* src/vm/extrap.c
* contrib/hbwin/win_prn3.c
+ Added TOFIXes for unsafe function usage:
IsBadWritePtr(), IsBadReadPtr(), lstrlen(), lstrcpy(), lstrcat()
; QUESTION: Any idea how to fix that? win_prn3.c can
probably be fixed with some good general coding
idea.
* contrib/hbwin/wce_smsc.c
! Fixed to not use unsafe CRTL functions.
* src/rtl/oemansi.c
! Rewritten HB_OEMTOANSI() and HB_ANSITOOEM() to not use
Windows API functions marked as unsafe. New version is a bit
slower but won't mark Harbour apps as unsafe in an audit.
; Please test and review.
* utils/hbmk2/hbmk2.prg
* config/wce/msvcarm.mk
* config/win/msvc.mk
% Deleted -Gs MSVC option. Testing with MSVC 2008 I've found
this have no effect on x86 builds, and it makes x64 builds
slighly less efficient by forcing stack checks in each function
call. This seems to contradict MSDN, which doesn't suggest such
difference between x86 and x64:
http://msdn.microsoft.com/en-us/library/9598wk25.aspx
% Deleted -GS- MSVC option used for wce targets for MSVC >= 8.00.
'-GS-' will disable stack cookies (on by default), thus trading
app size/speed for security. Maybe this is preferred for wce
users, but in Harbour I'd rather opt to pass this decision to users
for all targets.
Enable with: HB_USER_CFLAGS=-GS- / -cflag={allmsvc}-GS-
MSDN:
http://msdn.microsoft.com/en-us/library/8dbf701c.aspx
* INSTALL
* Updated valgrind example.
* contrib/hbnetio/utils/netiosrv.hbp
* utils/hbformat/hbformat.hbp
* utils/hbmk2/hbmk2.hbp
* utils/hbi18n/hbi18n.hbp
* utils/hbtest/hbtest.hbp
* utils/hbrun/hbrun.hbp
* config/wce/msvcarm.mk
* config/win/msvc.mk
+ Added MSVC linker switches to enhance security by
enabling NX compatibility and ASLR.
Especially important for server apps like netiosrv and hbrun.
; TODO: Enable the same for MinGW.
; TOCHECK: Please test this with old MSVC versions and MSVCARM,
maybe we will need to disable them in some cases.
; QUESTION: I'm unsure of -fixed:no MSVC option. Do we need it
to increase security?
* config/wce/msvcarm.mk
+ Changed to use link.exe for linking instead of cl[arm].exe.
This now syncs it with rest of Harbour.
* harbour/contrib/hbnetio/netiocli.c
! use hb_vmAtInit() to initialize NETIO - looks that winsock
initialization in startup code does not work - please test
* harbour/src/rtl/hbznet.c
* harbour/include/hbznet.h
* added new optional parameter to hb_znetWrite() function which allow
to catch result of last hb_socketSend() operation
* removed automatic flushing from hb_znetWrite() function,
it should help in implementing more efficient C code using
hb_znet*() functions
* harbour/src/rtl/hbinet.c
* flush HB_ZNETSTREAM output after write operation
% small optimization in internal inet error setting
* harbour/include/hbvmpub.h
* minor formatting
* harbour/contrib/hbnetio/netio.h
* harbour/contrib/hbnetio/netiosrv.c
+ added new PRG function:
NETIO_RPCFUNC( <pConnectionSocket> [, <sFuncSym> ] ) -> NIL
which allows to set user filter for RPC calls. <sFuncSym> is executed
on each RPC request instead of requested function and receives as
first parameter requested function symbol then function parameters.
A simple wrapper may look like:
static function rpc_filter( sFunc, ... )
? "DO", sFunc:name, "WITH", ...
return sFunc:exec( ... )
and can be activated by:
netio_rpcfunc( pConnectionSocket, @rpc_filter() )
When NETIO_RPCFUNC() is called without valid function symbol then
existing user RPC filter is cleared. Please remember that setting
user RPC filter does not automatically enable RPC support for
given connection socket. RPC support has to be enabled by NETIO_RPC()
function or by 4-th parameter of NETIO_LISTEN() socket.
* contrib/hbide/ideactions.prg
* contrib/hbide/ideprojmanager.prg
* contrib/hbide/idestylesheets.prg
! Fixed few more artifacts.
