* harbour-ce-spec
* harbour-w32-spec
* include/hbver.h
* harbour.spec
* RC1 -> RC2
* contrib/hbziparch/hbziparc.c
! Fixed potential GPF in HBZIPCLEANUP().
! HBZIPCLEANUP() changed to be a HB_FUNC() (from HB_EXIT_FUNC()),
so now it has to be called manually, but at least it's
now doing something useful. (HB_EXIT_FUNC() in .c code
never gets executed by itself).
* include/hbapi.h
* source/rtl/set.c
* source/vm/fm.c
! Fixed _SET_HBOUTLOG, _SET_HBOUTLOGINFO handling in fm.c.
hb_setget*() functions cannot be used here as the set
system is deinited at this point. So local buffers are
holding the values.
* include/set.ch
* include/hbset.h
* source/rtl/set.c
* source/vm/fm.c
* source/vm/extrap.c
+ Added: Set( _SET_HBOUTLOGINFO[, <cValue>] ) -> <cOldValue>
Default value: ""
This user defined text will be included in each hb_out.log
messages. This way it's possible to include custom information
in these entries to help identify and locate them.
! Fixed a few string not marked with HB_I_() translation
marker.
* contrib/Makefile
* After this change, HB_CONTRIBLIBS will _override_ the
default list of contribs, rather than adding them to
the default list (which BTW contains everything since
this week, so there's no much point anyway, besides if
someone was hosting local contribs in the Harbour contrib
dir).
* include/hbver.h
- Removed recenly added HB_VERSION 0x010000.
* Instead I've changed both .prg level and .c level __HARBOUR__
to contain this more granular (full) version number.
; NOTE: If someone is relying on the value of __HARBOUR__ in
.prg code, the values should be multiplied by 256.
* include/hbstdgen.ch
* include/hbver.h
* include/hbdefs.h
* source/pp/pplib.c
* source/pp/ppcore.c
* source/pp/hbpp.c
* source/common/hbver.c
* source/common/hbverdsp.c
* contrib/hbsqlit3/hbsqlit3.c
* contrib/examples/pp/hbppdef.h
* Changed .c level __HARBOUR__ to have the same value
as the existing .prg level one.
* PP now gets the __HARBOUR__ define directly from the .c
level instead of generating it locally.
; QUESTION: Should or shouldn't we change __HARBOUR__
to contain revision number too? (0x0100 -> 0x01000)
This would allow for a greater level of development
opportunities between minor revisions, more "agile"
development.
* include/hbpp.h
* source/pp/ppcore.c
- Two __XHARBOUR__ guards removed.
* include/hbapierr.h
* source/rtl/errorapi.c
* source/rdd/hbsix/sxtable.c
* contrib/hbct/charonly.c
* contrib/hbct/charevod.c
* contrib/hbct/charone.c
* contrib/hbct/pos1.c
* contrib/hbct/ascpos.c
* contrib/hbct/token1.c
* contrib/hbct/atnum.c
* contrib/hbct/count.c
* contrib/hbct/replace.c
* contrib/hbct/ctpad.c
* contrib/hbct/justify.c
* contrib/hbct/remove.c
* contrib/xhb/hbxml.c
* contrib/hbmzip/hbmzip.c
* contrib/hbtip/utils.c
* Changed remaining &hb_errFuncName -> HB_ERR_FUNCNAME
+ Added compatibility comments.
! Fixed hb_errFuncName value (changed to 1 to be the
same as HB_ERR_FUNCNAME).
; NOTE: 3rd party developers are encouraged to use
HB_ERR_FUNCNAME instead of hb_errFuncName.
There is no change in functionality.
[TOMERGE 1.0.0RC1]
* include/set.ch
* include/hbset.h
* source/rtl/set.c
* source/vm/fm.c
* source/vm/extrap.c
+ Added: Set( _SET_HBOUTLOG[, <cFileName>] ) -> <cOldFileName>
Default value: "hb_out.log"
This replaces hb_fm.log and hb_ex.log, and makes it possible
for the applications to control the filename where these
internal Harbour outputs are written to.
