You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hello @johnkerl .
I've finally managed to start taking a look at Miller 6.
First of all, it's rather thrilling as an evolution, although I've literally just started exploring it.
Looking at the new docs my eye fell on the REPL functionality, which I personally enjoy quite a lot, for reasons I shall explain in a second.
A few months ago I had started giving a few talks at work on data wrangling, with CSV data being one of the most relevant in day-to-day operations.
I had come across posts and code using fzf or up with jq to provide a live as-you-type REPL interface, with commands being evaluated as one typed on the keyboard; since that was, imho, ideal for data exploration and demonstrations (I don't know about you, but immediate feedback when trying out pipelines of commands is a whole new world as far as I am concerned).
Doing that with Miller was a bit complicated, because jq and many other programs have commands wrapped in a single string argument, while Miller takes (potentially) a thousand command-line arguments, so handling layers of quoting and the likes while abusing the hell out of the fzf preview function proved a little problematic but vaguely/partially workable in the end.
All of this to say that although the REPL is a significant improvement, providing an interface for as-you-type REPL, or an option/switch to facilitate this kind of use might be a good evolution, especially since it allows people to play around with data and reach a workable processing pipeline much more quickly. On top of that, it's really striking to see in demonstrations ;D.
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
Hello @johnkerl .
I've finally managed to start taking a look at Miller 6.
First of all, it's rather thrilling as an evolution, although I've literally just started exploring it.
Looking at the new docs my eye fell on the REPL functionality, which I personally enjoy quite a lot, for reasons I shall explain in a second.
A few months ago I had started giving a few talks at work on data wrangling, with CSV data being one of the most relevant in day-to-day operations.
I had come across posts and code using fzf or up with
jq
to provide a live as-you-type REPL interface, with commands being evaluated as one typed on the keyboard; since that was, imho, ideal for data exploration and demonstrations (I don't know about you, but immediate feedback when trying out pipelines of commands is a whole new world as far as I am concerned).Doing that with Miller was a bit complicated, because
jq
and many other programs have commands wrapped in a single string argument, while Miller takes (potentially) a thousand command-line arguments, so handling layers of quoting and the likes while abusing the hell out of thefzf
preview function proved a little problematic but vaguely/partially workable in the end.All of this to say that although the REPL is a significant improvement, providing an interface for as-you-type REPL, or an option/switch to facilitate this kind of use might be a good evolution, especially since it allows people to play around with data and reach a workable processing pipeline much more quickly. On top of that, it's really striking to see in demonstrations ;D.
Beta Was this translation helpful? Give feedback.
All reactions