| Commit message (Collapse) | Author | Age |
| |
|
|
|
|
|
|
|
| |
The hashtable has been slightly integration-tested, but not rigorously
-- even though the implementation is non-trivial enough that it could
warrant testing. And it's a completely separate module, so eminently
testable. Please test.
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
| |
I think writing a custom client may work better with the callback
structure, though I'm not sure
|
| |
|
|
|
|
| |
Thanks @lieuwex
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
|
| |
Previously, l(o(g(in?)?)?)? meant "login", since only the cmdlen
prefix of the registered command was matched against the sent command.
Discovered by surprisedly seeing "list_room" work.
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|