! Recent Files and Projects Menu was growing crazy,
was a result of not deleting the menu item in XbpMenu() class.
More attention is required yet.
! Fixed Viktor's reported bug.
* contrib/hbxbp/xbpmenubar.prg
! Fixed a very bad type, copy/paste syndrome.
* contrib/hbide/projects/hbide.hbi
! Deleted hard-coded -o directive, no longer needed.
! Project paths are now shown without meta-dat.
* contrib/hbide/resources/projectproperties.ui
! Changed components background behavior not to use gradients.
Now it is plain vanila white.
* contrib/hbide/hbide.prg
* contrib/hbide/idedocks.prg
* contrib/hbide/ideeditor.prg
* contrib/hbide/idemisc.prg
* contrib/hbide/ideobject.prg
* contrib/hbide/ideprojmanager.prg
* contrib/hbide/idesaveload.prg
* contrib/hbide/idestylesheets.prg
! Another round of reforms. And I am almost done with them. But wait...
! Started implementation of .env concept. What was missing from
hbIDE => hbMK2 were the shell commands minimum necessary to
let hbMK2 engine take control of. Now it can be done via
QProcess:setEnvironment() call. I have left under the commented
code in ideprojmanager.prg:
/* Mechanism to supply environment variables to called process */
/* I do not know nixes but assume that Qt must be issueing proper */
/* shell command for the target OS to set them. */
/* If I am not wrong, HBMK2 can have these variables alread set */
/* and hence developer can choose any compiler of his choice. */
/* */
/* Actually, this was the intension in hbIDE.env I commited in IDE root */
I am been able to compile/link/execute hbIDE.exe after setting ( WinXP )
qListSets := QStringList():new()
qListSets:append( "HB_WITH_QT=c:\qt\4.5.3\lib" )
::qProcess:setEnvironment( qListSets )
just before triggering the process.
Your input is desired as I know nothing about make system.
! Fixed many artifacts difficult to jot-down.
+ Implemented not to load any source until brought forward for view.
This has increased the loading speed extremely fast and always
consistent in timing. This feature was hanging on my drawing board
since begining.
This effectively means that sessions with any number of source tabs
will not pose speed panelty.
TODO: Syntax highlighting is a slow process, to speed-up.
* contrib/hbwin/win_err.c
* contrib/hbwin/hbwin.h
* contrib/hbwin/wapi_winbase_mutex.c
* Changed low-level hbwin_SetLastError() to accept lasterror
as parameter.
* contrib/hbwin/wapi_winbase.c
* contrib/hbwin/wapi_winuser.c
+ Added internal handling of GetLastError(). This fixes lost
GetLastError() values with apps linked with MT HVM.
[ I didn't review wapi_commctrl.c, but some lazy checking
on MSDN showed that these are generally not setting GetLastError(). ]
* WAPI_GETSCROLLRANGE() now set ref params even in case of failure.
; TODO: Do some final renaming and rearrangement on the low-level.
* contrib/hbwin/wapi_winbase_mutex.c
* Formatting.
* mpkg_tgz.sh
! Fixed to detect case when tar is mapped to bsdtar
(such is the case on Snow Leopard). Use slightly modified
tar command in this case.
Please review.
; TOFIX?: Unpacking stub script always uses 'tar', maybe this
needs to be fixed, although current command should
work on Snow Leopard.
- Deleted comment no longer valid.
* utils/hbmk2/hbmk2.prg
! Fixed C++ memory manager override to handle zero size memory
request and added protection when free operations are called
with NULL. Thanks Istvan Bisz and Przemek for input.
* harbour/src/rtl/hbznet.c
! do not use DEF_MEM_LEVEL to avoid potential problems when <zutil.h>
is not available
* harbour/contrib/hbnetio/netio.h
* harbour/contrib/hbnetio/netiocli.c
* harbour/contrib/hbnetio/netiosrv.c
+ implemented RPC in HBNETIO protocol
+ added the following client functions:
check if function/procedure exists on the server side:
NETIO_PROCEXISTS( <cProcName> ) -> <lExists>
execute function/procedure on server the side,
do not wait for confirmation:
NETIO_PROCEXEC( <cProcName> [, <params,...>] ) -> <lSent>
execute function/procedure on the server side and wait for
confirmation:
NETIO_PROCEXECW( <cProcName> [, <params,...>] ) -> <lExecuted>
execute function on the server side and wait for its return value:
NETIO_FUNCEXEC( <cFuncName> [, <params,...>] ) -> <xFuncRetVal>
All above functions use default connection set by NETIO_CONNECT()
for RPCs but it's also possible to specify server address and port
in <cProcName>/<cFuncName> just like in <cFileName> parameter in RDD
functions, i.e.:
NETIO_PROCEXISTS( "192.168.0.1:10005:MYFUNC" )
+ added new server side functions to enable/disable/check RPC support:
NETIO_RPC( <pListenSocket> | <pConnectionSocket> [, <lEnable> ] )
-> <lEnabled>
if RPC is enabled for listen socket then connection sockets inherit
this setting.
