rb_execarg_parent_start1): need to convert the encoding to
	  ospath's one.

Sun Dec 27 20:54:22 2015  NAKAMURA Usaku  <usa@ruby-lang.org>

	* process.c: use rb_w32_uchdir() instead of plain chdir() on Windows.
	  reported by naruse via twitter.

	* process.c (rb_execarg_addopt): need to convert the encoding to
	  ospath's one.

Sun Dec 27 20:00:31 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* enc/x_emoji.h: fix dead-link.

Sun Dec 27 19:55:55 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>

	* doc/NEWS-2.3.0: fix a typo.

Sun Dec 27 18:08:15 2015  Kuniaki IGARASHI  <igaiga@gmail.com>

	* string.c (rb_str_lstrip_bang, rb_str_rstrip_bang): [DOC] Fix
	  ruby-doc comments for String#rstrip! and lstrip!.  It looks like
	  dropped bang.  [Fix GH-1175]

Sun Dec 27 15:14:20 2015  Eric Wong  <e@80x24.org>

	* io.c (io_getpartial): remove unused kwarg from template
	* test/ruby/test_io.rb (test_readpartial_bad_args): new
	  [Bug #11885]

Sun Dec 27 11:50:53 2015  Kuniaki IGARASHI  <igaiga@gmail.com>

	* test/ruby/test_string.rb (test_rstrip, test_lstrip): Add tests
	  for String#lstrip and rstrip.  The test cases are used from
	  string.c ruby-doc comments.  [Fix GH-1174]

Sun Dec 27 11:47:46 2015  Kuniaki IGARASHI  <igaiga@gmail.com>

	* test/ruby/test_string.rb (test_insert): Add test for
	  String#insert.  The test cases are written in string.c
	  comments as a reference.  [Fix GH-1173]

Sun Dec 27 11:03:33 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* parse.y (show_bitstack): trace stack_type value if yydebug.

Sun Dec 27 10:03:14 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* enc/depend (enc, trans): fix version dependency, shared object
	  files depend on the RUBY_SO_NAME value for runtime link.

Sun Dec 27 09:47:20 2015  Masaki Suketa <masaki.suketa@nifty.ne.jp>

	* ext/win32ole/win32ole.c (ole_vstr2wc, ole_variant2val): fix blank
	  string conversion.
	  [Bug #11880]
	  Thanks Akio Tajima for the patch!

Sun Dec 27 09:34:53 2015  craft4coder  <yooobuntu@163.com>

	* doc/extension.rdoc: [DOC] `nul` should be uppercase.
	  change 'nul' => 'NUL'.  [Fix GH-1172]

Sat Dec 26 18:29:01 2015  Kouhei Sutou  <kou@cozmixng.org>

	* lib/xmlrpc/client.rb: Support SSL options in async methods of
	  XMLRPC::Client.
	  [Bug #11489]
	  Reported by Aleksandar Kostadinov. Thanks!!!

Sat Dec 26 11:26:38 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* miniinit.c (Init_enc): add some common aliases of built-in
	  encodings.  [ruby-core:72481] [Bug #11872]

Fri Dec 25 22:43:26 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>

	* configure.in: extract RUBY_RELEASE_DAY at generating Makefile.

	* version.h (RUBY_RELEASE_DATE): construct from RUBY_RELEASE_YEAR,
	  RUBY_RELEASE_MONTH, and RUBY_RELEASE_DAY.

Fri Dec 25 21:33:06 2015  Yukihiro Matsumoto  <matz@ruby-lang.org>

	* version.h (RUBY_VERSION): 2.4.0 development has started.

Fri Dec 25 14:12:12 2015  Martin Duerst  <duerst@it.aoyama.ac.jp>

	* doc/ChangeLog-2.3.0, ext/tk/lib/tkextlib/SUPPORT_STATUS,
	  include/ruby/version.h: minor grammar fixes [ci skip]

Fri Dec 25 08:23:22 2015  Tadashi Saito  <tad.a.digger@gmail.com>

	* compile.c, cont.c, doc, man: fix common misspelling.
	  [ruby-core:72466] [Bug #11870]


For the changes before 2.3.0, see doc/ChangeLog-2.3.0
For the changes before 2.2.0, see doc/ChangeLog-2.2.0
For the changes before 2.1.0, see doc/ChangeLog-2.1.0
For the changes before 2.0.0, see doc/ChangeLog-2.0.0
For the changes before 1.9.3, see doc/ChangeLog-1.9.3
For the changes before 1.8.0, see doc/ChangeLog-1.8.0

Local variables:
coding: us-ascii
add-log-time-format: (lambda ()
  (let* ((time (current-time))
	 (system-time-locale "C")
	 (diff (+ (cadr time) 32400))
	 (lo (% diff 65536))
	 (hi (+ (car time) (/ diff 65536))))
  (format-time-string "%a %b %e %H:%M:%S %Y" (list hi lo) t)))
indent-tabs-mode: t
tab-width: 8
change-log-indent-text: 2
end:
vim: tabstop=8 shiftwidth=2