Finding the latest screenshot in macOS Mojave

• Posted 11 March 2019
• Tagged with macos, screenshots, shell-scripting

One of the things that changed in macOS Mojave was the format of screenshot filenames. On older versions of macOS, the filename would be something like:

Screen Shot 2016-10-10 at 18.34.18.png

On Mojave, the first two words got collapsed into one:

Screenshot 2019-03-08 at 18.38.41.png

I have a handful of scripts for doing something with screenshots – and in particular, a shortcut that grabs the newest screenshot. When I started updating to Mojave, I had to update the shell snippet that powers that shortcut. Because I couldn’t update to Mojave on every machine immediately, it had to work with both naming schemes.

This is what I’ve been using for the last few months (bound to last_screenshot in my shell config):

find ~/Desktop -name 'Screen Shot*' -print0 -o -name 'Screenshot*' -print0
  | xargs -0 stat -f '%m %N'
  | sort --numeric-sort --reverse
  | head -1
  | cut -f "2-" -d " "

Let’s break it down:

• The find command looks for files in ~/Desktop (where my screenshots get saved) that match the filename Screen Shot* or Screenshot*.

  It prints the name of every matching file, separated by the null character (as set by the -print0 flag). Because most shell languages don’t have a proper list type, just strings, the null character is a way to shoehorn a list into a string. It’s less likely to appear in one of your list elements than, say, a newline or a space.

• xargs -0 unpacks the null character-separated string, and passes each filename to the stat command. This prints information about the file: %m is the last modified time, and %N is the filename.

  You get a list a bit like this:

  1551602947 /Users/alexwlchan/Desktop/Screenshot 2019-03-03 at 08.49.01.png
  1552070046 /Users/alexwlchan/Desktop/Screenshot 2019-03-08 at 18.34.00.png
  1552260786 /Users/alexwlchan/Desktop/Screenshot 2019-03-10 at 23.33.01.png
  1552070259 /Users/alexwlchan/Desktop/Screenshot 2019-03-08 at 18.37.33.png
  1552070326 /Users/alexwlchan/Desktop/Screenshot 2019-03-08 at 18.38.41.png
  1552070066 /Users/alexwlchan/Desktop/Screenshot 2019-03-08 at 18.34.19.png
  

• That entire string in turn gets passed to sort, which treats the strings as numeric (so 3 sorts below 10, for example), then reverses the order. This puts the biggest number – the newest file modification date – at the top.

• The sorted list is passed to head, which extracts the top line (the newest file).

• Finally, cut separates the string on spaces (-d " "), then prints the second element and everything after it – throwing away the timestamp, and leaving the filename.

It’s certainly possible to do this with a higher-level language like Python or Ruby, but I like the elegance of chaining together tiny utilities like this. For non-critical code, I enjoy the brevity.