scymtym_ has quit [Remote host closed the connection]
scymtym has joined #slime
scymtym has quit [Client Quit]
scymtym has joined #slime
prite has quit [Ping timeout: 268 seconds]
prite has joined #slime
prite has quit [Client Quit]
prite has joined #slime
prite has quit [Read error: Connection reset by peer]
prite has joined #slime
cage_ has quit [Remote host closed the connection]
makomo has joined #slime
prite has quit [Ping timeout: 250 seconds]
scymtym has quit [Ping timeout: 268 seconds]
scymtym has joined #slime
<fiddlerwoaroof>
luis: I don't think we have to worry about colliding with slime-repl because the functions in slime-repl special case the :repl-results target
<fiddlerwoaroof>
As far as the keyword leak issue goes, that's sort of by design: the keyword gets returned from make-buffer-stream and allows the user to "reattach" to a previously opened buffer
<fiddlerwoaroof>
e.g. after getting this result like this: (make-buffer-stream) => #<...> :target500 if you quit the current lisp and open a new one, you can reclaim that buffer with (make-buffer-stream :target500)
<fiddlerwoaroof>
I tend to think that people that mess with *gensym-counter* get to keep both pieces when things break
<fiddlerwoaroof>
I don't object to using eval-when to load the event handler, but I'm not sure that it's a great idea to have toplevel forms that aren't definitions
<fiddlerwoaroof>
Plus, if a lot of contribs did that, I suspect you'd end up with performance issues if they all automatically registered their event handlers