Sign in Join

Live chat for Free Pascal, Lazarus, Delphi and every Pascal programmer.

Mailing lists / lazarus / 2008-12

Making components "interpreted"?

8 messages · 3 people · –

Reply by email Subscribe to this list Your subscription

Read-only archive. Use j and k to move between messages.

Reenen Laurie

Hi,

I am totally just throwing an idea into the bush here (and don't understand the underlying difficulties)... At the moment Lazarus has to be recompiled if we want to add a new component. I tried that on an old laptop, and Lazarus is not a joke to recompile on older hardware (especially memory wise).

Can't we make components "interpreted" (ala Python) using Pascal Script? That'll make it a lot easier to add and play around with components.

Would it be doable?

Regards,
-Reenen

o__ ,_.>/ _ (_)_\(_)_______ ...speed is good _______________ I believe five out of four people have a problem with fractions. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.lazarus.freepascal.org/pipermail/lazarus/attachments/20081205/75beabe2/attachment.html

Reply by email

Marco van de Voort
to Reenen Laurie
Reenen Laurie wrote 7 lines
I am totally just throwing an idea into the bush here (and don't understand the underlying difficulties)... At the moment Lazarus has to be recompiled if we want to add a new component. I tried that on an old laptop, and Lazarus is not a joke to recompile on older hardware (especially memory wise). Can't we make components "interpreted" (ala Python) using Pascal Script?

It would require recoding all existing components, since pascal script is not compatible. Moreover, it only works for components where the designtime parts don't depend on the runtime parts.

Moreover, at the time this would be ready, 90% of the hardware in question would already have been thrashed.

That'll make it a lot easier to add and play around with components.

I think simply implementing packages would be easier, and also serves other purposes. (it's not just a plugin system for Lazarus, but at the same time for the generated apps)

It just needs sb that makes a study out of it, and starts testing/implementing.

Reply by email

Reenen Laurie
to Marco van de Voort
I think simply implementing packages would be easier, and also serves other purposes. (it's not just a plugin system for Lazarus, but at the same time for the generated apps)

Is this in the short / mid term pipeline?

Because I guess that's all I actually wanted, but thought that perhaps an interpreter would be a shortcut.

Regards,
-Reenen

o__ ,_.>/ _ (_)_\(_)_______ ...speed is good _______________ I believe five out of four people have a problem with fractions. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.lazarus.freepascal.org/pipermail/lazarus/attachments/20081205/d6510bf7/attachment.html

Reply by email

Thierry Coq
to Reenen Laurie: I am totally just throwing an idea into the bush here (and don't understand the…

It's probably not very far away from a technical point of view. Some time ago, I wrote a piece of code to dynamically load FPC code. So the basic mechanics are available.

check this page for an example
http://tcoq.free.fr/Lazarus.html

Best regards,
Thierry

Reenen Laurie wrote 33 lines
Hi, I am totally just throwing an idea into the bush here (and don't understand the underlying difficulties)... At the moment Lazarus has to be recompiled if we want to add a new component. I tried that on an old laptop, and Lazarus is not a joke to recompile on older hardware (especially memory wise). Can't we make components "interpreted" (ala Python) using Pascal Script? That'll make it a lot easier to add and play around with components. Would it be doable? Regards, -Reenen -- o__ ,_.>/ _ (_)_\(_)_______ ...speed is good _______________ I believe five out of four people have a problem with fractions. ------------------------------------------------------------------------ _______________________________________________ Lazarus mailing list http://www.lazarus.freepascal.org/mailman/listinfo/lazarus

Reply by email

Marco van de Voort
to Reenen Laurie: Is this in the short / mid term pipeline? Because I guess that's all I actually wanted,…
Reenen Laurie wrote 5 lines
> I think simply implementing packages would be easier, and also serves other > purposes. (it's not just a plugin system for Lazarus, but at the same time > for the generated apps) Is this in the short / mid term pipeline?

Nobody is currently working on it. There is even a trajectory of research before it. (to come up with the best approach to implement them on *nix)

http://wiki.freepascal.org/packages

Because I guess that's all I actually wanted, but thought that perhaps an interpreter would be a shortcut.

I think the interpreter wouldn't be that much more easy to get workable, its is not just the IDE, but you need to define script->binary interfaces anywhere.

But more importantly, it would break compatibility (with delphi, with the current situation) horribly, AND make packages dual language, dual-tool (pascalscript and native pascal). Despite being static, at least the packages system now is both versatile and Delphi pkgs can be ported reasonably fast.

IMHO the solution is worse than the problem here.

Reply by email

Thierry Coq
to Marco van de Voort

The research could be sped up quite fast by looking at how Kylix implemented package loading some time ago. I think it had dynamic loading on Unix.

Best regards,
Thierry

Marco van de Voort wrote 37 lines
On Fri, Dec 05, 2008 at 02:36:48PM +0200, Reenen Laurie wrote: >> I think simply implementing packages would be easier, and also serves other >> purposes. (it's not just a plugin system for Lazarus, but at the same time >> for the generated apps) >> > Is this in the short / mid term pipeline? > Nobody is currently working on it. There is even a trajectory of research before it. (to come up with the best approach to implement them on *nix) http://wiki.freepascal.org/packages > Because I guess that's all I actually wanted, but thought that perhaps an > interpreter would be a shortcut. > I think the interpreter wouldn't be that much more easy to get workable, its is not just the IDE, but you need to define script->binary interfaces anywhere. But more importantly, it would break compatibility (with delphi, with the current situation) horribly, AND make packages dual language, dual-tool (pascalscript and native pascal). Despite being static, at least the packages system now is both versatile and Delphi pkgs can be ported reasonably fast. IMHO the solution is worse than the problem here. _______________________________________________ Lazarus mailing list http://www.lazarus.freepascal.org/mailman/listinfo/lazarus

Reply by email

Marco van de Voort
to Thierry Coq
Thierry Coq wrote
The research could be sped up quite fast by looking at how Kylix implemented package loading some time ago. I think it had dynamic loading on Unix.

Maybe, but there are both license issues, and also the chance that Kylix solution is a Linux hack looming over htat.

Reply by email

Thierry Coq
to Marco van de Voort
Marco van de Voort wrote 10 lines
On Fri, Dec 05, 2008 at 11:35:26PM +0100, Thierry Coq wrote: > The research could be sped up quite fast by looking at how Kylix > implemented package loading some time ago. I think it had dynamic > loading on Unix. > Maybe, but there are both license issues, and also the chance that Kylix solution is a Linux hack looming over htat.

One just needs to look at which APIs Kylix used in Linux: there is no chance of a breach of licence there. I'm not suggesting to copy/paste code. After all, I have independtly reimplemented the library loading code in Windows, so it should be possible to do the same kind of "trick" in Linux.

Show quoted text 7 lines
_______________________________________________ Lazarus mailing list http://www.lazarus.freepascal.org/mailman/listinfo/lazarus

Reply by email