Re: Inserting all completions

[Peter Stephenson <pws@ibmth.df.unipi.it>]

Sven Wischnowsky wrote:
> 1) a way to get at the strings for the matches added
> 
> Unfortunately 1) is not trivial:
> 
> - we really have the information about the matches only *after* the
>   completion widget finished, because then the code that sorts the
>   matches and eliminates duplicates is run; changing that would make
>   adding matches much more expensive (checking for duplicates after
>   each added match)

Maybe a compstate parameter could be set to give a hook function to run
after completion is finished, which could simply be another completion
widget.  All the completion variables are available at that point and the
second widget could decide what to add based on the now-existing list.
Some fiddling round with compstate parameters in the first and second
widgets should get this to work.  It could be done as a command like
_expand, calling the normal completion but with some extra things set.  It
could even be done in such a way that the same widget was called again by
the hook mechanism and could detect the fact on the second call and perform
the insertion or whatever.

-- 
Peter Stephenson <pws@ibmth.df.unipi.it>       Tel: +39 050 844536
WWW:  http://www.ifh.de/~pws/
Dipartimento di Fisica, Via Buonarroti 2, 56127 Pisa, Italy