On Thu, Jan 17, 2008 at 05:36:43PM -0800, Larry Wall wrote:
> On Thu, Jan 17, 2008 at 12:34:23PM -0800, Ævar Arnfjörð Bjarmason via RT
> wrote:
>
> : Index: src/parser/grammar.pg
> : ===================================================================
> : --- src/parser/grammar.pg (revision 24905)
> : +++ src/parser/grammar.pg (working copy)
> : @@ -111,8 +111,8 @@
> : token pod_comment {
> : ^^ '=' <.unsp>?
> : [
> : - | begin <.ws> <ident> .*? \n
> : - '=' <.unsp>? 'end' <.ws> $<ident> \N* {*} #= block
> : + | begin <.ws> [ <ident> ]? .*? \n
> : + '=' <.unsp>? 'end' <.ws> [ $<ident> ]? \N* {*} #= block
> : | \N* {*} #= misc
> : ]
> : {*}
>
> That is problematic, since it allows
>
> =begin foo
> =end
>
> (Well, problematic unless you believe in that sort of thing...)
> Of course, both grammar.pg and STD.pm are just filling in for the
> missing S26 implementation there, so it's a stopgap in any case.
Agreed. Since neither S26 nor STD.pm currently allow =begin or
=end without an identifier, we won't implement it that way. (Of
course, if the change is useful we could ask the perl6-language
gods to allow this.)
So, I'll close this ticket for now, and when/if the spec or STD.pm
changes we can do an update then.
Pm
