0327226 · typeable/pandoc

git.iko.soy/typeable/pandoc

Man reader: don't parse \[ul] as unicode escape.

commit0327226d4c8976bd997fd00dea1faf4120905c27 authorJohn MacFarlane <jgm@berkeley.edu> dateMon 22 Oct 2018, 12:05 (8 years ago) parent7a111d2

2 files changed+4−2

unifiedsplit

src/Text/Pandoc/Readers/Man.hs+3−1

@@ -204,9 +204,11 @@ escapeLexer = try $ do

204204 <|> charCode `sepBy1` (many1 Parsec.space)

205205 ) <* char ']'

206206

207− ucharCode = do

207+ ucharCode = try $ do

208208 char 'u'

209209 cs <- many1 (satisfy isHexDigit)

210+ let lcs = length cs

211+ guard $ lcs >= 4 && lcs <= 6

210212 case chr <$> safeRead ('0':'x':cs) of

211213 Nothing -> escUnknown ("\\[u" ++ cs ++ "]") '\xFFFD'

212214 Just c -> return c

test/Tests/Readers/Man.hs+1−1

@@ -76,7 +76,7 @@ tests = [

7676 , "two letter escapes" =:

7777 "\\(oA\\(~O" =?> (para $ text "ÅÕ")

7878 , "bracketed escapes" =:

79− "\\[oA]\\[~O]\\[Do]\\[Ye]\\[product]" =?> (para $ text "ÅÕ$¥∏")

79+ "\\[oA]\\[~O]\\[Do]\\[Ye]\\[product]\\[ul]" =?> (para $ text "ÅÕ$¥∏_")

8080 , "unicode escapes" =:

8181 "\\[u2020]" =?> (para $ text "†")

8282 , "unicode escapes (combined)" =: