diff --git a/Data/ByteString.hs b/Data/ByteString.hs --- a/Data/ByteString.hs +++ b/Data/ByteString.hs @@ -1937,9 +1937,6 @@ -- files > half of available memory, this may lead to memory exhaustion. -- Consider using 'readFile' in this case. -- --- As with 'hGet', the string representation in the file is assumed to --- be ISO-8859-1. --- -- The Handle is closed once the contents have been read, -- or if an exception is thrown. -- diff --git a/Data/ByteString/Builder.hs b/Data/ByteString/Builder.hs --- a/Data/ByteString/Builder.hs +++ b/Data/ByteString/Builder.hs @@ -226,15 +226,13 @@ , doubleLE -- ** Character encodings + -- | Conversion from 'Char' and 'String' into 'Builder's in various encodings. -- *** ASCII (Char7) -- | The ASCII encoding is a 7-bit encoding. The /Char7/ encoding implemented here -- works by truncating the Unicode codepoint to 7-bits, prefixing it -- with a leading 0, and encoding the resulting 8-bits as a single byte. - -- For the codepoints 0-127 this corresponds the ASCII encoding. In - -- "Data.ByteString.Builder.ASCII", we also provide efficient - -- implementations of ASCII-based encodings of numbers (e.g., decimal and - -- hexadecimal encodings). + -- For the codepoints 0-127 this corresponds the ASCII encoding. , char7 , string7 @@ -242,19 +240,14 @@ -- | The ISO/IEC 8859-1 encoding is an 8-bit encoding often known as Latin-1. -- The /Char8/ encoding implemented here works by truncating the Unicode codepoint -- to 8-bits and encoding them as a single byte. For the codepoints 0-255 this corresponds - -- to the ISO/IEC 8859-1 encoding. Note that you can also use - -- the functions from "Data.ByteString.Builder.ASCII", as the ASCII encoding - -- and ISO/IEC 8859-1 are equivalent on the codepoints 0-127. + -- to the ISO/IEC 8859-1 encoding. , char8 , string8 -- *** UTF-8 -- | The UTF-8 encoding can encode /all/ Unicode codepoints. We recommend -- using it always for encoding 'Char's and 'String's unless an application - -- really requires another encoding. Note that you can also use the - -- functions from "Data.ByteString.Builder.ASCII" for UTF-8 encoding, - -- as the ASCII encoding is equivalent to the UTF-8 encoding on the Unicode - -- codepoints 0-127. + -- really requires another encoding. , charUtf8 , stringUtf8 diff --git a/Data/ByteString/Builder/ASCII.hs b/Data/ByteString/Builder/ASCII.hs --- a/Data/ByteString/Builder/ASCII.hs +++ b/Data/ByteString/Builder/ASCII.hs @@ -14,7 +14,13 @@ -- module Data.ByteString.Builder.ASCII ( - -- ** ASCII text + -- ** Formatting numbers as text + -- | Formatting of numbers as ASCII text. + -- + -- Note that you can also use these functions for the ISO/IEC 8859-1 and + -- UTF-8 encodings, as the ASCII encoding is equivalent on the + -- codepoints 0-127. + -- *** Decimal numbers -- | Decimal encoding of numbers using ASCII encoded characters. int8Dec diff --git a/Data/ByteString/Builder/Internal.hs b/Data/ByteString/Builder/Internal.hs --- a/Data/ByteString/Builder/Internal.hs +++ b/Data/ByteString/Builder/Internal.hs @@ -1,5 +1,5 @@ {-# LANGUAGE ScopedTypeVariables, CPP, BangPatterns, RankNTypes #-} -#if __GLASGOW_HASKELL__ >= 701 +#if __GLASGOW_HASKELL__ >= 703 {-# LANGUAGE Unsafe #-} #endif {-# OPTIONS_HADDOCK hide #-} diff --git a/Data/ByteString/Builder/Prim/Internal.hs b/Data/ByteString/Builder/Prim/Internal.hs --- a/Data/ByteString/Builder/Prim/Internal.hs +++ b/Data/ByteString/Builder/Prim/Internal.hs @@ -1,5 +1,5 @@ {-# LANGUAGE ScopedTypeVariables, CPP, BangPatterns #-} -#if __GLASGOW_HASKELL__ >= 701 +#if __GLASGOW_HASKELL__ >= 703 {-# LANGUAGE Unsafe #-} #endif {-# OPTIONS_HADDOCK hide #-} @@ -69,7 +69,7 @@ import Foreign import Prelude hiding (maxBound) -#if defined(__GLASGOW_HASKELL__) && __GLASGOW_HASKELL__ < 611 +#if !