> Brian's changes to HDS have never made it out of RAL, so only he can let
> us know what state that work is in (the issue was to do with the size of
> the pointers stored in an HDS file, these limit offsets within an HDS
> file to around 500Mb, with your full featured NDFs I expect you'd run
> into this problem quite quickly). But, once that was released and proved
> to work, the applications themselves will need work too, as they store
> addresses in Fortran integers, which are 32 bit. We developed a system
> to work around this (see CNF_PVAL), which Mark looked at using in an
> automated fixup script, but all that worked has been stalled for quite
> some time.
Yes, I vaguely remember all this now, I've just had a look through my
filing cabinet, and from the looks of it the last time we talked about
any of this stuff was at SMM63 in October 2000.
Al.
|