At 8:29 PM +0200 on 4/8/00, M. Uli Kusterer wrote:
> I'm talking about interruptions, not low level interrupts. I.e.
>command-period being pressed, a runtime error causing a script to exit
>prematurely etc.
So am I. That can't happen when a command add is in progress.
>
>>You shouldn't be sending random commands to unknown stacks which aren't
>>even in the hierarchy. The worst thing that would happen, though, would
>>be a syntax error.
>
> I beg to differ. Many of the stacks I'm using implement plug-in stacks in
>a way that would need something like this.
We could have a message sent to a stack on file open, so it could
install it's syntax then. But when do you need to send messages to
stacks which are not in use and which you don't control?
Besides, the feature is for major extensions to FreeCard -- not
everyone rewriting the syntax because they can.
> We know what linker we have on *our* computer. But how about some other
>user's computer? He might have an entirely different linker installed, and
>if the user happens to be a programmer, he'll be able to figure out how to
>compile it with his compiler and linker. However, the user would be less
>well off.
To compile:
./configure; make; make install
I think building small standalones would be a very advanced feature
anyway. The novice would not be doing it. It'd be the advanced software
developer creating a standalone to sell.
>
> I don't know, I just think that such a tool that relies on an external
>linker will both mean asking for something to go wrong as well as entirely
>miss the target audience of a FreeCard product. Those who can work with
>this can easily compiler FreeCard from the sources, while those who can't
>will probably have enough problems with the linker as is.
Using the linker would have to be an optional, not recommeneded for
novices feature.
