* harbour/include/hbapicls.h
* harbour/src/vm/classes.c
+ added new C function
HB_SIZE hb_clsGetVarIndex( HB_USHORT uiClass, PHB_DYNS pVarSym );
which can be used as hack to speedup accessing some object instance
variables
* harbour/include/hbclass.ch
+ added some class(y) compatible commands for delegated messages
* harbour/include/hboo.ch
* harbour/src/vm/classes.c
* harbour/src/rtl/tclass.prg
* native support for delegated messages - work in progress
* harbour/src/rtl/hbsocket.c
! fixed error checking to eliminate possible error code overload
* harbour/ChangeLog
* marked old TODO as [DONE]
* harbour/src/rtl/filesys.c
* unified hb_fsSeek[Large]() return value when seek operation cannot
be executed for given file handle. Now it's 0 for all platforms.
* harbour/include/hbapicls.h
* harbour/src/vm/classes.c
+ added new public C function
PHB_SYMB hb_clsFuncSym( HB_USHORT uiClass )
It returns class function symbol or NULL.
* harbour/ChangeLog
* harbour/config/os2/gccomf.mk
* harbour/include/hbpcode.h
* harbour/include/hbdebug.ch
* harbour/include/hbmath.ch
* harbour/include/hbgfx.ch
* harbour/src/vm/classes.c
* harbour/utils/hbmk2/hbmk2.prg
* removed dummy trailing spaces at EOLs
* harbour/src/vm/classes.c
+ extended error messages in __CLSINSTSUPER() function to show the name
of super class which cannot be created
* harbour/src/rtl/tclass.prg
* force RTE for superclasses defined using function declared as DYNAMIC
It should help in locating the problem in wrongly linked code, i.e.
without some superclasses.
* utils/hbrun/hbrun.prg
! Don't search passed program in PATH and hbrun dir, if
program name has path component.
* contrib/xhb/xhbenum.c
* contrib/xhb/xhbqself.c
* contrib/xhb/xhbwith.c
* contrib/xhb/xhberrc.c
! Fixed msvc64 warnings after HB_SIZE extension in stack API.
* contrib/make.hbs
+ Added logic to protect against case when 'clean install'
is used but hbrun or hbmk2 isn't available when launching
the make process. In this case 'clean' phase won't be executed
for contribs, which is solved by forcing a -rebuild in
'install' phase.
* src/vm/strapi.c
* src/vm/runner.c
* src/vm/itemapi.c
* src/vm/asort.c
* src/vm/hvm.c
* src/vm/arrays.c
* src/vm/memvars.c
* src/vm/set.c
* src/vm/classes.c
* src/rtl/lennum.c
* src/rtl/strcase.c
* src/rtl/replic.c
* src/rtl/padr.c
* src/rtl/padc.c
* src/rtl/strtran.c
* src/rtl/strtoexp.c
* src/rtl/padl.c
* src/rtl/saverest.c
* src/rtl/right.c
* src/rtl/philes.c
* src/rtl/mlcfunc.c
* src/rtl/left.c
* src/rtl/xsavescr.c
* src/rtl/setkey.c
* src/rtl/stuff.c
* src/rtl/trace.c
* src/rtl/samples.c
* src/rtl/inkeyapi.c
* src/rtl/inkey.c
* src/rtl/mtran.c
* src/rtl/colorind.c
* src/rtl/rat.c
* include/hbapi.h
* include/hbapistr.h
* Renamed HB_SIZE variables to have 'n' prefix.
(verified to generate the same objects as before)
* src/rtl/philes.c
* Changed hb_retnint()/hb_retni() to hb_retns().
* contrib/hbgd/gdwrp.c
* Formatting.
* src/vm/estack.c
* src/vm/debug.c
* src/vm/hvm.c
* src/vm/pvalue.c
* src/vm/arrays.c
* src/vm/proc.c
* src/vm/memvars.c
* src/vm/eval.c
* src/vm/classes.c
* include/hbstack.h
* include/hbapi.h
+ Win64 support for HVM stack and other related internals.
(HB_ULONG -> HB_SIZE, HB_LONG -> HB_ISIZ, long -> HB_ISIZ)
+ Using size API in debug functions.
; TOFIX: These two warnings remain in HVM (not counting dlmalloc):
---
hvm.c(10695) : warning C4244: 'argument' : conversion from 'HB_MAXINT' to 'double', possible loss of data
macro.c(1315) : warning C4310: cast truncates constant value
---
* src/vm/eval.c
* HB_FORNEXT() loop value can now be 64-bit values (was 32).
* src/vm/runner.c
* Formatting.
