revive zero and implicit precision support as a deprecated feature.
* test/bigdecimal/test_bigdecimal_util.rb: modify a test for the above
change.
* NEWS: describes the above change.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_9_3@32753 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* ext/openssl/ossl_cipher.c (ossl_cipher_initialize): Avoid possible
SEGV from AES encryption/decryption. Processing data by
Cipher#update without initializing key (meaningless usage of Cipher
object since we don't offer a way to export a key) could cause SEGV.
In OpenSSL, the EVP which has EVP_CIPH_RAND_KEY flag (such as DES3)
allows uninitialized key, but other EVPs (such as AES) does not
allow it. Calling EVP_CipherUpdate() without initializing key causes
SEGV so we set the data filled with "\0" as the key by default. See
#2768.
* test/openssl/test_cipher.rb: test it.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_9_3@32724 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* ext/bigdecimal/lib/bigdecimal/util.rb (Rational#to_d):
zero or negative precision is error. fixes#5098.
[ruby-dev:44210]
* ext/bigdecimal/lib/bigdecimal/util.rb (Float#to_d): modified for
specifying precision. fixes#5098. [ruby-dev:44210]
* ext/bigdecimal/lib/bigdecimal/util.rb (Integer#to_d): added
for symmetry to BigDecimal() function with an Integer.
fixes#5098. [ruby-dev:44210]
* ext/bigdecimal/lib/bigdecimal/util.rb (BigDecimal#to_d): added
for adapting other Numeric subclasses. [ruby-dev:44245]
* test/bigdecimal/test_bigdecimal_util.rb: add tests for the above
changes.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_9_3@32687 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* bigdecimal/bigdecimal.c (VpDup) a new function for duplicating
a BigDecimal.
* bigdecimal/bigdecimal.c (BigDecimal_new): support generating a new
BigDecimal from another BigDecimal using BigDecimal global function
or constructor. [ruby-dev:44245]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_9_3@32682 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* lib/xmlrpc/client.rb: Fix possible HTTP header formatting failure by
'Basic' header. Long username caused the base64 String truncation in
HTTP header which is not allowed. See #5046.
* test/xmlrpc/test_webrick_server.rb: test it.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_9_3@32668 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* test/rake/test_rake_functional.rb (setup): Use __FILE__ for the base
directory. Current directory is not the top source directory when
the building process runs on other than there.
* test/rake/test_rake_rake_test_loader.rb: ditto.
* test/rake/test_rake_task_argument_parsing.rb
(test_terminal_width_using_hardcoded_80): hardcoded 80 is used
when app.unix? is false.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_9_3@32653 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* lib/webrick/httprequest.rb (WEBrick::HTTPRequest#each):
Allow HTTP/0.9 request which doesn't has any header or body.
patched by Felix Jodoin. [ruby-core:38040] [Bug #5022]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_9_3@32625 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
------------------------------------------------------------------------
r32590 | naruse | 2011-07-21 00:20:14 +0900 (Thu, 21 Jul 2011) | 7 lines
Add test for r32586.
FreeBSD (at least 7.2 to 7.2) calls nsdispatch(3) when it calls
getaddrinfo(3). And nsdispatch(3) doesn't call dlerror(3) even if
it calls _nss_cache_cycle_prevention_function with dlsym(3).
So our DL::Handle#sym must call dlerror(3) before call dlsym.
In general uses of dlerror(3) should call it before use it.
------------------------------------------------------------------------
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_9_3@32624 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* ext/openssl/ossl_engine.c: Avoid double free of ENGINE reference.
* test/openssl/test_engine.rb: Add a test for it.
Thanks to Ippei Obayashi for providing the patch.
[ Ruby 1.9 - Bug #5062 ] [ruby-dev:44173]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_9_3@32605 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* thread.c (set_trace_func, thread_set_trace_func_m): reset tracing
state when set_trace_func hook is removed. This is workaround patch
to force to reset tracing state that is broken by continuation call.
a patch from James M. Lawrence. [Feature #4347] [ruby-core:34998]
* test/ruby/test_continuation.rb (class TestContinuation): add a test
for above. a patch from James M. Lawrence.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_9_3@32599 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* test/openssl/test_ssl_session.rb: add PEM SSL session without TLS
extensions. Use this as the default for the tests to ensure
compatibility with OpenSSL 0.9.7.
[ Ruby 1.9 - Bug #4961 ] [ruby-core:37726]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_9_3@32565 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* time.c (time_dup): used rb_obj_class() instead of CLASS_OF().
The patch is made by Kazuki Tsujimoto. [Bug #5012] [ruby-dev:44071]
* test/ruby/test_time.rb (TestTime#test_getlocal_dont_share_eigenclass):
added a new test for eigenclass of time object.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_9_3@32556 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* ext/digest/sha2/sha2.c (SHA256_Update, SHA512_Update): avoid Bus
Error caused by unalignment access on Sparc-Solaris (and possibly on
other similar environment.) This patch just do memcpy always instead
of checking architecture. I see no perf drop on my 64bit env. For
more details, see #4320.
* test/digest/test_digest.rb: add test for unalignment access.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_9_3@32547 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* ext/openssl/ossl.c (ossl_verify_cb): trap the exception from
verify callback of SSLContext and X509Store and make the
verification fail normally. Raising exception directly from callback
causes orphan resouces in OpenSSL stack. Patched by Ippei Obayashi.
See #4445.
* test/openssl/test_ssl.rb
(test_exception_in_verify_callback_is_ignored): test it.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_9_3@32538 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
precision treatment errors.
* test/bigdecimal/test_bigdecimal.rb: add tests for the above change.
fix precision treatment errors.
* ext/bigdecimal/bigdecimal.c (BigDecimal_power): precision argument
should be optional for its compatibility.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32500 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
references. based on a patch by Michael Edgar michael.j.edgar
AT dartmouth.edu. Bug #5002
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32498 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
modifiers. patched by michael.j.edgar AT dartmouth.edu at
[ruby-core:36248]. fixed#4716.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32428 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
time. The test fails if time past 1 second in line 136-145
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32415 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
Literal texts in HTTP ABNF is case-insensitive (RFC2616 2.1),
and a ample implementation in RFC2617 also ignores the case
of algorithms. So now this ignores those cases.
[ruby-dev:43965] [Feature #4936]
* lib/webrick/httpauth/digestauth.rb (initialize):
Because of above, opera_hack is useless and removed.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32410 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
some minor nonfree OS. I have no means of debugging the bug. My
personal opinion is that such OS should be unsupported unless there
is an active maintainer. [ruby-core:37352]
* test/coverage/test_coverage.rb: ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32401 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* test/openssl/test_x509_cert.rb: Perform SHA-256 tests only if
supported by the available OpenSSL version. Versions < 0.9.8 don't
support it. [ruby-core:37724]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@32366 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
