| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
| |
Otherwise, the written data to pty before the reading started may
be just lost.
|
| |
|
|
|
|
|
| |
to fix test failure on trunk-no-mjit
https://gist.github.com/ko1/32ab982ffd7555988818773c08f97123
|
|
|
|
|
| |
it fails like
https://ci.appveyor.com/project/ruby/ruby/builds/29235837/job/v0apdjj4qx8afars
|
| |
|
| |
|
| |
|
|
|
|
| |
to fix failure like https://github.com/ruby/ruby/runs/327745536
|
|
|
|
|
|
|
|
| |
This is a secret feature for me. It's only for testing and any behavior
with this flag override is unsupported.
I needed this because I sometimes want to add debug options but do not
want to disable optimizations, for using Linux perf.
|
| |
|
| |
|
|
|
|
|
|
| |
23d7f4c5e1ce164e78d77de694dccc0c41f6de3d breaks a test which expect
to match error message. To avoid this failure, use #strip for expect
and actual results.
|
| |
|
|
|
|
| |
At the moment, there are some problems with regard to bundler + did_you_mean because of did_you_mean being a bundled gem. Since the vendored version of thor inside bundler and ruby itself explicitly requires did_you_mean, it can become difficult to load it when using Bundler.setup. See this issue: https://github.com/yuki24/did_you_mean/issues/117#issuecomment-482733159 for more details.
|
|
|
|
| |
[Bug #16192]
|
|
|
|
|
|
| |
This reverts commit eab88d20eaa925d5e61a2a65820a099b46ccf3f8.
The some CI was broken with this.
|
|
|
|
| |
https://github.com/ruby/webrick/commit/79d7922de9
|
|
|
|
|
|
|
|
|
|
| |
is set
Patch from Leonard Garvey.
Fixes Ruby Bug 9986.
https://github.com/ruby/webrick/commit/8cff7f3995
|
|
|
|
|
|
|
|
|
|
| |
to be array
This way you don't need to escape each entry.
Implements Ruby Feature 15170.
https://github.com/ruby/webrick/commit/d8086e600c
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
`FileUtils#install` methed raises an unexpected `TypeError`, when
called with `mode:` option which has `"X"`.
```
$ ruby -rfileutils -e 'FileUtils.install("tmp/a", "tmp/b", mode: "o+X")'
/opt/local/lib/ruby/2.7.0/fileutils.rb:942:in `directory?': no implicit conversion of File::Stat into String (TypeError)
from /opt/local/lib/ruby/2.7.0/fileutils.rb:942:in `block (3 levels) in symbolic_modes_to_i'
from /opt/local/lib/ruby/2.7.0/fileutils.rb:933:in `each_char'
from /opt/local/lib/ruby/2.7.0/fileutils.rb:933:in `each'
from /opt/local/lib/ruby/2.7.0/fileutils.rb:933:in `inject'
from /opt/local/lib/ruby/2.7.0/fileutils.rb:933:in `block (2 levels) in symbolic_modes_to_i'
from /opt/local/lib/ruby/2.7.0/fileutils.rb:931:in `each'
from /opt/local/lib/ruby/2.7.0/fileutils.rb:931:in `each_slice'
from /opt/local/lib/ruby/2.7.0/fileutils.rb:931:in `block in symbolic_modes_to_i'
from /opt/local/lib/ruby/2.7.0/fileutils.rb:926:in `each'
from /opt/local/lib/ruby/2.7.0/fileutils.rb:926:in `inject'
from /opt/local/lib/ruby/2.7.0/fileutils.rb:926:in `symbolic_modes_to_i'
from /opt/local/lib/ruby/2.7.0/fileutils.rb:973:in `fu_mode'
from /opt/local/lib/ruby/2.7.0/fileutils.rb:883:in `block in install'
from /opt/local/lib/ruby/2.7.0/fileutils.rb:1588:in `block in fu_each_src_dest'
from /opt/local/lib/ruby/2.7.0/fileutils.rb:1604:in `fu_each_src_dest0'
from /opt/local/lib/ruby/2.7.0/fileutils.rb:1586:in `fu_each_src_dest'
from /opt/local/lib/ruby/2.7.0/fileutils.rb:877:in `install'
from -e:1:in `<main>'
```
In spite of that `symbolic_modes_to_i` considers the `File::Stat`
`path` case at the beginning, in `"X"` case, `path` is passed to
`FileTest.directory?` method which requires a `String`. In such
case, the mode in `path` should be examined instead.
https://github.com/ruby/fileutils/commit/2ea54ade2f
|
|
|
|
| |
https://github.com/ruby/fileutils/commit/bb10efe104
|
|
|
|
|
|
| |
Do so in a way that is also compatible with previous versions.
https://github.com/ruby/forwardable/commit/b2dd340988
|
|
|
|
|
| |
This issue was exposed by recent commits to better support including
refined modules.
|
|
|
|
|
| |
Instead of returning `nil`, raise a syntax error if its value is
used. [Feature #16355]
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Asynchronous events such as signal trap, finalization timing,
thread switching and so on are managed by "interrupt_flag".
Ruby's threads check this flag periodically and if a thread
does not check this flag, above events doesn't happen.
This checking is CHECK_INTS() (related) macro and it is placed
at some places (laeve instruction and so on). However, at the end
of C methods, C blocks (IMEMO_IFUNC) etc there are no checking
and it can introduce uninterruptible thread.
To modify this situation, we decide to place CHECK_INTS() at
vm_pop_frame(). It increases interrupt checking points.
[Bug #16366]
This patch can introduce unexpected events...
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
By this change, the following code prints only one warning.
```
def foo(**opt); end
100.times { foo({kw:1}) }
```
A global variable `st_table *caller_to_callees` is a map from caller to
a set of callee methods. It remembers that a warning is already printed
for each pair of caller and callee.
