* harbour/source/vm/garbage.c
- removed unnecessary memset()
* harbour/source/vm/hvm.c
% use startup initialized symbols: hb_symEnumIndex, hb_symEnumBase,
hb_symEnumValue instead of strcmp( pSym->szName, ... ) in hb_vmSend()
% remove unnecessary temporary item in hb_vmEnumStart() and use
hb_itemMove() instead of hb_itemCopy()+hb_itemClear()
% changed main VM loop to eliminate one comparision done on each
loop. Now HB_P_ENDBLOCK and HB_P_ENDPROC pcodes set HB_ENDPROC_REQUESTED
and the main loop can be exited only from one place. It gives some
small but noticeable speed improvement.
Ryszard please check this modification. AFAIK it should not cause
any bad side effect and IMHO the code is cleaner now.
* harbour/source/vm/itemapi.c
% removed two unnecessary hb_itemClear() before hb_itemRelease()
* include/hbapi.h
* include/hbapiitm.h
* include/hbcomp.h
* include/hberrors.h
* include/hbexpra.c
* include/hbpcode.h
* include/hbsetup.h
* source/common/expropt1.c
* source/compiler/expropta.c
* source/compiler/genc.c
* source/compiler/gencli.c
* source/compiler/harbour.c
* source/compiler/harbour.l
* source/compiler/harbour.y
* source/compiler/hbfix.c
* source/compiler/hbgenerr.c
* source/compiler/hbpcode.c
* source/macro/macroa.c
* source/vm/hvm.c
* source/vm/itemapi.c
+ tests/foreach.prg
+ added support for FOR EACH loop
FOR EACH var1 [,var255] IN expr1 [,expr255] [DESCEND]
NEXT
Note:
-expr can be a string or an array
if it is a string then assigments to the control
variable does not change the string
-after the loop the controlling variable(s) store the
value which they had before entering the loop
-the controlling variable supports the following properties
:__enumindex - the loop counter for variable
:__enumbase - the value that is being traversed
:__enumvalue - the value of variable
-see tests/foreach.prg for examples
+ added warnings in cases of duplicated loop variables
eg. FOR i:=1 TO 5
FOR i:=i TO i+5
-removed strong typing in the compiler (xHarbour too)
(reactivate it by compiling with -DHB_COMP_STRONG_TYPES)
NOTE:
The new pcodes were added - rebuild everything
* harbour/make_gnu.sh
+ harbour/make_xmingw.sh
* harbour/bin/hb-mkslib.sh
* harbour/bin/postinst.sh
* harbour/config/w32/mingw32.cf
* added support for cross compilation (Windows binaries at Linux) with
MinGW - borrowed from xHarbour Phil Krylov solution
* harbour/contrib/btree/hb_btree.c
* casting
* harbour/contrib/libct/files.c
! fixed iAttr initialization in SETFATTR()
* harbour/contrib/ole/ole2.c
! fixed names of included files
* harbour/contrib/rdd_ads/ace.h
* cover #pragma warning( error : 4706 ) by !defined( __GNUC__ )
* harbour/include/hbapi.h
* harbour/include/hbdefs.h
+ harbour/source/common/hbarch.c
* harbour/source/common/Makefile
* added functions for machine independent double and long long conversions
(my code borrowed from xHarbour)
* harbour/include/hbapifs.h
* synced file IO with xHarbour - it fixes some problems, adds some
missing functionality and long (64bit) file support for Windows.
For Linux I added it some time ago.
* harbour/include/hbcomp.h
* changed 'char cScope' to 'HB_SYMBOLSCOPE cScope'
* harbour/source/common/hbfsapi.c
! fixed some possible buffer overflow
* harbour/source/common/hbstr.c
* synced with xHarbour
* harbour/source/common/hbver.c
+ added hb_iswinnt() (borrowed from xHarbour)
* harbour/source/compiler/cmdcheck.c
+ added -undef: compiler switch (borrowed from xHarbour)
* harbour/source/compiler/gencobj.c
* cleanup
* harbour/source/pp/ppcore.c
! fixed path delimiters in included file names
* harbour/source/rtl/Makefile
+ harbour/source/rtl/fserror.c
+ added C -> OS file error trnalsations - not perfect but better then
the used hacks (borrowed from xHarbour)
* harbour/source/rtl/file.c
* use hb_fileNameConv() instead of hb_filecase() - hb_fileNameConv()
is the only one function to make file name conversions dependent on
some SETs.
* harbour/source/rtl/filesys.c
* synced file IO with xHarbour - it fixes some problems, adds some
missing functionality and long (64bit) file support for Windows.
For Linux I added it some time ago.
* harbour/source/rtl/fstemp.c
* synced with xHarbour
* harbour/source/rtl/strings.c
* use ULONG instead of size_t in hb_strnicmp declaration - we have to
decide what we should use. Using size_t or its Harbour version f.e.
