Subject: | |
From: | |
Reply To: | |
Date: | Thu, 1 Jun 1995 16:01:34 +1000 |
Content-Type: | text/plain |
Parts/Attachments: |
|
|
At 10:00 PM 31/5/95 pst, Mark Klein wrote:
>Jim Wowchuk wrote:
>>I've a simple program, written in 'C' failing to hpfopen a new temporary
>>file for short-mapped accesses. The following is the error message:
....
>What's the record size * # records max for the file? Even though the
>file is empty, it is the potential space that it can consume that could
>exceed the limit. When a file is opened, sm_map_in is called to allocate
>virtual space for the file. This is what is failing, and that is
>typically based on maximum potential space, not actual space in the
>file.
Of course... my program was so minimal, it was failing to describe a file
size when doing the hpfopen(). As a result the default size was 2 gb - far
greater than the 6MB permissible!! Corrected and working now.
Seems like something that should have been mentioned in the Intrinsics
Manual HPFOpen entry, though. Sort of getting like Unix (in the worst
possible way) with the "Gotcha!"s. Or possibly in the 'C' manual.
Better support for long pointers in 'C' would be handy as well - it appears
that none of the standard c calls (malloc, memcpy, strcpy, etc) support long
pointers. At least WinDoze provides near and far library support.
I presume Pascal must be the language of choice (design) for using long
pointers. Anyone confirm?
Thanks Mark! :)
----
Jim "seMPEr" Wowchuk Internet: [log in to unmask]
Vanguard Computer Services Compu$erve: 100036,106
_--_|\ Post: PO Box 18, North Ryde, NSW 2113
/ \ Phone: +61 (2) 888-9688
\.--.__/ <---Sydney NSW Fax: +61 (2) 888-3056
v Australia
|
|
|