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.
| Segment | % | Strings |
|---|---|---|
| ORTITLE.EXE | 100% | (18 / 18) |
| ORFIELD.EXE | 99% | (1322 / 1326) |
| ORBTL.EXE | 100% | (792 / 792) |
| Dialogue | 100% | (5601 / 5592) |
| Images | 94% | (52 / 55) |
| Total | 100% | (7785 / 7783) |
- A working hard-disk dump of Appareden
- romtools
- xlsxwriter
- openpyxl
- Bitstring
- nosetests
- ndcpy
nosetests test.py - a few sanity checks
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.
- 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.
- gem_codec.py - decoder and size-optimizing encoder for the .GEM image format, written from the game's own decoder. Its docstring is the format reference.
- ending_overline.py - ASM patch for ENDING.EXE so the credits show macrons (Jado~ -> Jadō), using user-defined characters; applied through asm.py's FD_EDITS/CD_EDITS by reinsert.py.
- spz.py - reader/writer for .SPZ sprite tables (which 16x16 tiles of a .GEM sheet make up each sprite), written from NEKORUN.EXE's sprite drawer.
- nrchr_text.py - renders the English "Ah!" / "There it is!!" / "Wait!!" sprites for the Neko Run minigame and repacks them into NRCHR_99.GEM/.SPZ (called from gem.py).
- automation/ - drives the game headless under np2core: replays scenes, renders every message block of every scene for text QA, and opens every shop to check description widths (see automation/README.md).
- 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.
- 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.
This project is licensed under the Creative Commons A-NC License - see the LICENSE.md file for details.