* include/hbcomp.h
* include/hbcompdf.h
* include/hbexpra.c
* include/hbexprb.c
* include/hbexprop.h
* include/hbmacro.h
* src/common/expropt1.c
* src/common/expropt2.c
* src/compiler/genc.c
* src/compiler/gencc.c
* src/compiler/harbour.y
* src/compiler/harbour.yyc
* src/compiler/hbdbginf.c
* src/compiler/hbdead.c
* src/compiler/hbident.c
* src/compiler/hblbl.c
* src/compiler/hbmain.c
* src/compiler/hbopt.c
* src/compiler/hbpcode.c
* src/vm/macro.c
+ Adding Win64 support for compiler/macro subsystems.
* HB_ULONG -> HB_SIZE where applicable.
* HB_LONG -> HB_ISIZ where applicable.
* Some now unnecessary cast deleted.
* hb_compExprAsStringLen() return value changed to HB_SIZE from int.
* hb_compNOOPfill() 3rd parameter changed to HB_ISIZ from int.
- Deleted unused hb_compExprMacroListLen().
; TOFIX: Use '$<lNumber>$' (or maybe newly added sNumber)
instead of '$<iNumber>$' in .y code.
Przemek, what do you think?
* src/compiler/Makefile
* C files sorted.
* src/vm/hvm.c
! hb_vmStaticsCount() changed to use HB_SIZE instead of
HB_ULONG. Fixing pending warnings.
* include/hbapi.h
* Changed hb_struEnum.offset structure member from HB_LONG
to HB_ISIZ, fixing one more pending Win64 type mismatch
problem. The change doesn't modify HB_ITEM size.
Please verify me.
* src/vm/classes.c
! Using size API after above change.
; QUESTION: In itemapi asEnum.offset is stored using
hb_itemPutNInt(). Should this be changed to
hb_itemPutNS()?
* contrib/hbct/ctstr.c
+ Cleaned away some HB_ULONG casts and using HB_PFS format
strings instead.
* contrib/hbct/tab.c
! Fixed regression from 2010-06-20 12:34 UTC+0200 Viktor Szakats,
by deleting now unnecessary casts. Shown by mingw64 4.6.0exp.
I'd appreciate if someone could check me!!
* src/vm/thread.c
* Added casts to pacify msvc64 warnings.
Of some the ones related to pMutex->events I'm unsure of,
probably it's better to use 'int' instead of HB_ULONG for
internal variable in some cases to make it look cleaner,
though I assume there was a reason for it, so I left it.
Please check me.
* src/vm/classes.c
* Added casts to pacify msvc64 warnings.
* harbour/ChangeLog
! fixed typo
* harbour/src/rtl/hbsocket.c
! fixed OS2 builds using TCPV40HDRS - please test
* harbour/src/vm/classes.c
+ added new PRG function:
__clsMsgType( <nClassH>, <cMsg> | <sMsg> ) -> <nType>
<nType> is one of HB_OO_MSG_* values defined in hboo.ch or -1 if
message is not supported.
* src/vm/runner.c
* src/vm/garbage.c
* src/vm/classes.c
* src/rtl/cdpapi.c
* src/rtl/cdpapihb.c
* src/rdd/dbcmd.c
* include/hbapicdp.h
* include/hbapiitm.h
* include/hbextern.ch
* include/hbdefs.h
* include/hbsetup.h
* include/hbapi.h
- Deleted code guarded with (previously disabled) HB_LEGACY_LEVEL2
* include/rdd.api
* include/clipdefs.h
+ Permanently enabled HB_LEGACY_LEVEL2 related changes.
* contrib/xhb/bkgtsks.c
- Deleted HB_OS_WIN_USED. It wasn't used.
* include/hbapi.h
* src/vm/cmdarg.c
* hb_winmainArgGet() changed to not use Windows types.
This function is declared in hbapi.h and this header
shouldn't depend on windows.h.
* include/hbwince.h
* include/hbsetup.h
* HB_OS_HAS_DRIVE_LETTER is now configured in central
place (hbsetup.h) for WinCE platform also.
* include/hbwince.h
* include/hbdefs.h
* Windows (CE and not-CE) specific declarations moved from
hbwince.h to hbdefs.h.
* include/hbdefs.h
% hbwince.h inclusion now better guarded to not include it
for non WinCE platforms.
+ HB_OS_WIN_USED macro will now autodetected by checking
whether windows.h was included before this header.
This is required for hbwince.h. I don't like it, since
hbapi.h still depends on windows.h for WinCE platform,
which still ties us to unnecessary windows.h inclusion
rules. Rather, hbwince.h should be included directly
by .c files when needed and when HB_OS_WIN_CE is defined.
After all this our little Windows API implementation
for compilers which need it.
* include/hbdefs.h
* include/hbwmain.c
+ hb_winmainArgInit() declaration moved to hbwmain.c, the
only place where this is used.
