Assume maintainership.
Changes since version 21:
V27
- Fixed REPOSITION-FILE FILE-SIZE and FILE-POSITION.
They used to use single precision offset. Now use double as specified.
- Delete object directories in Makefile clean.
- Fixed "Issue 4: Filehandle remains locked upon INCLUDE error".
http://code.google.com/p/pforth/issues/detail?id=4&can=1
- Fixed scrambled HISTORY on 64-bit systems. Was using CELL+ but really needed 4 +.
- Fixed floating point input. Now accepts "1E" as 1.0. Was Issue #2.
- Fixed lots of warning and made code compatible with C89 and ANSI. Uses -pedantic.
- Use fseek and ftell on WIN32 instead of fseeko and ftello.
- Makefile is now more standard. Builds in same dir as Makefile. Uses CFLAGS etc.
- Add support for console IO with _WATCOMC_
- Internal CStringToForth and ForthStringToC now take a destination size for safety.
- Run units tests for CStringToForth and ForthStringToC if PF_UNIT_TESTS is defined.
V26 5/20/2010
- 64-bit support for M* UM/MOD etc by Aleksej Saushev. Thanks Aleksej!
V25 5/19/2010
- Added 64-bit CELL support contributed by Aleksej Saushev. Thanks Aleksej!
- Added "-x c" to Makefile CCOPTS to prevent confusion with C++
- Allow space after -d command line option.
- Restore normal tty mode if pForth dictionary loading fails.
V24 2/20/09
- Fixed Posix IO on Mac. ?TERMINAL was always returning true.
- ACCCEPT now emits a space at end of line before output.
- Fixed RESIZE because it was returning the wrong address.
V23 8/4/2008
- Removed -v option from mkdir in build/unix/Makefile. It was not supported on FreeBSD.
Thank you Alexsej Saushev for reporting this.
V23 7/20/2008
- Reorganized for Google Code project.
V22 (unreleased)
- Added command line history and cursor control words.
- Sped up UM* and M* by a factor of 3. Thanks to Steve Green for suggested algorithm.
- Modified ACCEPT so that a line at the end of a file that does NOT have a line
terminator will now be processed.
- Use _getch(), _putch(), and _kbhit() so that KEY, EMIT and ?TERMINAL will work on PC.
- Fixed : foo { -- } 55 ; - Was entering local frame but not exiting. Now prints error.
- Redefined MAKE_ID to protect it from 16 bit ints
- John Providenza says "If you split local variables onto 2 lines, PForth crashes." Fixed. Also allow \
- Fixed float evaluation in EVALUATE in "quit.fth".
- Flush register cache for ffColon and ffSemiColon to prevent stack warnings from ;
developer is officially maintaining the package.
The rationale for changing this from "tech-pkg" to "pkgsrc-users" is
that it implies that any user can try to maintain the package (by
submitting patches to the mailing list). Since the folks most likely
to care about the package are the folks that want to use it or are
already using it, this would leverage the energy of users who aren't
developers.
V20
- Expand PAD for ConvertNumberToText so "-1 binary .s" doesn't crash.
Thank you Michael Connor of Vancouver for reporting this bug.
- Removed FDROP in REPRESENT to fix stack underflow after "0.0 F.".
Thank you Jim Rosenow of Minnesota for reporting this bug.
- Changed pfCharToLower to function to prevent macro expansion
bugs under VXWORKS
Thank you Jim Rosenow of Minnesota for reporting this bug.
- "0.0 F~" now checks actual binary encoding of floats. Before
this it used to just compare value which was incorrect. Now
"0.0 -0.0 0.0 F~" returns FALSE.
- Fixed definition of INPUT$ in tutorial.
Thank you Hampton Miller of California for reporting this bug.
- Added support for producing a target dictionary with a different
Endian-ness than the host CPU. See PF_BIG_ENDIAN_DIC and
PF_LITTLE_ENDIAN_DIC.
- PForth kernel now comes up in a mode that uses BASE for
numeric input when started with "-i" option. It used to
always consider numeric input as HEX. Initial BASE is decimal.
V21
- Fixed some compiler warnings.
