# Build/test pipeline for an RK86 disassembly project.

#

# Workflow:

# 1. Drop the original tape file into tape/<NAME>.<EXT>.

# 2. `just initial` — extract <NAME>.bin from the tape envelope.

# 3. `just disasm` — produce <NAME>.asm via linear disassembler.

# 4. Edit <NAME>.asm by hand, splitting code/data and adding labels.

# 5. `just ci` — assemble and byte-diff against tape/<NAME>.<EXT>.

# Empty diff = green. Run after every edit.

#

# Customise NAME and EXT for your tape file. ORG is the load

# address printed by tobin.py during `just initial`. PAD is the

# off-by-N count for this tape (see playbook/common-pitfalls.md

# "End-address off-by-N in tape headers"). Compute it from

# tobin.py's output: `wrote N bytes` minus `declared payload M

# bytes`. N=2 is the most common; N=1 also occurs; N=0 (no

# padding) for clean tapes.

NAME := "SPACE"

EXT := "RK" # one of: rk, rkr, pki, gam

ORG := "0x0100" # update to match the file header

PAD := "2" # off-by-N padding (default 2; set 1 / 0 if needed)

ext := lowercase(EXT)

ci: build test

build:

bunx asm8080 --split -l --trailer-padding {{ PAD }} --format {{ ext }} {{ NAME }}.asm

test:

xxd tape/{{ NAME }}.{{ EXT }} >tape/{{ NAME }}.{{ EXT }}.hex

xxd {{ NAME }}.{{ ext }} >{{ NAME }}.{{ ext }}.hex

diff tape/{{ NAME }}.{{ EXT }}.hex {{ NAME }}.{{ ext }}.hex

# Convert the tape file (envelope + trailer) to a raw binary file.

# Run this once when starting a new disassembly.

initial:

uv run tobin.py tape/{{ NAME }}.{{ EXT }}

mv tape/{{ NAME }}.bin {{ NAME }}.bin

# Linear disassembly. Re-run if the .bin changes (rarely needed).

# Beware: re-running overwrites your annotated .asm — commit first.

disasm:

python3 disasm.py {{ NAME }}.bin {{ ORG }} >{{ NAME }}.asm

clean:

rm -f {{ NAME }}.bin {{ NAME }}.{{ ext }} {{ NAME }}.bin.hex

rm -f {{ NAME }}.{{ ext }}.hex {{ NAME }}.lst {{ NAME }}.sym

rm -f {{ NAME }}.map {{ NAME }}.json

rm -f tape/{{ NAME }}.{{ EXT }}.hex