construction bits to Init_RandomSeed2. Random::DEFAULT
and Ruby internal hashes are no longer shared their seed.
* random.c (Init_RandomSeed2): ditto. And, kill evil
rb_obj_reveal() stuff.
* random.c (init_hashseed): add MT argument.
* random.c: (init_siphash): ditto.
* test/ruby/test_rand.rb (TestRand#test_default_seed): new
test for Random::DEFAULT::seed.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52815 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
be return -1 as an error. Therefore, added error handling.
* ext/openssl/ossl_pkey_dsa.c (dsa_generate): ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52810 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
(potential divide by zero) for VisualC++. It's meaningless.
Before r26197, there is ruby_div0() in this place and it
actually made divide by zero. But now it's just garbage.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52809 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
mean the result might not have an enough cryptic strength and
easy predictable. That's no good for SecureRandom.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52808 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* common.mk: add a rule for explicit.o.
* configure.in: detect explicit_bzero and memset_s.
* include/ruby/missing.h: add explicit_bzero.
* random.c (init_randomseed): use explicit_bzero() instead of
memset(). memset could be eliminated by compiler optimization.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52806 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
if --enable-libedit is spcified. [Bug #11751]
patched by John Hein
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52795 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* parse.y (ripper_dispatch_heredoc_end): at the end of here
document, the terminator always should present. has_scan_event
is true.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52794 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* vm_eval.c (rb_throw, rb_catch): get rid of making static IDs,
use dynamic symbos instead.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52793 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
Removing the indirection helps me with readability, at
least. It doesn't seem like there are many other places
in the Ruby code where macros are used like this.
[ruby-core:71735] [Feature #11749]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52790 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
because it will terminate the whole test process when it is
re-raised from this assertion.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52789 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* ruby.c (load_file_internal, ruby_process_options): share
ruby_engine instead of literal strings.
* version.c (Init_version): remove internal `ruby_engine_name`,
but set the VM program name in addition to the global constant.
* vm_backtrace.c (location_to_str, oldbt_init): use th eVM program
name always.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52782 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* parse.y (parser_here_document): store dispatched result of
on_tstring_content at the last fragment of a here document.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52777 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* ext/ripper/lib/ripper/lexer.rb: alias same methods instead of
eval for each events.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52776 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* ext/ripper/lib/ripper/sexp.rb (SexpBuilderPP): share scanner
event handlers and most of parser event handlers with
SexpBuilder.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52775 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* ext/ripper/lib/ripper/core.rb: alias same methods instead of
eval for each events.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52774 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
If the server closes a keep-alive http connection, the client socket
reaches EOF. To avoid an EOFError, detect the closed connection and
reconnect.
Added test to ensure HTTP#post succeeds even if the
keep-alive-connection has been closed by the server.
by Kristian Hanekamp <kris.hanekamp@gmail.com>
https://github.com/ruby/ruby/pull/1089 fix GH-1089
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52764 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* compile.c (iseq_peephole_optimize): enable tail call
optimization for specialized indexers.
* compile.c (iseq_compile_each): blockiseq should be NULL, but not
Qnil.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52759 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* tool/downloader.rb (Downloader.https): extract a method instead
of a global variable.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52758 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
OpenSSL's SHA1 struct name is SHA_CTX (which is also used for SHA-0),
but function name is SHA1_Transform.
http://openssl.org/docs/man0.9.8/crypto/SHA1.html
also note that LibreSSL 2.3 remove the SHA-0 support but still support SHA-1.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52755 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
I was worried r52750 would break IO.copy_stream with things like
OpenSSL sockets which wrap IOs, but require data to be run through
through encryption/decryption filters. Apparently my worry
was unfounded, but perhaps this test will ensure this case continues
to work.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52751 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* iseq.h (rb_compile_option_struct): rename the member
frozen_string_literal_debug as debug_frozen_string_literal.
[Feature #11725]
* ruby.c (proc_options): do not set $DEBUG and $VERBOSE only if no
arguments is given.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52749 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* configure.in: pass --color to ifchange from CONFIGURE_TTY for
reconfig.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52747 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
when _FILE_OFFSET_BITS=64 is defined (= when 32-bit compile).
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52736 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
kosaki
svn
kazu
naruse
nobu
normal
hsbt
ko1
shugo
ngoto