Irreal

If you enjoyed yesterday’s post on Mike Zamansky’s use of Clojure and Cider to solve an Advent of Code problem, he has a couple more offerings that you’ll probably like too. Like his last video, these two consider a current Advent of Code problem and apply Clojure and Cider to their solution.

If you don’t like Clojure, the first post above also considers the Python solution in the text but not the video. I don’t know any Clojure but it’s a Lisp and familiar enough that I had no problem following along. Even if you’re a complete Lisp n00b, Zamansky’s last 3 videos do a good job introducing some of the syntax and concepts.

As with yesterday’s post, these two approach the problems in an interactive or exploratory way. You can watch Zamansky build up the solution piece by piece and even do a bit of debugging when he gets stuck at one point.

I’ve always found the fact that Clojure is hosted on the JVM off-putting because of security and resource hog concerns. Still, I really enjoyed Zamansky’s videos and felt, despite my misgivings, an urge to try it out. Perhaps I will. Clojure is a far better way of dealing with the Java environment than Java itself.

Mike Zamansky has another video up. It’s not quite one of his Using Emacs Series videos but he says it could be as it discusses the Emacs Cider infrastructure. It’s ostensibly about day 3 of this year’s Advent of Code, and considers his solution to the day’s problem. To me, the interesting part was his introduction to and demonstration of the Clojure/Cider environment in Emacs.

Zamansky says that Cider is the best development environment he’s ever used. It reminds me of the Slime environment for Common Lisp or Geiser for Scheme. You have a REPL in one window and can enter code in another. You can try out bits of code in the code window as you enter it so Cider enables the interactive programming style that I find so attractive and useful.

Zamansky provides a nice example of the technique as he pieces together his solution to the Advent of Code problem. One nice bit is his use of a Clojure threading macro. The ->> macro that he uses doesn’t really introduce any new capabilities but it does make the code a little easier to read. I think it fits in nicely with the interactive programming paradigm because it makes it easy to use what’s already been calculated as input to another function.

The video is just short of 38 minutes so you’ll need to schedule some time but, as usual, it’s well worth watching. If you’re curious about Clojure, it’s also a nice introduction to it and Cider.

One of the things that Vi/Vim does much better than Emacs is repeat commands. Pressing . in Vim repeats the last native command. It simple and easy to use and when I was a Vim user, I did, in fact, use it all the time. Its real power, I think, comes from the fact that a single easy to reach key will repeat whatever you just did.

The situation in Emacs is not as nice. If you want to repeat the last simple command (one that didn’t involve the minibuffer) you can do so with Ctrl+x z. But that’s a lot harder to type than a single period and might even be harder to type than the command you want to repeat. As a result, I never bother with it. It’s just too much trouble.

Then there’s the repeat-complex-command function that you can use to repeat a command that involves the minibuffer. Marcin ‘mbork’ Borkowski, who has a knack for discovering obscure Emacs commands, has a nice post on repeating complex commands if you want more details. It has a horrible default key binding but even if you remap it, it’s still too much trouble to bother with. I know this is true because although I wrote about repeat-complex-command four and a half years ago, I had no memory of it or ever using it when I saw Borkowski’s post.

Still, if you’re looking for a way to repeat Emacs commands, take a look at Borkowski’s post. Perhaps you’ll find the Emacs repeat commands more useful than I do.

My day has been consumed by yak shaving so today’s offering is just a quickie pointing you to a very interesting post from Álvaro Ramírez. In a situation familiar to all of us, Ramírez noticed that every time he wanted to clone a repository, he had to go through the same set of simple but boring steps. Of course when you’re using the world’s most flexible editor that’s a problem crying out for the obvious solution.

Ramírez wanted a solution that would extract the repository URL from the clipboard and clone it into the proper directory. That’s pretty easy to do and he shows a nice bit of Elisp to take care of the task.

This sort of thing happens to me all the time: I’ll be doing some rote chore for the \(n\)^th time and suddenly realize that there’s no reason I should be doing this by hand yet again. It’s usually pretty easy to write a little Elisp and reduce the task to a key press.