Re: Resizing hash tables in Guile

[Roland Orre]

On Tue, 2003-02-11 at 14:59, Mikael Djurfeldt wrote:
> I've just committed resizing hash table functionality to the Guile
> core hashtable functions (in hashtab.c).
> 
> Currently, the only thing which has changed in the API is that the
> size argument to the hash table constructors is now optional.  If
> omitted, you get a resizing table.
> 
> Please comment on this API and also take a look at the implementation
> in hashtab.c.

I've studied the code and consider this a perfect minimalistic solution.
It works without disturbing any old code. I'll test drive the code
tonight.

> Questions:
> 
> Is this a good thing?  Should we keep it?
It's definitely a good thing. In our application we sometimes need just
a few tens of items hashed but often many millions of items without
really knowing beforehand.
It's  definitely a good thing to not need to bother about analyzing
optimal table sizes any longer.

We hope it will stay and we will start using it in our application as
soon as possible.

> I made the hash tables thread-safe (locking/unlocking a mutex at hash
> table access and rehashing).  Is that good?  An alternativ is to
> require of the programmer to make sure the hash tables aren't accessed
> in parallel.

In our case I think we will most likely set up the different processes
to use different hash tables but we have still not parallellized our
applications. Could/should the mutex protection maybe be optional?

        Best regards
        Roland