* contrib/hbwin/olecore.c
* pacified false MSVC warning. It's very seldom situation that I'm adding
code to pacify false compiler warnings anyhow in this case it cost in
practice nothing so I decided to do that.
* contrib/xhb/hboutdbg.c
* src/rtl/gtcrs/gtcrs.c
* NULL pointer casting to pacify warning
* src/pp/hbpp.c
% do not created new iLen variable but reuse the one previously declared
* src/vm/classes.c
% allocate puiMsgIdx as array of HB_USHORT items instead of HB_SYMCNT.
This modification only reduces the memory usage and has no effect on
number of accepted symbols.
* changed 'HB_SYMCNT uiHashKey' to 'HB_SIZE nHashKey' to avoid explicit
casting in the code
2023-11-11 18:38 UTC+0100 Phil Krylov (phil a t krylov.eu)
* include/hbwinuni.h
* src/vm/maindllp/dllpcode.c
* In `HB_WINAPI_GETPROCADDRESS[T]`, cast the return value to `void *`
to pacify further cast warnings, as the exact function signature is almost
never a `FARPROC` anyway.
* src/compiler/harbour.yyc
* src/macro/macro.yyc
* Added `HB_SYMBOL_UNUSED( yynerrs )` to pacify compiler warnings.
* src/rtl/replic.c
* Added a cast to a comparison to pacify compiler warning.
* src/vm/classes.c
! Fixed outdated format specifier in a trace message.
* contrib/xhb/hbcompat.ch
! fixed translations of xHarbour operators inside brackets,
thanks to Ron.
* contrib/xhb/xhb.hbx
* contrib/xhb/xcstr.prg
+ added support for timestamp translations
! fixed few wrong translations
! fixed serialization of hash items with cyclic references
in ValToPrg()
% optimized cyclic references detection in ValToPrg()
! generate dummy code which cannot be compiled for codeblocks serialized
by ValToPrg()
* contrib/xhb/xhb.ch
+ added SET TRACE ... command, thanks to Ron.
* contrib/xhb/xhberror.c
* do not show __ERRRT_BASE() and __ERRRT_SBASE() in error call stack.
* src/vm/classes.c
! added protection against possible GPF when manually created serialized object
data with super class symbols longer then HB_SYMBOL_NAME_LEN is deserialized
* src/rtl/arc4.c
! include hbarc4.h instead of arc4.h which does not contain
required for dynamic libraries export attributes in function
declaration
* src/vm/classes.c
! cleaned module symbol range checking
* src/vm/arrays.c
% optimized AClone() and hb_HClone()
* src/vm/hvmall.c
% changed the order of included files to keep alive function macros
undefined in garbage.c and fm.c
* src/vm/hvm.c
! fixed hb_vmGetRealFuncSym() to work with PRG functions written in C
and registered in HVM without their own symbol table
* src/vm/dynsym.c
; comment updated
* config/linux/clang.mk
! fixed rule for dynamic library
* src/3rd/png/Makefile
+ added -DPNG_ARM_NEON_OPT=0 to build flags
* contrib/3rd/sqlite3/sqlite3.c
* contrib/3rd/sqlite3/sqlite3.diff
! pacified warning
* contrib/gtwvg/gtwvgd.c
* contrib/gtwvg/wvgwing.c
! fixed missing break/return in case statements - please verify it.
* contrib/hbct/dattime3.c
* added #define _DEFAULT_SOURCE necessay in new Linux distors
* contrib/hblzf/3rd/liblzf/liblzf.diff
* contrib/hblzf/3rd/liblzf/lzfP.h
* do not use nested #define in #if statements - some C compilers do not
support it
* contrib/hbssl/bio.c
! tuned #if condition
* contrib/hbmisc/hbeditc.c
* simpliefied for condition and pacified warning
* contrib/hbodbc/hbodbc.hbp
* contrib/sddodbc/sddodbc.hbp
+ added check for iodbc library
* utils/hbmk2/hbmk2.prg
+ added support for clang in android builds
* include/hbdefs.h
+ added check for __BYTE_ORDER__ macro used in some new lib C
implementations
* include/hbapi.h
* include/hbdefs.h
* include/hbstack.h
* include/hbvmpub.h
* src/vm/classes.c
* src/vm/dynsym.c
* src/vm/estack.c
* src/vm/memvars.c
+ extended the size of dynamic symbol table from 65535 to 4294967295.
Adopting class code I decided to keep current algorithm of method indexes
hashing with only some minor modifications. It's very fast anyhow it may
cause noticeable (though static) quite big memory allocation for class
definitions in applications using millions of symbols and which increase
dynamic symbol table at runtime loading new classes dynamically form .hrb,
.dll, .so or other dynamic libraries supported by Harbour. It's random
and rather impossible to exploit situation in real life anyhow I cannot
exclude it so I'd like to report it in ChangeLog. The solution is very
simple, i.e. it's enough to use classic divide et impera algorithm using
symbol numbers to find method definition anyhow it will be slower then
current one and address only very seldom hypothetical situations so I
decided to not implement it. Such static memory cost begins to be
completely unimportant in the world of 64-bit architectures and extremely
big memory address space.
