is /#$/ a Ruby bug ?

[Dan Kohn]

p /#$/.match("#")[0] # syntax error: unterminated string meets eof

The interpreter seems to interpret the first # as a comment but not the
second.

p /\#$/.match("#")[0] # this works

This seems even odder:

p /[#$]/.match("#")[0] # syntax error: compile error

p /[#\$]/.match("#")[0] # we can escape the $ rather than the #

Shouldn't the interpreter ignore # within a Regexp /.../ unless there's
an x flag, a #{...} substitution, or a (?#...) comment? Bug or feature?

 

[Marcel Molina Jr.]

p /#$/.match("#")[0] # syntax error: unterminated string meets eof

The interpreter seems to interpret the first # as a comment but not the
second.

p /\#$/.match("#")[0] # this works

This seems even odder:

p /[#$]/.match("#")[0] # syntax error: compile error

p /[#\$]/.match("#")[0] # we can escape the $ rather than the #

Shouldn't the interpreter ignore # within a Regexp /.../ unless there's
an x flag, a #{...} substitution, or a (?#...) comment? Bug or feature?

String interpolation with #{} need not use the enclosing curly brackets when
interpolating instance variables or global variables:
=> "the var value is oops, gotcha"

marcel