* Deleted HB_EXPORT from hb_winmainArgInit. Pls speak up if
this is wrong. It's possible to move this back to hbdefs.h,
but parameter types should be changed to ANSI ones.
* contrib/gtwvg/gtwvg.c
* contrib/gtwvg/gtwvg.h
* contrib/gtwvg/wvggui.c
* contrib/gtwvg/wvgcuig.c
* contrib/gtwvg/wvggui.h
* contrib/gtwvg/wvgwin.c
* contrib/gtwvg/wvgutils.c
* contrib/gtwvg/wvgcore.c
* contrib/gtwvg/wvgwing.c
- Deleted HB_OS_WIN_USED. Now it should work without it.
windows.h was already included in non-official method, so
it's now okay also.
* harbour/src/vm/task.c
* harbour/src/vm/classes.c
* harbour/src/rtl/strpeek.c
* updated to use ANSI types
* harbour/src/rtl/hbmd5.c
% small optimization (more compiler friendly code due to fixed alignment)
* contrib/hbqt/hbqt_misc.prg
* contrib/hbide/hbide.prg
- Deleted hack which moved HBXBP specific functionality
into the core of HBQT.
I had to restore QT_PTROFXBP() macros inside hbide.
QT_PTROFXBP() has not much to do with QT_PTROF() as the
former serves to access oWidget member of passed class.
I see no need to move such logic into HBQT core code.
I restored QT_PTROF() also, but it's redefined as a dummy
in hbide.prg. I don't understand now why QT_PTROF() works
as dummy in hbide, but it doesn't when used as dummy in
HBXBP.
All in all, this means the .prg level :pPtr trick is still
needed in some places, while it's not needed in some others.
; TOFIX: HBQT has no pointer checking at all before accessing
C++ level objects, which means the simplest .prg level
error is instantly resulting in a GPF. All hb_par_*()
results must be checked for NULL before accessing them,
or better yet, and RTE should be generated right from
the hb_par_*() function.
* src/vm/classes.c
* Minor formatting.
* harbour/include/hbapiitm.h
* harbour/src/vm/classes.c
* added new internal HVM macro
* harbour/src/vm/thread.c
+ added new function
hb_threadOnceInit( @<item> <value> ) -> <lInitialized>
It assigns <value> to <item> if <item> is NIL in MT safe way so it's
assigned only once. It's reduced but faster version of hb_threadOnce()
* harbour/include/hbapicdp.h
* define HB_WCHAR as wchar_t on Windows platforms for compilers which
refuse to make conversions between types using the same base type.
* harbour/contrib/hbwin/win_dll.c
* code cleanup
* harbour/config/win/bcc.mk
* increased library page size for 64 to 96
* harbour/src/vm/thread.c
* harbour/include/hbthread.h
% removed 'BOOL fSync' parameter from hb_threadMutexCreate().
This parameter was added in the past to emulate xbase++ behavior
when oSignal:wait() is called from SYNC method but real life tests
shows that such feature works in xbase++ in different way then
it was initially described and Harbour code was updated to use
different solution making above parameter unusable.
* harbour/src/vm/hvm.c
* harbour/src/vm/thread.c
* harbour/src/vm/classes.c
* harbour/src/rtl/hbgtcore.c
* harbour/contrib/hbnetio/netiocli.c
* harbour/contrib/hbqt/hbqt_slots.cpp
* updated to use hb_threadMutexCreate() without parameters
* harbour/src/rdd/dbfnsx/dbfnsx1.c
! fixed bug reported by Jaroslav Janik (many thanks) - if index FOR
expression excluded all records from non empty table then root page
was not initialized and DBFNSX generated corruption RTE.
* harbour/utils/hbi18n/hbi18n.prg
* set default GT to CGI
* harbour/src/common/hbgete.c
! fixed memory leak in non UNICODE Windows hb_getenv_buffer()
! add C stack buffers to eliminate memory allocation in UNICODE
Windows hb_getenv_buffer() - they are used if envvar name and
passed buffer are not too large. It fixes HB_FM_STATISTICS
Windows builds.
