On Fri, Apr 7, 2023 at 8:29 PM Félix <[email protected]> wrote:

I don't see why vscode would not allow me to evaluate parts of those
> strings as expected by this feature, as I already have support of scripting
> , including offereing g, c, p etc. available in the scope of the running
> scripts.


Many years ago Paul Patternson convinced me (and all of Leo's other devs)
that we must not allow Leo to execute arbitrary scripts when loading a .leo
file. So:

1. leoSettings.leo contains: @bool scripting-at-script-nodes = False
2. Leo allows setting this setting to True *only* in myLeoSettings.leo.

I called this policy the "Lock on the H-Bomb".

Without this policy, opening any .leo file from someone else exposes the
user to malicious code in @script nodes.

Even with this policy, one should be wary of pushing any @button button in
an outline created by someone you don't trust. Ditto for executing any
command created by @command.

In short, .leo files are potential viruses that virus scanners will never
find.

Imo, the security implications of @script, @button and @command are Leo's
biggest unresolvable problem.

Path expressions bypass the lock on the H-Bomb. It's no good requiring a
new lock, as a few minutes thought should convince you.

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/CAMF8tS1-Ur_CnyO-NE_rB-H2VTw5UJv-tKnm7Gof2R%2BVY3_5jw%40mail.gmail.com.

Reply via email to