Timothy Donahue wrote:
> I have a program that I am writing that I need to accept input from
> either STDIN (for file redirections or pipes) or from the
> command-line. The program manipulates email addresses for our mail
> servers, so I should have the option to do either 'email_add
> [EMAIL PROTECTED]' or 'email_add < /path/to/file'.
>
> I thought that an algorithm similar to this should work.
>
> if (defined(@ARGV) {
> process the text from the command line here;
> } elsif (@EmailList = <STDIN>) {
Obviously you're going to block here, so the usage doesn't get printed.
> process the text from STDIN here;
> } else {
> print usage statement;
> }
>
> When I run the program without any input it sits at the command line
> until I either kill it, or I pass it some input from STDIN (typing a
> ^D (control + D) will make it continue its execution). Is there any
> way that I can test if there is information queued up in STDIN without
> actually requiring that there be information there (or specifically I
> believe that the test is looking for an EOF)?
You can do something like this:
if (@ARGV) {
... process from @ARGV
}
elsif (! -t) { # stdin not a tty
... process from stdin
}
else {
... print usage
}
This will print the usage if there are no arguments and stdin is not being
redirected from a file or pipe.
For documentation on -t, see 'perldoc -f -t'
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
<http://learn.perl.org/> <http://learn.perl.org/first-response>
