Bugs fixed and features added
Last revised: 5-May-2004
Release 0.8:
- Added 8085 support
- Fixed failure to complain about illegal addressing modes
- Fixed problem with initializing processor-specifics for GUI
- Accept binary (%) and octal (@) numbers for Motorola family processors
- Explain binary and octal number syntax in the docs
- Generate code for instructions that have bad labels
- Generate bytes of 00 for post-allocation errors
- Made GUI outputs the default when using the GUI.
Release 0.7:
- Added 8008 support
- Added 6502 support
- Got rid of the ton of generated instruction classes
- Fixed the problem with multiple outputs going to stdout; works OK now
- Fixed the problem with opcodes of 00 being treated as 0-length
- Added Intel-format numbers for 8080 and 8008 (suffix H=hex, etc)
- Improved the mechanism for finding processor-dependent code
- Fixed docs: run instructions, numerous small errors in syntax docs
Release 0.6:
- Efficiently recognize non-existant mnemonics.
- Flush the table of macros at the beginning.
- A macro/if without a endm/endif just emits an error message, doesn't die.
Release 0.5:
- Added macros - macro, endm, exitm
- Added conditionals - if, elseif, ifdef, ifndef, ifeq, ifneq, else, endif
- Added 'error' pseudo-op
- Added operators - (prefix) ! - ~ (infix) % << >> < <= > >= != == & ^ |
- Made s-record output all upper-case
- Fixed limitation of forward referencing depth of 1
- Fixed ArrayIndexOutOfBounds writing s-records when org != 0
- Fixed code generation for 16-bit indirect addressing (6809 only)
- Fixed recognition of trailing characters on an expression list
- Fixed handling of strings as numbers, and documentation about it
- Improved speed of output to GUI
Release 0.4:
- Added 8080 support
- Added GUI
- Added Applet
Release 0.3:
- Added 6800 support
- fixed checksums in s-record output
- added S5 and S9 records in s-record output
- added 'end' pseudo-op
- Included pathnames are relative to the directory of the including file
(unless the pathname is absolute)
- Allow for opcodes that match Java keywords
- Made 'include' 6809-dependent; pseudo-op may differ for other processors
Release 0.2:
- S-record output file
- Shows object code in output listing
- Better error reporting, somewhat more meaningful messages
- Command line options for listing, symbol table, binary/s-record/no output
- Changed 'incl' pseudo-op to 'include'
- 2-byte instruction operands no longer overwrite 2nd byte of opcode
- A label on a line by itself is OK
- The current-address symbol "*" is now recognized
- Illegal label names are now complained about
- Fixed bounds-checking for prettyShort and reallyShort index offsets
- "Really short" indirect mode now generates "pretty short" code
- Now generates the right opcodes for LDS/STS
- Now recognizes JSR
- Numerous other problems...
Known bugs: