Move repartitionMod8 function to internal ByteString module. Move table to internal module

Author: newhoggy

.vscode/tasks.json

@@ -5,7 +5,7 @@
       "label": "Build",
       "type": "shell",
       "command": "bash",
-      "args": ["-lc", "cabal new-build && echo 'Done'"],
+      "args": ["-lc", "cabal new-build --enable-tests --enable-benchmarks && echo 'Done'"],
       "group": {
         "kind": "build",
         "isDefault": true
@@ -37,7 +37,7 @@
       "label": "Test",
       "type": "shell",
       "command": "bash",
-      "args": ["-lc", "cabal new-test --enable-tests && echo 'Done'"],
+      "args": ["-lc", "cabal new-test --enable-tests --enable-benchmarks && echo 'Done'"],
       "group": {
         "kind": "test",
         "isDefault": true

bench/Main.hs

@@ -11,6 +11,7 @@ import HaskellWorks.Data.Bits.BitShown
 import HaskellWorks.Data.FromByteString
 import HaskellWorks.Data.Xml.Conduit
 import HaskellWorks.Data.Xml.Conduit.Blank
+import HaskellWorks.Data.Xml.Internal.Tables
 import HaskellWorks.Data.Xml.Succinct.Cursor
 import System.IO.MMap
 

hw-xml.cabal

@@ -2,8 +2,8 @@ cabal-version:      2.2
 
 name:               hw-xml
 version:            0.4.0.1
-synopsis:           Conduits for tokenizing streams.
-description:        Conduits for tokenizing streams. Please see README.md
+synopsis:           XML parser based on succinct data structures.
+description:        XML parser based on succinct data structures. Please see README.md
 category:           Data, XML, Succinct Data Structures, Data Structures
 homepage:           http://github.com/haskell-works/hw-xml#readme
 bug-reports:        https://github.com/haskell-works/hw-xml/issues
@@ -94,6 +94,8 @@ library
     HaskellWorks.Data.Xml.Grammar
     HaskellWorks.Data.Xml.Index
     HaskellWorks.Data.Xml.Internal.BalancedParens
+    HaskellWorks.Data.Xml.Internal.ByteString
+    HaskellWorks.Data.Xml.Internal.Tables
     HaskellWorks.Data.Xml.Internal.ToIbBp64
     HaskellWorks.Data.Xml.Lens
     HaskellWorks.Data.Xml.Succinct

project.sh

@@ -17,7 +17,7 @@ case "$cmd" in
 
   build)
     cabal new-build all -j8 \
-      --disable-tests --disable-benchmarks \
+      --enable-tests --enable-benchmarks \
       $CABAL_FLAGS "$@"
     ;;
 

src/HaskellWorks/Data/Xml/Conduit.hs

@@ -5,45 +5,23 @@ module HaskellWorks.Data.Xml.Conduit
   ( blankedXmlToInterestBits
   , byteStringToBits
   , compressWordAsBit
-  , interestingWord8s
-  , isInterestingWord8
   ) where
 
-import Data.ByteString                as BS
+import Data.ByteString                           (ByteString)
 import Data.Word
-import Data.Word8
-import HaskellWorks.Data.AtIndex      ((!!!))
 import HaskellWorks.Data.Bits.BitWise
-import Prelude                        as P
+import HaskellWorks.Data.Xml.Internal.ByteString
+import HaskellWorks.Data.Xml.Internal.Tables
+import Prelude
 
-import qualified Data.Bits            as BITS
-import qualified Data.Vector.Storable as DVS
+import qualified Data.Bits       as BITS
+import qualified Data.ByteString as BS
+import qualified Prelude         as P
 
-interestingWord8s :: DVS.Vector Word8
-interestingWord8s = DVS.constructN 256 go
-  where go :: DVS.Vector Word8 -> Word8
-        go v = if     w == _bracketleft
-                  ||  w == _braceleft
-                  ||  w == _parenleft
-                  ||  w == _bracketleft
-                  ||  w == _less
-                  ||  w == _a
-                  ||  w == _v
-                  ||  w == _t
-              then 1
-              else 0
-          where w :: Word8
-                w = fromIntegral (DVS.length v)
-{-# NOINLINE interestingWord8s #-}
-
-isInterestingWord8 :: Word8 -> Word8
-isInterestingWord8 b = fromIntegral (interestingWord8s !!! fromIntegral b)
-{-# INLINABLE isInterestingWord8 #-}
-
-blankedXmlToInterestBits :: [BS.ByteString] -> [BS.ByteString]
+blankedXmlToInterestBits :: [ByteString] -> [ByteString]
 blankedXmlToInterestBits = blankedXmlToInterestBits' ""
 
