[slime-devel] Re: Swank and *features*

[Ralf Mattes]

On Tue, 06 Mar 2007 02:23:52 +0300, Samium Gromoff wrote:

> At Mon, 05 Mar 2007 22:00:28 +0100 (CET),
> Harald Hanche-Olsen wrote:
>> + Samium Gromoff <_deepfire at feelingofgreen.ru>:
>> | As it stands now, it appears that Swank does not signal its own existance via *features*.
>> | 
>> | Is there another, more idiomatic way to check for RUNNING-UNDER-SWANK-iness,
>> | or did i actually spot something missing?
>> 
>> Um, what's wrong with (find-package :swank) ?
>> 
>> If you need to, you can always roll your own
>> (if (find-package :swank) (pushnew :swank *features*))
>> so that you can use #+swank or #-swank if that is what you need to do.
>> But if so, it makes me wonder why you want this.
> 
> Yes, i want #-/#+, and chiefly, just for the uniformness's sake.
> 
> (The actual use case involves trapping or not trapping the
> swank user interrupt condition, depending on whether we are
> running under swank.)
> 
> I'm under impression that *features* is the perfect place to
> publish such information.
>

Hmm, but isn't #-/#+ used only during  _read_ time? This only tells you
whether swank was present while you read code. So iff you compile a file
while  running under swank and then later you reload it outside of swank
you might be in deep trouble ...

 Cheers, Ralf Mattes


>> - Harald