HB_SIZE_T seems to be reasonable but it has to be global - redefining
single functions does not make sense and will create troubles only.
* harbour/include/hbapi.h
* harbour/include/hbapiitm.h
* harbour/include/hbapirdd.h
* harbour/include/hbdefs.h
* harbour/include/hbtypes.h
* harbour/source/rtl/empty.c
* harbour/source/vm/arrays.c
* harbour/source/vm/extend.c
* harbour/source/vm/hvm.c
* harbour/source/vm/itemapi.c
* harbour/source/vm/maindllp.c
+ added HB_TYPE to as HB_ITEM type marker.
* set HB_TYPE as UINT32
* updated to the above modifications
* cleaned some ~HB_IT_BYREF which has no effect in current VM
* harbour/harbour.spec
* harbour/make_tgz.sh
* harbour/bin/hb-func.sh
* harbour/bin/pack_src.sh
* harbour/config/global.cf
* harbour/include/hbapiitm.h
* harbour/include/hbdate.h
* harbour/include/hbextern.ch
* harbour/source/rdd/dbcmd.c
* harbour/source/rdd/dbf1.c
* harbour/source/rdd/workarea.c
* harbour/source/rdd/dbfntx/dbfntx1.c
* harbour/source/rtl/dates.c
* harbour/source/vm/itemapi.c
* updated for proper RDD synchronization with xHarbour
* harbour/source/vm/hvm.c
! fixed bug in divide by 0 substitution
* harbour/contrib/rdd_ads/ads0.prg
* harbour/contrib/rdd_ads/ads1.c
* harbour/contrib/rdd_ads/adsfunc.c
* harbour/contrib/rdd_ads/rddads.h
* synced base of ADS RDD with xHarbour - old code cannot be compiled
without it - external ADS*() functions not synced - I do not use them
and I do not want to break sth - Harbour ADS developers should check it
* harbour/harbour.spec
* harbour/make_tgz.sh
* harbour/bin/hb-func.sh
* harbour/bin/pack_src.sh
* harbour/config/global.cf
* harbour/include/hbapiitm.h
* harbour/include/hbdate.h
* harbour/include/hbextern.ch
* harbour/source/rdd/dbcmd.c
* harbour/source/rdd/dbf1.c
* harbour/source/rdd/workarea.c
* harbour/source/rdd/dbfntx/dbfntx1.c
* harbour/source/rtl/dates.c
* harbour/source/vm/hvm.c
* harbour/source/vm/itemapi.c
* updated for proper RDD synchronization with xHarbour
* harbour/contrib/rdd_ads/ads0.prg
* harbour/contrib/rdd_ads/ads1.c
* harbour/contrib/rdd_ads/adsfunc.c
* harbour/contrib/rdd_ads/rddads.h
* synced base of ADS RDD with xHarbour - old code cannot be compiled
without it - external ADS*() functions not synced - I do not use them
and I do not want to break sth - Harbour ADS developers should check it
* include/hbextern.ch
* removed reference to SDFC and DELIMC
* source/rdd/dbcmd.c
* removed definition of DBSKIPPER function (it is defined
already in rtl/browdbx.prg)
* source/rtl/disksphb.c
* hb_DiskSpace() changed to not give a runtime error in case
of any error, but set the FError() value instead.
* config/bsd/global.cf
* config/darwin/global.cf
* config/dos/global.cf
* config/dos/owatcom.cf
* config/hpux/global.cf
* config/linux/global.cf
* config/linux/owatcom.cf
* config/os2/dir.cf
* config/os2/global.cf
* config/os2/install.cf
* config/sunos/global.cf
* config/w32/global.cf
! Changed all line beginning spaces to Tabs.
* source/rtl/tget.prg
! :displat fixed (again) to always call VarGet block
to be CA-Cl*pper compatible.
* config/w32/global.cf
! Fixed to properly create the obj directories when compiling
using GCC under XP. This broke in the near past.
* source/pp/ppcore.c
! Fixed two compiler warnings.
; Please review one of them if you made the change, since
I'm not sure what the intention was, here's the code snippet:
...
lenreal = IsMacroVar( *ptri, com_or_tra ); /* THIS LINE IS NOW COMMENTED OUT */
*ptri +=1;
lenreal = getExpReal( expreal+2, ptri, FALSE, maxlenreal, FALSE );
...
* source/codepage/cdphu852.c
! HU sorting order fixed to be more compatible with CA-Cl*pper.
(Still missing some non-852 chars though)
* source/common/hbver.c
! Fixed forming NT version number (5.01.2600 -> 5.1.2600)
* source/compiler/hbusage.c
+ Copyright year updated.
* harbour/source/rtl/idle.c
* Cleaned up a platform dependent code part.