* source/rtl/gtwvt/gtwvt.c
% Minor redundancy removed.
* include/hbgtinfo.ch
* source/rtl/gtwvt/gtwvt.c
! Fixed stupid typo from me. This is why maximized size
didn't want to get right.
! Fixed to not reposition non-maximized windows while
resizing them and their top/left corner was off screen.
! Fixed to not allow font screen below a certain limit
(3x4 pixels). This fixes some of the more obvious cases
when window gets trashed.
! Fixed to not resize the windows when toggling the
resizing feature using HB_GTI_RESIZABLE.
+ Added to disable maximize icon after maximizing and
reenable it on the first resize attempt.
- Removed TIMER related stuff as it's very difficult
(or impossible) to support on other platforms.
Timer support could be added to hbw32.lib, though.
* Some cleanups.
; Remaining TOFIXes:
- Window trashing when settling on a size which
cannot be properly filled with the console. Try with
vertically or horizontally narrow window shapes.
- Mark&Copy is still trashing when swapping directions
with the selection compared to the starting position.
* tests/wvtext.prg
! Added missing SVN keywords properties.
! Added missing SVN eol property.
! Fixed to use the default font to have a more common
ground for testing.
+ Added markers to the corners to easily spot chopped
borders.
* contrib/xhb/Makefile
* contrib/xhb/common.mak
- Removed some xhb specific non-standard headers from
the list of those getting installed to the Harbour central
include dir. Only hbcompat.ch, hbcompat.h and xhb.ch
goes now to the official include dir. After this change
has been merged with RC1, xhb.lib may be included in
the default build.
[TOMERGE RC1]
* contrib/hbw32/tests/testdll.prg
+ Added some more DLL calling samples.
* harbour/include/hbdate.h
* harbour/source/rtl/dates.c
+ added hb_dateUnformat() - code by David G. Holm extracted from CTOD()
* updated copyright note to point original author of hb_dateFormat()
and hb_dateUnformat()
* harbour/source/rtl/dateshb.c
* use hb_dateUnformat() in CTOD()
* harbour/source/common/hbwince.c
! fixed typo in GetDriveTypeA()
* temporary disabled GetDriveTypeW() call because not all CRTLs for
WinCE support it
* harbour/include/hbexpra.c
* harbour/include/hbexprb.c
* harbour/source/common/expropt2.c
* harbour/source/compiler/harbour.y
* harbour/source/compiler/harbour.yyc
* added full expression list stripping - it enables optimizations
in few additional places
! fixed code to always refresh expression address after direct or
indirect reduction - seems that we have some memory leaks and
possible GPFs not located so far
Please make tests with your code.
* source/rtl/gtwvt/gtwvt.c
! Fixed windows size after maximize.
* Temporary restored "instant resize" on mouse move.
+ contrib/hbw32/dllcall.c
* contrib/hbw32/common.mak
* contrib/hbw32/Makefile
+ Added DLL interface. From xhb. Work of several authors.
; I've applied some modifications made by Alex Strickland.
; The API calls below are implemented:
XBase++ compatible:
- DLLLoad()
- DLLUnload()
- DLLPrepareCall()
- DLLExecuteCall()
- DLLCall()
Proprietary:
- LoadLibrary() - Same as DLLLoad()
- FreeLibrary() - Same as DLLUnload()
- GetProcAddress()
- CallDLL()
- CallDLLBool()
- GetLastError()
- SetLastError()
! Fixed GetProcAddress() passed ordinals to always be
WORDs casted to DWORD.
! Fixed a few possible buffer overruns.
* Cleanup, format, etc.
; Compile with HB_NO_ASM if you use BCC55 and don't have
TASM32 (or upgrade to free BCC58 which contains TASM32).
; "C struct" support was stripped (but that seems only a
small portion.
