26 points | by nateb2022 19 hours ago ago
11 comments
This is legend. I’ll be putting this to extensive use tomorrow.
You've always had this ability, but with more convoluted syntax:
osascript -e "tell application \"System Events\" to keystroke \"whatever\""
osascript -e "tell application \"Safari\" to activate" -e "tell application \"System Events\" to keystroke \"c\" using {command down}"
How do you stimulate delays, animation intervals, key down/ups a with System Events?
Apple Script is a programming language and you have full control flow at your disposal to wait and branch as needed.
This is too generic to answer this specific question. Key down, for example, is not an inherent property of a programming language.
(and neither is delay precision guarantees, which might be important for some key sequences)
Wonder how this compares to Autohotkey on Windows.
Ahk isn't cli, and also is a full-blown programming language, so not as limited in its input
Now to finetune my local LLM on the sendkeys syntax, add kyutai for streaming voice recognition, and my mac will be fully voice operable.
Nice. But why?
For example to automate what couldn't be automated otherwise.
For example to deal with software designers that think web browsers are people
This is legend. I’ll be putting this to extensive use tomorrow.
You've always had this ability, but with more convoluted syntax:
Specify the application too with multiple -e parameters. This will copy whatever is selected in Safari:How do you stimulate delays, animation intervals, key down/ups a with System Events?
Apple Script is a programming language and you have full control flow at your disposal to wait and branch as needed.
This is too generic to answer this specific question. Key down, for example, is not an inherent property of a programming language.
(and neither is delay precision guarantees, which might be important for some key sequences)
Wonder how this compares to Autohotkey on Windows.
Ahk isn't cli, and also is a full-blown programming language, so not as limited in its input
Now to finetune my local LLM on the sendkeys syntax, add kyutai for streaming voice recognition, and my mac will be fully voice operable.
Nice. But why?
For example to automate what couldn't be automated otherwise.
For example to deal with software designers that think web browsers are people