If I understood it right this would execute the current selection as a command. IMHO it could be dangerous to execute the selection without any sanitizing checks. A user could accidentally select the context action. In the good case it causes an error message because the selection does not include a meaningful command. But e.g. if the user is editing a bash script it could also be a rm command. Although this is not very likely the example points out that it can be dangerous to simply execute the current selection as-is without any checks.

So if this would be changed the question arises if we want to protect the user from the above by e.g. having a config option to enable/disable the selection as a command (in case that the action command is empty) or maybe add a whitelist of commands and only execute the selection if the first word is found in the whitelist.

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub, or mute the thread.