GitHub/gf-core
1
0
Fork 1
mirror of https://github.com/GrammaticalFramework/gf-core.git synced 2026-05-23 01:52:50 -06:00
Code Issues Packages Projects Releases Wiki Activity

converted Hindi to the revised encoding

Browse Source
This commit is contained in:
aarne
2008-06-15 15:55:49 +00:00
parent c2bbdc8a16
commit 72b4ba9639
1 changed files with 7 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
src-3.0/GF/Text/Transliterations.hs
View File

@@ -23,7 +23,8 @@ characterTable = unlines . map prOne . Map.assocs . trans_from_unicode where
data Transliteration = Trans { data Transliteration = Trans {
trans_to_unicode :: Map.Map String Int, trans_to_unicode :: Map.Map String Int,
trans_from_unicode :: Map.Map Int String trans_from_unicode :: Map.Map Int String,
invisible_chars :: [String]
} }
appTransToUnicode :: Transliteration -> String -> String appTransToUnicode :: Transliteration -> String -> String
@@ -32,6 +33,7 @@ appTransToUnicode trans =
map (\c -> maybe c (return . toEnum) $ map (\c -> maybe c (return . toEnum) $
Map.lookup c (trans_to_unicode trans) Map.lookup c (trans_to_unicode trans)
) . ) .
filter (flip notElem (invisible_chars trans)) .
unchar unchar
appTransFromUnicode :: Transliteration -> String -> String appTransFromUnicode :: Transliteration -> String -> String
@@ -46,9 +48,10 @@ appTransFromUnicode trans =
-- conventions: -- conventions:
-- each character is either [letter] or [letter+nonletter] -- each character is either [letter] or [letter+nonletter]
-- when using a sparse range of unicodes, mark missing codes as "-" in transliterations -- when using a sparse range of unicodes, mark missing codes as "-" in transliterations
-- characters can be invisible: ignored in translation to unicode
mkTransliteration :: [String] -> [Int] -> Transliteration mkTransliteration :: [String] -> [Int] -> Transliteration
mkTransliteration ts us = Trans (Map.fromList (tzip ts us)) (Map.fromList (uzip us ts)) mkTransliteration ts us = Trans (Map.fromList (tzip ts us)) (Map.fromList (uzip us ts)) []
where where
tzip ts us = [(t,u) | (t,u) <- zip ts us, t /= "-"] tzip ts us = [(t,u) | (t,u) <- zip ts us, t /= "-"]
uzip us ts = [(u,t) | (u,t) <- zip us ts, t /= "-"] uzip us ts = [(u,t) | (u,t) <- zip us ts, t /= "-"]
@@ -75,9 +78,9 @@ transThai = mkTransliteration allTrans allCodes where
allCodes = [0x0e00 .. 0x0e7f] allCodes = [0x0e00 .. 0x0e7f]
transDevanagari :: Transliteration transDevanagari :: Transliteration
transDevanagari = mkTransliteration allTrans allCodes where transDevanagari = (mkTransliteration allTrans allCodes){invisible_chars = ["a"]} where
allTrans = words $ allTrans = words $
"~ * - - " ++ "M N - - " ++
"a- A- i- I- u- U- R- - - - e- E- - - o- O- " ++ "a- A- i- I- u- U- R- - - - e- E- - - o- O- " ++
"k K g G N: c C j J n: t. T. d. D. n. t " ++ "k K g G N: c C j J n: t. T. d. D. n. t " ++
"T d D n - p P b B m y r - l - - v " ++ "T d D n - p P b B m y r - l - - v " ++