* harbourr/contrib/hbqt/QPrinter.cpp
* harbourr/contrib/hbqt/QTextCursor.cpp
* harbourr/contrib/hbqt/TQPrinter.prg
* harbourr/contrib/hbqt/TQTextCursor.prg
* harbourr/contrib/hbqt/qth/QPrinter.qth
! Commented out setWinPage() and winPage() calls - Windows only.
* harbourr/contrib/hbqt/qth/QTextCursor.qth
! changed the order of insertImage() calls.
* INSTALL
+ Added information about creating Windows / Windows CE
builds on Linux. Corresponding scripts references.
! Minor correction to OS/2 on Windows cross-build example.
* contrib/hbct/Makefile
- contrib/hbct/blank.prg
+ contrib/hbct/blank.c
+ Applied patch by Pavel Tsarenko (xhb) (with some little modifications):
BLANK() function rewritten in C.
* contrib/rddads/adsfunc.c
! Reimplemented fix from Pavel Tsarenko (xhb):
Removed unnecessary OEM to ANSI convertation in SQL functions.
* harbour/contrib/hbxbp/Makefile
* harbour/contrib/hbxbp/xbp.ch
* harbour/contrib/hbxbp/xbpdialog.prg
* harbour/contrib/hbxbp/xbpgeneric.prg
* harbour/contrib/hbxbp/xbpprinter.prg
+ harbour/contrib/hbxbp/xbprtf.prg
+ Implemented XbpRtf() class - the toughest so far.
Example code lifted from Xbase++ documentaion is working as is
except for one feature. Still more work is required. I see a
possibility of lot more features in this class - later.
* harbour/contrib/hbxbp/tests/demoxbp.prg
+ Demonstrated the XbpRtf() class implementation.
Click on <Rtf> tab-page and view the results. Also examine the code.
* INSTALL
+ Added OS/2 cross-build examples for both Windows and Linux
hosts (Thanks Przemek for the Linux one).
+ Added information that DOS and OS/2 builds can be created
on other host OSes.
* config/os2/global.cf
* config/os2/install.cf
+ Added support for non-bash non-OS/2 shell for
Windows NT cross compilation.
! Fixed syntax error in one of the OS/2 specific rules.
* config/dos/dir.cf
* config/win/global.cf
* config/win/dir.cf
* config/wce/dir.cf
* Some more sync. Now os2/win/wce dir.cfs are the same,
and they're almost the same as the dos one.
* harbour/config/dos/watcom.cf
* harbour/config/win/watcom.cf
* harbour/config/os2/watcom.cf
! hacked to not revert slashes to backslashes for wcc in Linux host OS
It's not clean solution because it will have to be adopted also for
other OS2 ports
* harbour/config/dos/watcom.cf
! fixed passing parameters to wcc by environment variable
* harbour/external/sqlite3/Makefile
* set SQLITE_OS_OTHER for all DOS builds
* harbour/config/dos/watcom.cf
* harbour/config/win/watcom.cf
* modified to work with *nix like shells so it can be used in cross build
environment
Now in my Linux box I can create Windows, DOS and OS2 harbour builds
using OpenWatcom 1.8 and native Linux binaries. Final Windows executables
works correctly. DOS one doesn't and it's probably problem with
OpenWatcom Linux linker and DOS DPMI stubs - I'll check it yet.
I cannot tested generated OS2 executable files. If someone has
OS2 and can try to run it then I can send them to him for test.
* make_gnu.bat
+ Minor adjustment to support OS/2 cross-builds.
* source/rtl/gtwvt/gtwvt.c
* Minor adjustment to casting to Windows type.
* contrib/hbfimage/Makefile
- Disabled for OS/2.
* bin/postinst.cmd
+ Synced with .bat. 'copy' replaced by 'xcopy'.
* config/os2/watcom.cf
+ Added fix (from win/watcom.cf) to swap dirseps
for wcc386 as it cannot accept forward slashes.
Przemek, please adapt this for Linux, as it's
probably not needed there.
* harbour/config/os2/watcom.cf
* harbour/config/os2/dir.cf
* harbour/config/os2/install.cf
* small cleanup for builds with *nix like shells and cross builds
* harbour/contrib/rddads/Makefile
! disabld from OS2 builds - if ADS supports OS2 then please speak-up
* harbour/include/hbthread.h
* added small hack as workaround for missing OS2 declarations
* harbour/source/rtl/filesys.c
! fixed typo in OS2 builds
* harbour/source/rtl/hbproces.c
! fixed non GCC OS2 builds
* harbour/contrib/hbtpathy/tpos2.c
! fixed casting in OS2 builds
* harbour/contrib/xhb/hbserv.c
! disabled from OS2 non GCC builds
Now I can create OS2 OpenWatcom build in my Linux box using this simple
script:
#!/bin/sh
if [ -z "${WATCOM}" ]; then
export WATCOM="/opt/lng/watcom"
export INCLUDE="${WATCOM}/h:${WATCOM}/h/os2"
export PATH="${WATCOM}/binl:$PATH"
fi
export HB_ARCHITECTURE=os2
export HB_COMPILER=watcom
export HB_INSTALL_PREFIX="$(pwd)/hb-os2/watcom"
#cross build settings
# this should be fixed and we should use HB_PPGEN_PATH as direct
# reference to hbpp binaries and HB_COMP_PATH as direct reference
# to harbour binaries in all builds. Otherwise we have to create
# some hacks to create platform specific names for above binaries
# i.e. in *nixes links with .exe extension.
export HB_PPGEN_PATH="/tmp/cros-hb"
export HB_BIN_COMPILE="/tmp/cros-hb"
export HB_USER_PRGFLAGS="${HB_USER_PRGFLAGS} -D__PLATFORM__OS2"
export HB_USER_PRGFLAGS="${HB_USER_PRGFLAGS} -undef:__PLATFORM__UNIX"
export HB_USER_PRGFLAGS="${HB_USER_PRGFLAGS} -undef:__PLATFORM__LINUX"
# eliminate libraries which scans for local host header files
export HB_XBUILD=yes
./make_gnu.sh "$@"
It should work also in Windows with MSys.
This should resolve the problem with testing OS2 builds, at least
for OpenWatcom. I do not know if GCC for OS2 is available as cross
build environment.
I think that Viktor can quite easy adopt hbmk2 for OS2 cross compilation.
* source/common/hbver.c
% Deleted unnecessary casts.
* examples/hbmsql/msql.c
* Fixed const warnings.
* examples/hbmake/hbmake.hbp
- examples/hbmake/hbmakec.c
* examples/hbmake/hbmake.prg
* Using portable HB_USERLANG() instead of local Windows-only solution.
- examples/hboleold
- Deleted. It's no longer easy to fix with new const usage.
- examples/hbgf
- Deleted. We have much better GUI examples these days then these
very early ones.
* examples/httpsrv/uhttpdc.c
* Minor formatting.
* include/hbextern.ch
+ Added HB_SETCLSHANDLE().
* source/vm/evalhb.c
* source/vm/eval.c
* source/vm/dynlibhb.c
! Fixed old int <-> USHORT conversion MSVC warnings.
; TOFIX: ? There is a disagreement between different APIs
and Harbour program parts on the size and type of
number of parameters, it can be int, USHORT, ULONG,
and in some places there is also BYTE limit.
* source/common/hbver.c
* Minor formatting.
* source/vm/fm.c
! Fixed bcc C++ build errors by enabling USE_DL_PREFIX in this case.
Only this warning remain:
Warning W8008 ../../fm.c 386: Condition is always true in function dlmalloc_destroy()
* contrib/hbxbp/xbpdev.ch
+ Added SVN ID.
* source/rtl/hbinet.c
+ Added hbU16 casts to silence very old warnings.
* source/rtl/gtapiu.c
* Tweaked type and constant usage.
* include/hbgtcore.h
* include/hbapigt.h
* source/rtl/gtapi.c
* source/rtl/hbgtcore.c
* contrib/hbct/ctwin.c
* contrib/hbct/screen1.c
* contrib/hbct/ctwfunc.c
* GetClearChar/SetClearChar changed to use USHORT for char
(was int). This resolves all remaining MSVC warnings.
* _HB_GT_BASE structure adapted to latest changes, three
vars changed from USHORT to int, one renamed to reflect
int type.
+ Added var names to a few functions declarations.
% Minor optimizations.
