Documentation improvements

.github/workflows/ci.yml

@@ -18,7 +18,7 @@ jobs:
       fail-fast: true
       matrix:
         os: [ubuntu-latest]
-        ghc: ['8.4', '8.6', '8.8', '8.10', '9.0', '9.2', '9.4', '9.6', '9.8', '9.10', '9.12']
+        ghc: ['8.6', '8.8', '8.10', '9.0', '9.2', '9.4', '9.6', '9.8', '9.10', '9.12']
         include:
         - os: macOS-latest
           ghc: 'latest'

Data/ByteString.hs

@@ -1558,6 +1558,13 @@ isValidUtf8 (BS ptr len) = accursedUnutterablePerformIO $ unsafeWithForeignPtr p
 
 -- | Break a string on a substring, returning a pair of the part of the
 -- string prior to the match, and the rest of the string.
+-- If the substring is not found, return the entire string in the first component
+-- and an empty string in the second component.
+--
+-- >>> breakSubstring "needle" "hayneedlestraw"
+-- ("hay","needlestraw")
+-- >>> breakSubstring "needle" "hay"
+-- ("hay","")
 --
 -- The following relationships hold:
 --
@@ -1576,7 +1583,7 @@ isValidUtf8 (BS ptr len) = accursedUnutterablePerformIO $ unsafeWithForeignPtr p
 --
 -- > fst (breakSubstring x y)
 --
--- Note that calling `breakSubstring x` does some preprocessing work, so
+-- Note that calling 'breakSubstring' @x@ does some preprocessing work, so
 -- you should avoid unnecessarily duplicating breakSubstring calls with the same
 -- pattern.
 --
@@ -2017,8 +2024,7 @@ hGetContentsSizeHint hnd =
              -- we grow the buffer sizes, but not too huge
              -- we concatenate in the end anyway
 
--- | getContents. Read stdin strictly. Equivalent to hGetContents stdin
--- The 'Handle' is closed after the contents have been read.
+-- | Equivalent to 'hGetContents' 'stdin', reading it /strictly/.
 --
 getContents :: IO ByteString
 getContents = hGetContents stdin

Data/ByteString/Lazy.hs

@@ -1597,11 +1597,21 @@ illegalBufferSize handle fn sz =
       msg = fn ++ ": illegal ByteString size " ++ showsPrec 9 sz []
 
 -- | Read entire handle contents /lazily/ into a 'ByteString'. Chunks
--- are read on demand, using the default chunk size.
+-- are read on demand, using the default chunk size ('defaultChunkSize').
 --
 -- File handles are closed on EOF if all the file is read, or through
 -- garbage collection otherwise.
 --
+-- Beware of using this function inside of 'Control.Exception.bracket' \/
+-- 'System.IO.withFile' \/ 'System.IO.withBinaryFile':
+-- lazy I/O can easily escape it, causing 'Control.Exception.bracket' to close a handle prematurely:
+--
+-- >>> withBinaryFile "foo.txt" ReadMode BL.hGetContents >>= print
+-- "*** Exception: foo.txt: hGetBufSome: illegal operation (handle is closed)
+--
+-- The expected way of using 'hGetContents' is to open a handle,
+-- pass it to 'hGetContents', and leave 'hGetContents' to close the handle when it sees fit.
+--
 hGetContents :: Handle -> IO ByteString
 hGetContents = hGetContentsN defaultChunkSize
 
@@ -1644,7 +1654,7 @@ writeFile = modifyFile WriteMode
 appendFile :: FilePath -> ByteString -> IO ()
 appendFile = modifyFile AppendMode
 
--- | getContents. Equivalent to hGetContents stdin. Will read /lazily/
+-- | Equivalent to 'hGetContents' 'stdin', reading it /lazily/.
 --
 getContents :: IO ByteString
 getContents = hGetContents stdin

Data/ByteString/Lazy/Internal.hs

@@ -78,9 +78,14 @@ import Control.Exception (assert)
 -- 8-bit characters.
 --
 #ifndef HS_BYTESTRING_ASSERTIONS
-data ByteString = Empty | Chunk  {-# UNPACK #-} !S.StrictByteString ByteString
-  -- INVARIANT: The S.StrictByteString field of any Chunk is not empty.
-  -- (See also the 'invariant' and 'checkInvariant' functions.)
+data ByteString
+  = Empty
+  | Chunk
+    {-# UNPACK #-} !S.StrictByteString
+    -- ^ Must be nonempty. Consider using
+    -- the smart constructor 'chunk' to ensure this invariant.
+    -- See also the 'invariant' and 'checkInvariant' functions.
+    LazyByteString
 
