Added char-sets for text.

2 files changed, 180 insertions, 6 deletions

diff --git a/stdlib/source/lux/data/text/encoding.lux b/stdlib/source/lux/data/text/encoding.lux
index bd1a255ec..6e617e75b 100644
--- a/stdlib/source/lux/data/text/encoding.lux
+++ b/stdlib/source/lux/data/text/encoding.lux
@@ -1,7 +1,11 @@
 (.module:
   [lux #*
+   [control
+    [codec (#+ Codec)]]
    [data
     ["." error (#+ Error)]]
+   [type
+    abstract]
    [world
     [binary (#+ Binary)]]
    [platform
@@ -9,19 +13,181 @@
      ["_" host]]]
    [host (#+ import:)]])
 
-(`` (for {(~~ (static _.jvm))
-          (as-is (def: utf8 Text "UTF-8")
+## https://docs.oracle.com/javase/8/docs/technotes/guides/intl/encoding.doc.html
+
+(abstract: #export Char-Set
+  {}
+
+  Text
+
+  (do-template [<name> <charset>]
+    [(def: #export <name> Char-Set (:abstraction <charset>))]
+
+    [ascii "ASCII"]
+
+    [ibm-37 "IBM037"]
+    [ibm-273 "IBM273"]
+    [ibm-277 "IBM277"]
+    [ibm-278 "IBM278"]
+    [ibm-280 "IBM280"]
+    [ibm-284 "IBM284"]
+    [ibm-285 "IBM285"]
+    [ibm-290 "IBM290"]
+    [ibm-297 "IBM297"]
+    [ibm-300 "IBM300"]
+    [ibm-420 "IBM420"]
+    [ibm-424 "IBM424"]
+    [ibm-437 "IBM437"]
+    [ibm-500 "IBM500"]
+    [ibm-737 "IBM737"]
+    [ibm-775 "IBM775"]
+    [ibm-833 "IBM833"]
+    [ibm-834 "IBM834"]
+    [ibm-838 "IBM-Thai"]
+    [ibm-850 "IBM850"]
+    [ibm-852 "IBM852"]
+    [ibm-855 "IBM855"]
+    [ibm-856 "IBM856"]
+    [ibm-857 "IBM857"]
+    [ibm-858 "IBM00858"]
+    [ibm-860 "IBM860"]
+    [ibm-861 "IBM861"]
+    [ibm-862 "IBM862"]
+    [ibm-863 "IBM863"]
+    [ibm-864 "IBM864"]
+    [ibm-865 "IBM865"]
+    [ibm-866 "IBM866"]
+    [ibm-868 "IBM868"]
+    [ibm-869 "IBM869"]
+    [ibm-870 "IBM870"]
+    [ibm-871 "IBM871"]
+    [ibm-874 "IBM874"]
+    [ibm-875 "IBM875"]
+    [ibm-918 "IBM918"]
+    [ibm-921 "IBM921"]
+    [ibm-922 "IBM922"]
+    [ibm-930 "IBM930"]
+    [ibm-933 "IBM933"]
+    [ibm-935 "IBM935"]
+    [ibm-937 "IBM937"]
+    [ibm-939 "IBM939"]
+    [ibm-942 "IBM942"]
+    [ibm-942c "IBM942C"]
+    [ibm-943 "IBM943"]
+    [ibm-943c "IBM943C"]
+    [ibm-948 "IBM948"]
+    [ibm-949 "IBM949"]
+    [ibm-949c "IBM949C"]
+    [ibm-950 "IBM950"]
+    [ibm-964 "IBM964"]
+    [ibm-970 "IBM970"]
+    [ibm-1006 "IBM1006"]
+    [ibm-1025 "IBM1025"]
+    [ibm-1026 "IBM1026"]
+    [ibm-1046 "IBM1046"]
+    [ibm-1047 "IBM1047"]
+    [ibm-1097 "IBM1097"]
+    [ibm-1098 "IBM1098"]
+    [ibm-1112 "IBM1112"]
+    [ibm-1122 "IBM1122"]
+    [ibm-1123 "IBM1123"]
+    [ibm-1124 "IBM1124"]
+    [ibm-1140 "IBM01140"]
+    [ibm-1141 "IBM01141"]
+    [ibm-1142 "IBM01142"]
+    [ibm-1143 "IBM01143"]
+    [ibm-1144 "IBM01144"]
+    [ibm-1145 "IBM01145"]
+    [ibm-1146 "IBM01146"]
+    [ibm-1147 "IBM01147"]
+    [ibm-1148 "IBM01148"]
+    [ibm-1149 "IBM01149"]
+    [ibm-1166 "IBM1166"]
+    [ibm-1364 "IBM1364"]
+    [ibm-1381 "IBM1381"]
+    [ibm-1383 "IBM1383"]
+    [ibm-33722 "IBM33722"]
+    
+    [iso-2022-cn "ISO-2022-CN"]
+    [iso2022-cn-cns "ISO2022-CN-CNS"]
+    [iso2022-cn-gb "ISO2022-CN-GB"]
+    [iso-2022-jp "ISO-2022-JP"]
+    [iso-2022-jp-2 "ISO-2022-JP-2"]
+    [iso-2022-kr "ISO-2022-KR"]
+    [iso-8859-1 "ISO-8859-1"]
+    [iso-8859-2 "ISO-8859-2"]
+    [iso-8859-3 "ISO-8859-3"]
+    [iso-8859-4 "ISO-8859-4"]
+    [iso-8859-5 "ISO-8859-5"]
+    [iso-8859-6 "ISO-8859-6"]
+    [iso-8859-7 "ISO-8859-7"]
+    [iso-8859-8 "ISO-8859-8"]
+    [iso-8859-9 "ISO-8859-9"]
+    [iso-8859-11 "iso-8859-11"]
+    [iso-8859-13 "ISO-8859-13"]
+    [iso-8859-15 "ISO-8859-15"]
 
