"""
services/kb_service.py（瘦身版）

仅保留报告生成「附图提取」所需的知识库文档磁盘路径解析助手：
从 eval_report 的完整 kb_service.py 中抽取，去除知识库 CRUD / 上传 / worker 等无关逻辑。
"""

from __future__ import annotations

from pathlib import Path
from typing import List, Optional

from config import settings
from database.models import KbDocument as KbDocumentModel


def _normalize_rel_path(path: str) -> str:
    """将 'a\\b\\c' 规范为 'a/b/c'，并去掉前导 '/'。"""
    s = str(path or "").replace("\\", "/").strip()
    while s.startswith("./"):
        s = s[2:]
    return s.lstrip("/")


def _kb_doc_storage_rel_path(
    file_path_dir: Optional[str],
    basename: str,
    storage_rel_path: Optional[str] = None,
) -> str:
    """项目目录下的相对存储路径（含文件名）。优先 storage_rel_path（confirm 时写入）。"""
    stored = _normalize_rel_path(str(storage_rel_path or ""))
    if stored:
        return stored
    d = _normalize_rel_path(str(file_path_dir or ""))
    bn = str(basename or "").strip()
    if d and bn:
        return f"{d}/{bn}"
    return bn or d


def _kb_doc_path_candidates_for_model(doc_root: Path, doc: KbDocumentModel) -> List[Path]:
    """解析磁盘路径时的候选列表（按优先级）。"""
    rel = _kb_doc_storage_rel_path(
        doc.file_path,
        doc.name,
        getattr(doc, "storage_rel_path", None),
    )
    candidates: List[Path] = []
    if rel:
        candidates.append((doc_root / doc.project_id / rel).resolve())
    name = str(doc.name or "").strip()
    fp_dir = _normalize_rel_path(str(doc.file_path or ""))
    if fp_dir and name:
        candidates.append((doc_root / doc.project_id / fp_dir / name).resolve())
    if name:
        candidates.append((doc_root / doc.project_id / name).resolve())
    if not candidates:
        candidates.append((doc_root / doc.project_id / "_missing_").resolve())
    deduped: List[Path] = []
    seen: set[str] = set()
    for p in candidates:
        key = str(p)
        if key in seen:
            continue
        seen.add(key)
        deduped.append(p)
    return deduped


def _kb_doc_absolute_file_path_for_model(doc_root: Path, doc: KbDocumentModel) -> Path:
    for p in _kb_doc_path_candidates_for_model(doc_root, doc):
        if p.is_file():
            return p
    return _kb_doc_path_candidates_for_model(doc_root, doc)[0]


def _kb_doc_file_exists_for_model(doc: KbDocumentModel) -> bool:
    """文档在磁盘上是否可读（多路径回退，兼容历史 file_path/name 组合）。"""
    doc_root = Path(settings.DOC_PAT).resolve()
    return any(p.is_file() for p in _kb_doc_path_candidates_for_model(doc_root, doc))