Differences

This shows you the differences between two versions of the page.

--- codesnippets:octets [2021/04/17 07:30] – f2b216
+++ codesnippets:octets [2025/10/08 00:48] (current) – external edit 127.0.0.1
@@ Line 1: / Line 1: @@
 ====== Octets ======
-~~DISCUSSION~~
   * type ''class'' ''Octetable''
-    * with type ''Octets''
+    * with type ''Octet'', and ''Octets''
-    * with instances for ''Word32'', ''Char'', ''Integer'', and ''Int''
+    * with instances for ''Word32'', ''Char'', ''Utf8Char'', ''Integer'', and ''Int''
-  * as an example
+  * example
-  * with test execution
     * code, of module ''Octetable'':<code Haskell>
-{-# LANGUAGE FlexibleInstances #-}
+{-# LANGUAGE BinaryLiterals #-}
 module Octetable
     (
         Octetable(..),
+        Octet,
         Octets,
         getOctets,
@@ Line 23: / Line 22: @@
 import qualified System.IO as SysIo
 import qualified Data.ByteString as BS
+import           Data.Bits ((.|.), (.&.))
-type Octets = [W.Word8]
+type Octet = W.Word8
+type Octets = [Octet]
 getOctets :: SysIo.Handle -> IO [W.Word8]
@@ Line 35: / Line 37: @@
     toOctets :: a -> Octets
     fromOctets :: Octets -> a
 instance Octetable W.Word32 where
@@ Line 51: / Line 52: @@
     fromOctets lw = Chr.chr (integralFromOctets lw)
-instance Octetable [Char] where
+{-| UTF-8 character
-    toOctets lch = foldr (\ch lw8 -> (toOctets ch) ++ lw8) [] lch
-    fromOctets [] = []
+* prefix: cu8
-    fromOctets lw = fmap (Chr.chr . fromIntegral) lw
+* represent on UTF-8 character
+* keeps one, two, three or four character
+-}
+data Utf8Char =
+      OneByte (W.Word8)
+    | TwoBytes (W.Word8, W.Word8)
+    | ThreeBytes (W.Word8, W.Word8, W.Word8)
+    | FourBytes (W.Word8, W.Word8, W.Word8, W.Word8)
+    deriving (Show, Eq)
+instance Octetable Utf8Char where
+    toOctets (OneByte w81) = [w81]
+    toOctets (TwoBytes (w81,w82)) = [w81,w82]
+    toOctets (ThreeBytes (w81,w82,w83)) = [w81,w82,w83]
+    toOctets (FourBytes (w81,w82,w83,w84)) = [w81,w82,w83,w84]
+    fromOctets (w81:w82:w83:w84:_)
+        | (w81 .&. 0b10000000) == 0b00000000 = OneByte w81
+        | (w81 .&. 0b11100000) == 0b11000000 = TwoBytes (w81,w82)
+        | (w81 .&. 0b11110000) == 0b11100000 = ThreeBytes (w81,w82,w83)
+        | otherwise                          = FourBytes (w81,w82,w83,w84)
+    fromOctets (w81:w82:w83:[])
+        | (w81 .&. 0b10000000) == 0b00000000 = OneByte w81
+        | (w81 .&. 0b11100000) == 0b11000000 = TwoBytes (w81,w82)
+        | otherwise                          = ThreeBytes (w81,w82,w83)
+    fromOctets (w81:w82:[])
+        | (w81 .&. 0b10000000) == 0b00000000 = OneByte w81
+        | otherwise                          = TwoBytes (w81,w82)
+    fromOctets (w81:[]) = OneByte w81
+    fromOctets [] = (OneByte 0)
 instance Octetable Integer where
@@ Line 80: / Line 109: @@
     fIntegralToOctets' (n `div` 256, (fromIntegral (n `mod` 256)) : lw)
-keepOctetsPositive :: Octets-> Octets
+keepOctetsPositive :: Octets -> Octets
 keepOctetsPositive [] = []
 keepOctetsPositive lw8
@@ Line 86: / Line 115: @@
     | otherwise                =     lw8 -- highest bit is NOT set, hence positive -> keep as is
-keepOctetsNegative :: Octets-> Octets
+keepOctetsNegative :: Octets -> Octets
 keepOctetsNegative [] = []
 keepOctetsNegative lw8
@@ Line 96: / Line 125: @@
     | Bts.testBit (head lw8) 7 = (negate . (+1) . integralFromOctets . (fmap Bts.complement)) lw8
     | otherwise                = foldl (\n w8 -> (n * 256) + (fromIntegral w8))  0 lw8
-</code>
-    * code, of module ''Main'', to get and put octets from streams:<code Haskell>
-import qualified Octetable as Oct
-import qualified Data.Word as W
-import qualified System.IO as SysIo
-main :: IO ()
-main =
-    do
-        o1 <- Oct.getOctets SysIo.stdin
-        Oct.putOctets SysIo.stdout o1
-        Oct.putOctets SysIo.stdout o2
-o2 :: [W.Word8]
-o2 = Oct.toOctets "Hallo!"
