Courage++ I think you have explained the fgetpos() fsetpos() connundrum.

As both calls are called with the address of the fpos_t parameter, if you pass the address of a 64-bit space to the function that only uses 32-bits of it, and then you pass that back to another function that only inspects the first 32-bits of it, then neither will care whether the space pointed at was 32-bits or 64-bits in size. Hence, that line from config_H.vc

#define Fpos_t fpos_t        /* File position type */
[download]

which as you say means that the underlying API will be called with a 64-bit chunck of memory regardless explains why it works.

Thanks :) You put that one to bed nicely and (hee hee!) now I don't have to bother p5p with a wild goose chase:)

As for submitting the patches, the only fly in the ointment there is that whilst I know which 3 files and 56 lines (all inserts) are effected, I am having trouble producing the patches in the form required by p5p. The diff utility I have (from The Unix in Perl project unfortunately cannot handle diff'ing the 5100 lines of win32.c. So it will have to wait until I find an alternative. Also, I need to (find where to) download bleedperl? and ensure that my patches are applied to that base rather then rc4.

RE: The Borland's professional CRT not supporting huge files. Frankly I am surpised and disappointed by this. Shame on them:)

Thanks for your help. It's good to know I am not the only one who finds the perl source codebase somewhat confusing:)

---

Examine what is said, not who speaks.

"Efficiency is intelligent laziness." -David Dunham
"When I'm working on a problem, I never think about beauty. I think only how to solve the problem. But when I have finished, if the solution is not beautiful, I know it is wrong." -Richard Buckminster Fuller
If I understand your problem, I can solve it! Of course, the same can be said for you.