The modification was sponsored by TRES company.
* src/vm/estack.c
! fixed __mvClear() in MT builds - due to stupid typo GetList variable
was removed in MT programs by CLEAR MEMORY command (__mvClear())
So far noone reported it and I've found it analyzing the code before
increasing symbol table size.
* contrib/hbwin/hbolesrv.c
* updated for new size of dynamic symbol table
* *
* partial sync with the 3.4 fork codebase. These are the things
synces for the most part:
- copyright headers
- grammar/typos in comments and some readmes
- comment/whitespace/decorations
- variable scoping in C files
- DO CASE/SWITCH and some other alternate syntax usage
- minimal amount of human readable text in strings
- minor code updates
- HB_TRACE() void * casts for pointers and few other changes to
avoid C compiler warnings
- various other, minor code cleanups
- only Harbour/C code/headers were touched in src, utils, contrib,
include. No 3rd party code, no make files, and with just a few
exceptions, no 'tests' code was touched.
- certain components were not touched were 3.4 diverged too much
already, like f.e. hbmk2, hbssl, hbcurl, hbexpat
- the goal was that no actual program logic should be altered by
these changes. Except some possible minor exceptions, any such
change is probably a bug in this patch.
It's a massive patch, if you find anything broken after it, please
open an Issue with the details. Build test was done on macOS.
The goal is make it easier to see what actual code/logic was changed
in 3.4 compared to 3.2 and to make patches easier to apply in both
ways.
* *
% remove brandings and homepage [1] from copyright header. Pass 1 - using script.
[1] nobody has access to it anymore AFAIK - and it's also just
a redirect since long
! update url in copyright header
; this should make the diff between 3.4 and 3.2 easier to manage
* include/hbapicls.h
* src/vm/classes.c
* minor modification in variable name
* src/rtl/filebuf.c
* src/rtl/fscopy.c
* src/rtl/fslink.c
* src/rtl/vfile.c
* cleaned FError() setting. It should not be changed by C file IO
functions because they can be activated from RDD code so accessing
work area field could change the FError() result, i.e. this code
does not work when FError() is changed by C level IO operations:
AnyFunc( FSeek( hFile, 0, FS_END ), field->MEMOVAL, FError() )
Even simple field access may cause file IO operations in RDD code
when relations are set because records in child area are repositioned
only when necessary (when user tries to make some operations on this
area). In the past RDD code was full of constructions like:
uiSaveError = hb_fsFError();
// make some IO operations
hb_fsSetFerror( uiSaveError );
to eliminate such overhead I separated PRG level FError() from
C level IOError(). Please keep it.
* ChangeLog.txt
! few typos
* src/vm/classes.c
! fixed __objGetIVars() to correctly serialize objects without any own
and inherited instance variables
* contrib/hbhpdf/3rd/libhpdf/hpdfimac.c
! include string.h
* contrib/hbhpdf/3rd/libhpdf/libhpdf.dif
* rediffed
* include/hbserial.ch
* src/rtl/itemseri.c
+ added HB_SERIALIZE_IGNOREREF flag.
This flag fully disables logic used to detect multireferences to the
same complex (sub)items like arrays or hashes. It increses the speed
of serialization but serialized data does not contain any information
about refences, i.e. aVal[ 1 ] and aVal[ 2 ] in code below:
aSub := { 1, 2, 3 }
aVal := { aSub, aSub }
are serialized as separated arrays. Additionally items with cyclic
references like:
aSub[ 2 ] := aSub
cannot be serialized at all with HB_SERIALIZE_IGNOREREF flag because
it will create infinite serialization loop and crash with out of
memory message.
* src/rtl/itemseri.c
% rewritten algorithm used to detect cyclic and multi references in
serialized items. The original algorithm has very high overhead when
huge arrays were serialized, i.e. serialization of array with 1'000'000
of subarrays needed about 30 minutes on my i5@3.30GHz. Now it needs
less then a second and this time is only a little bit bigger then
used by serialization with HB_SERIALIZE_IGNOREREF flag.
This modification improve performance also in other code using Harbour
serialization mechanism like I18N files, HBNETIO, GTNET, ... when large
arrays or hashes are saved or transmitted.
* include/hbapi.h
* include/hbapicls.h
* src/vm/arrays.c
* src/vm/classes.c
* src/vm/hashes.c
* src/vm/itemapi.c
* replaced algorithm used to detect cyclic and multi references in
array and hash clone operations with new one similar to current
item serial code. The speed improvement for very large arrays is
the same as in case of serialization code.
