Experimental Physics and
Industrial Control System

Benjamin Franksen <[email protected]> · Mon, 24 Jun 2013 16:14:38 +0200

Am Samstag, 22. Juni 2013, 00:01:34 schrieb Andrew Johnson:
> On 2013-06-21 Benjamin Franksen wrote:
> > Cool. However, for 3.15 I propose to change the existing
> >
> >  errlogRemoveListener, instead of adding a new function; I was
> >  speculating on the patch being accepted for 3.14.12.4, which is why I
> >  added a new function.
>
> I added the ability to delete multiple listeners, and it now returns a
> count of the number actually removed.
>
> This version (against 3.15) renames the function and adds a flags argument,
> but provides an API-compatible macro for code written against 3.14, plus
> some others to give it a simple API.
>
> Comments welcome.

Hi Andrew

your proposal tries to reify all the possible ways to compare two closures
(listener code + environment) as data. Another approach, which I personally
prefer, is to let users define their own matching criteria, i.e.

        typedef struct {
                void (*code)(char *,void*);
                void *env;
        } errlogListener;

        typedef int errlogMatchListener(errlogListener*, errlogListener*);

        void errlogAddListener(errlogListener *this);
        void errlogRemoveListeners(errlogListener *this, errlogMatchListener *eq);

with the informal spec

* errlogAddListener(this) stores its argument 'this' in the list

* errlogRemoveListeners(this,eq) removes all listeners 'l' for which

        eq(this,l)

  returns true (non-zero)

Note that this approach opens up the possibility to match using pointer
equality of the listener struct, which is not covered by your enumeration.

IMO, removing only the first matching list element is unsafe and should not be
supported. If a user has called errlogAddListener(l) with the "same" l
(according to some errlogMatchListener) more than once, removing it more than
once will do no harm (subsequent calls are just no-ops).

One or more pre-defined matchers can be supplied for convenience, similar as
in your proposal, possibly as static (i.e. inlinable) functions defined in the
header, e.g.

        static int errlogMatchListenerDefault(
                errlogListener *x, errlogListener *y)
        {
                return (x->code == y->code && x->env == y->env);
        }

I am willing to flesh out this proposal and send a patch if teh general idea
meets approval.

Cheers
--
Ben Franksen
()  ascii ribbon campaign - against html e-mail
/\  www.asciiribbon.org   - against proprietary attachm€nts

________________________________

Helmholtz-Zentrum Berlin für Materialien und Energie GmbH

Mitglied der Hermann von Helmholtz-Gemeinschaft Deutscher Forschungszentren e.V.

Aufsichtsrat: Vorsitzender Prof. Dr. Dr. h.c. mult. Joachim Treusch, stv. Vorsitzende Dr. Beatrix Vierkorn-Rudolph
Geschäftsführung: Prof. Dr. Anke Rita Kaysser-Pyzalla, Thomas Frederking

Sitz Berlin, AG Charlottenburg, 89 HRB 5583

Postadresse:
Hahn-Meitner-Platz 1
D-14109 Berlin

http://www.helmholtz-berlin.de

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System