Are you referring to this code in storage/gdbm/gdbm.c::gdbm_init():
/* check record lengths. This is the only spot where these
* values are hard-coded.
*/
if (sizeof(gdbm_timestamp) != 5
|| sizeof(gdbm_data_value_header_t) != 16
|| sizeof(gdbm_data_value_record_t) != 20
|| sizeof(gdbm_lookup_t) != 8)
{
fprintf(stderr, "%s: bad record sizes in internal data "
"definition structures\n", me);
exit(1);
}
That, to me, looks like a fairly simple (if ill-advised) sanity check. I'm
not sure what's it's purpose would be -- perhaps in the past (or even now),
ipac-ng is doing grubby things with the internals of GDBM structs. Barring
that, I don't see what purpose those code is serving (although I've never
used GDBM, so perhaps there's a good reason for it).
Did you try commenting out this sanity check and seeing if ipac-ng ran
correctly?
If you're referring to other sections of the code where sizes are
hard-coded, could you please give some indication as to where in the system
the hard-coding is taking place?
- Matt
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