* hb_gtScrollEx() modified to not do parameter evaluation,
just pass them to low-level.
; Przemek, please review me, I've again encountered UCHARs
in hbct / SETCLEARB().
* source/rtl/idle.c
* SHORT -> int.
* Minor formatting.
* harbour/contrib/hbxbp/Makefile
* harbour/contrib/hbxbp/xbp.ch
* harbour/contrib/hbxbp/xbpgeneric.prg
* harbour/contrib/hbxbp/xbpwindow.prg
! Cleanups ad tweaks.
+ harbour/contrib/hbxbp/xbppresspace.prg
+ Started XbpPresSpace() class - skeleton only.
+ harbour/contrib/hbxbp/xbpprintdialog.prg
+ Implemented XbpPrintDialog() class.
Mostly working but has some differences with Xbase++ - a work-in-progress.
+ harbour/contrib/hbxbp/xbpprinter.prg
+ Implemented XbpPrinter() class.
Mostly working but has some differences with Xbase++ - a work-in-progress.
Integrated with XbpPrintDialog()
* harbour/contrib/hbxbp/tests/demoxbp.prg
+ Demonstration of XbpPrintDialog() and XbpPrinter() classes.
Click on <Print Dialog> toolbar button.
/* I would request Xbase++ users to test the code under different
permutations and combinations and please report the differences.
So far I have not received any support in this direction.
*/
* examples/terminal/trm_cli.prg
* examples/terminal/trm_cli.hbp
* examples/terminal/terminal.prg
* examples/terminal/trm_srv.prg
* examples/terminal/trm_srv.hbp
* examples/terminal/terminal.hbc
* examples/terminal/trm_app.prg
* examples/terminal/trm_app.hbp
+ Changed to use portable solutions instead of GTWVG/Windows facilites.
This means that these programs should now build and run on
all OSes with MT support and they will work with any full screen
GTs. I didn't test yet on other platforms and stopping the server
is a bit messy yet due to the newly added threads.
* contrib/hbct/ctwin.c
! Silenced MSVC warning.
* harbour/include/hbgtcore.h
* harbour/source/rtl/hbgtcore.c
* changed BYTE to USHORT in few functions which expected character
for GTs with UNICODE screen buffer
* harbour/source/rtl/gtcgi/gtcgi.c
* harbour/source/rtl/gtstd/gtstd.c
* harbour/source/rtl/gttrm/gttrm.c
* updated for USHORT used as character holder
* harbour/source/rtl/gtwvt/gtwvt.c
% minor modification
* harbour/source/rtl/gtxwc/gtxwc.c
* harbour/contrib/gtwvg/gtwvg.c
* harbour/contrib/gtwvg/wvgcore.c
* harbour/contrib/gtalleg/gtalleg.c
* updated for BYTE to int as color holder modification
* harbour/contrib/hbct/screen1.c
* harbour/contrib/hbct/screen2.c
* harbour/contrib/hbct/ctwfunc.c
* harbour/contrib/hbnf/ftattr.c
! fixed casting changing values or added casting to force strict
translation
* modified to accept color and character values with wider ranges
* harbour/contrib/hbct/ctwin.c
! removed unnecessary stripping to BYTE in character and color values
* harbour/source/rtl/console.c
* pacified warning
* include/hbgtcore.h
* include/hbapigt.h
* source/rtl/gtdos/gtdos.c
* source/rtl/gtwin/gtwin.c
* source/rtl/gtxwc/gtxwc.c
* source/rtl/gtcrs/gtcrs.c
* source/rtl/gtstd/gtstd.c
* source/rtl/gttrm/gttrm.c
* source/rtl/gtcgi/gtcgi.c
* source/rtl/gtapi.c
* source/rtl/gtos2/gtos2.c
* source/rtl/hbgtcore.c
* source/rtl/gtsln/gtsln.c
* source/rtl/gtpca/gtpca.c
* source/rtl/gtwvt/gtwvt.c
* contrib/hbct/screen2.c
* contrib/hbct/ctwin.c
* contrib/hbct/screen1.c
* contrib/hbnf/ftattr.c
* Changed color representation to be 'int' in all places.
This means changes in high level and low level GT interfaces.
; QUESTION: In hbct UCHAR is used in a few places, I didn't want to
touch it.
