Cleanups for new version release.

Author: ondrap

Data/JsonStream/Parser.hs

@@ -1,7 +1,7 @@
 {-# LANGUAGE BangPatterns      #-}
 {-# LANGUAGE OverloadedStrings #-}
-{-# LANGUAGE TupleSections     #-}
 {-# LANGUAGE PatternGuards     #-}
+{-# LANGUAGE TupleSections     #-}
 
 -- |
 -- Module : Data.JsonStream.Parser
@@ -43,7 +43,6 @@ module Data.JsonStream.Parser (
     -- * FromJSON parser
   , value
   , string
-  , bytestring
     -- * Constant space parsers
   , safeString
   , number
@@ -82,8 +81,7 @@ import qualified Data.HashMap.Strict         as HMap
 import           Data.Scientific             (Scientific, isInteger,
                                               toBoundedInteger, toRealFloat)
 import qualified Data.Text                   as T
-import           Data.Text.Encoding          (encodeUtf8)
-import           Data.Text.Encoding     (decodeUtf8')
+import           Data.Text.Encoding          (decodeUtf8')
 import qualified Data.Vector                 as Vec
 
 import           Data.JsonStream.CLexer
@@ -377,25 +375,13 @@ longString mbounds = Parser $ moreData (handle [] 0)
           | otherwise -> Failed "Error decoding UTF8"
         _ ->  callParse ignoreVal tok
 
--- | Match string as a ByteString without decoding the data from UTF8 and unescaping the contents
--- (strings larger than input chunk, small get always decoded).
-bytestring :: Parser BL.ByteString
-bytestring = Parser $ moreData (handle [])
-  where
-    handle acc tok el ntok =
-      case el of
-        JValue (AE.String str) -> Yield (BL.fromChunks [encodeUtf8 str]) (Done "" ntok)
-        StringContent str -> moreData (handle (str:acc)) ntok
-        StringEnd -> Yield (BL.fromChunks $ reverse acc) (Done "" ntok)
-        _ -> callParse ignoreVal tok
-
-
 -- | Parse string value, skip parsing otherwise.
 string :: Parser T.Text
 string = longString Nothing
 
 -- | Stops parsing string after the limit is reached. The string will not be matched
--- if it exceeds the size.
+-- if it exceeds the size. The size is the size of escaped string including escape
+-- characters.
 safeString :: Int -> Parser T.Text
 safeString limit = longString (Just limit)
 

benchmarks/aeson-benchmarks.cabal

@@ -4,12 +4,6 @@ build-type:          Simple
 
 cabal-version:       >=1.8
 
-flag old-locale
-  description: If false then depend on time >= 1.5.
-               .
-               If true then depend on time < 1.5 together with old-locale.
-  default: False
-
 library
   hs-source-dirs: ..
   c-sources:           ../c_lib/lexer.c, ../c_lib/unescape_string.c
@@ -27,7 +21,7 @@ library
   build-depends:         base >=4.7 && <4.9
                        , bytestring
                        , text
-                       , aeson (>=0.9)
+                       , aeson (>=0.7)
                        , vector
                        , unordered-containers
                        , scientific
@@ -39,7 +33,7 @@ executable aeson-benchmark-jstream-parse
   ghc-options: -Wall -O2 -rtsopts
   build-depends:
     aeson-benchmarks,
-    aeson (>=0.9),
+    aeson (>=0.7),
     base,
     bytestring,
     time

changelog.md

@@ -1,3 +1,7 @@
+# 0.3.2.0
+- Changed string parsing; parsing of escaped strings is now very vast
+- Removed bytestring parser
+
 # 0.3.0.4
 - Fixed bug in safestring
 - Fixed test so it doesn't depend on versions of other packages

json-stream.cabal

@@ -1,5 +1,5 @@
 name:                json-stream
-version:             0.3.1.0
+version:             0.3.2.0
 synopsis:            Incremental applicative JSON parser
 description:         Easy to use JSON parser fully supporting incremental parsing.
                      Parsing grammar in applicative form.
@@ -36,12 +36,13 @@ library
   build-depends:         base >=4.7 && <4.9
                        , bytestring
                        , text
-                       , aeson
+                       , aeson >= 0.7
                        , vector
                        , unordered-containers
                        , scientific
   default-language:    Haskell2010
 
+
 test-suite spec
   main-is:             Spec.hs
   other-modules:       Data.JsonStream.CLexType, ParserSpec
@@ -51,6 +52,7 @@ test-suite spec
   type:                exitcode-stdio-1.0
   hs-source-dirs:      test, .
   default-language:    Haskell2010
+  ghc-options:         -Wall
   build-depends:         base >=4.7 && <4.9
                        , bytestring
                        , text

test/ParserSpec.hs

@@ -8,16 +8,13 @@ import Data.Aeson (Value(..))
 import qualified Data.ByteString.Char8 as BS
 import qualified Data.ByteString.Lazy as BL
 import qualified Data.Text as T
-import Data.Text.Encoding (encodeUtf8)
 import Control.Monad (forM_)
 import Data.Text.Encoding (encodeUtf8)
 import qualified Data.Vector as Vec
 import qualified Data.HashMap.Strict as HMap
 import System.Directory (getDirectoryContents)
 
 import Data.JsonStream.Parser
-import Data.JsonStream.TokenParser
-import Data.JsonStream.CLexer
 
 -- During the tests the single quotes are replaced with double quotes in the strings as
 -- otherwise it would be unreadable in haskell
@@ -28,7 +25,7 @@ todquotes = BS.map squotes
     squotes '\'' = '"'
     squotes x = x
 
-
+parse :: Parser a -> BS.ByteString -> [a]
 parse parser text = parseByteString parser (todquotes text)
 
 testRemaining :: Parser a -> BS.ByteString -> BS.ByteString
@@ -331,10 +328,11 @@ aeCompare = describe "Compare parsing of strings aeason vs json-stream" $ do
     let Just resAeson = AE.decode (BL.fromChunks [test])
     resStream `shouldBe` resAeson
 
+readBenchFiles :: FilePath -> IO [BS.ByteString]
 readBenchFiles dirname =
-    getDirectoryContents dirname >>= return . (filter isJson) >>= mapM readFile
+    getDirectoryContents dirname >>= return . (filter isJson) >>= mapM readFile'
     where
-      readFile fname = BS.readFile (dirname ++ "/" ++ fname)
+      readFile' fname = BS.readFile (dirname ++ "/" ++ fname)
       isJson fname = take 5 (reverse fname) == "nosj."
 
 aeCompareBench :: Spec