jos@idca.tds.PHILIPS.nl (Jos Vos) (07/07/89)
I need the ftruncate(2) function from BSD4.3 UNIX on System V.3.2. For non-BSD-manual-owners, here's the description of ftruncate(2): * ftruncate (fd, length) * int fd; * long length; * * Ftruncate causes the file referenced by fd (a file descriptor of an * file open for writing) to be truncated to at most length bytes in size. * If the file previously was larger than the size, the extra data is lost. Hint: I do *not* know the filename of the open file, of course :-) Because I don't even know if there exists a solution, also a *proof* :-) that no solution exists will be very helpfull to me (than I can start rewriting the code around ftruncate() in my application :-( ). -- -- ###### Jos Vos ###### Internet jos@idca.tds.philips.nl ###### -- ###### ###### UUCP ...!mcvax!philapd!jos ######
kucharsk@uts.amdahl.com (William Kucharski) (07/08/89)
In article <1132@ssp15.idca.tds.philips.nl> jos@idca.tds.PHILIPS.nl (Jos Vos) writes: >I need the ftruncate(2) function from BSD4.3 UNIX on System V.3.2. >For non-BSD-manual-owners, here's the description of ftruncate(2)... I can see that this one is going to make it into the "frequently asked question" section... ========[Cut Here]======== #include <sys/types.h> #include <sys/stat.h> #include <errno.h> #include <fcntl.h> #include <unistd.h> int ftruncate(fd, length) int fd; /* file descriptor */ off_t length; /* length to set file to */ { extern long lseek(); struct flock fl; struct stat filebuf; if (fstat(fd, &filebuf) < 0) return(-1); if (filebuf.st_size < length) { /* extend file length */ if ((lseek(fd, (length - 1), SEEK_SET)) < 0) return(-1); /* write a "0" byte */ if ((write(fd, "", 1)) != 1) return(-1); } else { /* truncate length */ fl.l_whence = 0; fl.l_len = 0; fl.l_start = length; fl.l_type = F_WRLCK; /* write lock on file space */ /* * This relies on the UNDOCUMENTED F_FREESP argument to * fcntl(2), which truncates the file so that it ends at the * position indicated by fl.l_start. * * Will minor miracles never cease? */ if (fcntl(fd, F_FREESP, &fl) < 0) return(-1); } return(0); } -- William Kucharski ARPA: kucharsk@uts.amdahl.com UUCP: ...!{ames,decwrl,sun,uunet}!amdahl!kucharsk Disclaimer: The opinions expressed above are my own, and may not agree with those of any other sentient being, not to mention those of my employer. So there.
chip@ateng.com (Chip Salzenberg) (07/11/89)
According to jos@idca.tds.PHILIPS.nl (Jos Vos):
>I need the ftruncate(2) function from BSD4.3 UNIX on System V.3.2.
Use the Xenix-derived chsize() function; it does the same thing.
--
You may redistribute this article only to those who may freely do likewise.
Chip Salzenberg | <chip@ateng.com> or <uunet!ateng!chip>
A T Engineering | Me? Speak for my company? Surely you jest!pat@rwing.UUCP (Pat Myrto) (07/14/89)
In article <02KJ02SJ3cms01@amdahl.uts.amdahl.com>, kucharsk@uts.amdahl.com (William Kucharski) writes:
<
< ... [ Most of posted source deleted ] ...
< /*
< * This relies on the UNDOCUMENTED F_FREESP argument to
< * fcntl(2), which truncates the file so that it ends at the
< * position indicated by fl.l_start.
< *
< * Will minor miracles never cease?
< */
<
< if (fcntl(fd, F_FREESP, &fl) < 0)
< return(-1);
This ioctl() call looks like something that would be interesting to
try, however, the value of F_FREESP is not to be found in *ANY* of the
includes on my site - could someome perhaps indicate what numerical
value F_FREESP is supposed to be on USG type systems?
Thanks...
--
pat@rwing ...!nwnexus!mltco!camco!happym!\ (Pat Myrto), Seattle, WA
...!uunet!pilchuck!rwing!pat
...!uw-beaver!sumax!polari!/
WISDOM: "Travelling unarmed is like boating without a life jacket" guy@auspex.auspex.com (Guy Harris) (07/18/89)
>This ioctl() call looks like something that would be interesting to >try, however, the value of F_FREESP is not to be found in *ANY* of the >includes on my site - could someome perhaps indicate what numerical >value F_FREESP is supposed to be on USG type systems? The value that's defined in <fcntl.h>. If it's not defined there, your system may not *have* such an "fcntl()" call (not "ioctl()") - it first appeared in some flavor of S5R3.1, isn't in S5R3 (or S5R2 or earlier), and in S5R3.1 the code that implements it has the comments: /* * This is an experimental facility for SVR3.1 and WILL be changed * in the next release. ... * ... * Bug: unused bytes in the last retained block are not cleared. * This may result in a "hole" in the file that does not read as zeroes. */ so beware - if it's not documented, it may be because it's not considered "done" and may not be supported or even work correctly in all cases....