appareden

Romhacking utilities and notes developed as part of 46 OkuMen's English translation patch for Appareden - Fukuryuu no Shou (あっぱれ伝 ー伏龍の章ー) (PC-98), a traditional and colorful JRPG based on Japanese history and myth.

Draft Reinsertion Progress

Segment	%	Strings
ORTITLE.EXE	100%	(18 / 18)
ORFIELD.EXE	100%	(1326 / 1326)
ORBTL.EXE	100%	(792 / 792)
Dialogue	100%	(5601 / 5592)
Images	89%	(49 / 55)
Total	100%	(7786 / 7783)

Project notes

• This game's script is longer than our three released projects's scripts combined.
• The game is available on both floppy and CD versions, and our patch targets both.
• A few hacks unique to this project:
  • Dictionary compression - the patch implements compression of the most-common words, and adds ASM to decompress these words at display time.
  • .GEM encoder - this project re-encodes a fairly complex file format used for all iamges in the game.
• Because the game is long and testing is time-consuming, we spent some time on stuff to make it a little easiser:
  • Deciphering save-file format
  • Developing a few cheats
  • Generating a world map from lots of game screenshots
• File types:
  • .MSG - game script file
  • .EXE - system text
  • .GEM - custom image format
  • .SPZ - sprite sheet format
  • .COD - dialogue event routine file format
  • Some of these formats may exist in other TGL games, such as the Farland Story games.
• This is a fullscreen 16-color RPG, so it is an absolute memory hog. It was fairly prone to memory issues upon release, and is still fairly unstable even after the developer's patch.

Development

Requirements

• A working hard-disk dump of Appareden
• romtools
• xlsxwriter
• openpyxl
• Bitstring
• nosetests
• ndcpy

Running Tests

nosetests test.py - a few sanity checks

Building

This requires an existing HDI of Appareden, such as the one from the Neo Kobe pack. The utilities should work on the (developer-) patched or unpatched versions. Place your dump of Appareden.hdi in the subfolder original. Then run these scripts to dump the system text, dialogue, and pointers:

python sys_dump.py
python msg_dump.py
python find_pointers.py

Translate the script, which is dumped into appareden_sys_dump.xlsx and appareden_msg_dump.xlsx.

Now create a subfolder patched, where another copy of Appareden.hdi should go. Now run these to reinsert text and images:

python typeset_script.py
python reinsert.py
python gem.py

Now, patched/Appareden.hdi should be playable in English as far as you've translated.

Main scripts

• sys_dump.py - dumps text from system files (ORTITLE.EXE, ORFIELD.EXE, ORBTL.EXE) into appareden_sys_dump.xlsx.
• msg_dump.py - dumps text from dialogue files (a few hundred .MSG files) into appareden_msg_dump.xlsx.
• find_pointers.py - dumps pointers into appareden_pointer_dump.xlsx.
• typeset_script.py - adds line breaks where necessary into the translated column of the .MSG files.
• reinsert.py - Replaces the JP strings with the translated ones in the dump excel, adjusts pointers, and reinserts the translated files into the HDI.

Other scripts (used by the above)

• rominfo.py - maps out many properties of the game files, including text blocks, pointer blocks, control codes, etc.
• asm.py - holds the ASM strings used for various text/game modification hacks. Some will be in the final patch (text compression routines, bugfixes), others are just for testing ("turn off random battles" cheat, etc).
• pointer_info.py - list of pointers to reassign. There are lots of duplicate strings - we can save space by re-pointing duplicates to the same text location.
• portraits.py - list of portraits and their character/emotion.
• utils.py - helper functions.

One-time utility scripts

• dictionary.py - counts the words in ORFIELD/ORBTL and generates a compression dictionary for each. The dictionary is then inserted into the first string in each file.
• replace_title.py - for easily viewing a GEM image. Give it the filename of a GEM image in patched/, and it will replace the title screen with it.
• portraitcrop.py - crop the fullscreen portraits.
• tools/find_cd_diffs.py - determine the text-block location differences between the FD and CD versions.
• n.py - explorative script for trying different control-code values for a line break.

License

This project is licensed under the Creative Commons A-NC License - see the LICENSE.md file for details.