GitLab

more Ptr/FunPtr documentation

Change the interface (but not the implementation) to match the FFI spec:

* added charIsRepresentable, *CAString and *CWString

* currently (and in violation of the spec), fooCString = fooCAString

* The WString versions use UTF-16 under Windows and UTF-32 elsewhere
  (valid only if the compiler defines __STDC_ISO_10646__).

Add missing instances of Data.Bits.Bits for nhc98.

Changes related to arithmetic types:

* Renamed macros NUMERIC_FOO to ARITHMETIC_FOO to match C99-speak

* ARITHMETIC_TYPEs now have a Real instance, otherwise they are quite useless.
  Note that this differs from the FFI spec, but the spec should very probably
  changed in this respect.

* Some changes to fix the wrong assumption that CTime/CClock are integral types,
  C99 in fact guarantees only that they are arithmetic types. This has been
  accomplished by using

     realToInteger = round . realToFrac :: Real a => a -> Integer

  instead of fromIntegral for CTime/CClock. I'm not sure if we could do better,
  going via Double seems to be overkill, but I couldn't think of a better way.
  GHC could e.g. use RULES here. Improvements welcome.

The FFI report does not mention that CClock and CTime have instances for
Bounded, Real, Integral, and Bits.

Added withArrayLen and withArrayLen0

Added finalizeForeignPtr function

Unbreak for nhc98 after recent errno changes.

fix for Hugs

Fix, and simplify, the getting/setting of errno in Foreign.C.Error.

It was previously wrong: although we called a C function to get the
location of errno (correct), we cached the result in a CAF which was
wrong because the location is OS thread dependent.  We must call the C
function every time we need the value of errno.

I also simplified things by making the getter/setter functions inlined
in the same way as the other C fragments in this library, and putting
them in HsBase.h.

MERGE TO STABLE

Annual copyright update

Updated my email address

Hugs-only tweak

Doc wibbles

doc updates for mallocForeignPtr and friends

Hugs only: add finalizers with environments, as per the latest FFI draft.

Make peekArray0 run in constant stack-space by testing the length of
the array first, then calling peekArray (which works backwards from
the end so it can be tail-recursive).

Make the GHC implementation of peekCString run in constant stack-space
by checking the length of the array first and then working backwards
from the end.  Interestingly, this version is faster than the original.

peekCString{Len}: tweak GHC impls (why are these provided anyway?)
  - Do as for Foreign.Marshal.Array.peekArray and read in length-bounded
    string back-to-front.
  - force evaluation of CChar->Char conversions.

merge to STABLE.

Add some realToFrac rules for CFloat,CDouble and CLDouble, so that
eg. realToFrac :: CDouble -> Double turns into a no-op.

Unbreak Show instance for Ptr caused by the changes to showHex

merge to STABLE

* Cleaned up FP_CHECK_ALIGNMENT test a bit.

* Nuked FPTOOLS_CHECK_CCONST test in favour of a more general one
  which handles cross compilation and languages different from C.

Swapped argument order of `newForeignPtr' and `addForeignPtrFinalizer'
to track FFI spec.  (Maybe the Conc ones should do the same?)

This will break NHC.

Merge Foreign.C.TypesISO into Foreign.C.Types

trim imports

Major refactoring of Data/Generics.
This also affects the compiler (because of deriving issues).
This is an intermediate commit.
The library is supposed to compile fine.
But the deriving stuff for Data needs to be revised.
The testsuite for Data/Generics will not pass.
gread is broken at the moment.
So it is strongly recommended not to cvs upd for a few hours or a day.

Detailed description of changes:
- Split up Data/Dynamic into Data/Typeable and Data/Dynamic.
  (This makes clear what part is about TypeReps and cast vs.
   dynamics. The latter is not needed by Data/Generics.)
- Renamed Data/include/Dynamic.h -> Typeable.h to end confusion.
- Split up Data/Generics.hs in a set of modules.
- Revised class Data:
  - Got rid of gunfold but added fromConstr as more primtive one
  - Revised representations of constructors
  - Revised treatment of primitive types
- Revised type of gmapQ; preserved old gmapQ as gmapL
- Added a module Data/Types.hs for treatment of types as values.

This is going somewhere.

Fix reallocBytes: when size is zero, it is supposed to act like free.
Which it did, except that we were complaining about the NULL return
value from C's realloc and reporting it as an error.

MERGE TO STABLE

move portable newForeignPtr from GHC.ForeignPtr to Foreign.ForeignPtr

... and add newForeignPtr_ for nhc98.

For nhc98, define mallocForeignPtr and friends.

Update to latest revision of the FFI spec:

  - foreignPtrToPtr is now unsafeForeignPtrToPtr
  - newForeignPtr_ added.

For nhc98, temporarily make foreignPtrToPtr an alias for unsafeForeignPtrToPtr
until ghc and hugs catch up.

For nhc98 only, export the basic C types non-abstractly.
This is due to a deficiency in the way newtypes are handled in
interface files - the compiler needs full information about the
newtype in order to pass values across the FFI.

Unbreak for nhc98, after the addition of FinalizerPtr as a type synonym,
and the change of Foreign.C.Types to export all newtypes abstract.

Hide this module in the Haddock docs.

Remove the words "A cool hack..." from the documentation of this
module, since they refer to the implementation, not the API :-)

unbreak for Hugs

export architecture-dependent types opaquely.

Added a `FinalizerPtr' synonym as in the FFI Addendum (RC 10).

** ATTENTION **

This will break the "#ifdef __NHC__" and "#ifdef __HUGS__" variants of
`ForeignPtr'.  See `GHC/ForeignPtr.hs' for the definitions.

trim import