from __future__ import annotations

import json
import re
from pathlib import Path
from typing import Any

from app.settings import Settings


TOKEN_PATTERN = re.compile(r"[a-zA-Z0-9_]+")
CN_TOKEN_PATTERN = re.compile(r"[\u4e00-\u9fff]")

TERM_ALIASES = {
    "法兰": {"flange", "mounting_plate"},
    "套筒": {"sleeve", "tube", "cylinder"},
    "孔": {"hole", "circle", "circles"},
    "沉头": {"countersunk", "counterbore"},
    "沉孔": {"counterbore", "counterbored"},
    "槽": {"slot", "obround"},
    "底板": {"plate", "rectangle", "mounting_plate"},
    "轴": {"shaft", "cylinder"},
}


def tokens(value: str) -> set[str]:
    text = str(value or "").lower()
    result = {token.lower() for token in TOKEN_PATTERN.findall(text)}
    result.update(CN_TOKEN_PATTERN.findall(text))
    for term, aliases in TERM_ALIASES.items():
        if term in text:
            result.update(aliases)
    return result


class CdslLibrary:
    def __init__(self, settings: Settings) -> None:
        self.settings = settings
        self.index_path = settings.library_root / "index" / "catalog.json"

    def _records(self) -> list[dict[str, Any]]:
        if not self.index_path.is_file():
            return []
        payload = json.loads(self.index_path.read_text(encoding="utf-8"))
        return payload.get("samples", [])

    def count(self) -> int:
        return len(self._records())

    def search(self, query: str, limit: int = 5) -> list[dict[str, Any]]:
        query_tokens = tokens(query)
        if not query_tokens:
            return [
                {
                    "part_id": record["part_id"],
                    "profiles": record.get("profiles", []),
                    "features": record.get("features", []),
                    "summary": record.get("summary", ""),
                    "family": record.get("family", ""),
                    "atomics": record.get("features", []),
                }
                for record in self._records()[:limit]
            ]
        scored: list[tuple[int, dict[str, Any]]] = []
        for record in self._records():
            corpus = " ".join([
                record.get("part_id", ""),
                record.get("source_name", ""),
                record.get("summary", ""),
                record.get("family", ""),
                " ".join(record.get("profiles", [])),
                " ".join(record.get("features", [])),
                " ".join(record.get("parameters", [])),
            ])
            score = len(query_tokens & tokens(corpus))
            if score:
                scored.append((score, record))
        scored.sort(key=lambda item: (-item[0], item[1]["part_id"]))
        return [
            {
                "part_id": record["part_id"],
                "profiles": record.get("profiles", []),
                "features": record.get("features", []),
                "summary": record.get("summary", ""),
                "family": record.get("family", ""),
                "atomics": record.get("features", []),
            }
            for _, record in scored[:limit]
        ]

    def read_sample(self, part_id: str) -> dict[str, Any]:
        for record in self._records():
            if record.get("part_id") == part_id:
                source = self.settings.library_root / "samples" / part_id / "model.cdsl.json"
                if not source.is_file():
                    break
                return json.loads(source.read_text(encoding="utf-8"))
        raise ValueError(f"CDSL sample not found: {part_id}")