Instead of having anagramSolutionsX.json and dictionary.json checked into github we should have:
the source text files (lists of words and phrases) checked into GitHub in the scripts directory
git rm the .json files
build the .json files on deployment (in Travis) - i.e. generate the .json files from the .txt files at deployment-time, just prior to creating the Lambda zip files
This will give us a very simple way of adding things to our dictionaries:
Instead of having
anagramSolutionsX.json
anddictionary.json
checked into github we should have:git rm
the .json filesThis will give us a very simple way of adding things to our dictionaries: