LaTeX writer: ignore languages with no babel equivalent...

instead of generating an invalid command in the preamble.
Closes #8325.

3 files changed, 115 insertions, 117 deletions

diff --git a/src/Text/Pandoc/Writers/LaTeX.hs b/src/Text/Pandoc/Writers/LaTeX.hs
index 2090e7bcd..cd1f59056 100644
--- a/src/Text/Pandoc/Writers/LaTeX.hs
+++ b/src/Text/Pandoc/Writers/LaTeX.hs
@@ -208,9 +208,9 @@ pandocToLaTeX options (Pandoc meta blocks) = do
           maybe id (\l -> defField "lang"
                       (literal $ renderLang l)) mblang
         $ maybe id (\l -> defField "babel-lang"
-                      (literal $ toBabel l)) mblang
+                      (literal l)) (mblang >>= toBabel)
         $ defField "babel-otherlangs"
-             (map (literal . toBabel) docLangs)
+             (map literal $ mapMaybe toBabel docLangs)
         $ defField "latex-dir-rtl"
            ((render Nothing <$> getField "dir" context) ==
                Just ("rtl" :: Text)) context
@@ -738,10 +738,9 @@ inlineToLaTeX (Span (id',classes,kvs) ils) = do
       kvToCmd ("dir","ltr") = Just "LR"
       kvToCmd _ = Nothing
       langCmds =
-        case lang of
-           Just lng -> let l = toBabel lng
-                       in  ["foreignlanguage{" <> l <> "}"]
-           Nothing  -> []
+        case lang >>= toBabel of
+           Just l  -> ["foreignlanguage{" <> l <> "}"]
+           Nothing -> []
   let cmds = mapMaybe classToCmd classes ++ mapMaybe kvToCmd kvs ++ langCmds
   contents <- inlineListToLaTeX ils
   return $
diff --git a/src/Text/Pandoc/Writers/LaTeX/Lang.hs b/src/Text/Pandoc/Writers/LaTeX/Lang.hs
index 3c2bfd262..60a2ea8d8 100644
--- a/src/Text/Pandoc/Writers/LaTeX/Lang.hs
+++ b/src/Text/Pandoc/Writers/LaTeX/Lang.hs
@@ -21,125 +21,125 @@ import Text.Collate.Lang (Lang(..))
 -- http://mirrors.ctan.org/macros/latex/required/babel/base/babel.pdf
 -- List of supported languages (slightly outdated):
 -- http://tug.ctan.org/language/hyph-utf8/doc/generic/hyph-utf8/hyphenation.pdf
-toBabel :: Lang -> Text
+toBabel :: Lang -> Maybe Text
 toBabel (Lang "de" _ (Just "AT") vars _ _)
-  | "1901" `elem` vars                  = "austrian"
-  | otherwise                           = "naustrian"
+  | "1901" `elem` vars                  = Just "austrian"
+  | otherwise                           = Just "naustrian"
 toBabel (Lang "de" _ (Just "CH") vars _ _)
-  | "1901" `elem` vars                  = "swissgerman"
-  | otherwise                           = "nswissgerman"
+  | "1901" `elem` vars                  = Just "swissgerman"
+  | otherwise                           = Just "nswissgerman"
 toBabel (Lang "de" _ _ vars _ _)
-  | "1901" `elem` vars                  = "german"
-  | otherwise                           = "ngerman"
-toBabel (Lang "dsb" _ _ _ _ _)          = "lowersorbian"
+  | "1901" `elem` vars                  = Just "german"
+  | otherwise                           = Just "ngerman"
+toBabel (Lang "dsb" _ _ _ _ _)          = Just "lowersorbian"
 toBabel (Lang "el" _ _ vars _ _)
-  | "polyton" `elem` vars               = "polutonikogreek"
-toBabel (Lang "en" _ (Just "AU") _ _ _) = "australian"
-toBabel (Lang "en" _ (Just "CA") _ _ _) = "canadian"
-toBabel (Lang "en" _ (Just "GB") _ _ _) = "british"
-toBabel (Lang "en" _ (Just "NZ") _ _ _) = "newzealand"
-toBabel (Lang "en" _ (Just "UK") _ _ _) = "british"
-toBabel (Lang "en" _ (Just "US") _ _ _) = "american"
-toBabel (Lang "fr" _ (Just "CA") _ _ _) = "canadien"
+  | "polyton" `elem` vars               = Just "polutonikogreek"
+toBabel (Lang "en" _ (Just "AU") _ _ _) = Just "australian"
+toBabel (Lang "en" _ (Just "CA") _ _ _) = Just "canadian"
+toBabel (Lang "en" _ (Just "GB") _ _ _) = Just "british"
+toBabel (Lang "en" _ (Just "NZ") _ _ _) = Just "newzealand"
+toBabel (Lang "en" _ (Just "UK") _ _ _) = Just "british"
+toBabel (Lang "en" _ (Just "US") _ _ _) = Just "american"
+toBabel (Lang "fr" _ (Just "CA") _ _ _) = Just "canadien"
 toBabel (Lang "fra" _ _ vars _ _)
-  | "aca" `elem` vars                   = "acadian"
-toBabel (Lang "grc" _ _ _ _ _)          = "ancientgreek"
-toBabel (Lang "hsb" _ _ _ _ _)          = "uppersorbian"
+  | "aca" `elem` vars                   = Just "acadian"
+toBabel (Lang "grc" _ _ _ _ _)          = Just "ancientgreek"
+toBabel (Lang "hsb" _ _ _ _ _)          = Just "uppersorbian"
 toBabel (Lang "la" _ _ vars _ _)
-  | "x-classic" `elem` vars             = "classiclatin"
-toBabel (Lang "pt" _ (Just "BR") _ _ _) = "brazilian"
-toBabel (Lang "sl" _ _ _ _ _)           = "slovene"
+  | "x-classic" `elem` vars             = Just "classiclatin"
+toBabel (Lang "pt" _ (Just "BR") _ _ _) = Just "brazilian"
+toBabel (Lang "sl" _ _ _ _ _)           = Just "slovene"
 toBabel x                               = commonFromBcp47 x
 
 -- Takes a list of the constituents of a BCP47 language code
 -- and converts it to a string shared by Babel and Polyglossia.
 -- https://tools.ietf.org/html/bcp47#section-2.1
-commonFromBcp47 :: Lang -> Text
-commonFromBcp47 (Lang "sr" (Just "Cyrl") _ _ _ _)      = "serbianc"
+commonFromBcp47 :: Lang -> Maybe Text
+commonFromBcp47 (Lang "sr" (Just "Cyrl") _ _ _ _)      = Just "serbianc"
 commonFromBcp47 (Lang "zh" (Just "Latn") _ vars _ _)
-  | "pinyin" `elem` vars                               = "pinyin"
+  | "pinyin" `elem` vars                               = Just "pinyin"
 commonFromBcp47 (Lang l _ _ _ _ _) = fromIso l
   where
-    fromIso "af"  = "afrikaans"
-    fromIso "am"  = "amharic"
-    fromIso "ar"  = "arabic"
-    fromIso "as"  = "assamese"
-    fromIso "ast" = "asturian"
-    fromIso "bg"  = "bulgarian"
-    fromIso "bn"  = "bengali"
-    fromIso "bo"  = "tibetan"
-    fromIso "br"  = "breton"
-    fromIso "ca"  = "catalan"
-    fromIso "cy"  = "welsh"
-    fromIso "cs"  = "czech"
-    fromIso "cop" = "coptic"
-    fromIso "da"  = "danish"
-    fromIso "dv"  = "divehi"
-    fromIso "el"  = "greek"
-    fromIso "en"  = "english"
-    fromIso "eo"  = "esperanto"
-    fromIso "es"  = "spanish"
-    fromIso "et"  = "estonian"
-    fromIso "eu"  = "basque"
-    fromIso "fa"  = "farsi"
-    fromIso "fi"  = "finnish"
-    fromIso "fr"  = "french"
-    fromIso "fur" = "friulan"
-    fromIso "ga"  = "irish"
-    fromIso "gd"  = "scottish"
-    fromIso "gez" = "ethiopic"
-    fromIso "gl"  = "galician"
-    fromIso "gu"  = "gujarati"
-    fromIso "he"  = "hebrew"
-    fromIso "hi"  = "hindi"
-    fromIso "hr"  = "croatian"
-    fromIso "hu"  = "magyar"
-    fromIso "hy"  = "armenian"
-    fromIso "ia"  = "interlingua"
-    fromIso "id"  = "indonesian"
-    fromIso "ie"  = "interlingua"
-    fromIso "is"  = "icelandic"
-    fromIso "it"  = "italian"
-    fromIso "ja"  = "japanese"
-    fromIso "km"  = "khmer"
-    fromIso "kmr" = "kurmanji"
-    fromIso "kn"  = "kannada"
-    fromIso "ko"  = "korean"
-    fromIso "la"  = "latin"
-    fromIso "lo"  = "lao"
-    fromIso "lt"  = "lithuanian"
-    fromIso "lv"  = "latvian"
-    fromIso "ml"  = "malayalam"
-    fromIso "mn"  = "mongolian"
-    fromIso "mr"  = "marathi"
-    fromIso "nb"  = "norsk"
-    fromIso "nl"  = "dutch"
-    fromIso "nn"  = "nynorsk"
-    fromIso "no"  = "norsk"
-    fromIso "nqo" = "nko"
-    fromIso "oc"  = "occitan"
-    fromIso "or"  = "oriya"
-    fromIso "pa"  = "punjabi"
-    fromIso "pl"  = "polish"
-    fromIso "pms" = "piedmontese"
-    fromIso "pt"  = "portuguese"
-    fromIso "rm"  = "romansh"
-    fromIso "ro"  = "romanian"
-    fromIso "ru"  = "russian"
-    fromIso "sa"  = "sanskrit"
-    fromIso "se"  = "samin"
-    fromIso "sk"  = "slovak"
-    fromIso "sq"  = "albanian"
-    fromIso "sr"  = "serbian"
-    fromIso "sv"  = "swedish"
-    fromIso "syr" = "syriac"
-    fromIso "ta"  = "tamil"
-    fromIso "te"  = "telugu"
-    fromIso "th"  = "thai"
-    fromIso "ti"  = "ethiopic"
-    fromIso "tk"  = "turkmen"
-    fromIso "tr"  = "turkish"
-    fromIso "uk"  = "ukrainian"
-    fromIso "ur"  = "urdu"
-    fromIso "vi"  = "vietnamese"
-    fromIso _     = ""
+    fromIso "af"  = Just "afrikaans"
+    fromIso "am"  = Just "amharic"
+    fromIso "ar"  = Just "arabic"
+    fromIso "as"  = Just "assamese"
+    fromIso "ast" = Just "asturian"
+    fromIso "bg"  = Just "bulgarian"
+    fromIso "bn"  = Just "bengali"
+    fromIso "bo"  = Just "tibetan"
+    fromIso "br"  = Just "breton"
+    fromIso "ca"  = Just "catalan"
+    fromIso "cy"  = Just "welsh"
+    fromIso "cs"  = Just "czech"
+    fromIso "cop" = Just "coptic"
+    fromIso "da"  = Just "danish"
+    fromIso "dv"  = Just "divehi"
+    fromIso "el"  = Just "greek"
+    fromIso "en"  = Just "english"
+    fromIso "eo"  = Just "esperanto"
+    fromIso "es"  = Just "spanish"
+    fromIso "et"  = Just "estonian"
+    fromIso "eu"  = Just "basque"
+    fromIso "fa"  = Just "farsi"
+    fromIso "fi"  = Just "finnish"
+    fromIso "fr"  = Just "french"
+    fromIso "fur" = Just "friulan"
