On 27/04/2024 17:31, Max Nikulin wrote:
On 26/04/2024 18:49, Ihor Radchenko wrote:
+ shell-file-name
...
+ (list shell-command-switch
+ (concat (file-local-name
script-file) " " cmdline))))
Max Nikulin writes:
Using `shell-command-switch' unconditionally may lead to executing
/bin/sh instead of shell specified by `shell-file-name' for script
files
having no shebang, see
https://superuser.com/questions/502984/writing-shell-scripts-that-will-run-on-any-shell-using-multiple-shebang-lines
I conclude that your concern, while being valid, is a _different_ bug.
Thus, I do not see it as a blocker for my patch - my patch will fix the
*original bug reported on top of this thread*.
My concern is that your patch trying to fix one bug (I am not convinced
it is an improvement despite it is a step toward consistency) introduces
another one that is not currently present in the code.
I read current variant of code once more. -c is unconditionally added,
so your variant just makes the bug apparent. I still do not like behavior of
:cmdline 1 ; touch /tmp/not-an-arg
as I stated in
Re: [DISCUSSION] The meaning of :cmdline header argument across babel
backends. Sat, 27 Apr 2024 17:53:25 +0700.
https://list.orgmode.org/v0ilf6$34l$1...@ciao.gmane.io
