image downloader service added

modules/image_storage/app/main.py

@@ -0,0 +1,110 @@
+from __future__ import annotations
+
+import hashlib
+import os
+from pathlib import Path
+from typing import Tuple
+from urllib.parse import urlparse
+
+import httpx
+from fastapi import FastAPI, UploadFile, File, HTTPException
+from fastapi.responses import FileResponse
+from pydantic import BaseModel
+
+# Хранилище картинок только для этого сервиса
+MEDIA_ROOT = Path(os.getenv("NYANIMEDB_MEDIA_ROOT", "media")).resolve()
+MEDIA_ROOT.mkdir(parents=True, exist_ok=True)
+
+app = FastAPI()
+
+
+def _guess_ext_from_url(url: str) -> str:
+    path = urlparse(url).path
+    _, ext = os.path.splitext(path)
+    if ext and len(ext) <= 5:
+        return ext
+    return ".jpg"
+
+
+def _guess_ext_from_name(name: str) -> str:
+    _, ext = os.path.splitext(name)
+    if ext and len(ext) <= 5:
+        return ext
+    return ".jpg"
+
+
+def _build_rel_path_from_hash(h: str, ext: str, subdir: str = "posters") -> Tuple[str, Path]:
+    """
+    Строим путь вида subdir/ab/cd/<hash>.ext по sha1-хешу содержимого.
+    """
+    level1 = h[:2]
+    level2 = h[2:4]
+    rel = f"{subdir}/{level1}/{level2}/{h}{ext}"
+    abs_path = MEDIA_ROOT / rel
+    return rel, abs_path
+
+
+async def _save_bytes(data: bytes, *, filename_hint: str, subdir: str = "posters") -> str:
+    # sha1 по содержимому
+    h = hashlib.sha1(data).hexdigest()
+    # расширение либо из имени, либо .jpg
+    ext = _guess_ext_from_name(filename_hint)
+
+    rel, abs_path = _build_rel_path_from_hash(h, ext, subdir=subdir)
+
+    if abs_path.exists():
+        return rel
+
+    abs_path.parent.mkdir(parents=True, exist_ok=True)
+    tmp_path = abs_path.with_suffix(abs_path.suffix + ".tmp")
+
+    with open(tmp_path, "wb") as f:
+        f.write(data)
+    os.replace(tmp_path, abs_path)
+
+    return rel
+
+
+class DownloadByUrlRequest(BaseModel):
+    url: str
+    subdir: str = "posters"
+
+
+@app.post("/upload")
+async def upload_image(
+    file: UploadFile = File(...),
+    subdir: str = "posters",
+):
+    """
+    Загрузка файла от клиента (ETL/админка/бекенд).
+    Возвращает относительный путь (subdir/ab/cd/hash.ext).
+    """
+    data = await file.read()
+    rel = await _save_bytes(data, filename_hint=file.filename or "", subdir=subdir)
+    return {"path": rel}
+
+
+@app.post("/download-by-url")
+async def download_by_url(payload: DownloadByUrlRequest):
+    """
+    Скачивает картинку по URL и сохраняет по тем же правилам, что downloader.py.
+    """
+    async with httpx.AsyncClient(timeout=20.0) as client:
+        r = await client.get(payload.url)
+        r.raise_for_status()
+        data = r.content
+
+    # filename_hint берём из URL
+    rel = await _save_bytes(data, filename_hint=payload.url, subdir=payload.subdir)
+    return {"path": rel}
+
+
+@app.get("/media/{path:path}")
+async def get_image(path: str):
+    """
+    Отдаёт файл по относительному пути (например, posters/ab/cd/hash.jpg).
+    """
+    abs_path = MEDIA_ROOT / path
+    if not abs_path.is_file():
+        raise HTTPException(status_code=404, detail="Image not found")
+    return FileResponse(abs_path)