Note that I barely tested this , so you'll be the official tester now.
It should be very fast (depending on your Everything config, of course, but it's designed with speed in mind), both for recursive and non-recursive columns.
I tested with Everything v1.5 on my machine. The script is very basic, so it assumes you already have Everything properly configured (created/modified properties indexed, etc.).
It's designed to reuse the data as much as possible, so it doesn't query Everything for each column. Sometimes this can work against it, (I think it's some quirks with multi-column behavior).
Currently subfolders are ignored when picking the data. You can change that in the code if you want, by editing the Everything query.
As for filenames related columns, it currently shows the full path. Edit as needed.
Let me know how it goes (maybe this one deserves its own thread here ?).
Itās working as expected in my first tests. I have several folders that only contain subfolders and no files, so I removed the file: filter from the query to also consider folders (esp. relevant in the non-recursive version). There seems to be some caching going on, as the removed file: filter does not seem to be respected for folders that Iāve tested your script against before. (Restarting DOpus hasnāt helped, restarting the Everything service / computer might help, will test that.)
Btw, I donāt think this is actually a ānicheā use case: having folders for each project with many sub-files and sub-folders, the modified date of the project-folder itself isnāt of much value to actually sort by āwhich project have I last worked onā. So looking forward, this might as well become a native feature at some point. (CC: @Leo@Jon)
That said, I think this script definately deserves its own thread. Again, thanks so much.
I'll just test it myself this time. I found some quirks too. The good news is that all of them are fixed now. I'm going to post a formal version later.
Your example is the definition of "niche"
I'm not sure if something like this would make sense to be built-in, I'm not the gatekeeper for that type of requests, but FWIW, with a script column I wrote, I can get size/file counts faster than the built-in columns. The "fixed" version of this one is also really fast.
Do you mean relative date columns like āX days agoā? Definitely helpful for me, though Iām also fine by solving this via an evaluator column for consistency. (I also have eval. cols. configured for the regular modifieddate column.)
, so you'll be the official tester now.
?).
Again, thanks so much.
