Instead of incrementing the depth using call by reference as we're
recursing up the stack we could instead store an offset for each known
scope where we know the depth is going to represented differently in the
Prism ast.
(https://github.com/ruby/irb/pull/784)
* Page evaluation result's output
This will make it easier to work with long output that exceeds the terminal's height.
* Use consistent TERM in rendering tests
This makes sure we get consistent result on all platforms.
4fedce93d3
It's possible for `defined?(A::B::C)` to raise an exception. `defined?`
must swallow the exception and return nil, so this commit adds a rescue
entry for `defined?` expressions on constant paths
This commit fixes a bug with locals in ensure nodes by setting
the local tables correctly. It also changes accessing locals to
look at local tables in parent scopes, and account for this
correctly on depths of get or setlocals.
Ruby allows for 0 or negative line start, this is often used
with `eval` calls to get a correct offset when prefixing a snippet.
e.g.
```ruby
caller = caller_locations(1, 1).first
class_eval <<~RUBY, caller.path, caller.line - 2
# frozen_string_literal: true
def some_method
#{caller_provided_code_snippet}
end
RUBY
```
0d14ed1452
Fundamentally, `foo { |bar,| }` is different from `foo { |bar, *| }`
because of arity checks. This PR introduces a new node to handle
that, `ImplicitRestNode`, which goes in the `rest` slot of parameter
nodes instead of `RestParameterNode` instances.
This is also used in a couple of other places, namely:
* pattern matching: `foo in [bar,]`
* multi target: `for foo, in bar do end`
* multi write: `foo, = bar`
Now the only splat nodes with a `NULL` value are when you're
forwarding, as in: `def foo(*) = bar(*)`.
dba2a3b652
Prior to this commit, we were conflating the size of the locals
list with the number of parameters. This commit distinguishes
the two, and fixes a related bug which would occur if we set a local
that was not a parameter
We are aware at parse time how many numbered parameters we have
on a BlockNode or LambdaNode, but prior to this commit, did not
store that information anywhere in its own right.
The numbered parameters were stored as locals, but this does not
distinguish them from other locals that have been set, for example
in `a { b = 1; _1 }` there is nothing on the AST that distinguishes
b from _1.
Consumers such as the compiler need to know information about how
many numbered parameters exist to set up their own tables around
parameters. Since we have this information at parse time, we should
compute it here, instead of deferring the work later on.
bf4a1e124d
Previously, we didn't invalidate the method entry wrapped by
VM_METHOD_TYPE_REFINED method entries which could cause calls to
land in the wrong method like it did in the included test.
Do the invalidation, and adjust rb_method_entry_clone() to accommodate
this new invalidation vector.
Fix: cfd7729ce7
See-also: e201b81f79
This impacted other tests. Please mind the commons.
/home/runner/work/ruby/ruby/src/test/ruby/test_compile_prism.rb:394: warning: already initialized constant Bar
/tmp/test_reline_config_60145/bazbarbob.rb:6: warning: previous definition of Bar was here
1) Failure:
TestModule#test_const_get_evaled [/home/runner/work/ruby/ruby/src/test/ruby/test_module.rb:1239]:
NameError expected but nothing was raised.
* Remove trailing spaces
* Migrate show_source tests to integration tests
Because show_source tests often need to define class and/or methods,
they can easily leak state to other tests. Changing them to integration
tests will ensure that they are run in a clean environment.
* Fix NoMethodError caused by SourceFinder#method_target
3c39f13397
prompt
(https://github.com/ruby/irb/pull/791)
Currently, IRB just terminates if `main.to_s` raises while IRB
constructs the prompt string. This can easily happen if the user wants
to start an IRB session in the instance scope of an uninitialized
object, for example:
```
class C
def initialize
binding.irb
@values = []
end
def to_s = @values.join(',') # raises if uninitialized
end
C.new
```
This patch makes IRB rescue from such an exception and displays the
class name of the exception instead of `main.to_s` to indicate some
error has occurred.
We may display more detailed information about the exception, but this
patch chooses not to do so because 1) the prompt has limited space,
2) users can evaluate `to_s` in IRB to examine the error if they want,
and 3) obtaining the details can also raise, which requires nested
exception handling and can be complicated.
412ab26067
I don't prefer this style, but it appears that a plurality of syntax
error messages between with un-capitalized messages in CRuby, so
we'll go with that for consistency, for now.
b02df68954
