Skip to content

Latest commit

 

History

History

scripts

README.md

Helper scripts

This hodge-podge collection of scripts is used to extract raw game data from the binary image files and help enrich the disassembler output.

The main entrypoint is extractdata.py, which uses fonts.py and diskimage.py. This runs in a conda environment which is created by:

conda env update -f environment.yml

Then run the script via:

conda activate ef1941
python extractdata.py

This creates a number of data files in extract/ as well as the APX binary image in reference/apxdump.dat read from the disk image. These files are incorporated into the game engine, e.g. in src/engine/map-data.ts, with the font image masks copied directly to static/.

Other scripts

  • apxsrcmap.py: Generates a cross-reference in apxsrc.map from APX listing line numbers to the annotated APX disassembly in apxdump.asm for the howitworks.md document. I think apxsrcrefs.py an earlier version?
  • atascii.py: Dumps a binary file with the Atascii font
  • matchmaps.py: A failed attempt to look for common binary code sequences between the APX and cartridge images.
  • asciimap.html: old helper to render an ascii map as HTML (see unused fonts.py:asciimapposter)
  • apx.cmd, card.cmd: example command files for driving my fork of the py65 6502 emulator

Other notes

Dissemblies were produced using my fork of the pydisass6502 disassembler, which includes significant changes and few bugfixes. The typical recipe is:

python <pydisass6502>/disass.py -i apxdump.dat -o apxdump.asm -e apxdump.map.json -c apxdump.stats -m <pydisass6502>/lib/atari-mapping.json

python <pydisass6502>/disass.py -i cartridge.rom -o cartridge.asm -e cartridge.map.json -c cartridge.stats -m <pydisass6502>/lib/atari-mapping.json

Convert the black and white fontmap images to transparent background with a command like:

convert extract/fontmap-custom-apx.png -background black -alpha remove -sample 1600% -negate doc/fontapx.png

Extract pages from PDF document to individual JPG images for OCR via:

convert -density 300 APX_Source_Code_for_Eastern_Front_1941_rev_2.pdf[10-58] -quality 90 apxsrc%02d.jpg

Run OCR on each page with tesseract like:

for i in {10..58}; do echo $i; echo $i >> apxsrc.txt; tesseract apxsrc$i.jpg - >> apxsrc.txt; done

Pad cover image for GH aspect ratio:

convert preview.png -gravity center -background black -extent 1536x768 preview-padded.png

Check broken markdown links with markdown-link-check