; Please test. I doesn't work me for simple test code,
I have no idea why.
; TOFIX: Replace MessageBox() calls with RTEs.
; TOFIX: Solve any name clashing problems and/or use namespace
for the proprietary interface.
* include/hbapierr.h
+ HB_ERR_FUNCNAME() added to replace &hb_errFuncName
* tests/wvtext.prg
* include/hbgtinfo.ch
* include/hbgtcore.h
* source/rtl/hbgtcore.c
* source/rtl/gtwvt/gtwvt.c
* Changed the way notifications work. It's now platform
neutral and more simple.
hb_gtInfo( HB_GTI_NOTIFIERBLOCK, {|nEvent, ...| myhandler( nEvent, mycustomdata, ... ) })
may be used to set it. Use NIL to disable it. The purpose of
this is to be able to get notifications from GT related
user interactions not initiated inside the apps UI world.
F.e. to save screen size after the user has changed it,
or to have a last chance to commit tables, set flags, etc
when user (or OS) closed the app window.
+ Added basic support to handle return values from notifier.
! Added missing SVN header to wvtext.prg
; TOFIX: New TIMER stuff in GTWVT should be replace by
a platform neutral solution, which is probably MT.
(see also "background events")
; TOFIX: Is there a way to force "non-full content" resize
on the window level? This way we could avoid
current trashing while resizing.
* tests/wvtext.prg
* include/hbgtinfo.ch
* include/hbgtcore.h
* source/rtl/gtwvt/gtwvt.h
* source/rtl/gtwvt/gtwvt.c
! Fixed a few warnings.
! Fixed bad selection behaviour when starting the
selection (to not let it "stick" to the upper left corned).
- Removed predefined symbols as a way to control GT
behaviour.
* HB_GTI_MARKCOPY -> HB_GTI_SELECTCOPY to use more generic
terminology.
* HB_GTI_RESIZEABLE -> HB_GTI_RESIZABLE
* harbour/source/rtl/gtwvt/gtwvt.c
* harbour/source/rtl/gtwvt/gtwvt.h
* harbour/source/rtl/hbgtcore.c
* harbour/include/hbgtinfo.ch
+ HB_GTI_MARKCOPY
Added feature to toggle "Mark and Copy" Sysmenu option with:
Hb_GtInfo( HB_GTI_MARKCOPY, TRUE|FALSE ) -> lPrevState
Removed necessity to have Hb_NoCopyConsole() to not include it.
Now appln always HAS this SysMenu option.
+ HB_GTI_RESIZEABLE
Added feature to toggle resizing of appln window as:
Hb_GtInfo( HB_GTI_RESIZEABLE, TRUE|FALSE ) -> lPrevState
Function Hb_NoResizeableWindow() still hold good for startup.
You always can activate/deactivate this feature from within appln.
Few more optimizations while resizing, still work in progress.
* harbour/tests/wvtext.prg
* Updated to demonstrate above functionality.
* include/hbgtinfo.ch
+ Added comment to compatibility #defines, to not use
them, since they deprecated.
[TOMERGE RC1]
* HB_GTI_CB_* -> HB_GTE_* (stands for GT Event). Removed
plain GTI_ versions, GTI_* range is frozen, pls don't
and new codes to them and especially don't use them in
core code.
* include/hbgfx.ch
* include/hbgfxdef.ch
* contrib/gtwvg/gtwvg.c
* source/rtl/gtxwc/gtxwc.c
* source/rtl/gtwvt/gtwvt.c
* GFX_* constants renamed to HB_GFX_*. The old ones are
still working but it's recommended to update to use HB_GFX_*.
#define HB_GT_NO_XHB to disable the compatibility #defines,
if it collides with your code or you want to test after
transition.
; [TOMERGE RC1]
* source/rtl/dirdrive.c
! Guarded C53 undoc functions with HB_C52_UNDOC, which is enabled
so it won't make a difference in default builds.