* config/dos/global.cf
*restored creation of subdirectiories under plain DOS
* include/hbapi.h
* include/hbcomp.h
* include/hbdefs.h
* include/hbvm.h
* source/compiler/genc.c
* source/compiler/harbour.c
* source/compiler/hbpcode.c
* source/vm/dynsym.c
* source/vm/hvm.c
* source/vm/macro.c
*modified creation of symbols table
*symbol of function name never share a symbol of variable
*symbol for INIT/EXIT procedure has now '$' suffix - this means
that such procedures cannot be called from user code
(Clipper compatible)
See the following code:
PROCEDURE MAIN
aaa()
RETURN
INIT PROCEDURE aaa
? "In INIT procedure", PROCNAME(0)
RETURN
STATIC PROCEDURE aaa
? "In STATIC procedure", PROCNAME(0)
RETURN
It will print:
In INIT procedure aaa$
In STATIC procedure aaa
*fixed access to static functions in a macro compiler (symbols for
static functions never goes into dynamic symbols table)
* source/pp/ppcore.c
*increased numer of nested optional clauses (max 5 currently - work
in progress)
* harbour/source/vm/memvars.c
* fixed public releasing - fix borrowed from xHarbour
Ryszard please check if it doesn't cause any bad side effects.
The following code demonstrate the problem.
===========================================
memvar x, y, z
proc main()
local l := { || x }
public x := "QWE"
private z := "asd"
Errorblock( { | oError | errHandler( oError ) } )
? eval( l )
test()
? x
? z
? eval( l )
return
proc test()
private y := "xyz"
release x, y, z
? x
? y
? z
return
func errHandler( oErr )
local varName := oErr:operation()
public &varName := oErr:description() + ": " + varName
return .t.
===========================================
* xharbour/contrib/rdd_ads/ace.h
* small modification for OpenWatcom on Linux
* xharbour/source/rdd/dbfntx/dbfntx1.c
* minor fix: use *_GOTO( pArea, 0 ) instead
of *_GOTO( pArea, pArea->ulRecCount+1000 )
+ use hb_fsSeekLarge() - for DBFs longer then 2GB
* xharbour/include/hbrddcdx.h
* xharbour/source/rdd/dbfcdx/dbfcdx1.c
+ use hb_fsSeekLarge() - for DBFs longer then 2GB
* my last xHarbour modifications:
improved indexing speed of DBFs with very big number of
records. In small DBFs less then 100'000 records the new
algorithm can even decrease the speed by ~0.5% due to static
costs but in bigger it becomes faster. For 1'000'000 records
DBFs is about 20% faster, for 10'000'000 records about twice
faster. With bigger DBFs more. Please test it. If you will
have any troubles (I hope not) then please comment out in
dbfcdx.c:
#define HB_CDX_NEW_SORT
I should done it before when I was rewriting the indexing but
sorry I was too lazy. Now I've found a while. Current cost of
key comparisons is always n*log2(n) and cannot be more improved.
Some small optimizations still can be done but the static cost
for small database will be bigger then ~0.5% as in this case
so I do not plan to do it in the nearest future.
* harbour/include/dbinfo.ch
* harbour/source/rdd/dbf1.c
+ use hb_fsSeekLarge() - for DBFs longer then 2GB
* synched with xHarbour (added DBRI_RAW*)
* do not call GOCOLD() inside CHILDSYNC() but do it in FORCEREL()
the modification based on CL5.3 NG.
Below is a note I lef in dbf1.c:
/*
* !!! The side effect of calling GOCOLD() inside CHILDSYNC() is
* evaluation of index expressions (index KEY and FOR condition)
* when the pArea is not the current one - it means that the
* used RDD has to set proper work area before eval, DBFCDX does
* but DBFNTX not yet - it should be changed.
* IMHO GOCOLD() could be safely removed from this place but I'm not
* sure it's Clipper compatible - I will have to check it, Druzus.
*/
/*
* I've checked in CL5.3 Technical Reference Guide that only
* FORCEREL() should ensure that the work area buffer is not hot
* and then call RELEVAL() - I hope it describes the CL5.3 DBF* RDDs
* behavior so I replicate it - the GOCOLD() is moved from CHILDSYNC()
* to FORCEREL(), Druzus.
*/
* harbour/source/rdd/dbf0.prg
* harbour/source/rdd/delim0.prg
* harbour/source/rdd/rddsys.prg
* harbour/source/rdd/sdf0.prg
* harbour/source/rdd/dbfntx/dbfntx0.prg
* formatting (adding EOL at EOF)
* harbour/include/hbapifs.h
* harbour/source/rtl/filesys.c
* added hb_fsLockLarge() and hb_fsSeekLarge() which uses 64bit ofsets
* harbour/source/rdd/workarea.c
* set current work area to the used one in EVALBLOCK RDD method
before code block evaluation
* harbour/source/vm/harbinit.prg
* harbour/source/vm/hvm.c
+ added hb_vmDoInitClip() which calls CLIPINIT function to initialize
ErrorBlock() and __SetHelpK()
Because on some platform the execution order of init functions
is out of Harbour control then CLIPINIT has to be called explicitly
in VM initialization process before hb_vmDoInitFunctions() and do not
depends on INIT clause.
