2008-03-14
| 14:26 | Chouser | This should be legal, right? (defn f [a & [b & c]] (when nil (recur a b c))) |
| 14:26 | Chouser | I get: Mismatched argument count to recur, expected: 2 args, got: 3 |
| 14:36 | Chouser | looks like recur wants the "rest" arg passed in as an actual seq? |
| 14:41 | rhickey | yes, recur must match actual parameter count, which is 2 |
| 17:22 | rhickey | cgrand: thanks for the details on LNR |
| 17:24 | rhickey | I'm thinking about changing the Repl purge-on-error to: |
| 17:25 | rhickey | while(rdr.ready()) rdr.readLine(); |
| 17:25 | rhickey | but don't have an WinXP box to try it on |
| 17:34 | Chouser | heh. no destructuring in "binding" yet. |
| 17:34 | rhickey | did you really need it? |
| 17:34 | Chouser | nope. easy to work around with an extra let |
| 17:34 | Chouser | or "first" and "second" I suppose. |
| 17:35 | rhickey | I am not opposed, but would like to see some real-world need |
| 17:37 | Chouser | I'm don't know that what I'm doing it correct enough to justify any support. |
| 17:54 | bgeron | I think destructuring is not something you "need", as are all other layers around assembly |
| 17:55 | bgeron | but they're convenient, make programs more readable and let you grasp a program / come up with one easier |
| 19:25 | rhickey | specifically destructuring in 'binding' though, is less compelling |