Re: PATCH: completion

[Sven Wischnowsky <wischnow@informatik.hu-berlin.de>]

Bart Schaefer wrote:

> On Oct 27, 10:42am, Sven Wischnowsky wrote:
> } Subject: Re: PATCH: completion
> }
> } No, the output of `functions' isn't changed (of course).
> 
> Oh, I thought you said it was.  You meant the value of $functions, then?
> That's not so bad, but you could still get unexpected effects.

Yep, I have to start putting `$'s before parameter names...

> } Right. Hadn't thought about that either. But in a fully static shell,
> } wouldn't it a bit surprising when, for example, parameters just appear 
> } because they are used (because they were `autoloaded')?
> 
> That's what I meant about "require some changes to the way zmodload and
> static modules work."  Basically, I want the modules linked in, but not
> initialized until an actual "zmodload" is attempted on them.  This is
> not the same as the current auto-autoloading.

I almost thought so. What I was worrying about is exactly these auto-
autoloaded modules (and we where talking about those). People who
don't want to use the new completion stuff probably don't want to see
the parameters from the modules either. But to get the same behavior
in dynamic and static shells we have to make them available without an 
extra call to `zmodload'.
But still, we could try to make `zmodload module' and `zmodload -u module'
work in static shells, making them show and hide the stuff the module
offers. Auto-autoloaded modules would then start up in the `show' state.

Bye
 Sven


--
Sven Wischnowsky                         wischnow@informatik.hu-berlin.de