[ruby-core:91711] [Ruby trunk Bug#15620] Block argument usage affects lambda semantic

Issue #15620 has been updated by marcandre (Marc-Andre Lafortune).


Eregon (Benoit Daloze) wrote:
> `lambda(&proc{}).lambda?` returns false because I think the rule is:
> once a Proc is created, it never changes its lambda-ness.

Right. I was wondering why this is the "rule", what's the rationale. It makes `lambda(&...)` equivalent to `...to_proc` but less clear. I was thinking only in terms of parameter passing (in which case the rule feels counter productive), but there's also handling of `break` and `return` (in which case the rule makes more sense I guess, still not convinced).

Note that the rule can be broken, for example with:

```
def transform_proc_into_lambda(&proc)
  o = Object.new
  o.singleton_class.define_method(:foo, &proc)
  o.method(:foo).to_proc
end

transform_proc_into_lambda{}.lambda? # => true
```

----------------------------------------
Bug #15620: Block argument usage affects lambda semantic
https://bugs.ruby-lang.org/issues/15620#change-76981

* Author: alanwu (Alan Wu)
* Status: Open
* Priority: Normal
* Assignee: matz (Yukihiro Matsumoto)
* Target version: 
* ruby -v: 
* Backport: 2.4: UNKNOWN, 2.5: UNKNOWN, 2.6: UNKNOWN
----------------------------------------
The following snippet demonstrate the issue:

``` ruby
def pass_after_use(&block)
  raise unless block
  lambda(&block).call
end

def direct_pass(&block)
  lambda(&block).call
end

pass_after_use do |_arg|
  puts "fine, because block is materialized into a Proc before it is passed to #lambda"
end

direct_pass do |_arg|
  puts "Raises because all args are required. This is not printed"
end

```

Output:
```
fine, because block is materialized into a Proc before it is passed to #lambda
Traceback (most recent call last):
	2: from lambda-block-pass.rb:14:in `<main>'
	1: from lambda-block-pass.rb:7:in `direct_pass'
lambda-block-pass.rb:14:in `block in <main>': wrong number of arguments (given 0, expected 1) (ArgumentError)
```

I think having the line `raise unless block` affect `Kenrel#lambda`'s  semantic is pretty surprising. Note that if I do `raise unless block_given?`, call to the lambda without arg also raises.

If I was to decide, I would always have the resulting lambda have required arguments even after multiple levels of block pass. That is, as long as the original block is a literal block.

This is either a breaking change or a regression from 2.4. The same script executes without raising in 2.4.5 (block arguments are always materialized).





-- 
https://bugs.ruby-lang.org/

Unsubscribe: <mailto:[e-mail]?subject=unsubscribe>
<http://lists.ruby-lang.org/cgi-bin/mailman/options/ruby-core>