commit 7719734dd7 org-compat.el: New compatibility function `org-random'

commit 7719734dd7 org-compat.el: New compatibility function `org-random'

[Achim Gratz]

Sorry, but that macro doesn't call `random´ appropriately at all.  No
package should ever use `(random t)´ at all and especially not
repeatedly.  Please make org-random an alias to random.


Regards,
Achim.
-- 
+<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+

Factory and User Sound Singles for Waldorf Blofeld:
http://Synth.Stromeko.net/Downloads.html#WaldorfSounds

Re: commit 7719734dd7 org-compat.el: New compatibility function `org-random'

[Bastien]

Achim Gratz <Stromeko@nexgo.de> writes:

> Sorry, but that macro doesn't call `random´ appropriately at all.  

The real problem is (random t) in previous versions of org-id.el.

Do you mean this should be (random) instead of (random t) in those
versions?  Why?  If so, we can simply get rid of the compatibility
macro and use (random).

> No package should ever use `(random t)´ at all and especially not
> repeatedly.

I thought (random t) was okay especially because related functions
(e.g. org-id-new) are *never* repeated at regular intervals.

Besides, (random t) is documented in Emacs <24.3 and XEmacs.

-- 
 Bastien

Re: commit 7719734dd7 org-compat.el: New compatibility function `org-random'

[Achim Gratz]

Bastien writes:
> The real problem is (random t) in previous versions of org-id.el.

Yes.

> Do you mean this should be (random) instead of (random t) in those
> versions?  Why?  If so, we can simply get rid of the compatibility
> macro and use (random).

That would be the correct thing to do.

>> No package should ever use `(random t)´ at all and especially not
>> repeatedly.
>
> I thought (random t) was okay especially because related functions
> (e.g. org-id-new) are *never* repeated at regular intervals.

That unfortunate misconception is built into many Emacs packages, hence
the effort to finally make this operation a no-op and move the seeding
of the PRNG to the Emacs init code.

> Besides, (random t) is documented in Emacs <24.3 and XEmacs.

Sure, but you still aren't supposed to re-seed a PRNG each time you want
a random number from it.  Especially not with the seeding method that
Emacs used to use.  Each Emacs session should call `(random t)´ exactly
once, in the user init file.


Regards,
Achim.
-- 
+<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+

Factory and User Sound Singles for Waldorf rackAttack:
http://Synth.Stromeko.net/Downloads.html#WaldorfSounds

Re: commit 7719734dd7 org-compat.el: New compatibility function `org-random'

[Nick Dokos]

Bastien <bzg@altern.org> wrote:

> Achim Gratz <Stromeko@nexgo.de> writes:
> 
> > Sorry, but that macro doesn't call `random´ appropriately at all.  
> 
> The real problem is (random t) in previous versions of org-id.el.
> 
> Do you mean this should be (random) instead of (random t) in those
> versions?  Why?  If so, we can simply get rid of the compatibility
> macro and use (random).
> 
> > No package should ever use `(random t)´ at all and especially not
> > repeatedly.
> 
> I thought (random t) was okay especially because related functions
> (e.g. org-id-new) are *never* repeated at regular intervals.
> 
> Besides, (random t) is documented in Emacs <24.3 and XEmacs.
> 

My understanding is that (random t) is called *once* to seed the RNG -
after that you use (random) or (random N) - where N is a positive integer -
to get the next (pseudo) random number in the sequence (scaled to fit in
[0, N) in the second case).

Nick

Re: commit 7719734dd7 org-compat.el: New compatibility function `org-random'

[Achim Gratz]

Bastien writes:
> Do you mean this should be (random) instead of (random t) in those
> versions?  Why?  If so, we can simply get rid of the compatibility
> macro and use (random).

Do you mind if I revert that commit and implement the change?


Regards,
Achim.
-- 
+<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+

SD adaptation for Waldorf rackAttack V1.04R1:
http://Synth.Stromeko.net/Downloads.html#WaldorfSDada

Re: commit 7719734dd7 org-compat.el: New compatibility function `org-random'

[Bastien]

Achim Gratz <Stromeko@nexgo.de> writes:

> Bastien writes:
>> Do you mean this should be (random) instead of (random t) in those
>> versions?  Why?  If so, we can simply get rid of the compatibility
>> macro and use (random).
>
> Do you mind if I revert that commit and implement the change?

Absolutely not, on the contrary.

Thanks in advance,

PS: I've been busy today with some Emacs issues I thought would 
take less time.  And I've been busy with the website last week,
it always take more time than it seems... and maybe deserves.
But I'm back focusing on Org.

-- 
 Bastien

Re: commit 7719734dd7 org-compat.el: New compatibility function `org-random'

[Achim Gratz]

Bastien writes:
>> Do you mind if I revert that commit and implement the change?
>
> Absolutely not, on the contrary.

Done.


Regards,
Achim.
-- 
+<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+

Factory and User Sound Singles for Waldorf Q+, Q and microQ:
http://Synth.Stromeko.net/Downloads.html#WaldorfSounds