Backport #2233: [PATCH] Ripper parsing Proc call as: f.() or f::() crashes

• Status: Closed
• Assignee: yugui (Yuki Sonoda)

[akeep (Andy Keep)]

=begin
Ripper seems to crash (with YARV and C stack trace) when trying to parse the two Proc call conveniences added to Ruby 1.9.x where the 'call' keyword is eliminated.

Ripper.sexp("f.()")

Ripper.sexp("f::()")

The problem seems to be that Ripper is inserting 'call' in the parse stream but was using rb_intern instead of ripper_intern. The included patch calls the ripper_intern and fixes the crash.

With the patch:

Ripper.sexp("f.()")
=> [:program, [[:method_add_arg, [:call, [:paren, [:var_ref, [:ident, "f", [1, 0]]]], :".", :call], [:arg_paren, nil]]]]
Ripper.sexp("f::()")
=> [:program, [[:method_add_arg, [:call, [:paren, [:var_ref, [:ident, "f", [1, 0]]]], :"::", :call], [:arg_paren, nil]]]]
=end

[akeep (Andy Keep)]

• File parse.y.callless_call_patch parse.y.callless_call_patch added

=begin
A small addenda, it seems like the current version of the ripper code was (needlessly?) adding an extra set of parenthesis to the parser results. So f.() was parsed as (f).call() and f::() was parsed as (f).call(). I've re-patched it to remove the extra (). (The original patch fixed f::() being parsed as f.call() instead of f::call()).
=end

[nobu (Nobuyoshi Nakada)]

• Status changed from Open to Closed
• % Done changed from 0 to 100

=begin
This issue was solved with changeset r25405.
Andy, thank you for your reporting of the issue.
You have greatfully contributed toward Ruby.
May Ruby be with you.

=end

[nobu (Nobuyoshi Nakada)]

• Status changed from Closed to Assigned
• Assignee set to yugui (Yuki Sonoda)

=begin

=end

[yugui (Yuki Sonoda)]

• Status changed from Assigned to Closed

=begin
This issue was solved with changeset r26013.
Andy, thank you for reporting this issue.
Your contribution to Ruby is greatly appreciated.
May Ruby be with you.

=end