* harbour/contrib/hbwin/wapi_winuser.c
+ added WAPI_DESTROYWINDOW function
* harbour/contrib/xhb/traceprg.prg
* harbour/contrib/xhb/hbcompat.ch
! fixed typo in TraceLog() condition - was always disabled
+ added emulation for xHarbour Set( _SET_TRACE* [, <params,...> ] )
settings
* harbour/src/vm/hvmall.c
! define _GNU_SOURCE to fix compilation with HB_FM_DL_ALLOC in Linux
builds
* harbour/include/hbdefs.h
* declare HB_FOFFSET as LONG instead of LONGLONG on platforms where
LONG is 64bit integer to be compatible with C RTL file API
* harbour/src/vm/maindllh.c
* harbour/src/vm/maindllp.c
* harbour/contrib/hbct/ctstr.c
! casting in HB_TRACE() massages
* harbour/include/hbapi.h
* harbour/include/hbapiitm.h
* harbour/src/vm/arrays.c
* harbour/src/vm/hashes.c
* harbour/src/vm/itemapi.c
+ added hb_arrayCloneTo(), hb_hashCloneTo(), hb_itemCloneTo()
* harbour/src/vm/arrayshb.c
* harbour/src/vm/classes.c
* harbour/src/vm/hashes.c
* harbour/src/vm/hashfunc.c
* harbour/src/vm/hvm.c
% use hb_*CloneTo() instead of hb_*Clone()
* harbour/include/hbapi.h
* harbour/src/vm/garbage.c
- removed hb_gcRegisterSweep() function and support for user defined
sweep mark functions in GC blocks
- removed hb_gcAlloc() function
+ added new function hb_gcAllocate() which work in similar way to removed
hb_gcAlloc() function but it accepts as second parameter HB_GC_FUNCS
structure instead of HB_GARBAGE_FUNC_PTR and returned GC blocks are
locked so programmer does not have to worry that they can be removed
by GC.
+ added new internal function hb_gcAllocRaw() which works like
hb_gcAllocate() but returned GC blocks are not locked. They can
be used only in HVM in places well known that cannot activate
automatic GC scan. It should not be used by 3-rd party code.
* harbour/include/hbapi.h
* harbour/include/hbapiitm.h
* harbour/src/vm/extend.c
* harbour/src/vm/itemapi.c
* modified hb_parptrGC(), hb_parvptrGC(), hb_arrayGetPtrGC(),
hb_itemGetPtrGC() functions to use HB_GC_FUNCS structure instead
of HB_GARBAGE_FUNC_PTR
* modified hb_itemPutPtrGC() and indirectly related functions like
hb_arraySetPtrGC() or hb_stor[v]ptrGC() to automatically unlock
passed GC block just after attaching to known HVM item.
* harbour/include/hbregex.h
* harbour/src/rtl/hbregexc.c
* declare GC block detractor as static function
+ added hb_regexIs() function
* updated to work with new GC API
* removed unnecessary in new GC API hb_gcLock() call - GC blocks
allocated by hb_gcAllocate() are already locked and they mustn't
be locked explicitly by programmer (until he does not exactly knows
what he does ;))
* harbour/src/rtl/hbregex.c
* use hb_regexIs() instead of using regex GC destructor
* removed unnecessary in new GC API hb_gcUnlock() call - GC blocks
stored in HB_IT_POINTER item are automatically unlocked
* harbour/src/rtl/hbinet.c
* harbour/contrib/hbcurl/hbcurl.c
* updated to work with new GC API
% unlock items storing callback codeblocks and use GC mark functions
to mark them as used in GC passes
* harbour/src/pp/pplib.c
* harbour/src/vm/runner.c
* harbour/src/vm/codebloc.c
* harbour/src/vm/hashes.c
* harbour/src/vm/hvm.c
* harbour/src/vm/arrays.c
* harbour/src/vm/thread.c
* harbour/src/vm/dynlibhb.c
* harbour/src/rtl/hbzlibgz.c
* harbour/src/rtl/hbregex.c
* harbour/src/rtl/hbregexc.c
* harbour/src/rtl/hbgtcore.c
* harbour/src/rtl/hbi18n1.c
* harbour/src/rdd/wacore.c
* harbour/contrib/hbsqlit3/hbsqlit3.c
* harbour/contrib/hbnetio/netiosrv.c
* harbour/contrib/hbmzip/hbmzip.c
* harbour/contrib/hbhpdf/harupdf.c
* harbour/contrib/hbwin/olecore.c
* harbour/contrib/hbwin/win_prn1.c
* harbour/contrib/hbwin/win_dll.c
* updated code which was using hb_gcAlloc() to work with new GC API.
* harbour/contrib/hbwin/axcore.c
* updated syntax description
TODO: update contrib code: hbqt, hbssl, ... to work with new GC API.
I would like to ask authors of above libraries to do it.
TODO: defined when and how we should free AX control and user callback
handler and update AX code to respect new definition
- source
+ src
* src/Makefile
* src/dynlib/mt/Makefile
* src/dynlib/Makefile
* INSTALL
* Makefile
* ChangeLog
* harbour.spec
* mpkg_tgz.sh
* Renamed 'source' dir to 'src' to move closer to other FOSS
projects. It's also easier to type, plus it has the side
effect that some cmdlines will be shorter in the
build process.
