I usually copy text to the clipboard, run
$ pbpaste | hunspell -d en_QS | pbcopy, and then paste the spell-check results into a new document.
pbpaste is built into OS X, while
hunspell comes from homebrew.
I use lint.py to make sure that I’m not accidentally violating spelling rules in Read’s own manual. I use pragmas in the comments to tell the automated linter that some constructions, in some words, are OK. These are:
- utter-low-ok — this word ends in •utter•low, and that’s OK
- et-no-tea-ok (e.g. “spent”)
- it-ing-ok (e.g. “bring”) — also automatically OK for nouns (po:noun)
- may-utter-ok (e.g. “demon”)
If you want to check for duplicate lines in BBEdit or TextWrangler, I use the following pattern:
I then match using the specific subsets “\1 \4”. Duplicates get shunted to a new document so I can check and resolve differences between, say, foo, foo/z, and foo/zvs.