apr-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Greg Stein <gst...@lyra.org>
Subject Re: cvs commit: apr/tables apr_tables.c
Date Sat, 20 Jan 2001 22:21:30 GMT
It wasn't clear from my checkin message :-) ... I just removed the "btable"
option. Binary tables are not useful, given our hash table implementation.

What features of a table are you looking for, which hash tables do not
provide?

Cheers,
-g

On Sat, Jan 20, 2001 at 09:15:30AM -0800, rbb@covalent.net wrote:
> 
> > > >   yes, you *will* use a hash instead. any questions can be directed to
Bubba.
> > > 
> > > ?????  A hash and a table are two different things and both are
> > > useful.  How do I insert data into a hash and iterate over it in the exact
> > > same order?
> > 
> > apr_table_t is a list, but we don't insert.  Hashed tables can't be iterated.
> > There needs to be something in between for some classes of data.
> 
> Hmmm...  What do you mean we don't insert into a table?  There is an
> apr_hash_next function to allow us to iterate over the hash table, but of
> course the data is now out of order.
> 
> > > I would really have appreciated if this change had been discussed at some
> > > point before an entire type was just removed.
> > 
> > Ack, although we aren't using it right -now-, I'm -0.5 on this change
> 
> -0.5 from me too.  We may not use it, but this API has been being talked
> about since I joined this list over two years ago.  By removing it I
> believe we have taken a step backwards.  :-(
> 
> Ryan
> 
> _______________________________________________________________________________
> Ryan Bloom                        	rbb@apache.org
> 406 29th St.
> San Francisco, CA 94131
> -------------------------------------------------------------------------------

-- 
Greg Stein, http://www.lyra.org/

Mime
View raw message