; [TOMERGE RC1]
* source/rtl/filesys.c
! Fixed hb_fsIsDrv() to use GetDriveTypeA() API, to avoid some
win32 system MessageBox() to be shown when certain types of
drives (f.e. USB reader devices) are queried.
; Please test.
; TOFIX: hb_fsChDrv() suffers from the same problem.
; [TOMERGE RC1 - after testing]
* contrib/hbcurl/hbcurl.c
* contrib/rddads/adsfunc.c
* contrib/rddads/ads1.c
% Optimized out explicit function name strings in RTE calls.
! The above is actually a fix (I hope) for the one in
hb_adsCheckLock(). Error text also made more consistent.
; [TOMERGE RC1 - after testing]
* contrib/hbw32/w32_prn.c
% Minor optimizations, cleanups.
* harbour/source/rtl/hbgtcore.c
* harbour/include/hbgtcore.h
* harbour/include/hbgtinfo.ch
* harbour/source/rtl/gtwvt/gtwvt.c
* harbour/source/rtl/gtwvt/gtwvt.h
+ Added building blocks for CallBacks to be set with Hb_GtInfo( GTI_* ).
* harbour/tests/wvtext.prg
+ Test program to taste few of the callbacks.
/* This also sets the ground for futuristic multi-window GT
I have kept MW GT code out of current implementation.
This implementation does not interfere nor affects current functionality.
A work in progress.
Viktor, I have found a way to implement 'consol shearing, working on it
and for that to happen these building blocks are a must.
Przemek, please review the structure of <PHB_PGT>. I have added a few members.
Please bear with my meagre knowledge if I did something stupid, just correct it.
*/
* harbour/include/hbpp.h
* harbour/source/pp/ppcore.c
+ added hb_pp_errorCount() which returns total number of errors
which appeared during preprocessing
* harbour/source/pp/hbpp.c
* added new switch:
-u[<file>] use command def set in <file> (or none)
* if neither -o nor -v is specified then -w is default action
+ added to default search paths INCLUDE envvar
* set application exit code to non 0 value if errors appear
during preprocessing
* harbour/source/compiler/harbour.yyc
* harbour/source/compiler/harbour.y
! added support for expression list without parenthesis used in
statements - Clipper compatibility fix.
* include/hbver.h
+ Added HB_VERSION unified version identifier:
three bytes: major + minor + revision (0x000909)
This is to aid 3rd parties to detect version.
[RC1]
* contrib/hbsqlit3/make_b32.bat
* contrib/hbsqlit3/hbsqlit3.c
! SQLITE3_OPEN(): Fixed potential leak.
+ BCC warnings are now suppressed for the foreign code only,
and this is now effective in all build systems.
* Code formatting.
* harbour/include/hbexprb.c
! do not optimize HB_BIAND(), HB_BITOR() and HB_BITXOR() when only
one parameter is passed to keep RT error compatibility
* use hb_compExprListStrip() to optimize function calls also when
parameters are passed in parenthesis, f.e.:
? CHR(65), CHR((65))
It's not Clipper compatible and disabled by -kc switch.
Viktor it will interact with some hbtest tests when compile time
optimization used by Clipper gives different results then RT function
f.e. CHR( ( 256 ) ). Please think if we should keep it.
It's important optimization if someone uses constant values or
expressions defined in parentheses.
* harbour/include/hbexprop.h
* harbour/include/hbexprb.c
* harbour/source/common/expropt2.c
+ added hb_compExprReducePower() and compile time optimization for:
<NumConst1> ^ <NumConst2>
+ added compile time optimization for:
<NumConst1> % <NumConst2>
when <NumConst1> or <NumConst2> is not integer value
NOTE: both optimizations are not Clipper compatible and disabled
when -kc compiler switch is used
* include/hbvm.h
* source/vm/hvm.c
* source/vm/extrap.c
! Fixed OS/2 problem reported by David.
(BTW, do we need really need to remove the exception
handler before exiting? If yes, shouldn't we do the
same under Win32?)