[Feature #16289]
|
|
|
|
|
|
| |
This is a top-level version of Module#ruby2_keywords.
It can be used for functions (top-level methods) that delegates
arguments. [Feature #16364]
|
| |
|
|
|
|
|
| |
`JSONGeneratorTest#test_remove_const_seg` is meaningful only for
the extension library version, but nonsense for pure ruby version.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
vm_getivar() provides fastpath for T_OBJECT by caching an index
of ivar. This patch also provides fastpath for FL_EXIVAR objects.
FL_EXIVAR objects have an each ivar array and index can be cached
as T_OBJECT. To access this ivar array, generic_iv_tbl is exposed
by rb_ivar_generic_ivtbl() (declared in variable.h which is newly
introduced).
Benchmark script:
Benchmark.driver(repeat_count: 3){|x|
x.executable name: 'clean', command: %w'../clean/miniruby'
x.executable name: 'trunk', command: %w'./miniruby'
objs = [Object.new, 'str', {a: 1, b: 2}, [1, 2]]
objs.each.with_index{|obj, i|
rep = obj.inspect
rep = 'Object.new' if /\#/ =~ rep
x.prelude str = %Q{
v#{i} = #{rep}
def v#{i}.foo
@iv # ivar access method (attr_reader)
end
v#{i}.instance_variable_set(:@iv, :iv)
}
puts str
x.report %Q{
v#{i}.foo
}
}
}
Result:
v0.foo # T_OBJECT
clean: 85387141.8 i/s
trunk: 85249373.6 i/s - 1.00x slower
v1.foo # T_STRING
trunk: 57894407.5 i/s
clean: 39957178.6 i/s - 1.45x slower
v2.foo # T_HASH
trunk: 56629413.2 i/s
clean: 39227088.9 i/s - 1.44x slower
v3.foo # T_ARRAY
trunk: 55797530.2 i/s
clean: 38263572.9 i/s - 1.46x slower
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
After the previous commit, this was still broken. The reason it
was broken is that a refined module that hasn't been prepended to
yet keeps the refined methods in the module's method table. When
prepending, the module's method table is moved to the origin
iclass, and then the refined methods are moved from the method
table to a new method table in the module itself.
Unfortunately, that means that if a class has included the module,
prepending breaks the refinements, because when the methods are
moved from the origin iclass method table to the module method
table, they are removed from the method table from the iclass
created when the module was included earlier.
Fix this by always creating an origin class when including a
module that has any refinements, even if the refinements are
not currently used. I wasn't sure the best way to do that.
The approach I choose was to use an object flag. The flag is
set on the module when Module#refine is called, and if the
flag is present when the module is included in another module
or class, an origin iclass is created for the module.
Fixes [Bug #13446]
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This previously did not work, and the reason it did not work is
that:
1) Refining a module or class that prepends other modules places
the refinements in the class itself and not the origin iclass.
2) Inclusion of a module that prepends other modules skips the
module itself, including only iclasses for the prepended modules
and the origin iclass.
Those two behaviors combined meant that the method table for the
refined methods for the included module never ends up in the
method lookup chain for the class including the module.
Fix this by not skipping the module itself when the module is
included. This requires some code rearranging in
rb_include_class_new to make sure the correct method tables and
origin settings are used for the created iclass.
As origin iclasses shouldn't be exposed to Ruby, this also
requires skipping modules that have origin iclasses in
Module#ancestors (classes that have origin iclasses were already
skipped).
Fixes [Bug #16242]
|
| |
|
|
|
|
|
|
| |
IRB completion logic always needed exponential notation for complex literal
such as 3e6i but it's bug. I fixed to support complex literal without
exponential notation such as 3i.
|
|
|
|
|
|
|
|
| |
* `expr in pattern` should raise `NoMatchingError` when unmatched
* `expr in pattern` should return `nil`. (this is unspecified, but
this feature is experimental, at all)
[Feature #16355]
|
| |
|
|
|
|
|
|
| |
Previously, the rest array was modified, but it turns out that is
not necessary. Not modifying the rest array fixes cases when the
rest array is used more than once.
|
|
|
|
|
|
| |
Check keys
* by an internal table, instead of unstable dispatched results
* and by parsed key values, instead of escaped forms in the source
|
|
|
|
|
|
|
|
|
| |
The difference from 0b8c73aa65add5c57b46b0cfdf4e661508802172 is to
add the result of `string_add` event to marking objects.
```C
RNODE($1)->nd_rval = add_mark_object(p, $$);
```
|
| |
|
|
|
|
|
|
| |
Original Tracer.set_get_line_procs is implemented by
"def set_get_line_procs(p = proc)". It means that original
Tracer.set_get_line_procs supports block and Proc object.
|
|
|
|
|
| |
Original Tracer.add_filter is implemented by "def add_filter(p = proc)". It
means that original Tracer.add_filter supports block and Proc object.
|
|
|
|
|
|
| |
`git --version` failed as expected when git is not installed,
but unexpectedly pass when git installed and pwd is not in git working directory.
So use `git rev-parse` instead, and it failed when git installed too.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Most of these formats were documented as supported, but were not
actually supported. Document that %g and %G are supported.
If %U/%W is specified without yday and mon/mday are not specified,
then Date.strptime is used to get the appropriate yday.
If cwyear is specifier without the year, or cwday and cweek are
specified without mday and mon, then use Date.strptime and convert
the resulting value to Time, since Time.make_time cannot handle
those conversions
Fixes [Bug #9836]
Fixes [Bug #14241]
|
| |
|
| |
|
| |
|
|
|
|
| |
This is the same as the behavior of Readline.
|