#!/usr/bin/env python3
"""SCICRAFT — story generator.

Reads STORY/*.md, merges every ```lang JSON block (in filename order) into
en_us.json, writes it to BOTH packs, and validates that every lang key
referenced from the KubeJS scripts and advancement JSONs resolves.

Usage:
    python tools/gen_story.py

Edit the story by editing STORY/*.md, then run this script. Never hand-edit
en_us.json — it is a generated artifact.
"""

import json
import re
import sys
from pathlib import Path

ROOT = Path(__file__).resolve().parent.parent
STORY_DIR = ROOT / "STORY"
LANGS = [
    ROOT / "server_pack" / "kubejs" / "assets" / "scicraft" / "lang" / "en_us.json",
    ROOT / "client_pack" / "kubejs" / "assets" / "scicraft" / "lang" / "en_us.json",
]
KUBEJS_DIRS = [ROOT / "server_pack" / "kubejs"]
ADVANCEMENT_DIR = ROOT / "server_pack" / "kubejs" / "data" / "scicraft" / "advancements"

LANG_BLOCK = re.compile(r"```lang\s*\n(.*?)```", re.S)


def load_lang_files():
    merged = {}
    order = []
    for path in sorted(STORY_DIR.glob("*.md")):
        text = path.read_text(encoding="utf-8")
        for i, match in enumerate(LANG_BLOCK.finditer(text)):
            try:
                data = json.loads(match.group(1))
            except json.JSONDecodeError as e:
                sys.exit(f"BAD JSON in {path.name} block #{i + 1}: {e}")
            for key, value in data.items():
                if key in merged:
                    print(f"WARN: duplicate key {key!r} ({path.name} overrides)")
                merged[key] = value
                order.append(key)
    return merged, order


def write_lang(merged, order):
    payload = {k: merged[k] for k in order}
    for target in LANGS:
        target.parent.mkdir(parents=True, exist_ok=True)
        target.write_text(
            json.dumps(payload, ensure_ascii=False, indent="\t") + "\n",
            encoding="utf-8",
        )
        print(f"wrote {target.relative_to(ROOT)} ({len(payload)} keys)")


def validate(merged):
    missing = set()

    def check_key(key, source):
        if key not in merged:
            missing.add((key, source))

    for root_dir in KUBEJS_DIRS:
        for script in root_dir.rglob("*.js"):
            for key in re.findall(r"Text\.translate\(\s*['\"]([^'\"]+)['\"]\s*(?:\+|\))", script.read_text(encoding="utf-8")):
                # Keys followed by ` + msg` are dynamic templates — skip.
                if key.endswith("."):
                    continue
                check_key(key, str(script.relative_to(ROOT)))

    if ADVANCEMENT_DIR.exists():
        for jf in ADVANCEMENT_DIR.rglob("*.json"):
            for key in re.findall(r'"translate"\s*:\s*"([^"]+)"', jf.read_text(encoding="utf-8")):
                check_key(key, str(jf.relative_to(ROOT)))

    if missing:
        print("\nMISSING LANG KEYS:")
        for key, source in sorted(missing):
            print(f"  {key}  (used in {source})")
        sys.exit(1)
    print("all referenced lang keys resolve")


if __name__ == "__main__":
    merged, order = load_lang_files()
    write_lang(merged, order)
    validate(merged)