(__GLASGOW_HASKELL__ >= 612) -- ghc-6.10 and older do not support {-# INLINE CONLIKE #-} #define CONLIKE #endif diff --git a/Data/ByteString/Internal.hs b/Data/ByteString/Internal.hs --- a/Data/ByteString/Internal.hs +++ b/Data/ByteString/Internal.hs @@ -2,7 +2,7 @@ #if __GLASGOW_HASKELL__ {-# LANGUAGE UnliftedFFITypes, MagicHash, UnboxedTuples, DeriveDataTypeable #-} -#if __GLASGOW_HASKELL__ >= 701 +#if __GLASGOW_HASKELL__ >= 703 {-# LANGUAGE Unsafe #-} #endif #endif @@ -530,12 +530,21 @@ concat [bs] = bs concat bss0 = unsafeCreate totalLen $ \ptr -> go bss0 ptr where - totalLen = List.sum [ len | (PS _ _ len) <- bss0 ] + totalLen = checkedSum "concat" [ len | (PS _ _ len) <- bss0 ] go [] !_ = return () go (PS fp off len:bss) !ptr = do withForeignPtr fp $ \p -> memcpy ptr (p `plusPtr` off) len go bss (ptr `plusPtr` len) +-- | Add a list of non-negative numbers. Errors out on overflow. +checkedSum :: String -> [Int] -> Int +checkedSum fun = go 0 + where go !a (x:xs) + | ax >= 0 = go ax xs + | otherwise = overflowError fun + where ax = a + x + go a _ = a + ------------------------------------------------------------------------ -- | Conversion between 'Word8' and 'Char'. Should compile to a no-op. @@ -579,13 +588,35 @@ c == '\xa0' {-# INLINE isSpaceChar8 #-} +overflowError :: String -> a +overflowError fun = error $ "Data.ByteString." ++ fun ++ ": size overflow" + ------------------------------------------------------------------------ --- | Just like unsafePerformIO, but we inline it. Big performance gains as --- it exposes lots of things to further inlining. /Very unsafe/. In --- particular, you should do no memory allocation inside an --- 'inlinePerformIO' block. On Hugs this is just @unsafePerformIO@. +-- | This \"function\" has a superficial similarity to 'unsafePerformIO' but +-- it is in fact a malevolent agent of chaos. It unpicks the seams of reality +-- (and the 'IO' monad) so that the normal rules no longer apply. It lulls you +-- into thinking it is reasonable, but when you are not looking it stabs you +-- in the back and aliases all of your mutable buffers. The carcass of many a +-- seasoned Haskell programmer lie strewn at its feet. -- +-- Witness the trail of destruction: +-- +-- * +-- +-- * +-- +-- * +-- +-- * +-- +-- * +-- +-- Do not talk about \"safe\"! You do not know what is safe! +-- +-- Yield not to its blasphemous call! Flee traveller! Flee or you will be +-- corrupted and devoured! +-- {-# INLINE inlinePerformIO #-} inlinePerformIO :: IO a -> a #if defined(__GLASGOW_HASKELL__) @@ -657,4 +688,3 @@ foreign import ccall unsafe "static fpstring.h fps_count" c_count :: Ptr Word8 -> CULong -> Word8 -> IO CULong - diff --git a/Data/ByteString/Lazy.hs b/Data/ByteString/Lazy.hs --- a/Data/ByteString/Lazy.hs +++ b/Data/ByteString/Lazy.hs @@ -14,7 +14,7 @@ -- Maintainer : dons00@gmail.com, duncan@community.haskell.org -- Stability : stable -- Portability : portable --- +-- -- A time and space-efficient implementation of lazy byte vectors -- using lists of packed 'Word8' arrays, suitable