3 ms·

Agreed. It could be written much more clearly in my opinion like this: [varname] = [x for x in l if predicate_with_single_truth_value(x)]

• Estragon 14y ago
  Good idea!
• euccastro 14y ago
  I had to check that on the REPL. I'm surprised that even works and I can't think of a good reason why should list syntax be allowed as a lvalue, in addition to tuple syntax.
  • Estragon 14y ago
    It's called destructuring assignment. It's been around for a while. http://dunsmor.com/lisp/onlisp/onlisp_22.html http://dunsmor.com/lisp/onlisp/onlisp_22.html
    • euccastro 14y ago
      I mean why would you want to allow both list and tuple syntax for exactly the same semantics, when either of them would be enough.
      • almost 14y ago
        I just wish we had destructing assignment for other types as well. And maybe a proper pattern matching!