A prerequisite to fix https://bugs.ruby-lang.org/issues/15589 with JIT.
This commit alone doesn't make a significant difference yet, but I thought
this commit should be committed independently.
This method override was discussed in [Misc #16961].
This problem exists because Marshal.load calls Date.allocate, which
uses a SimpleDateData. There doesn't seem to be any support for
taking an existing Date instance and converting it from SimpleDateData
to ComplexDateData. Work around this issue by making Date.allocate
use a ComplexDateData. This causes problems in Date#initialize,
so remove the Date#initialize method (keeping the date_initialize
function, used internally for Date.civil). Alias Date.new to
Date.civil, since they do the same thing.
https://github.com/ruby/date/commit/6bb8d8fa0f
Formerly, branch coverage measurement counters are generated for each
compilation traverse of the AST. However, ensure clause node is
traversed twice; one is for normal-exit case (the resulted bytecode is
embedded in its outer scope), and the other is for exceptional case (the
resulted bytecode is used in catch table). Two branch coverage counters
are generated for the two cases, but it is not desired.
This changeset revamps the internal representation of branch coverage
measurement. Branch coverage counters are generated only at the first
visit of a branch node. Visiting the same node reuses the
already-generated counter, so double counting is avoided.
Previously, these were not implemented, and Object#== and #eql?
were used. This tries to check the proc internals to make sure
that procs created from separate blocks are treated as not equal,
but procs created from the same block are treated as equal, even
when the lazy proc allocation optimization is used.
Implements [Feature #14267]
instead of `@root.encoding`.
And fallback to ASCII-8BIT when filesystem encoding is US-ASCII.
When `@root.encoding` is not compatible filesystem encoding,
`Encoding::CompatibilityError` raised at `webrick/httpservlet/filehandler.rb:341`.
So `DocumentRoot` must be compatible with filesystem encoding.
https://rubyci.org/logs/mswinci.japaneast.cloudapp.azure.com/vc12-x64/ruby-master/log/20200619T054159Z.fail.html.gz
```
1) Failure:
WEBrick::TestFileHandler#test_cjk_in_path [D:/tmp/mswin-build20200619-14304-utgij/ruby/test/webrick/utils.rb:72]:
exceptions on 2 threads:
webrick log start:
[2020-06-19 16:28:42] ERROR `/あ.txt' not found.
webrick log end
Filesystem encoding is Windows-31J.
<"200"> expected but was
<"404">.
---
<[]> expected but was
<["[2020-06-19 16:28:42] ERROR `/\xE3\x81\x82.txt' not found.\n"]>.
```
`prevent_directory_traversal` treats `path_info` as filesystem encoding.
So path_info should be filesystem encoding in request URL.
On some environments, fallback to ASCII-8BIT when EncodingError.
http://rubyci.s3.amazonaws.com/ubuntu2004/ruby-master/log/20200619T003004Z.log.html.gz
```
/home/chkbuild/chkbuild/tmp/build/20200619T003004Z/ruby/test/ruby/test_refinement.rb:2428: warning: method redefined; discarding old foo
/home/chkbuild/chkbuild/tmp/build/20200619T003004Z/ruby/test/ruby/test_refinement.rb:2418: warning: previous definition of foo was here
```
"hack" was too short. The tests checks the error html, which includes
hostname. GitHub Actions hostname seems randomly generated, and it may
include the indicator string.
https://github.com/ruby/ruby/runs/784896235
```
1) Failure:
WEBrick::TestHTTPResponse#test_prevent_response_splitting_cookie_headers_cr [/home/runner/work/ruby/ruby/src/test/webrick/test_httpresponse.rb:71]:
Expected /hack/ to not match "HTTP/1.1 500 Internal Server Error\r\nContent-Type: text/html; charset=ISO-8859-1\r\n\r\n<!DOCTYPE HTML PUBLIC \"-//W3C//DTD HTML 4.0//EN\">\n<HTML>\n <HEAD><TITLE>Internal Server Error</TITLE></HEAD>\n <BODY>\n <H1>Internal Server Error</H1>\n WEBrick::HTTPResponse::InvalidHeader\n <HR>\n <ADDRESS>\n WEBrick/1.6.0 (Ruby/2.8.0/2020-06-18) at\n fv-az89.dddawhu1s14uzpjkiz5fhackre.cx.internal.cloudapp.net:80\n </ADDRESS>\n </BODY>\n</HTML>\n".
```
This change uses longer indicator "cracked_indicator_for_test" instead.
Doing so modifies the class's method table, but not in a way that should
be detectable from Ruby, so it may be safe to avoid checking if the
class is frozen.
Fixes [Bug #11669]
900e83b501 changed from a warning
to an error in this case, but the warning was only issued in
verbose mode, and therefore the error was only raised in verbose
mode. That was not intentional, verbose mode should only change
whether warnings are emitted, not other behavior. This issues
the RuntimeError in all cases.
This change broke a couple tests, as the tests actually issued
the warning and therefore now raise an error. This wasn't caught
earlier as test_variable suppressed the warning in this case,
effectively setting $VERBOSE = false around the code that warned.
basictest isn't run in verbose mode and therefore didn't expose
the issue previously. Fix these tests.
Fixes [Bug #14541]
This makes:
```ruby
args = [1, 2, -> {}]; foo(*args, &args.pop)
```
call `foo` with 1, 2, and the lambda, in addition to passing the
lambda as a block. This is different from the previous behavior,
which passed the lambda as a block but not as a regular argument,
which goes against the expected left-to-right evaluation order.
This is how Ruby already compiled arguments if using leading
arguments, trailing arguments, or keywords in the same call.
This works by disabling the optimization that skipped duplicating
the array during the splat (splatarray instruction argument
switches from false to true). In the above example, the splat
call duplicates the array. I've tested and cases where a
local variable or symbol are used do not duplicate the array,
so I don't expect this to decrease the performance of most Ruby
programs. However, programs such as:
```ruby
foo(*args, &bar)
```
could see a decrease in performance, if `bar` is a method call
and not a local variable.
This is not a perfect solution, there are ways to get around
this:
```ruby
args = Struct.new(:a).new([:x, :y])
def args.to_a; a; end
def args.to_proc; a.pop; ->{}; end
foo(*args, &args)
# calls foo with 1 argument (:x)
# not 2 arguments (:x and :y)
```
A perfect solution would require completely disabling the
optimization.
Fixes [Bug #16504]
Fixes [Bug #16500]
3556a834a2 added support for
Module#include to affect the iclasses of the module. It didn't add
support for Module#prepend because there were bugs in the object model
and GC at the time that prevented it. Those problems have been
addressed in ad729a1d11 and
98286e9850, and now adding support for
it is straightforward and does not break any tests or specs.
Fixes [Bug #9573]
Since we enabled parallel installation by default, we've fixed all the
issues related to it that have come up, except for a Windows issue that
we haven't yet figured out. This issue is hit by our specs on a daily
basis and there's no reason to believe that it won't be hit by end users
in a similar way.
So, both to stop the testing flakyness and to prevent regressions in the
default behavior on Windows, I'd rather leave the default as it was
before on Windows for now.
https://github.com/rubygems/rubygems/commit/5d8ab57131
Takashi Kokubun
Nobuyoshi Nakada
git
Hiroshi SHIBATA
Jeremy Evans
Kazuki Tsujimoto
Yusuke Endoh
Kazuhiro NISHIYAMA
Jean Boussier
Timo Schilling
Burdette Lamar
David Rodríguez