Hi Blake,
I see, thanks.
However that method seems to have undesirable side effects.
For example the *tty* of the shell that starts a script is messed up.
And the interpreter ID is is not properly unregistered.
My script is:
*
#!/usr/bin/env -S apl "--script --id 1010"
⊃⎕ARG ⍝ show command line options
)OFF ⍝ leave the interpreter
*
and executing it twice gives:*
*
*eedjsa@server68:~/apl-1.9/workspaces$ ./SCRIPT.apl
apl
--script
--id
1010
./SCRIPT.apl
eedjsa@server68:~/apl-1.9/workspaces$ ./SCRIPT.apl
*** Another APL interpreter with --id 1010 is already running
eedjsa@server68:~/apl-1.9/workspaces$ *
Best regards,
Jürgen
On 3/9/26 11:41, Blake McBride wrote:
Greetings,
APL scripts start with: #!/usr/local/bin/apl --script
It is my understanding that that is not a portable way to start scripts.
For maximum portability on a modern system, the following should be
used instead:
#!/usr/bin/env -S apl --script
For maximum historical portability:
#!/bin/sh
exec apl --script "$0" "$@"
Blake McBride