* src/rtl/gtsln/mousesln.c
! typo in while loop - synced with Viktor's branch
* include/hbapifs.h
* src/rtl/filebuf.c
* src/rtl/filebufd.c
* contrib/hbcomio/comio.c
* contrib/hbmemio/memio.c
* contrib/hbnetio/netiocli.c
* contrib/hbtcpio/tcpio.c
* changed internal FILE IO API. Now all functions which do
not operate on PHB_FILE receive pointer to function table
(PHB_FILE_FUNCS) used by given redirector.
* src/rtl/vfile.c
* open wirtual files in READWRITE mode by default
* translate FO_* attributes to FXO_* ones
* src/vm/classes.c
! fixed typo in RTE message
* src/vm/garbage.c
* use different error code (1301/1302) for destructor RTE
* src/rtl/gtxwc/gtxwc.c
* src/rtl/gtxwc/gtxwc.h
+ added support for HB_GTI_FONTWEIGHT with HB_GTI_FONTW_* actions
* contrib/hbpgsql/rddcopy.c
! fixed memory leak
% encapsulate PQ calls in hb_vmUnlock()/hb_vmLock() to not stop
other threads when GC is activated
* src/compiler/complex.c
+ added support for AS HASH, POINTER, SYMBOL, TIMESTAMP
+ added support for AS BLOCK as synonym of AS CODEBLOCK
+ added support for AS DATETIME as synonym of AS TIMESTAMP
* src/vm/classes.c
+ added support for "DATETIME" type uses as synonym of "TIMESTAMP"
* contrib/hbct/dattime2.c
! fixed potential GPF trap due to wrong protection condition
and change internal variable name (synced with Viktor's branch)
* src/vm/classes.c
! fixed typo in alternative code - thanks to Lorenzo for the info
* contrib/hbhpdf/3rd/libhpdf/*
* updated HARUPDF library 2.3.0RC2 -> 2.3.0RC3
(the same version is used in Viktor's branch)
* .gitignore
! removed executable file permission
* package/mpkg_src.sh
! added executable file permission
* contrib/hbct/screen2.c
% use STR API for parameters instead of using local conversions
* contrib/hbct/token1.c
% small simplification
* contrib/hbnetio/netiosrv.c
+ respect timeout parameter also in send operation
* contrib/make.hb
! moved project name normalization and directory verification to
AddProject() function. It fixes HB_BUILD_ADDONS envvar functionality.
* doc/xhb-diff.txt
+ added new paragraph: DECLARATION AND INITIALIZATION OF VARIABLES
* src/rtl/hbcom.c
* added 3-rd parameter to TIOCEXCL and TIOCNXCL ioctl() codes
to pacify valgrind warnings
* include/hbexpra.c
! fixed compilation with HB_USE_ENUM_FUNCTIONS macro
* include/hbapicls.h
* include/hbcompdf.h
* include/hbexpra.c
* src/compiler/harbour.y
* src/vm/classes.c
+ added support for :__enumIsFirst() iterator message. It's opposite
to recently added :__enumIsLast()
+ added support for overloading :__enumIsFirst() and :__enumIsLast()
functionality in custom FOR EACH implementations
* include/hbcompdf.h
* include/hbexpra.c
* src/compiler/harbour.y
+ added support for reverting :__enumIsFirst() and :__enumIsLast()
messages in descendant FOR EACH loops. It's disabled now by 2 #if 0
but I think it should be discussed. Should we keep it enable it?
The answer is not trivial when FOR EACH is used to iterate some
objects. In general such names are confusing.
* src/compiler/harbour.yyc
* src/compiler/harbour.yyh
* regenerated using bison 2.5
* tests/foreach.prg
+ added :__enumIsFirst() to test code
* tests/foreach2.prg
! typo in comment
* utils/hbmk2/hbmk2.es_PE.po
* utils/hbmk2/hbmk2.hu_HU.po
* utils/hbmk2/hbmk2.pt_BR.po
! added standard .po meta headers (manually for now)
* include/hbexpra.c
* src/compiler/harbour.y
* src/compiler/harbour.yyc
* src/vm/classes.c
* tests/foreach.prg
+ added :__enumIsLast() enum method. It returns .T. for
last relement, .F. else. It's useful to replace conditions
like: 'item:__enumIndex() < Len( item:__enumBase() )'
* tests/lang2po.hb
+ added standard .po meta header
; use __enumIsLast()
* contrib/hbnetio/utils/hbnetio/_console.prg
! fixed wrongly initialized management plugin
(missed to update it when adapted the plugin
for hbrun last year)
* contrib/hbnetio/utils/hbnetio/netiomgm.hb
! fixed to show proper error instead of RTE, if RPC calls
fail to return expect values
* contrib/hbnetio/utils/hbnetio/hbnetio.prg
* formatting
* (all files)
* stripped svn header
* minor cleanups
; use following command to find out the history of files:
git log
git log --follow
git blame
git annotate