+ added 4-th parameter <lRPC> to NETIO_LISTEN() function. .T. enable
RPC support in returned listen socket which is later inherited by
connection sockets
* changed protocol version ID - current NETIO clients and servers
cannot be used with old code
* harbour/contrib/hbnetio/utils/netiosrv.prg
* added option to enable RPC support in NETIO server
If you want to make some test then you can execute netiosrv with
non empty 4-th parameter as server, i.e.:
./netiosrv "" "" "" 1
and try this code as client:
proc main()
// pass server address to netio_connect() for non localhost tests
? "NETIO_CONNECT():", netio_connect()
?
? "DATE() function is supported:", netio_procexists( "DATE" )
? "QOUT() function is supported:", netio_procexists( "DATE" )
? "HB_DATETIME() function is supported:", ;
netio_procexists( "HB_DATETIME" )
?
? netio_procexec( "QOUT", repl( "=", 50 ) )
? netio_procexec( "QOUT", "This is RPC TEST", date(), hb_datetime() )
? netio_procexecw( "QOUT", repl( "=", 50 ) )
?
? "SERVER DATE:", netio_funcexec( "DATE" )
? "SERVER TIME:", netio_funcexec( "TIME" )
? "SERVER DATETIME:", netio_funcexec( "HB_DATETIME" )
?
? netio_funcexec( "upper", "hello world !!!" )
return
Please remember that only functions linked with server are available.
If you want to enabled all core functions in netiosrv then please
uncomment this line in netiosrv.prg:
REQUEST __HB_EXTERN__
Have a fun with a new toy. I hope that many Harbour user will find it
very interesting. Please only be careful !!!. This feature allows to
execute remotely _ANY_ code on the server side. _NEVER_ leave open ports
with RPC support for untrusted access.
* contrib/hbide/hbide.prg
* contrib/hbide/idedocks.prg
* contrib/hbide/idemisc.prg
* contrib/hbide/ideprojmanager.prg
! Many artifacts corrected.
! Build process is now very satisfying if selected an
option with right-click on project node in Project Tree window.
! Diabled gradients in any window. Is it ok to fine tune
the windows with plain colors ?
* harbour/src/rtl/gtwin/gtwin.c
+ added support for HB_GTI_ISUNICODE
* harbour/include/hbapifs.h
* added missing 'extern' in some declarations
* harbour/include/hbzlib.ch
+ added HB_ZLIB_STRATEGY_* constants
+ added HB_ZLIB_COMPRESSION_DISABLE
* harbour/include/Makefile
+ harbour/include/hbznet.h
* harbour/include/hbextern.ch
* harbour/src/rtl/Makefile
* harbour/src/rtl/hbinet.c
+ harbour/src/rtl/hbznet.c
+ added support for ZLIB compression in stream sockets.
+ added .prg function:
HB_INETCOMPRESS( <pSocket>, [<nCompressionLevel>], [<nStrategy>] )
which enables ZLIB compression for given HB_INET*() socket.
<pSocket> is a socket created by one of HB_INET*() functions
<nCompressionLevel> is compression factor between 1 (fastest) and
9 (best) (see HB_ZLIB_COMPRESSION_*)
0 (none) disable compression on output data
but decompression is still working.
<nStrategy> is used to tune compression algorithm,
see HB_ZLIB_STRATEGY_*
The compression must be enabled on both connection sides, i.e.
on the server side:
conn := hb_inetAccept( sock )
hb_inetCompress( conn )
and on the client side:
sock := hb_inetConnect( cServer, nPort )
hb_inetCompress( sock )
in the same moment but it's not necessary to enable it at the
beginning of connection. It can be done later, i.e. when both
sides agree to enable connection using some custom protocol.
