On Wed, 10 Oct 2007, David Korn wrote:
> > Is there a fast ksh method (builtin, operator, etc.) for getting
> > the size of a file in bytes?
> >
> > I generally use $(wc -c < file) because I think it's more portable than
> > parsing 'ls' output,
> > but it's expensive just from the fork&exec alone -- maybe even more
> > expensive than 'ls'
> > if wc is actually *counting* bytes as opposed to getting the size from
> > the i-node.
> >
> > But it would sure be handy if I could use something like $(#<file).
> >
> > Regards,
> > Mario DeFazio
> >
>
> The ast wc command doesn't count bytes with wc -c < file. In addition,
> you can do
> builtin wc
> to get the libcmd version.
>
> However, with ksh93, you can get the file size with mechanisms
> that are built-in.
>
> function fsize # file
> {
> command exec 5< $1 5<#((EOF)) # open file and seek to EOF
> print -r -- $(5<#) # output file offset
> command exec 5<&- #close the file
> }
> ...
I think it's still unfortunate that there is no portable** stat(2)
equivalent in ksh93.
** portable, as in, no need to compile/link a special library, or
other extraordinary means; ie, standard builtin.
(sorry, I have to whine about this every so often; note: Xmas is
approaching...)
--
Daniel E. Singer, System Administrator
Dept. of Computer Science, Duke University, Durham NC 27708 USA
_______________________________________________
ast-users mailing list
[email protected]
https://mailman.research.att.com/mailman/listinfo/ast-users
