(WIP) Stardict converter. Pseudo code. To support making PWA Japanese dictionary. Progress depend on laziness and the desire to procrastinate...

dict_converter.py

#!/usr/bn/env python3
# coding: utf-8
import json
import struct

DICT_NAMES = ["star_nhatviet", "star_vietnhat"]
DICT_JSON = {
    name: {
        "bookname": "",
        "wordcount": "",
        "synwordcount": "",
        "idxfilesize": "",
        "idxoffsetbits": "",
        "author": "",
        "email": "",
        "website": "",
        "description": "",
        "date": "",
        "sametypesequence": "",
        "dicttype": "",
        "data": [],
    }
    for name in DICT_NAMES
}

for name in DICT_NAMES:
    print(f"Processing dictionary: {name}...")
    with open(f"{name}.ifo", "rt", encoding="utf-8", newline=None) as finfo:
        for line in finfo:
            if "=" in line:
                k, v = line.split("=", 1)
                DICT_JSON[name][k] = v.strip()
    print(f"{name}: {DICT_JSON[name]}")
    with open(f"{name}.idx", "rb") as fidx:
        with open(f"{name}.dict", "rb") as fdict:
            i = 0

            while fidx.read(1):
                fidx.seek(-1, 1)
                try:
                    word = b"".join(iter(lambda: fidx.read(1),
                                         b"\0")).decode("utf-8",
                                                        errors="ignore")
                    start = struct.unpack("!I", fidx.read(4))[0]
                    length = struct.unpack("!I", fidx.read(4))[0]
                    # print(f"word: {word}\nstart: {start}\nlength: {length}")
                except Exception:
                    exit(1)
                fdict.seek(start)
                definition = fdict.read(length).decode("utf-8",
                                                       errors="ignore")
                # print(definition)
                DICT_JSON[name]["data"].append({
                    "word": word,
                    "definition": definition
                })
                i += 1
                print(f"\r{i} word(s) processed so far...", end="")
    print(f"\nDumping to {name}.json...")
    with open(f"{name}.json", "w+", encoding="utf-8", newline="\n") as fjson:
        json.dump(DICT_JSON[name], fjson)