The compression has effect only on stream connections, i.e.
TCP and it's ignored in datagram connections like UDP.
This function can be executed more then once changing the compression
parameters but it causes that all data in readahead decompression
buffer is discarded. When called with HB_ZLIB_COMPRESSION_DISABLE
as <nCompressionLevel> then support for stream compression is removed
and sockets works again in raw mode.
The compression level and strategy do not have to be the same on both
connection sides. Each side can chose the best settings for data it's
going to send.
This code was written in a way which allows to easy implement
alternative compression methods or other extensions like encryption
in existing HB_INET*() sockets also by non core code. The public C
functions declared in hbznet.h allows to use this extension with raw
harbour sockets two.
* contrib/hbxbp/tests/demoxbp.prg
! Changed to HBQT_SET_RELEASE_METHOD( HBQT_RELEASE_WITH_DELETE )
* contrib/hbide/hbide.prg
* contrib/hbide/ideactions.prg
* contrib/hbide/idedocks.prg
* contrib/hbide/ideeditor.prg
* contrib/hbide/idemisc.prg
* contrib/hbide/ideprojmanager.prg
! Another round of reforms.
Vailton, I had to delete some of your code causing the recursions.
Please note that never ever you should call the methods like
a->b->a. Please check the code and its usability as per
your line-of-thinking. More refinements are under-way.
* INSTALL
+ Added new section: 7. DEBUGGING OPTIONS
So far covered: tracing, memory statistics, valgrind and codeguard.
I'd like to ask our experts to review and if needed extend
these informations (send me suggestions and I'll update it).
* bin/postinst.bat
+ Enabling -debug option on shared tool hbmk2 builds if HB_BUILD_DEBUG=yes.
* contrib/hbide/ideactions.prg
! Typo.
* contrib/hbide/projects/hbide.hbi
! Deleted windows and local environment specific hbmk2
options, plus a few ones which are automatically pulled anyway
via hbqt.hbc (which is in turn pulled by hbxbp.hbc).
Please set HB_WITH_QT to make it find the QT libs. This is
the universal method.
* contrib/hbqt/detect.mk
+ Will now verify if HB_QT_MOC_BIN is pointing to an existing filename.
If not, an error message will be shown, trying to make it clear how
to set it up properly. Many users like to set the dir only, which
makes the build process fail.
* contrib/hbide/hbide.prg
+ Switched to HBQT_RELEASE_WITH_DELETE as suggested by Istvan Bisz.
* contrib/hbqt/generator/hbqtgen.prg
* Minor cosmetic.
* utils/hbmk2/hbmk2.prg
+ Added disctinctive static text which is always linked in
when C++ mm override is active. This makes it easy to
verify if C++ mm override is activated in an executable.
* ChangeLog
! Fixed messed text in latest hbmk2 changes.
2010-01-05 18:48 UTC+0100 Viktor Szakats (harbour.01 syenar.hu)
* utils/hbmk2/hbmk2.pt_BR.po
* utils/hbmk2/hbmk2.hu_HU.po
* utils/hbmk2/hbmk2.prg
+ Added support for C++ input files. Now C++ files will be compiled
in a separate pass where native C++ compiler is used, which
means that C++ compilation mode is now enforced by using .cpp
extension. This also means that it's now possible to build
mixed mode projects with both C and C++ input files.
In general, this feature makes -cpp option unnecessary, its
only purpose now is to override default mode.
+ Added option to override standard C++ memory management functions
with Harbour ones.
Enable with option: -hbcppmm (cmdline) or hbcppmm=yes (in .hbc files)
NOTE1: Suggestions for better names are welcome.
NOTE2: hbide will now crash with mingw if this option is enabled.
Any idea why?
; Please test.
* utils/hbmk2/hbmk2.pt_BR.po
* utils/hbmk2/hbmk2.hu_HU.po
* utils/hbmk2/hbmk2.prg
+ Added support for C++ input files. Now C++ files will be compiled
in a separate pass where native C++ compiler is used, which
means that C++ compilation mode is now enforced by using .cpp
extension. This also means that it's now possible to build
mixed mode projects with both C and C++ input files.
Enable with option: -hbcppmm (cmdline) or hbcppmm=yes (in .hbc files)
NOTE: Suggestions for better names are welcome.
In general, this feature makes -cpp option unnecessary, its
only purpose now is to override default mode.
; Please test.