* small modification in VM loop which gives noticeable speed-up
* xharbour/source/rtl/mod.c
! fixed mod() function to be Clipper/DBASE III compatible
(not the % operator) for combination of psitive and negative
numbers
+ harbour/include/hbfixdj.h
+ I missed to add this file in my last commit - it's borrowed from
xHarbour.
* harbour/source/rtl/filesys.c
! fixed DJGPP compilation I broke
* harbour/hbgtmk.sh
* harbour/make_rpm.sh
* harbour/harbour.spec
* added flex to dependences
* added --without gpl swich to exclude code which needs 3-rd party
GPL libs (GPM, SLANG) - necessary for people who want to create
commercial closed source application
* harbour/make_gnu.sh
* modified DJGPP detection - try to use env var first
* set HB_GT_LIB depending on platform information
* check PREFIX env var for default installation
* harbour/make_tgz.sh
* modified to work with DJGPP and 8.3 file names
* harbour/bin/hb-func.sh
* DJGPP support
* harbour/bin/pack_src.sh
* autodetect Harbour version number
* harbour/config/linux/gcc.cf
* added small comment
+ harbour/config/linux/owatcom.cf
+ OpenWatcom support on Linux
* harbour/config/w32/global.cf
* harbour/config/w32/mingw32.cf
* small modification in clean procedure
+ added Phil's modification for MINGW cross compilation in Linux and BSD
* harbour/source/compiler/gencobj.c
+ added Phil's modification for MINGW cross compilation in Linux and BSD
* harbour/include/hbsetup.h
* added OS_HAS_DRIVE_LETTER, OS_PATH_DELIMITER_STRING, OS_FILE_MASK,
OS_DRIVE_DELIMITER and HOST_OS_UNIX_COMPATIBLE
* modified for OW on Linux and MINGW cross compilation
* added snprintf macro for compilers which do not support it
* harbour/source/rtl/mod.c
! fixed bug in mod(a,b) when a/b exceeds LONG limit
* harbour/source/rtl/filesys.c
* removed HB_FS_DRIVE_LETTER - use OS_HAS_DRIVE_LETTER
* modified for OW on Linux and MINGW cross compilation
* harbour/contrib/libct/files.c
* harbour/include/hb_io.h
* harbour/include/hbdefs.h
* harbour/source/rtl/diskspac.c
* harbour/source/rtl/disksphb.c
* harbour/source/rtl/fssize.c
* harbour/source/rtl/fstemp.c
* harbour/source/rtl/hbffind.c
* harbour/source/rtl/net.c
* harbour/source/rtl/seconds.c
* harbour/source/rtl/gtcrs/Makefile
* harbour/source/rtl/gtcrs/kbdcrs.c
* harbour/source/rtl/gtpca/gtpca.c
* harbour/source/rtl/gtsln/Makefile
* harbour/source/rtl/gtsln/gtsln.c
* harbour/source/rtl/gtsln/kbsln.c
* harbour/source/rtl/gtstd/gtstd.c
* harbour/source/vm/hvm.c
* harbour/source/vm/mainstd.c
* harbour/source/vm/mainwin.c
* modified for OW on Linux and MINGW cross compilation
* harbour/source/vm/memvars.c
- removed unnecessary checking for s_globalTable == NULL in
hb_memvarValueDecRef() - in Harbour it cannot happen
* harbour/source/vm/itemapi.c
- removed snprintf macros (see hbsetup.h)
* fixed possible GPF in hb_itemCopyC()
* modified hb_itemGetPtr() to check item type
* include/hbexprb.c
* source/compiler/exproptb.c
* source/macro/macrob.c
* fixed to compile with SIMPLEX defined
Notice: Simplex version is not updated for a long time and
it doesn't support all fixes from FLEX version
* include/hbcomp.h
* include/hbexprb.c
* include/hbmacro.h
* source/compiler/exproptb.c
* source/compiler/harbour.c
* source/compiler/hbgenerr.c
* source/macro/macrob.c
* source/vm/macro.c
* compiler reports error on the following syntax now (Clipper
compatibility):
{|| ¯ovar <operator> anysymbol}
"codeblock contains both macro and declared symbol reference"
Notice, that Clipper reports error even in this case:
MEMVAR mvar
{|| &mvar,mvar}
* source/vm/memvars.c
* restored initial and expand sizes for memvars table
* source/pp/ppcore.c
* some empty lines added