+    fromIso "ga"  = Just "irish"
+    fromIso "gd"  = Just "scottish"
+    fromIso "gez" = Just "ethiopic"
+    fromIso "gl"  = Just "galician"
+    fromIso "gu"  = Just "gujarati"
+    fromIso "he"  = Just "hebrew"
+    fromIso "hi"  = Just "hindi"
+    fromIso "hr"  = Just "croatian"
+    fromIso "hu"  = Just "magyar"
+    fromIso "hy"  = Just "armenian"
+    fromIso "ia"  = Just "interlingua"
+    fromIso "id"  = Just "indonesian"
+    fromIso "ie"  = Just "interlingua"
+    fromIso "is"  = Just "icelandic"
+    fromIso "it"  = Just "italian"
+    fromIso "ja"  = Just "japanese"
+    fromIso "km"  = Just "khmer"
+    fromIso "kmr" = Just "kurmanji"
+    fromIso "kn"  = Just "kannada"
+    fromIso "ko"  = Just "korean"
+    fromIso "la"  = Just "latin"
+    fromIso "lo"  = Just "lao"
+    fromIso "lt"  = Just "lithuanian"
+    fromIso "lv"  = Just "latvian"
+    fromIso "ml"  = Just "malayalam"
+    fromIso "mn"  = Just "mongolian"
+    fromIso "mr"  = Just "marathi"
+    fromIso "nb"  = Just "norsk"
+    fromIso "nl"  = Just "dutch"
+    fromIso "nn"  = Just "nynorsk"
+    fromIso "no"  = Just "norsk"
+    fromIso "nqo" = Just "nko"
+    fromIso "oc"  = Just "occitan"
+    fromIso "or"  = Just "oriya"
+    fromIso "pa"  = Just "punjabi"
+    fromIso "pl"  = Just "polish"
+    fromIso "pms" = Just "piedmontese"
+    fromIso "pt"  = Just "portuguese"
+    fromIso "rm"  = Just "romansh"
+    fromIso "ro"  = Just "romanian"
+    fromIso "ru"  = Just "russian"
+    fromIso "sa"  = Just "sanskrit"
+    fromIso "se"  = Just "samin"
+    fromIso "sk"  = Just "slovak"
+    fromIso "sq"  = Just "albanian"
+    fromIso "sr"  = Just "serbian"
+    fromIso "sv"  = Just "swedish"
+    fromIso "syr" = Just "syriac"
+    fromIso "ta"  = Just "tamil"
+    fromIso "te"  = Just "telugu"
+    fromIso "th"  = Just "thai"
+    fromIso "ti"  = Just "ethiopic"
+    fromIso "tk"  = Just "turkmen"
+    fromIso "tr"  = Just "turkish"
+    fromIso "uk"  = Just "ukrainian"
+    fromIso "ur"  = Just "urdu"
+    fromIso "vi"  = Just "vietnamese"
+    fromIso _     = Nothing
diff --git a/src/Text/Pandoc/Writers/LaTeX/Util.hs b/src/Text/Pandoc/Writers/LaTeX/Util.hs
index a0d207a70..9ff46f587 100644
--- a/src/Text/Pandoc/Writers/LaTeX/Util.hs
+++ b/src/Text/Pandoc/Writers/LaTeX/Util.hs
@@ -237,9 +237,8 @@ wrapDiv (_,classes,kvs) t = do
                   Just "rtl" -> align "RTL"
                   Just "ltr" -> align "LTR"
                   _          -> id
-      wrapLang txt = case lang of
-                       Just lng -> let l = toBabel lng
-                                   in  inCmd "begin" "otherlanguage"
+      wrapLang txt = case lang >>= toBabel of
+                       Just l -> inCmd "begin" "otherlanguage"
                                             <> (braces (literal l))
                                        $$ blankline <> txt <> blankline
                                        $$ inCmd "end" "otherlanguage"