* utils/hbmk2/hbmk2.prg
- Deleted support for link-related macros in C compilation phase,
these weren't used anymore inside current hbmk2, so it seems safe,
but it's possible nevertheless that it will cause regressions in
some exotic use cases.
* utils/hbmk2/hbmk2.prg
! Fixed parsing -warn option on the commandline and .hbm/.hbp files.
[TOMERGE 2.0]
* contrib/hbqt/generator/hbmk.hbm
! Deleted xhb lib. It's not required and was probably added for
"outdebug" stuff.
* contrib/hbqt/hbqt.hbc
* contrib/hbqt/hbqts.hbc
* contrib/gtqtc/gtqtcs.hbc
* contrib/gtqtc/gtqtc.hbc
+ Added hbcppmm=yes option. This will force C++ memory allocation
functions to be overridden by Harbour mm functions.
* src/vm/fm.c
- Deleted C++ memory allocation functions. This feature now
lives inside hbmk2, and it now doesn't require whole Harbour
to be built in C++ mode.
* contrib/hbqt/*
* Regenerated.
* contrib/hbct/dummy.c
+ Added dummy stubs for all CT functions which are not yet
implemented in Harbour. This file is intentionally not
built, because a link-time error is better than a runtime
one, anyhow we could convert them to real dummies emulating
error behavior and add them to build process.
* contrib/hbqt/generator/hbqtgen.prg
% Optimization to parameter list ptr conversion loop.
; TODO: Regenerate QT sources.
* harbour/contrib/hbct/video.c
% use hb_gtColorToN() function instead of own code to decode Clipper
color in VGAPALETTE()
! added protection against possible transfer buffer overflow in
SETFONT() function
! return -2 when SETFONT() is not implemented
TODO: clean these functions, add support for other DOS compilers,
add some missing functions and redirect few of them like
VGAPALETTE() to hb_gtInfo() in non DOS builds
* harbour/src/rtl/hbgtcore.c
! fixed missing ';' => CHR( 13 ) translation in __KEYBOARD() function
[TOMERGE 2.0]
* contrib/hbxbp/xbpmenubar.prg
* contrib/hbxbp/xbptoolbar.prg
+ Implemented to accept QAction() as menu constructor.
Before only cCaption or XbpMenu() were accepted.
* contrib/hbide/hbide.ch
* contrib/hbide/hbide.prg
* contrib/hbide/ideactions.prg
* contrib/hbide/idedocks.prg
* contrib/hbide/idemisc.prg
* contrib/hbide/ideobject.prg
* contrib/hbide/ideprojmanager.prg
! Next round of reforms.
+ Implemented IdeActions() class which forms the basis
of IDE's menus and toolbars. Now we have greater control
over the uniformity of different actions hooked into
different interface elements, i.e., menu option, toolbar
option or popup option. The same visual experience is
presented in all the three flavours.
It also implies that we can switch on/off any action
only at one place. For example, if you issue this command:
::oAC:getAction( "Save" ):setEnabled( .f. )
then the option will be grayed in any flavour of user
interaction, viz., menu, toolbar, or context menu.
The code is much cleaner and paves the way to be reloaded
with various means.
* contrib/hbide/hbide.prg
* contrib/hbide/ideactions.prg
* contrib/hbide/ideeditor.prg
* contrib/hbide/idemisc.prg
* contrib/hbide/ideobject.prg
* contrib/hbide/ideprojmanager.prg
* contrib/hbide/idesaveload.prg
! Next round of reforms, stay patient.
! Implemented to hold last working project.
! Fixed to not generate error if an open project is tried
to reopen.
! Many small artifacts correctd.
* harbour/ChangeLog
! fixed typo in my recent ChangeLog entry
* harbour/src/vm/fm.c
% minor simplification
* harbour/src/common/hbtrace.c
! fixed possible recursive call reported by Istvan
* contrib/hbgd/hbgd.hbc
+ examples/httpsrv/modulesg.hbp
* examples/httpsrv/uhttpdgd.hbp
* examples/httpsrv/readme.txt
* Merged gd linking logic from uhttpdgd.hbp to hbgd.hbc.
(contains little hack, too) (untested)
+ Added .hbp file to build GD enabled modules.
* Modified readme to reflect above change.
+ Changed .hbp files to use hbgd.hbc and hbct.hbc to
build with hbgd and hbct libs. This is much better
method, since hbmk2 will locate them automatically
plus set header path as well.
