513058a · typeable/pandoc

git.iko.soy/typeable/pandoc

XML: change toEntities to emit numerical hex character references.

Previously decimal references were used. But Polyglot Markup prefers hex. See #5718. This affects the output of pandoc with `--ascii`.

commit513058a24eebb97a849e7eb5051e29ce2f6642c6 authorJohn MacFarlane <jgm@berkeley.edu> dateTue 3 Sep 2019, 11:28 (7 years ago) parent0fe635d

2 files changed+5−4

compare with master

unifiedsplit

src/Text/Pandoc/XML.hs+2−1 test/command/ascii.md+3−3

src/Text/Pandoc/XML.hs+2−1

@@ -26,6 +26,7 @@ import Data.Text (Text)

2626import qualified Data.Text as T

2727import Text.HTML.TagSoup.Entity (lookupEntity, htmlEntities)

2828import Text.DocLayout

29+import Text.Printf (printf)

2930import qualified Data.Map as M

3031import Data.String

3132

@@ -89,7 +90,7 @@ inTagsIndented tagType = inTags True tagType []

8990toEntities :: Text -> Text

9091toEntities = T.concatMap go

9192 where go c | isAscii c = T.singleton c

92− | otherwise = T.pack ("&#" ++ show (ord c) ++ ";")

93+ | otherwise = T.pack (printf "&#x%X;" (ord c))

9394

9495-- | Escape all non-ascii characters using HTML5 entities, falling

9596-- back to numerical entities.

test/command/ascii.md+3−3

@@ -2,7 +2,7 @@

22pandoc -t html --ascii

33äéıå

44^D

5−äéıå

5+äéıå

66```

88```

@@ -33,7 +33,7 @@ pandoc -t docbook --ascii

3333äéıå

3434^D

3535<para>

36− äéıå

36+ äéıå

3737</para>

3838```

3939

@@ -41,7 +41,7 @@ pandoc -t docbook --ascii

4141pandoc -t jats --ascii

4242äéıå

4343^D

44−äéıå

44+äéıå

4545```

4646

4747```