Hello to all. I'm new to this forum, so please go easy on me. =)

I am working on a script to send two e-mail attachments in a single e-mail, and am running into a little bit of an issue when using process substitution. I am using the following:

cat <(uuencode $1 <(basename $1)) <(uuencode $2 <(basename $2)) | mailx -s $3 $4

uuencode /apps/out/report.html report.html

cat <(basename /apps/out/report.html)

report.html

echo <(basename /apps/out/report.html)

/dev/fd/63

What does echo filename do? It prints the string 'filename'.

What does cat filename do? It reads from the file named filename.

So the 'echo' just prints the path to the pipe it ought to be reading from. cat actually reads from the pipe, and shows the output of the program.

I see - Ok - that makes perfect sense

I now understand why the output is what it is. Now I just need to know if there's a way I can work around it. For the sake of simplicity, how could I make echo display the results of the basename command?

You don't need echo to do so. basename is fully capable of doing so all by itself. All you're doing with all this redirection is changing where it ends up.

I get it now. You want backticks. And also, a subshell, to combine their results into one pipe -- you can group processes together with ( ), and separate them with ; inside.

( uuencode $1 `basename $1` ; uuencode $2 `basename $2` ) | mailx -s $3 $4

Take it to the next example:

uuencode /apps/out/report.html basename /apps/out/report.html

This doesn't work because uuencode reads basename as its second parameter, and /apps/out/report.html as a third parameter.

Now of course I'm not using "/apps/out/report.html", my script is using $1.

---------- Post updated at 05:53 PM ---------- Previous update was at 05:50 PM ----------

Aha! That's it! I have now learned about backticks! That works!

Thanks Corona!