-                 (import: java/lang/String
+    [mac-arabic "MacArabic"]
+    [mac-central-europe "MacCentralEurope"]
+    [mac-croatian "MacCroatian"]
+    [mac-cyrillic "MacCyrillic"]
+    [mac-dingbat "MacDingbat"]
+    [mac-greek "MacGreek"]
+    [mac-hebrew "MacHebrew"]
+    [mac-iceland "MacIceland"]
+    [mac-roman "MacRoman"]
+    [mac-romania "MacRomania"]
+    [mac-symbol "MacSymbol"]
+    [mac-thai "MacThai"]
+    [mac-turkish "MacTurkish"]
+    [mac-ukraine "MacUkraine"]
+    
+    [utf-8 "UTF-8"]
+    [utf-16 "UTF-16"]
+    [utf-32 "UTF-32"]
+
+    [windows-31j "windows-31j"]
+    [windows-874 "windows-874"]
+    [windows-949 "windows-949"]
+    [windows-950 "windows-950"]
+    [windows-1250 "windows-1250"]
+    [windows-1252 "windows-1252"]
+    [windows-1251 "windows-1251"]
+    [windows-1253 "windows-1253"]
+    [windows-1254 "windows-1254"]
+    [windows-1255 "windows-1255"]
+    [windows-1256 "windows-1256"]
+    [windows-1257 "windows-1257"]
+    [windows-1258 "windows-1258"]
+    [windows-iso2022jp "windows-iso2022jp"]
+    [windows-50220 "windows-50220"]
+    [windows-50221 "windows-50221"]
+    
+    [cesu-8 "CESU-8"]
+    [koi8-r "KOI8-R"]
+    [koi8-u "KOI8-U"]
+    )
+
+  (def: #export name
+    (-> Char-Set Text)
+    (|>> :representation))
+  )
+
+(`` (for {(~~ (static _.jvm))
+          (as-is (import: java/lang/String
                    (new [(Array byte) String])
                    (getBytes [String] (Array byte))))}))
 
 (def: #export (to-utf8 value)
   (-> Text Binary)
   (`` (for {(~~ (static _.jvm))
-            (String::getBytes ..utf8 (:coerce String value))})))
+            (String::getBytes (..name ..utf-8) (:coerce String value))})))
 
 (def: #export (from-utf8 value)
   (-> Binary (Error Text))
   (`` (for {(~~ (static _.jvm))
-            (#error.Success (String::new value ..utf8))})))
+            (#error.Success (String::new value (..name ..utf-8)))})))
+
+(structure: #export UTF-8 (Codec Binary Text)
+  (def: encode ..to-utf8)
+  (def: decode ..from-utf8))
diff --git a/stdlib/source/lux/world/net/http/mime.lux b/stdlib/source/lux/world/net/http/mime.lux
index 4f090a54f..191a998ff 100644
--- a/stdlib/source/lux/world/net/http/mime.lux
+++ b/stdlib/source/lux/world/net/http/mime.lux
@@ -1,5 +1,8 @@
 (.module:
   [lux #*
+   [data
+    [text
+     ["." encoding (#+ Char-Set)]]]
    [type
     abstract]])
 
@@ -68,7 +71,6 @@
   [tiff "image/tiff"]
   [typescript "application/typescript"]
   [truetype-font "font/ttf"]
-  [text "text/plain"]
   [microsoft-visio "application/vnd.visio"]
   [wav "audio/wav"]
   [webm-audio "audio/webm"]
@@ -88,3 +90,9 @@
   [!3gpp2 "video/3gpp2"]
   [!7z "application/x-7z-compressed"]
   )
+
+(def: #export (text char-set)
+  (-> Char-Set MIME)
+  (..mime (format "text/plain; charset=" text.double-quotes (encoding.name char-set) text.double-quotes)))
+
+(def: #export utf-8 MIME (..text encoding.utf-8))