Álvaro Ramírez has a nice post that expands on his theme of gluing Emacs functions together. It’s a nice example in part because it’s so out of the ordinary and seemingly impossible from within Emacs. It turns out that Ramírez is learning Japanese and wanted to import some vocabulary from a class handout into Emacs/Org-mode for further processing and eventual export to Anki so that he can employ spaced repetition learning on the vocabulary. I can sympathize. I’m learning Spanish and getting word lists into a usable form is a pain.
The process is instructive. He begins by calling the macOS screen capture utility from Emacs to get the data into Emacs. The he uses his chatgpt-shell that I wrote about earlier to get the data into an Org table. Once you have that, you can do almost anything.
In his case that means exporting it to Anki so that he can have it on his iPhone. He uses a function, org-table-get-field, that I didn’t know about to extract data from the rows of the table. Take a look at his short code snippet to see how it works.
Ramírez’s post is a great example how you can import the most improbable of work flows into Emacs by using just a bit of Elisp as glue code. Most of us aren’t gong to need to do what Ramírez is doing, let alone use Chatgpt, in our workflows but his post is a great example of how to do whatever you need doing from within Emacs.