for high performance -- use, both in terms of large data quantities, or high speed @@ -263,7 +263,7 @@ singleton w = Chunk (S.singleton w) Empty {-# INLINE singleton #-} --- | /O(n)/ Convert a '[Word8]' into a 'ByteString'. +-- | /O(n)/ Convert a '[Word8]' into a 'ByteString'. pack :: [Word8] -> ByteString pack = packBytes @@ -295,7 +295,7 @@ toStrict (Chunk c Empty) = c toStrict cs0 = S.unsafeCreate totalLen $ \ptr -> go cs0 ptr where - totalLen = foldlChunks (\a c -> a + S.length c) 0 cs0 + totalLen = checkedSum "Lazy.toStrict" . L.map S.length . toChunks $ cs0 go Empty !_ = return () go (Chunk (S.PS fp off len) cs) !destptr = @@ -303,6 +303,18 @@ S.memcpy destptr (p `plusPtr` off) len go cs (destptr `plusPtr` len) +-- just here in 0.10.4.x, in later version it's exported from .Internal +checkedSum :: String -> [Int] -> Int +checkedSum fun = go 0 + where go !a (x:xs) + | ax >= 0 = go ax xs + | otherwise = overflowError fun + where ax = a + x + go a _ = a + +overflowError :: String -> a +overflowError fun = error $ "Data.ByteString." ++ fun ++ ": size overflow" + ------------------------------------------------------------------------ {- @@ -479,7 +491,7 @@ foldl' f z = go z where go a _ | a `seq` False = undefined go a Empty = a - go a (Chunk c cs) = go (S.foldl f a c) cs + go a (Chunk c cs) = go (S.foldl' f a c) cs {-# INLINE foldl' #-} -- | 'foldr', applied to a binary operator, a starting value @@ -689,7 +701,7 @@ splitAt' _ Empty = (Empty, Empty) splitAt' n (Chunk c cs) = if n < fromIntegral (S.length c) - then (Chunk (S.take (fromIntegral n) c) Empty + then (Chunk (S.take (fromIntegral n) c) Empty ,Chunk (S.drop (fromIntegral n) c) cs) else let (cs', cs'') = splitAt' (n - fromIntegral (S.length c)) cs in (Chunk c cs', cs'') @@ -738,7 +750,7 @@ -- | 'breakByte' breaks its ByteString argument at the first occurence -- of the specified byte. It is more efficient than 'break' as it is -- implemented with @memchr(3)@. I.e. --- +-- -- > break (=='c') "abcd" == breakByte 'c' "abcd" -- breakByte :: Word8 -> ByteString -> (ByteString, ByteString) @@ -798,12 +810,12 @@ -- > split '\n' "a\nb\nd\ne" == ["a","b","d","e"] -- > split 'a' "aXaXaXa" == ["","X","X","X",""] -- > split 'x' "x" == ["",""] --- +-- -- and -- -- > intercalate [c] . split c == id -- > split == splitWith . (==) --- +-- -- As for all splitting functions in this library, this function does -- not copy the substrings, it just constructs new 'ByteStrings' that -- are slices of the original. @@ -821,7 +833,7 @@ {- -- | Like 'splitWith', except that sequences of adjacent separators are -- treated as a single separator. eg. --- +-- -- > tokens (=='a') "aabbaca" == ["bb","c"] -- tokens :: (Word8 -> Bool) -> ByteString -> [ByteString] @@ -888,13 +900,13 @@ index cs0 i = index' cs0 i where index' Empty n = moduleError "index" ("index too large: " ++ show n) index' (Chunk c cs) n - | n >= fromIntegral (S.length c) = + | n >= fromIntegral (S.length c) = index' cs (n - fromIntegral (S.length c)) | otherwise = S.unsafeIndex c (fromIntegral n) -- | /O(n)/ The 'elemIndex' function returns the index of the first -- element in the given 'ByteString' which is equal to the query --- element, or 'Nothing' if there is no such element. +-- element, or 'Nothing' if there is no such element. -- This implementation uses memchr(3). elemIndex :: Word8 -> ByteString -> Maybe Int64 elemIndex w cs0 = elemIndex' 0 cs0 @@ -910,7 +922,7 @@ -- element, or 'Nothing' if there is no such element. The following -- holds: -- --- > elemIndexEnd c xs == +-- > elemIndexEnd c xs == -- > (-) (length xs - 1) `fmap` elemIndex c (reverse xs) -- elemIndexEnd :: Word8 -> ByteString -> Maybe Int @@ -1057,7 +1069,7 @@ -- | /O(n)/ The 'isSuffixOf' function takes two ByteStrings and returns 'True' -- iff the first is a suffix of the second. --- +-- -- The following holds: -- -- > isSuffixOf x y == reverse x `isPrefixOf` reverse y @@ -1142,7 +1154,7 @@ -- Lazy ByteString IO -- -- Rule for when to close: is it expected to read the whole file? --- If so, close when done. +-- If so, close when done. -- -- | Read entire handle contents /lazily/ into a 'ByteString'. Chunks diff --git a/Data/ByteString/Lazy/Internal.hs b/Data/ByteString/Lazy/Internal.hs --- a/Data/ByteString/Lazy/Internal.hs +++ b/Data/ByteString/Lazy/Internal.hs @@ -1,7 +1,7 @@ {-# LANGUAGE CPP, ForeignFunctionInterface, BangPatterns #-} #if __GLASGOW_HASKELL__ {-# LANGUAGE DeriveDataTypeable #-} -#if __GLASGOW_HASKELL__ >= 701 +#if __GLASGOW_HASKELL__ >= 703 {-# LANGUAGE Unsafe #-} #endif #endif diff --git a/Data/ByteString/Short/Internal.hs b/Data/ByteString/Short/Internal.hs --- a/Data/ByteString/Short/Internal.hs +++ b/Data/ByteString/Short/Internal.hs @@ -2,7 +2,7 @@ ForeignFunctionInterface, MagicHash, UnboxedTuples, UnliftedFFITypes #-} {-# OPTIONS_GHC -fno-warn-name-shadowing #-} -#if __GLASGOW_HASKELL__ >= 701 +#if __GLASGOW_HASKELL__ >= 703 {-# LANGUAGE Unsafe #-} #endif {-# OPTIONS_HADDOCK hide #-} diff --git a/Data/ByteString/Unsafe.hs b/Data/ByteString/Unsafe.hs --- a/Data/ByteString/Unsafe.hs +++ b/Data/ByteString/Unsafe.hs @@ -224,7 +224,7 @@ -- collected by Haskell. This operation has /O(1)/ complexity as we -- already know the final size, so no /strlen(3)/ is required. -- --- This funtion is /unsafe/. If the original @CStringLen@ is later +-- This function is /unsafe/. If the original @CStringLen@ is later -- modified, this change will be reflected in the resulting @ByteString@, -- breaking referential transparency. -- @@ -236,7 +236,7 @@ -- | /O(n)/ Build a @ByteString@ from a malloced @CString@. This value will -- have a @free(3)@ finalizer associated to it. -- --- This funtion is /unsafe/. If the original @CString@ is later +-- This function is /unsafe/. If the original @CString@ is later -- modified, this change will be reflected in the resulting @ByteString@, -- breaking referential transparency. -- @@ -250,10 +250,10 @@ len <- c_strlen cstr return $! PS fp 0 (fromIntegral len) --- | /O(n)/ Build a @ByteString@ from a malloced @CStringLen@. This +-- | /O(1)/ Build a @ByteString@ from a malloced @CStringLen@. This -- value will have a @free(3)@ finalizer associated to it. -- --- This funtion is /unsafe/. If the original @CString@ is later +-- This function is /unsafe/. If the original @CString@ is later -- modified, this change will be reflected in the resulting @ByteString@, -- breaking referential transparency. -- diff --git a/README b/README deleted file mode 100644 --- a/README +++ /dev/null @@ -1,205 +0,0 @@ ------------------------------------------------------------------------- - ByteString : Fast, packed strings of bytes ------------------------------------------------------------------------- - -This library provides the Data.ByteString library -- strict and lazy -byte arrays