On Thu, Jul 28, 2011 at 1:04 PM, David Chelimsky <dchelim...@gmail.com>wrote:
> On Jul 28, 2011, at 11:40 AM, Roger Pack wrote:
>
> >>> 'abc' ~! /def/
> > => true
> >
> > 'abc'.should !~ /def/
> >
> > fails though. Seemed unexpected...
> > -roger-
>
> This comes up from time to time but it's a bitch to google for. It boils
> down to this: the only way to support "actual.should != unexpected" in Ruby
> 1.8 is to go back and parse the file. This is because == is a method but !=
> is not a method: it's handled by the parser. What that means is this:
>
> 5.should == 5
> # becomes
> 5.should.==(5)
>
> 5.should != 4
> # becomes
> !(5.should.==(4))
>
> In the latter case the code evaluating 5 == 4 has no way to know that it's
> been negated.
>
> HTH,
> David
>
All true David, but you might want to get your eyeglass prescription
checked, or maybe you glossed over the difference between = and ~.
Roger is using != but !~ which is the negated form of ~=, although I think
!~ uses the same kind of compile time expansion as !=.
And then, Roger is trying to transpose the characters to ~! which is
syntactically incorrect sugar. Although this might be a typo in the post.
→ irb
>> 'abc' !~ /def/
=> true
>> 'abc' ~! /def/
SyntaxError: compile error
(irb):2: syntax error, unexpected '~', expecting $end
'abc' ~! /def/
^
from (irb):2
And of course the solution is to use should_not
'abc'.should_not =~ /def/
or
'abc'.should_not match(/def/)
--
Rick DeNatale
Blog: http://talklikeaduck.denhaven2.com/
Github: http://github.com/rubyredrick
Twitter: @RickDeNatale
WWR: http://www.workingwithrails.com/person/9021-rick-denatale
LinkedIn: http://www.linkedin.com/in/rickdenatale
_______________________________________________
rspec-users mailing list
rspec-users@rubyforge.org
http://rubyforge.org/mailman/listinfo/rspec-users
