Jukka Marin <jmarin@pyy.jmp.fi> writes:

> 
> On Thu, Apr 24, 1997 at 02:26:41AM -0400, Charles M. Hannum wrote:
> > I've added code to truncate the IDs to 65535.  Nothing appears to
> > actually *use* the stored IDs, other than to print them out, so this
> > shouldn't be a big deal.
> 
> Uh.  Can't ar be fixed instead?

That requires inventing a new header format.

Like I said, nothing really uses the stored IDs, so it's really no
more than a minor annoyance.  There are a few things that break much
worse than this (like cpio, which already has a similar truncation
hack).

> Is anyone going to improve the pw system to work faster with huge
> password files (ie. not to rebuild the whole database when one password
> is changed)?

Incremental password updates, as well as some other interesting
changes, are on my todo list, but it's really hard for me to find time
to work on it.