manipulable as strings -- providing very time and space -efficient string and IO operations. - -For very large data requirements, or constraints on heap size, -Data.ByteString.Lazy is provided, a lazy list of bytestring chunks. -Efficient processing of multi-gigabyte data can be achieved this way. - -Requirements: - > Cabal - > GHC 6.4 or greater, or hugs - -Building: - > runhaskell Setup.lhs configure --prefix=/f/g - > runhaskell Setup.lhs build - > runhaskell Setup.lhs install - -After installation, you can run the testsuite as follows: - - > cd tests ; make - or - > cd tests ; make hugs - -For the full test and benchmark suite, you need GHC and Hugs: - - > cd tests ; make everything - -Authors: - ByteString was derived from the GHC PackedString library, - originally written by Bryan O'Sullivan, and then by Simon Marlow. - It was adapted, and greatly extended for darcs by David Roundy, and - others. Don Stewart cleaned up and further extended the implementation. - Duncan Coutts wrote much of the .Lazy code. Don, Duncan and Roman - Leshchinskiy wrote the fusion system. - ------------------------------------------------------------------------- - -Performance, some random numbers (with GHC): - -This table compares the performance of common operations ByteString, -from various string libraries. - -Size of test data: 21256k, Linux 3.2Ghz P4 - - FPS7 SPS PS [a] -++ 0.028 ! ! 1.288 -length 0.000 0.000 0.000 0.131 -pack 0.303 0.502 0.337 - -unpack 3.319* 1.630 7.445 - -compare 0.000 0.000 0.000 0.000 -index 0.000 0.000 0.000 0.000 -map 2.762* 2.917 4.813 7.286 -filter 0.304 2.805 0.954 0.305 -take 0.000 0.000 0.024 0.005 -drop 0.000 0.000 11.768 0.130 -takeWhile 0.000 1.498 0.000 0.000 -dropWhile 0.000 1.985 8.447 0.130 -span 0.000 9.289 11.144 0.131 -break 0.000 9.383 11.268 0.133 -lines 0.052 1.114 1.367 2.790 -unlines 0.048 ! ! 10.950 -words 1.344 2.128 5.644 4.184 -unwords 0.016 ! ! 1.305 -reverse 0.024 12.997 13.018 1.622 -concat 0.000 12.701 11.459 1.163 -cons 0.016 2.064 8.358 0.131 -empty 0.000 0.000 0.000 0.000 -head 0.000 0.000 0.000 0.000 -tail 0.000 0.000 14.490 0.130 -elem 0.000 1.490 0.001 0.000 -last 0.000 - - 0.143 -init 0.000 - - 1.147 -inits 0.414 - - ! -tails 0.460 - - 1.136 -intersperse 0.040 - - 10.517 -any 0.000 - - 0.000 -all 0.000 - - 0.000 -sort 0.168 - - ! -maximum 0.024 - - 0.183 -minimum 0.025 - - 0.185 -replicate 0.000 - - 0.053 -findIndex 0.096 -find 0.120 - - 0.000 -elemIndex 0.000 - - 0.000 -elemIndicies 0.008 - - 0.314 -foldl 0.148 -spanEnd 0.000 -snoc 0.016 -filterChar 0.031 -filterNotChar 0.124 -join 0.016 -split 0.032 -findIndices 0.408 -splitAt 0.000 -lineIndices 0.029 -breakOn 0.000 -breakSpace 0.000 -splitWith 0.329 -dropSpace 0.000 -dropSpaceEnd 0.000 -joinWithChar 0.017 -join / 0.016 -zip 0.960 -zipWith 0.892 -isSubstringOf 0.039 -isPrefixOf 0.000 -isSuffixOf 0.000 -count 0.021 - -Key: FPS6 = FPS 0.6 - SPS = Simon Marlow's packedstring prototype - PS = Data.PackedString - [a] = [Char] - - - = no function exists - ! = stack or memory exhaustion - ------------------------------------------------------------------------- - -== Stress testing really big strings - -Doing some stress testing of FPS, here are some results for 0.5G strings. - -3.2Ghz box, 2G physical mem. - -Size of test data: 524288k -Size of test data: 524288k - Char8 Word8 - -Effectively O(1) or O(m) where m < n - all 0.000 0.000 - any 0.000 0.004 - break 0.000 0.000 - breakChar 0.000 0.000 - breakSpace 0.000 - compare 0.000 - concat 0.000 - drop 0.000 - dropSpace 0.000 - dropSpaceEnd 0.000 - dropWhile 0.000 0.000 - elem 0.000 0.000 - elemIndex 0.000 0.000 - elemIndexLast 0.000 0.000 - empty 0.000 - head 0.000 0.000 - index 0.000 0.000 - init 0.000 - last 0.000 0.000 - length 0.000 - notElem 0.000 0.000 - span 0.000 0.000 - spanChar 0.000 0.000 - spanEnd 0.000 0.000 - splitAt 0.000 - tail 0.000 - take 0.000 - takeWhile 0.000 0.000 - isPrefixOf 0.000 - isSuffixOf 0.000 - addr1 0.000 - addr2 0.000 - -O(n) - ++ 0.676 - map 6.080 5.868 - cons 0.396 0.396 - snoc 0.400 0.400 - find 3.240 - split 1.204 1.200 - lines 2.000 - foldl 3.804 - unwords 0.552 - reverse 0.884 - findIndex 3.128 - filterChar 0.756 0.732 - filter/='f' 8.265 7.012 - filterNotChar 4.456 3.388 - join 0.400 - sort 4.344 - maximum 0.776 0.764 - minimum 0.772 0.776 - replicate 0.008 0.000 - elemIndices 0.240 0.240 - lineIndices 1.092 - joinWithChar 0.400 0.400 - isSubstringOf 0.052 - count 0.748 - -slow O(n) - words 38.722 - group 77.261 - groupBy 96.226 - inits 32.430 - tails 23.225 - findIndices 13.841 15.825 - splitWith 18.445 19.225 - zip 33.926 - zipWith 33.562 - - diff --git a/README.md b/README.md new file mode 100644 --- /dev/null +++ b/README.md @@ -0,0 +1,38 @@ +## ByteString: Fast, Packed Strings of Bytes + +[![Build Status](https://secure.travis-ci.org/haskell/bytestring.png?branch=master)](http://travis-ci.org/haskell/bytestring) + +This library provides the `Data.ByteString` module -- strict and lazy +byte arrays manipulable as strings -- providing very time/space-efficient +string and IO operations. + +For very large data requirements, or constraints on heap size, +`Data.ByteString.Lazy` is provided, a lazy list of bytestring chunks. +Efficient processing of multi-gigabyte data can be achieved this way. + +The library also provides `Data.ByteString.Builder` for efficient construction +of `ByteString` values from smaller pieces during binary serialization. + +Requirements: + + * Cabal 1.10 or greater + * cabal-install + * GHC 6.12 or greater + +Building: +``` +cabal install +``` + +You can run the testsuite as follows: +``` +cabal test +``` + +### Authors +`ByteString` was derived from the GHC `PackedString` library, +originally written by Bryan O'Sullivan, and then by Simon Marlow. +It was adapted and greatly extended for darcs by David Roundy and +others. Don Stewart and Duncan Coutts cleaned up and further extended +the implementation and added the `.Lazy` code. Simon Meier contributed +the `Builder` feature. diff --git a/bytestring.cabal b/bytestring.cabal --- a/bytestring.cabal +++ b/bytestring.cabal @@ -1,5 +1,5 @@ Name: bytestring -Version: 0.10.4.0 +Version: 0.10.4.1 Synopsis: Fast, compact, strict and lazy byte strings with a list interface Description: An efficient compact, immutable byte string type (both strict and lazy) @@ -51,14 +51,13 @@ Author: Don Stewart, Duncan Coutts -Maintainer: Don Stewart , - Duncan Coutts +Maintainer: Duncan Coutts Homepage: https://github.com/haskell/bytestring Bug-reports: https://github.com/haskell/bytestring/issues Tested-With: GHC==7.8.1, GHC==7.6.3, GHC==7.4.2, GHC==7.0.4, GHC==6.12.3 Build-Type: Simple Cabal-Version: >= 1.10 -extra-source-files: README TODO +extra-source-files: README.md TODO source-repository head type: git @@ -67,7 +66,8 @@ source-repository this