This program takes a list of words, looks up their definitions in the Oxford dictionary API, formats the resulting data into HTML that looks like the iOS 'look up' feature, and saves it to an output file. This file can then be imported into a flashcard program that supports HTML/CSS like Anki.
I made this project because I had just finished reading the book Dune (awesome book, by the way!), and I kept track of all the words I didn't know yet. I wanted to study the words in this flashcard program called Anki. However that's where the headaches began. It required me to manually get all the definitions of the words from somewhere and then format them into HTML myself. On top of that there were quite a few words (see inputData/sampleWordList.txt), so I didn't want to bother doing manual labor. Programming to the rescue!
This project parses all the words (separated by a newline) from the inputData/wordList.txt file. It then goes through each word and gets its lemmas from the Oxford dictionary API (https://www.oxfordlearnersdictionaries.com/), in order to get all forms of the root of the provided word ('swimming' returns both 'swimming' and 'swim'). It filters all duplicate lemmas and then goes though each lemma and gets its definition from the Oxford dictionary API. It proceeds to parse the data and format it into HTML (I tried to emulate the iOS/macOS 'look up' style) before finally appending that data to an output file in the outputData folder.
- First you need Node, preferably v12.22.5.
- Get access to your API credentials through https://developer.oxforddictionaries.com/
- Create a
.envfile in the root and add the following credentials to it:
APP_ID=<your app_id>
APP_KEY=<your app_key>- Run
npm install - Fill the
inputData/wordList.txtwith the words you seek to learn, separated by a newline (use the 'enter' or 'return' keys) - Run
npm start
If using Anki:
- Finally, you can simply import that file into Anki by going to Decks -> Import File and then navigating to the correct file in
outputData.
In order to get the proper styling, you need to use CSS in styling.css.
If you're using Anki, go to Decks -> your deck -> Study now -> edit (bottom left) -> Cards... (top left) -> styling -> (you can comment out your current styling by putting it between /* and */) paste the styling from styling.css in there -> save -> close.
If you need to make this styling work with any existing styling, then you'll need to manually change the class names in the code yourself to something that won't overlap with your own styles, and then rerun the program.
This program uses the en-us dictionary. If you wish to use the gb one, change it in the url on line 61 of src/prepWordListForAnki.ts, and rerun the program.
The free tier of the Oxford dictionary allows only 60 requests per minute. That means we would have to limit our wordlist to +- 20 entries when running the program, considering each word requires 1 request to get the lemmas, and then 1 request per non-duplicate lemma. But because we don't like extra effort, there is a circumventable 2s timeout after every definition request in order to stay within the limits.
Be prepared to let your computer run a while. On my laptop it took 21 minutes to run the program for 421 words.
If you prefer not to wait 2s before each definition request, perhaps because you have a small word list, then set SKIP_DEFINITION_TIMEOUT=true, in your .env. Once it finishes you can load the next +- 20 words, etc. a minute later. If you're a big shot and you've got a paid plan without limitations, then by all means throw in all your words and get your results fast!
Enoy :)