; Please test and review.
; TODO: Clean char type usage (from int, UCHAR, BYTE, USHORT to what?)
; TODO: Clean attr type usage to be hbU8, or HB_ATTR, or else?
* include/hbapigt.h
* source/rtl/gtapi.c
! Fixed a few byAttr parameter names which should really be named bColor.
; TOFIX: ? hb_gtSetAttribute() and *SETATTRIBUTE interfaces
have a misleading name, as they're really setting
color, not attribute. (we called color as attr in
the beginning when no separate attr info was
implemented)
* harbour/source/rtl/gtwvt/gtwvt.c
% small optimization in non unicode builds - do not divide text
output and reduce number of hdc settings if the same font is
used for box and normal characters
* include/hbdefs.h
+ Added HB_RESULT. This is meant to replace current USHORT used
to store result (or "error" values).
* source/rtl/console.c
* USHORT -> int (printer coordinates).
* source/rtl/gtwin/gtwin.c
* source/rtl/gtwvt/gtwvt.h
* source/rtl/gtwvt/gtwvt.c
* Updated to use ints instead of [U]SHORTs internally.
; Please review me and correct any possible mistakes.
* contrib/xhb/xhbscr.c
% Deleted explicit casts.
* harbour/include/hbapigt.h
* harbour/source/rtl/gtapi.c
* changed last parameter of hb_gtRepChar() from USHORT uiCount to
ULONG ulCount
* removed basic parameter validation from hb_gtScroll() - such
things should be done by GT driver
* minor cleanup
* harbour/source/rtl/hbgtcore.c
% moved pbyBuffer initialization outside the loop in Rest/Save methods
* minor cleanup in casting
* harbour/source/rtl/mouse53.c
% minor optimization
* harbour/source/rtl/gtdos/gtdos.c
* changed buffer type in Mouse{Save,Restore}State methods from char *
to void *
* harbour/source/rtl/gtxwc/gtxwc.c
* harbour/source/rtl/gtcrs/gtcrs.c
! fixed casting
* harbour/source/rtl/xhelp.c
* harbour/source/rtl/console.c
% use 'int' instead of 'USHORT'
* harbour/source/rtl/saverest.c
* removed explicit casting
* harbour/source/rtl/scroll.c
* harbour/source/rtl/console.c
* harbour/contrib/hbct/screen1.c
* harbour/contrib/hbct/screen2.c
! fixed casting from 'char *' items which was wrongly changing
the results.
* harbour/contrib/hbct/screen1.c
* harbour/contrib/hbct/screen2.c
% eliminated not longer necessary intermediate variables used for
passing by reference in previous version with [U]SHORT casting
* harbour/contrib/hbnf/dispc.c
* harbour/contrib/hbnf/ftattr.c
* harbour/contrib/hbnf/ftshadow.c
* changed screen coordinates from [U]SHORT to int
* include/hbgtcore.h
* include/hbapigt.h
* include/hbapi.h
* source/rtl/scroll.c
* source/rtl/maxrow.c
* source/rtl/gtapi.c
* source/rtl/saverest.c
* source/rtl/console.c
* source/rtl/oldclear.c
* source/rtl/hbgtcore.c
* source/rtl/scrrow.c
* source/rtl/shadow.c
* source/rtl/setcurs.c
* source/rtl/mouse53.c
* source/rtl/gx.c
* source/rtl/mouseapi.c
* contrib/hbct/cursor.c
* contrib/hbct/ctwfunc.c
* contrib/hbclipsm/gauge.c
* Changed all USHORT to int in high-level GT API layer which
already had an int representation on the low-level.
Please review me. It's possible that some extra checks (or else)
need to be added since signedness has changed.
* Change mouse state buffer from char * to void *.
; TODO: Review remaining BYTE/USHORT usage for attr, color
and char. Char is sometimes represented by UCHAR,
sometimes by char and sometimes by USHORT. Color is
represented by both int and BYTE. attr is BYTE, which
can be changed to hbU8.
* include/hbgtcore.h
* source/rtl/gtapi.c
* source/rtl/hbgtcore.c
* Changed low level GT SAVE/REST method buffer parameter type
to void * from BYTE *. Some explicit casts removed.
* Few more BYTE *s to char *.