type: git location: https://github.com/haskell/bytestring - tag: 0.10.4.0 + branch: 0.10.4.x + tag: 0.10.4.1 flag integer-simple description: Use the simple integer library instead of GMP @@ -158,11 +158,14 @@ TestFramework hs-source-dirs: . tests build-depends: base, ghc-prim, deepseq, random, directory, - QuickCheck >= 2.3 && < 3 + test-framework, test-framework-quickcheck2, + QuickCheck >= 2.3 && < 2.7 c-sources: cbits/fpstring.c include-dirs: include ghc-options: -fwarn-unused-binds -fno-enable-rewrite-rules + -threaded -rtsopts + cpp-options: -DHAVE_TEST_FRAMEWORK=1 default-language: Haskell98 -- older ghc had issues with language pragmas guarded by cpp if impl(ghc < 7) @@ -170,6 +173,26 @@ DeriveDataTypeable, BangPatterns, NamedFieldPuns +test-suite regressions + -- temporarily disabled as it allocates too much memory + buildable: False + type: exitcode-stdio-1.0 + main-is: Regressions.hs + hs-source-dirs: . tests + build-depends: base, ghc-prim, deepseq, random, directory, + test-framework, test-framework-hunit, HUnit + c-sources: cbits/fpstring.c + include-dirs: include + ghc-options: -fwarn-unused-binds + -fno-enable-rewrite-rules + -threaded -rtsopts + default-language: Haskell98 + -- older ghc had issues with language pragmas guarded by cpp + if impl(ghc < 7) + default-extensions: CPP, MagicHash, UnboxedTuples, + DeriveDataTypeable, BangPatterns, + NamedFieldPuns + test-suite test-builder type: exitcode-stdio-1.0 hs-source-dirs: . tests tests/builder @@ -181,13 +204,18 @@ build-depends: base, ghc-prim, deepseq, - QuickCheck >= 2.4 && < 3, + QuickCheck >= 2.4 && < 2.7, byteorder == 1.0.*, dlist == 0.5.*, directory, - mtl >= 2.0 && < 2.2 + mtl >= 2.0 && < 2.2, + HUnit, + test-framework, + test-framework-hunit, + test-framework-quickcheck2 - ghc-options: -Wall -fwarn-tabs + ghc-options: -Wall -fwarn-tabs -threaded -rtsopts + cpp-options: -DHAVE_TEST_FRAMEWORK=1 default-language: Haskell98 -- older ghc had issues with language pragmas guarded by cpp @@ -201,4 +229,3 @@ include-dirs: include includes: fpstring.h install-includes: fpstring.h - diff --git a/tests/Regressions.hs b/tests/Regressions.hs new file mode 100644 --- /dev/null +++ b/tests/Regressions.hs @@ -0,0 +1,25 @@ +{-# LANGUAGE ScopedTypeVariables #-} + +import Control.Exception (SomeException, handle) +import Test.HUnit (assertBool, assertEqual, assertFailure) +import qualified Data.ByteString as B +import qualified Test.Framework as F +import qualified Test.Framework.Providers.HUnit as F + +-- Try to generate arguments to concat that are big enough to cause an +-- Int to overflow. +concat_overflow :: IO () +concat_overflow = + handle (\(_::SomeException) -> return ()) $ + B.concat (replicate lsize (B.replicate bsize 0)) `seq` + assertFailure "T.replicate should crash" + where + (lsize, bsize) | maxBound == (2147483647::Int) = (2^14, 2^18) + | otherwise = (2^34, 2^29) + +tests :: [F.Test] +tests = [ + F.testCase "concat_overflow" concat_overflow + ] + +main = F.defaultMain tests diff --git a/tests/builder/Data/ByteString/Builder/Tests.hs b/tests/builder/Data/ByteString/Builder/Tests.hs --- a/tests/builder/Data/ByteString/Builder/Tests.hs +++ b/tests/builder/Data/ByteString/Builder/Tests.hs @@ -1,4 +1,4 @@ -{-# LANGUAGE CPP, ScopedTypeVariables, BangPatterns #-} +{-# LANGUAGE CPP, FlexibleContexts, ScopedTypeVariables, BangPatterns #-} {-# OPTIONS_GHC -fno-warn-orphans #-} -- |