"Uri Guttman" <[email protected]> writes:
>>>>>> "HP" == Harry Putnam <[email protected]> writes:
>
> HP> open(SENDMAIL,"|$sendmail $recip") or die "Can't open <$sendmail>:
> $!";
> HP> while (<SENDMAIL>) {
>
> what is that line supposed to do? it reads from the SENDMAIL handle
> which is opened for WRITING. why do you think you need that or even to
> loop over that handle? just print to it.
I see now I could just write to the handle in one blob ($emailmsg)
Just that change; from a while loop to simple print to handle, makes it
do what I was after.
Thanks...
> and yes, this is garbage code. there are many easy to use mail modules
> that will use sendmail or any available mail service for you.
With such a simple call to sendmail, is it still better to involve
extra modules as you suggest above.
Are these modules you mention easier to use than:
open(SENDMAIL,"|$sendmail $recip") or die "Can't open <$sendmail>: $!";
print SENDMAIL $blob;
close(SENDMAIL);
> HP> print SENDMAIL "$emailmsg\n";
> HP> print "$emailmsg\n";
> HP> }
> HP> close(SENDMAIL);
> when you open a piped process, check the close as that is where you will
> actually find any errors from the open. since open | does a fork/exec,
> it can't tell you if the exec fails until you close the handle.
Not sure what you are suggesting to do there. I mean other than ditch
the code in favor of a module. Do you mean to put the `die' after the
CLOSE?
Even without that though, wouldn't any errors at CLOSE still print to
stderr?
> HP> } else {
>
> why the else? $MAIL is 'TRUE' so this will never execute.
>
> HP> for (@body) {
> HP> print "$_\n";
> HP> }
I did mention in the first few lines of OP that I had snipped the code
from a larger script I'm working on, to make it work standalone for
testing... in the larger script there is a reason for the `else'.
Mail being true really depends on a getopts option having been
selected. else.... just print.
--
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
http://learn.perl.org/