From: "mame (Yusuke Endoh)" Date: 2012-03-27T01:36:33+09:00 Subject: [ruby-core:43695] [ruby-trunk - Feature #5006][Rejected] Allow ~> as a method Issue #5006 has been updated by mame (Yusuke Endoh). Status changed from Assigned to Rejected Hello, This syntax would bring ambiguity. Consider "0.~>1". This can parse as: 0.send("~>", 1) as well as: 0.send("~") > 1 . So I close this ticket as Rejected. There are some incomplete or arguable points in this proposal. - Doesn't Ruby define builtin Integer#~> ? And what's the behavior? If Ruby does not provide Integer#~>, this feature will encourage users to define their own Integer#~>, which will bring collision and confusion. - An inversion method <~ will be needed. This also brings additional parsing ambiguities. - The looking is awkward with Japanese font, because a tilde character is printed as superscript ;-) -- Yusuke Endoh ---------------------------------------- Feature #5006: Allow ~> as a method https://bugs.ruby-lang.org/issues/5006#change-25193 Author: trans (Thomas Sawyer) Status: Rejected Priority: Normal Assignee: matz (Yukihiro Matsumoto) Category: Target version: It occurs to me that since RubyGems uses ~> as a recognized version constraint operator (along with >, <, >= and <=) it would be nice if Ruby supported it as a valid method. Ultimately RubyGems could use the new method in it's code. But more generally it could be useful too, as it has a couple of nice features, namely the '~' part can be thought of as meaning "approximate", and as a whole it looks like a pointing arrow. -- http://bugs.ruby-lang.org/