From 30a1c1af573820b5008d8d535e44772d4cab61a7 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?=E5=88=98=E8=88=AA=E5=AE=87?= <3364451258@qq.com>
Date: Sun, 5 Jul 2026 01:17:13 +0800
Subject: [PATCH] fix: SSL certificate issue and Windows terminal encoding

---
 src/cli/main.py      |  5 +++++
 src/core/embedder.py | 22 +++++++++++++++++++++-
 2 files changed, 26 insertions(+), 1 deletion(-)

diff --git a/src/cli/main.py b/src/cli/main.py
index 8aff262..4944172 100644
--- a/src/cli/main.py
+++ b/src/cli/main.py
@@ -1,10 +1,15 @@
 """命令行工具入口."""
 from pathlib import Path
 import sys
+import io
 
 import typer
 import uvicorn
 
+# Windows 终端默认用 GBK, 无法输出 emoji, 强制 UTF-8
+if sys.stdout.encoding != "utf-8":
+    sys.stdout = io.TextIOWrapper(sys.stdout.buffer, encoding="utf-8", errors="replace")
+
 # 确保 src 在路径中
 sys.path.insert(0, str(Path(__file__).parent.parent))
 
diff --git a/src/core/embedder.py b/src/core/embedder.py
index 0bcd300..cfa5ba4 100644
--- a/src/core/embedder.py
+++ b/src/core/embedder.py
@@ -1,4 +1,11 @@
 """嵌入模型抽象层."""
+import os
+
+# 解决 Windows SSL 证书问题: 强制使用本地缓存模型, 避免联网验证
+# 模型首次下载需要先临时取消这些环境变量 (或手动运行一次下载脚本)
+os.environ.setdefault("HF_HUB_OFFLINE", "1")
+os.environ.setdefault("TRANSFORMERS_OFFLINE", "1")
+
 from sentence_transformers import SentenceTransformer
 
 from src.core.config import EmbedConfig
@@ -10,7 +17,20 @@ class Embedder:
     def __init__(self, config: EmbedConfig):
         self._config = config
         if config.mode == "local":
-            self._model = SentenceTransformer(config.local_model)
+            # 优先从本地缓存加载, 若模型未下载则通过镜像下载
+            try:
+                self._model = SentenceTransformer(
+                    config.local_model, local_files_only=True
+                )
+            except Exception:
+                # 模型未缓存: 临时开启网络, 使用国内镜像
+                os.environ.pop("HF_HUB_OFFLINE", None)
+                os.environ.pop("TRANSFORMERS_OFFLINE", None)
+                os.environ["HF_ENDPOINT"] = "https://hf-mirror.com"
+                self._model = SentenceTransformer(config.local_model)
+                # 恢复离线设置, 后续加载走缓存
+                os.environ["HF_HUB_OFFLINE"] = "1"
+                os.environ["TRANSFORMERS_OFFLINE"] = "1"
         elif config.mode == "api":
             self._model = None  # 延迟初始化, 需要 openai 包
             self._api_base = config.api_base