feat: ✨ docker supports custom config.

Dockerfile

@@ -6,6 +6,8 @@ WORKDIR /app
 EXPOSE 7860
 
 ENV PYTHONUNBUFFERED=1
+ENV NOTO_FONT_PATH=/app
+ENV DOCKER_CONFIG=1
 
 # Download all required fonts
 ADD "https://github.com/satbyy/go-noto-universal/releases/download/v7.0/GoNotoKurrent-Regular.ttf" /app/
@@ -26,4 +28,4 @@ COPY . .
 
 RUN uv pip install --system --no-cache .
 
-CMD ["pdf2zh", "-i"]
+CMD ["pdf2zh", "--config", "/app/config.json", "-i"]

pdf2zh/config.py

@@ -81,7 +81,10 @@ def custome_config(cls, file_path):
         """使用自定义路径加载配置文件"""
         custom_path = Path(file_path)
         if not custom_path.exists():
-            raise ValueError(f"Config file {custom_path} not found!")
+            if "DOCKER_CONFIG" not in os.environ:
+                raise ValueError(f"Config file {custom_path} not found!")
+            with open("config.json", "w") as file:
+                json.dump({}, file, indent=4, ensure_ascii=False)
         # 加锁
         with cls._lock:
             instance = cls()

pdf2zh/high_level.py

@@ -380,7 +380,7 @@ def translate(
 
 
 def download_remote_fonts(lang: str):
-    URL_PREFIX = "https://github.com/timelic/source-han-serif/releases/download/main/"
+    URL_PREFIX = ConfigManager.get("FONT_URL_PREFIX")
     LANG_NAME_MAP = {
         **{la: "GoNotoKurrent-Regular.ttf" for la in noto_list},
         **{
@@ -396,14 +396,32 @@ def download_remote_fonts(lang: str):
     }
     font_name = LANG_NAME_MAP.get(lang, "GoNotoKurrent-Regular.ttf")
 
-    cache_folder = os.path.join(os.path.expanduser("~"), ".cache", "pdf2zh")
-    os.makedirs(cache_folder, exist_ok=True)
     # docker
-    font_path = ConfigManager.get("NOTO_FONT_PATH", Path("/app", font_name).as_posix())
-    if not Path(font_path).exists():
-        font_path = Path(cache_folder, font_name).as_posix()
-    if not Path(font_path).exists():
+    font_path = ConfigManager.get(
+        "NOTO_FONT_PATH", os.path.join(os.path.expanduser("~"), ".cache", "pdf2zh")
+    )
+    if not Path(font_path, font_name).exists():
+        font_path = Path(font_path, font_name).as_posix()
         print(f"Downloading {font_name}...")
-        urllib.request.urlretrieve(f"{URL_PREFIX}{font_name}", font_path)
-
-    return font_path
+        with tqdm.tqdm(
+            unit="B", unit_scale=True, leave=False, unit_divisor=1024, desc=font_path
+        ) as t:
+            last_downloaded = [0]  # 使用列表保存上一次的下载量
+
+            def reporthook(block_num, block_size, total_size):
+                if total_size > 0:
+                    t.total = total_size
+                downloaded = block_num * block_size
+                # 计算本次的增量
+                delta = downloaded - last_downloaded[0]
+                t.update(delta)
+                last_downloaded[0] = downloaded
+
+            urllib.request.urlretrieve(
+                f"{URL_PREFIX}{font_name}", font_path, reporthook=reporthook
+            )
+
+        print(f"Downloaded {font_path}...")
+        return font_path
+    print(font_path)
+    return Path(font_path, font_name).as_posix()

pdf2zh/pdf2zh.py

@@ -174,6 +174,13 @@ def create_parser() -> argparse.ArgumentParser:
         help="Use experimental backend babeldoc.",
     )
 
+    parse_params.add_argument(
+        "--CN",
+        default=False,
+        action="store_true",
+        help="download in mainland China.",
+    )
+
     return parser
 
 
@@ -229,6 +236,18 @@ def main(args: Optional[List[str]] = None) -> int:
     if parsed_args.debug:
         log.setLevel(logging.DEBUG)
 
+    if parsed_args.CN:
+        ConfigManager.set(
+            "FONT_URL_PREFIX",
+            "https://gitee.com/xzk1234/source-han-serif/releases/download/0.1/",
+        )
+        os.environ["HF_ENDPOINT"] = "https://hf-mirror.com"
+    else:
+        ConfigManager.set(
+            "FONT_URL_PREFIX",
+            "https://github.com/timelic/source-han-serif/releases/download/main/",
+        )
+
     if parsed_args.onnx:
         ModelInstance.value = OnnxModel(parsed_args.onnx)
     else: