[codespell]

ignore-words = .github/linters/codespell.txt

# corpus.rb is generated from the Unicode Character Database and carries the

# database's own names for the scripts, which read as misspelled English: Vai

# as "via". Nothing in it is prose. re_prop.h is generated from it too, and

# names a codepoint in hex beside every entry, U+AADD read as a misspelling.

skip = ./build,./doc/api,./doc/capi,*/difftest/corpus.rb,*/src/re_prop.h