From 9c21a5240fce1fc7e7fb1b8c8a2f03d192233601 Mon Sep 17 00:00:00 2001
From: xunbu <xunbu3@qq.com>
Date: Thu, 18 Dec 2025 00:02:56 +0800
Subject: [PATCH] fix

---
 full.spec | 18 +++++++++++++-----
 lite.spec | 22 ++++++++++++++++------
 2 files changed, 29 insertions(+), 11 deletions(-)

diff --git a/full.spec b/full.spec
index 24390b2..dcf394d 100644
--- a/full.spec
+++ b/full.spec
@@ -9,10 +9,14 @@ binaries = []
 hiddenimports = ['markdown.extensions.tables', 'pymdownx.arithmatex',
                 'pymdownx.superfences', 'pymdownx.highlight', 'pygments',
                 'docling_ibm_models',
-                'docling_parse','cv2']
+                'docling_parse', 'cv2']
 
-# 先收集第三方包的资源
-for package in ['easyocr', 'docling', 'pygments', 'docling_ibm_models']:
+# --- 核心修改：在列表中添加 'tiktoken' ---
+# tiktoken: 解决 Unknown encoding cl100k_base 问题
+# easyocr, docling 等: 收集其他依赖
+packages_to_collect = ['easyocr', 'docling', 'pygments', 'docling_ibm_models', 'tiktoken']
+
+for package in packages_to_collect:
     try:
         tmp_ret = collect_all(package)
         datas += tmp_ret[0]
@@ -21,6 +25,7 @@ for package in ['easyocr', 'docling', 'pygments', 'docling_ibm_models']:
     except Exception as e:
         print(f"Warning: Failed to collect resources for {package}: {e}")
 
+
 try:
     datas += copy_metadata('docling-ibm-models') # 这里必须用连字符(pip名)
     datas += copy_metadata('docling-parse')      # 预防性添加
@@ -28,6 +33,7 @@ except Exception as e:
     print(f"Warning: Failed to copy metadata: {e}")
 
 # 然后添加您的自定义资源（避免重复）
+# 注意：确保 .venv 路径在您当前的构建环境中是存在的
 custom_datas = [
     ('./.venv/Lib/site-packages/docling_parse/pdf_resources_v2', 'docling_parse/pdf_resources_v2'),
     ('./docutranslate/static', 'docutranslate/static'),
@@ -36,12 +42,13 @@ custom_datas = [
 
 # 避免添加重复的数据
 for data in custom_datas:
+    # 简单的去重检查，防止完全相同的源路径和目标路径被再次添加
     if data not in datas:
         datas.append(data)
 
 a = Analysis(
-    ['docutranslate/app.py'],  # 使用正斜杠
-    pathex=[],  # 添加当前工作目录到 pathex
+    ['docutranslate/app.py'],
+    pathex=[],
     binaries=binaries,
     datas=datas,
     hiddenimports=list(set(hiddenimports)),  # 去重
@@ -52,6 +59,7 @@ a = Analysis(
     noarchive=False,
     optimize=0,
 )
+
 pyz = PYZ(a.pure)
 
 exe = EXE(
diff --git a/lite.spec b/lite.spec
index 88f58b1..e91e876 100644
--- a/lite.spec
+++ b/lite.spec
@@ -1,12 +1,21 @@
 # -*- mode: python ; coding: utf-8 -*-
 import os
-from PyInstaller.utils.hooks import collect_data_files
+from PyInstaller.utils.hooks import collect_data_files, collect_all
 import docutranslate
 
+# --- 核心修改开始：收集 tiktoken 的所有依赖 ---
+# tiktoken 依赖动态加载的插件和二进制文件，必须全部收集
+tmp_ret = collect_all('tiktoken')
+tik_datas = tmp_ret[0]
+tik_binaries = tmp_ret[1]
+tik_hiddenimports = tmp_ret[2]
+# --- 核心修改结束 ---
+
 datas = [
     ('docutranslate/static', 'docutranslate/static'),
     ('docutranslate/template', 'docutranslate/template'),
-    *collect_data_files('pygments')  # 直接展开
+    *collect_data_files('pygments'),
+    *tik_datas
 ]
 
 hiddenimports = [
@@ -14,13 +23,14 @@ hiddenimports = [
     'pymdownx.arithmatex',
     'pymdownx.superfences',
     'pymdownx.highlight',
-    'pygments'
+    'pygments',
+    *tik_hiddenimports # 合并 tiktoken 的隐式导入 (包含 tiktoken_ext.openai_public 等)
 ]
 
 a = Analysis(
-    ['docutranslate/app.py'],  # 使用正斜杠，Windows 也支持
+    ['docutranslate/app.py'],
     pathex=[],
-    binaries=[],
+    binaries=tik_binaries, # 确保包含了 tiktoken 的二进制文件
     datas=datas,
     hiddenimports=hiddenimports,
     hookspath=[],
@@ -51,5 +61,5 @@ exe = EXE(
     target_arch=None,
     codesign_identity=None,
     entitlements_file=None,
-    icon='DocuTranslate.ico',  # 修正为字符串
+    icon='DocuTranslate.ico',
 )
\ No newline at end of file