* source/common/hbstr.c
! Silenced one 'loss of significant digit' BCC warning.
* include/hbsetup.h
* include/hbinit.h
* source/compiler/harbour.yyc
* source/compiler/harbour.y
* HARBOUR_YYDEBUG -> HB_YYDEBUG
(do we need this? it's a synomym for YYDEBUG)
* HARBOUR_STRICT_ANSI_C -> HB_STRICT_ANSI_C
(do we need this?)
+ HB_PCRE_REGEX added.
+ HB_PCRE_REGEX_BCC added.
+ HB_POSIX_REGEX added.
- HB_MACRO_STATEMENTS
* include/hbdefs.h
- Removed WIN32_LEAN_AND_MEAN to make things more standard.
We're losing some compile time performance for Windows build,
but it's marginal. Runtime performance is not affected.
* contrib/hbgd/gdwrp.c
* Removed BCC55 hack to workaround WIN32_LEAN_AND_MEAN
problem.
* include/hbsetup.ch
* include/hbsetup.h
* source/common/hbverdsp.c
* source/vm/arrays.c
- HB_FUNC_CALLCONV - no longer used.
- HARBOUR_OBJ_GENERATION - no longer used.
+ Added profiler setting to build info.
- YACC debug setting removed frmo build info.
* Build info strict compatibility line now says 5.2e/5.3b.
* HB_EXTENSION renamed to "non-portable extension"
- HB_COMP_STRONG_TYPES - no longer used.
- HB_FIX_ACOPY_BUG
Replaced with a HB_C52_STRICT guard, but locally
overridden to always be strict (thus replicating
the bug). This means no change in functionality
for default builds. In any ways this is not a decision
to make by the Harbour users. If anyone had this
locally #defined anyway, pls shout.
; NOTE: It's nice that I'm removing unused stuff, but
my impression is that by time we have added
quite some new such settings burried in the code,
so if you're aware of any such ones, please
add them to hbsetup.h (or let's discuss which
ones to add).
* harbour/include/hbvm.h
* harbour/source/vm/hvm.c
* harbour/source/vm/extrap.c
* moved code to set exception handler from hvm.c to extrap.c
+ added void hb_vmSetExceptionHandler( void ) to set exception
filter
! fixed problem with C++ builds caused by missing extern "C"
declaration for exception handlers in extrap.c
* harbour/harbour-ce-spec
* harbour/harbour-w32-spec
* harbour/harbour.spec
* harbour/make_rpmce.sh
* harbour/make_rpmw32.sh
* harbour/make_rpm.sh
* removed --with zlib switch - now zlib will be created in all builds
* added hbzlib library
* added hbmzip library - MiniZIP wrapper to manage ZIP files
* harbour/contrib/Makefile
* added hbmzip library to default builds
* harbour/include/hbzlib.h
! changed \ in include path
! added relative paths for LCC and XCC
* harbour/bin/hb-func.sh
* updated for hbzlib library in core code and hbmzip in contrib
* harbour/utils/hbdot/Makefile
* harbour/utils/hbrun/Makefile
* added hbzlib library to linked library list
* harbour/include/hbzlib.h
* harbour/bin/hb-func.sh
* harbour/make_gcc.mak
* harbour/config/darwin/gcc.cf
* harbour/config/hpux/gcc.cf
* harbour/config/linux/gcc.cf
* harbour/config/sunos/gcc.cf
* harbour/config/bsd/gcc.cf
* harbour/utils/hbdot/Makefile
* harbour/utils/hbrun/Makefile
+ added support for compilation with HB_EXT_ZLIB macro
It forces using external (OS) ZLIB library instead of hbzlib.
Still some other make files should be updated to respect it.
* include/hbapi.h
* source/vm/hvm.c
* source/vm/cmdarg.c
+ Added support for internal switch format:
--hb:<internal switch>
//hb:<internal switch>
Both are equivalent to original CA-Cl*pper:
//<internal switch>