   -- To make testing of this invariant convenient, we add an
   -- assertion to that effect when the HS_BYTESTRING_ASSERTIONS

Data/ByteString/Short/Internal.hs

@@ -1296,6 +1296,13 @@ isSuffixOf sbs1 = \sbs2 -> do
 
 -- | Break a string on a substring, returning a pair of the part of the
 -- string prior to the match, and the rest of the string.
+-- If the substring is not found, return the entire string in the first component
+-- and an empty string in the second component.
+--
+-- >>> breakSubstring "needle" "hayneedlestraw"
+-- ("hay","needlestraw")
+-- >>> breakSubstring "needle" "hay"
+-- ("hay","")
 --
 -- The following relationships hold:
 --

README.md

@@ -16,7 +16,7 @@ of `ByteString` values from smaller pieces during binary serialization.
 Requirements:
 
   * Cabal 2.2 or greater
-  * GHC 8.4 or greater
+  * GHC 8.6 or greater
 
 ### Authors
 

bench/BenchReadInt.hs

@@ -18,9 +18,6 @@ import Test.Tasty.Bench
 import Data.Int
 import Data.Word
 import Numeric.Natural
-#if !(MIN_VERSION_base(4,11,0))
-import Data.Semigroup (Semigroup((<>)))
-#endif
 import Data.Monoid (mconcat)
 
 ------------------------------------------------------------------------------

bytestring.cabal

@@ -57,17 +57,16 @@ Author:              Don Stewart,
 Maintainer:          Haskell Bytestring Team <[email protected]>, Core Libraries Committee
 Homepage:            https://github.com/haskell/bytestring
 Bug-reports:         https://github.com/haskell/bytestring/issues
-Tested-With:         GHC==9.12.1,
-                     GHC==9.10.1,
+Tested-With:         GHC==9.12.2,
+                     GHC==9.10.2,
                      GHC==9.8.4,
-                     GHC==9.6.6,
+                     GHC==9.6.7,
                      GHC==9.4.8,
                      GHC==9.2.8,
                      GHC==9.0.2,
                      GHC==8.10.7,
                      GHC==8.8.4,
-                     GHC==8.6.5,
-                     GHC==8.4.4
+                     GHC==8.6.5
 Build-Type:          Simple
 extra-source-files:  README.md Changelog.md include/bytestring-cpp-macros.h
 
@@ -110,7 +109,7 @@ common language
 
 library
   import:          language
-  build-depends:   base >= 4.11 && < 5, ghc-prim, deepseq, template-haskell
+  build-depends:   base >= 4.12 && < 5, ghc-prim, deepseq, template-haskell
 
   if impl(ghc < 9.4)
     build-depends: data-array-byte >= 0.1 && < 0.2
@@ -232,6 +231,10 @@ test-suite bytestring-tests
   if !arch(wasm32)
     ghc-options:    -threaded
 
+  -- https://github.com/haskellari/splitmix/issues/101
+  if os(openbsd)
+    build-depends: splitmix < 0.1.3 || > 0.1.3.1
+
 benchmark bytestring-bench
   import:           language
   main-is:          BenchAll.hs
@@ -245,8 +248,7 @@ benchmark bytestring-bench
   hs-source-dirs:   bench
 
   ghc-options:      -O2 "-with-rtsopts=-A32m"
-  if impl(ghc >= 8.6)
-    ghc-options:    -fproc-alignment=64
+                    -fproc-alignment=64
   build-depends:    base,
                     bytestring,
                     deepseq,

include/bytestring-cpp-macros.h

@@ -23,8 +23,7 @@ which are known not to trap (either to the kernel for emulation, or crash).
 
 
 #define HS_UNALIGNED_ByteArray_OPS_OK \
-  MIN_VERSION_base(4,12,0) \
-  && (MIN_VERSION_base(4,16,1) || HS_UNALIGNED_POKES_OK)
+  MIN_VERSION_base(4,16,1) || HS_UNALIGNED_POKES_OK
 /*
 The unaligned ByteArray# primops became available with base-4.12.0/ghc-8.6,
 but require an unaligned-friendly host architecture to be safe to use