* contrib/hbide/idethemes.prg
* contrib/hbide/ideeditor.prg
* contrib/hbide/hbide.prg
* contrib/hbide/ideactions.prg
* contrib/hbide/ideprojmanager.prg
! Reapplied my File() -> hb_FileExists() pacth.
Please note that it's not my favourite pass-time to
find out which of my changes were overwritten and such
practice pretty much defeats the purpose of version
tracking systems like SVN altogther. It's committers
job to properly resolve conflicts, unless there is
as advance request for commit-stop.
Since it's also not very efficient for me to refix all
future File() usages to hb_FileExists(), it would
be much more beneficiary for the whole project if coding
practices like that would be adopted by other developers
and taken care of in future updates. Thanks in advance.
* harbour/src/rtl/gtxwc/gtxwc.c
! cleaned a little bit hack with euro sign support
+ added optional support for input respecting locale settings
To enable it it's necessary to recompile GTXWC with HB_XWC_USE_LOCALE
macro. It can be done globally using
export UH_USER_CFLAGS=-DHB_XWC_USE_LOCALE
and recompiling Harbour core code.
When enabled GTXWC initialize locale at startup (LC_CTYPE only) what
enable national keyboard layout support in default input context
(XLookupString()). Current Harbour core code does not use any
<ctype.h> functions so enabling locale should not effect user
national settings in HVM.
Please test.
* contrib/hbide/hbide.hbp
* contrib/hbide/hbide.prg
* contrib/hbide/ideactions.prg
* contrib/hbide/idedocks.prg
* contrib/hbide/ideeditor.prg
* contrib/hbide/idefindreplace.prg
* contrib/hbide/idemisc.prg
* contrib/hbide/ideobject.prg
* contrib/hbide/ideprojmanager.prg
* contrib/hbide/idesaveload.prg
* contrib/hbide/idethemes.prg
* contrib/hbide/projects/hbide.hbi
+ contrib/hbide/resources/insert-datetime.png
+ contrib/hbide/resources/insert-external-file.png
+ contrib/hbide/resources/insert-procname.png
+ contrib/hbide/resources/readonly.png
* contrib/hbide/resources/tabreadonly.png
* contrib/hbxbp/xbpqtuiloader.prg
A lot of changes are in effect including the patch
sent by Vailton. The code is in the middle of restructing
phase and may take several days to finish. Till the I request
you all to bear with me and may expect some bumps too.
Please do not update this code until I finish preliminaries.
To sumup this commit includes:
! Thourough reclassification of classes and separation of
code fragments in a way that not two developers collide
with each other. One may takeup one segment and the other another.
+ Used hbide_* namespace and have changed almost every function.
+ Implemented, though not in a way I wanted, building a project
via Qt's process classes. You will see a refreshing difference.
+ A major breakthrough in XbpQtUiLoader() class implementation.
Now you can access the child widgets of a .ui component per
this syntax:
oUI:qObj[ "checkMatchCase" ]:setChecked( .t. )
=>
oUI:q_checkMatchCase:seChecked( .t. )
^^
just prefix object name with "q_" ( Q + Underscore ).
This greatly simmplifies coding and allows to exploit child
widgets as real-time objects.
* contrib/hbqt/hbqt_hbevents.cpp
* contrib/hbqt/hbqt_hbslots.cpp
- Deleted old method of slots/events handling.
* contrib/hbide/idethemes.prg
* contrib/hbide/ideeditor.prg
* contrib/hbide/hbide.prg
* contrib/hbide/idemisc.prg
* contrib/hbide/idesaveload.prg
% Using hb_FileExists() instead of File().
! Fixed to not force lowercase on filenames.
* contrib/hbqt/qth/QApplication.qth
* Minor optimizations.
% One static var deleted.
; TOFIX/TOCHECK: argv/argc usage.
* contrib/hbqt/qtgui/QApplication.cpp
* contrib/hbqt/hbqt_garbage.h
* Regenerated.
* contrib/hbxbp/xbpgeneric.prg
- Deleted unused statics.
; TOFIX: HBQT has an irregularity which I couldn't find a
reason yet. It maintains a pointer to QApplication
in a static variable on C level, plus every HBQT
app maintains a copy of the QApp object in a .prg
level STATIC var. Both are inited using C/prg
level init/exit functions.
