Changelog
0.10.0
- Support for half-precision floating point 
__fp16, _Float16, and _Float16x. 
- Support for bfloat16 
__bf16. 
- Support alignment specifier 
_Alignas in struct declatations. 
0.9.4
0.9.3
- Add support for 
__builtin_bit_cast to support clang toolchain,
enabling builds with GHC 9.4+ on Windows. 
0.9.2
- Add support for 128 bit builtin integer types
 
0.9.1
- Support parsing an empty list of clobbers after a colon in 
asm statements 
0.9.0.2
- Correct lower bound on GHC
 
0.9.0.1
Thanks to @vmchale
0.9
- Make 
Trav a monad transformer 
- Add 
runTravTWithTravState 
- Tests now use Cabal-3
 
- Implement alignment calculation and correct size calculation
 
- Add 
Show instance for Type 
- Add machine descriptions for x86_64 and armv7l
 
- Small correction in output of prettyprinter
 
- Allow zeros as line numbers in preprocessed directives
 
Thanks to @aweinstock314, @ersran9, @expipiplus1, @jkoppel, @trofi and @visq
Changes since 0.7.0
Wed May 23 07:08:02 2018 +0200
- Support IEC 60559 types and literals. Closes #52.
 
Wed May 16 07:08:15 2018 +0200
- introduce flag for optional IEC 60559 type support (see #43)
 
Wed May 16 06:36:04 2018 +0200
- WIP: towards a complete Export (#51)
 
Thu Feb 22 13:22:50 2018 -0500
- [syntax] GHC.Generics.Generic and Control.DeepSeq.NFData instances (#49)
(.cabal) drop splitBase flag
 
Tue Feb 20 08:34:39 2018 +0300
- Selectively hide Prelude.<>:  GHC 8.4 compat (#45)
 
Tue Feb 20 00:09:16 2018 -0500
- (Analysis) Minimal support for 
__float128 (#47) 
Sat Nov 25 08:59:59 2017 +0100
- Consider builtin_convertvector from #34 for pretty printing
 
Sat Nov 25 00:46:33 2017 -0600
- add support for 
_Float128 (#41) 
Changes since 0.6.0
Mon Jul 31 2017
- Improve TypeCheck output (#31)
 
Sun Jul 30 2017
- Add 
__float128 (fixes #33) 
Mon Maz 22 2017
- Support clang’s 
__builtin_convertvector (Fixes #34) 
Wed Mar 8 2017
- Move file and parent information to new datatype FilePosition for performance reasons
 
Changes since 0.5.0
Thu Feb 23 2017
- Make gccParseCPPArgs grab the arg to -MF -MT and -MQ
 
Wed Feb 15 2017
- In enums, allow multiple attribute specifiers per enumerator
 
Tue Feb 21 2017
- Allow typedef redefinition if it denotes the same type
 
Tue Feb 21 2017
- Change TypeDefRef to store Type, not Maybe Type
 
Tue Feb 14 2017
- Parse (and ignore) Clang attribute((availability(id=major.minor.rev)))
 
Sun Sep 11 2016
- Add 
__builtin_bswap32/64. 
Wed Jun 22 2016
- Add 
_Alignof to Lexer.x (fixes #7) 
Mon Jun 27 2016
- Updates for C11 (part 1)
 
_Nullable and _Nonnull support as well as Warnings/lint fixes provided by Anthony Cowley (https://github.com/acowley) 
Wed Mar 16 2016
- Consider storage specifier “ThreadSpec” for global and local declarations
 
Tue Mar 15 2016
- Support C11 
_NoReturn, genearlize is_inline to FunSpecs (Syntax) / FunAttrs (SemRep) 
Changes since 0.4.3
Wed Mar 2 2016
- Add direct base type BaseInt128 (complements previous 
__int128 patch)
[email protected] 
Sun Feb 28 2016
- Parse gcc-specific 
__int128 type 
Thu Dec 4 2014
Changes since 0.4.2
Sat Jan 11 2014
- Allow unicode characters in string/char literals and filenames
 
Mon Oct 27 2014
Tue Aug 13 2013
- Do not derive Error instances for newtypes (type parameter has non-parametric role)
 
Mon Aug 12 2013
- Fix bug caused by applying posFile to nopos (reported by Mikhail Sosonkin)
 
Changes since 0.4.1
Tue Mar 19 2013
- TypeCheck: Return Left str instead of fail str (do not rely on MonadError instance of Either)
 
- Improve printing of SUERefs and Ident
 
Thu Feb 28 2013
- Fix parsing and printing of octal character escapes.
 
Tue Jun 12 2012
- Export Annotated type class from AST module
 
Wed Aug 24 2012
Changes 0.3.1 - 0.4.1
Tue Aug 16 2011:
Fri Apr 15 2011: Alexander Bernauer [email protected]
- Show instances (popular request) for AST types, DumpAst demo
 
Changes 0.3 - 0.3.1
Thu Aug 21 [email protected]
- add aliases for exposed parsers, in order to document them
 
Fri Aug 15 [email protected]
- Remove NameMap from Data.Name. We will do this right when neccessary.
 
- Parser public API: expose parsers and the Parser Monad
 
- ParserMonad: Return updated name supply when executing parser
 
- Parser: Expose expression, statement, declaration and file parsers
 
- Data: Add newNameSupply ~ (namesStartingFrom 0)
 
Thu Aug 14 17:13:29 CEST 2008  [email protected]
- Add a utility function to create a “blank” set of cpp arguments.
 
- Make that analysis traversal monad abstract.
 
- Export the type synonym “Register” (and bump version)
 
Wed Aug 13 12:00:57 CEST 2008  [email protected]
- add Data.Position: internalIdentAt
 
Old Changes
Mon Jun  9 23:12:46 CEST 2008  [email protected]
- License switched to 3-clause BSD
$