 </code>
     * code, of module ''Main'', to test fromOctets and toOctets:<code Haskell>
@@ Line 317: / Line 331: @@
    -2299819935653645082894452792461391785316838819634854499582649773064497657338657176722815066047012042970707985364005424194851744592589349276289737892620644753316103773332601282923749730891215409124729751179101572661546264659399035941959433049031612193674931427369793499974748103941854396793949658196683481215335075331190973321382630193047486040570710476538476805752828595396636100262237982462365668148474400831819729686520016831142034423035989126884829798702369001522511603033547628403647714677841344568950917375570064143430404597420770773439867113060747581916408770350655859231601822693794335340024703291277047383009545489812044665828872069515086161470043404212964143599392663515383545253369964683097804303493196001346972875807092655666775774755153966987089557598493216241473998446786105682286003208993649406238722161069532282900618217219492148604518557515368241354830826940556257234738134897451091801018568418762856248825737587480173029024436395984067723608292575876261695468817805283649009112581009389828573386359739601662047431757990274166003464056141974382728940618484203465938650973604530113187308736921734424683365243118100998126580691247572322148932020120247574456331416847236654458804902164954387120540919072050362384906504090722699288613319388277341509353032905698955788548406169450791248838342383275106318381042808878054409211203094805561198699897359477344778815969939545770630197346441568662820604992751625101251259774130554173720020589092927339037514717894135290456906563866080001
 </code>
+  * example, used with module FileSystem
+    * code, of module ''Main'':<code Haskell>
+import qualified FileSystem as FS
+import qualified System.IO as SysIo
+import qualified Octetable as Oct
+import qualified Data.Char as Chr
+main :: IO ()
+main =
+        FS.processData
+            (FS.DataReader (FS.FileInp ".\\test\\a.txt") (hGetNReplace  octBlank octUnderscore))
+            [
+                FS.DataWriter (FS.FileOut ".\\test\\b.txt") Oct.putOctets,
+                FS.DataWriter FS.StdOut (hReplaceNPut octUnderscore octBlank),
+                FS.DataWriter FS.StdErr Oct.putOctets
+            ]
+octBlank :: Oct.Octet
+octBlank = ((fromIntegral . Chr.ord) ' ')
+octUnderscore :: Oct.Octet
+octUnderscore = ((fromIntegral . Chr.ord) '_')
+hGetNReplace :: Oct.Octet -> Oct.Octet -> SysIo.Handle -> IO Oct.Octets
+hGetNReplace a b h =
+    do
+        s <- Oct.getOctets h
+        return (fmap (\x -> if x == a then b else x) s)
+hReplaceNPut :: Oct.Octet -> Oct.Octet -> SysIo.Handle -> Oct.Octets -> IO ()
+hReplaceNPut a b h s = Oct.putOctets h (fmap (\x -> if x == a then b else x) s)
+</code>
+    * compiles, error and warning free, with compiler: GHC 8.10.4, using compiler option -Wall
+    * input, file "a.txt":<code>
+The quick brown fox jumps over the lazy dog.
+</code>
+    * executes, with output, first line stdout, second line stderr:<code>
+The quick brown fox jumps over the lazy dog.
+The_quick_brown_fox_jumps_over_the_lazy_dog.
+</code>
+    * executes, with output file "b.txt":<code>
+The_quick_brown_fox_jumps_over_the_lazy_dog.
+</code>
+===== ✎ =====
+~~DISCUSSION~~