diff options
Diffstat (limited to 'src/GF')
| -rw-r--r-- | src/GF/Compile/Coding.hs | 7 | ||||
| -rw-r--r-- | src/GF/Text/CP1251.hs | 18 | ||||
| -rw-r--r-- | src/GF/Text/Lexing.hs | 16 |
3 files changed, 20 insertions, 21 deletions
diff --git a/src/GF/Compile/Coding.hs b/src/GF/Compile/Coding.hs index 9d271964f..a226265c5 100644 --- a/src/GF/Compile/Coding.hs +++ b/src/GF/Compile/Coding.hs @@ -3,6 +3,7 @@ module GF.Compile.Coding where import GF.Grammar.Grammar import GF.Grammar.Macros import GF.Text.UTF8 +import GF.Text.CP1251 import GF.Infra.Modules import GF.Infra.Option import GF.Data.Operations @@ -47,9 +48,3 @@ codeSourceModule co (id,moi) = case moi of PSeq p q -> PSeq (codp p) (codp q) PAlt p q -> PAlt (codp p) (codp q) _ -> p - ----- from Text.Lexing -decodeCP1251 = map convert where - convert c - | c >= '\192' && c <= '\255' = chr (ord c + 848) - | otherwise = c diff --git a/src/GF/Text/CP1251.hs b/src/GF/Text/CP1251.hs new file mode 100644 index 000000000..af5c829a2 --- /dev/null +++ b/src/GF/Text/CP1251.hs @@ -0,0 +1,18 @@ +module GF.Text.CP1251 where
+
+import Data.Char
+
+decodeCP1251 = map convert where
+ convert c
+ | c >= '\192' && c <= '\255' = chr (ord c + 848)
+ | c == '\168' = chr 1025 -- cyrillic capital letter lo
+ | c == '\184' = chr 1105 -- cyrillic small letter lo
+ | otherwise = c
+
+encodeCP1251 = map convert where
+ convert c
+ | oc >= 1040 && oc <= 1103 = chr (oc - 848)
+ | oc == 1025 = chr 168 -- cyrillic capital letter lo
+ | oc == 1105 = chr 184 -- cyrillic small letter lo
+ | otherwise = c
+ where oc = ord c
diff --git a/src/GF/Text/Lexing.hs b/src/GF/Text/Lexing.hs index 2c6b417b8..38018ff54 100644 --- a/src/GF/Text/Lexing.hs +++ b/src/GF/Text/Lexing.hs @@ -2,6 +2,7 @@ module GF.Text.Lexing (stringOp) where import GF.Text.Transliterations import GF.Text.UTF8 +import GF.Text.CP1251 import Data.Char import Data.List (intersperse) @@ -98,18 +99,3 @@ unlexMixed = concat . alternate False where isPunct = flip elem ".?!,:;" isParen = flip elem "()[]{}" isClosing = flip elem ")]}" - - --- might be in a file of its own: Windows Cyrillic, used in Bulgarian resource - -decodeCP1251 = map convert where - convert c - | c >= '\192' && c <= '\255' = chr (ord c + 848) - | otherwise = c - -encodeCP1251 = map convert where - convert c - | oc >= 1040 && oc <= 1103 = chr (oc - 848) - | otherwise = c - where oc = ord c - |
