[Metacard] Re: Darwin, CGI Question

[Richard MacLemale]

Andu wrote,
> The more empirical way: put home, tools and help stacks in the same
> folder with the engine and do ./mc and see if metacard starts in gui.
> Also try .cgi instead of .mt for the cgi file.
> Test with a cgi known to work just in case the server may not have the
> proper settings.

That works.  I wrote a text file named "test.mt" with a simple script which
created a text file and put it in the same directory (in this case
CGI-Executables) as mc and the other stacks.  The first line of the file is
"#!mc" followed by "on startup" and then the script.  Then in command line,
I typed ./mc test.mt, and wonder of wonders, the script was instantly run
and the text file was created.  Cool!

Making the jump to being able to active the test.mt script via CGI, however,
is not working.  Both the test.mt file and mc have proper permissions (755).
I've got a test shell script which works fine when called from a browser.
But when I attempt to hit the test.mt script, no dice.  Renaming the script
to test.cgi does not help.

One thing - I can run my test shell script (named echo.sh) by typing:
sh echo.sh

in the command line.  But typing

mc test.mt

in the command line will not run the test.mt script, yet typing

./mc test.mt

in the command line does.  Being a UNIX newbie, I don't know what the ./
does, but you can't execute without it.  So perhaps this has something to do
with not being able to execute a script from a browser?

I gotta believe that at least one MetaCard developer has already done all
this.  I can't be the first person to try to run MetaCard Darwin engine on
OS X as a replacement for Perl, can I?  :)

This is SOOOOOOOOO close and yet so far.  The idea of, on OS X, being able
to run METACARD scripts is so cool I can hardly sleep at night.

Anyone have any further suggestions, or is this the end of the line?

:)
Richard MacLemale
Instructional Technology Specialist
James W. Mitchell High School