-blankedXmlToInterestBits' :: BS.ByteString -> [BS.ByteString] -> [BS.ByteString]
+blankedXmlToInterestBits' :: ByteString -> [ByteString] -> [ByteString]
 blankedXmlToInterestBits' rs is = case is of
   (bs:bss) -> do
     let cs = if BS.length rs /= 0 then BS.concat [rs, bs] else bs
@@ -63,16 +41,10 @@ blankedXmlToInterestBits' rs is = case is of
                     , BS.drop 8 as
                     )
 
-repartitionMod8 :: BS.ByteString -> BS.ByteString -> (BS.ByteString, BS.ByteString)
-repartitionMod8 aBS bBS = (BS.take cLen abBS, BS.drop cLen abBS)
-  where abBS = BS.concat [aBS, bBS]
-        abLen = BS.length abBS
-        cLen = (abLen `div` 8) * 8
-
-compressWordAsBit :: [BS.ByteString] -> [BS.ByteString]
+compressWordAsBit :: [ByteString] -> [ByteString]
 compressWordAsBit = compressWordAsBit' BS.empty
 
-compressWordAsBit' :: BS.ByteString -> [BS.ByteString] -> [BS.ByteString]
+compressWordAsBit' :: ByteString -> [ByteString] -> [ByteString]
 compressWordAsBit' aBS iBS = case iBS of
   (bBS:bBSs) -> do
     let (cBS, dBS) = repartitionMod8 aBS bBS
@@ -103,7 +75,7 @@ yieldBitsOfWord8 w =
 yieldBitsofWord8s :: [Word8] -> [Bool]
 yieldBitsofWord8s = P.foldr ((++) . yieldBitsOfWord8) []
 
-byteStringToBits :: [BS.ByteString] -> [Bool]
+byteStringToBits :: [ByteString] -> [Bool]
 byteStringToBits is = case is of
   (bs:bss) -> yieldBitsofWord8s (BS.unpack bs) ++ byteStringToBits bss
   []       -> []

src/HaskellWorks/Data/Xml/Internal/ByteString.hs

@@ -0,0 +1,14 @@
+module HaskellWorks.Data.Xml.Internal.ByteString
+  ( repartitionMod8
+  ) where
+
+import Data.ByteString (ByteString)
+
+import qualified Data.ByteString as BS
+
+repartitionMod8 :: ByteString -> ByteString -> (ByteString, ByteString)
+repartitionMod8 aBS bBS = (BS.take cLen abBS, BS.drop cLen abBS)
+  where abBS = BS.concat [aBS, bBS]
+        abLen = BS.length abBS
+        cLen = (abLen `div` 8) * 8
+{-# INLINE repartitionMod8 #-}

src/HaskellWorks/Data/Xml/Internal/Tables.hs

@@ -0,0 +1,32 @@
+module HaskellWorks.Data.Xml.Internal.Tables
+  ( interestingWord8s
+  , isInterestingWord8
+  ) where
+
+import Data.Word
+import Data.Word8
+import HaskellWorks.Data.AtIndex ((!!!))
+import Prelude                   as P
+
+import qualified Data.Vector.Storable as DVS
+
+interestingWord8s :: DVS.Vector Word8
+interestingWord8s = DVS.constructN 256 go
+  where go :: DVS.Vector Word8 -> Word8
+        go v = if     w == _bracketleft
+                  ||  w == _braceleft
+                  ||  w == _parenleft
+                  ||  w == _bracketleft
+                  ||  w == _less
+                  ||  w == _a
+                  ||  w == _v
+                  ||  w == _t
+              then 1
+              else 0
+          where w :: Word8
+                w = fromIntegral (DVS.length v)
+{-# NOINLINE interestingWord8s #-}
+
+isInterestingWord8 :: Word8 -> Word8
+isInterestingWord8 b = fromIntegral (interestingWord8s !!! fromIntegral b)
+{-# INLINABLE isInterestingWord8 #-}