增加对mineru的支持

.idea/workspace.xml

@@ -5,7 +5,16 @@
   </component>
   <component name="ChangeListManager">
     <list default="true" id="6b18b44a-df57-4212-a857-9e291ebe5dd2" name="更改" comment="">
-      <change beforePath="$PROJECT_DIR$/app.spec" beforeDir="false" afterPath="$PROJECT_DIR$/app.spec" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/docutranslate/converter/__init__.py" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/docutranslate/converter/converter.py" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/docutranslate/converter/converter_docling.py" afterDir="false" />
+      <change afterPath="$PROJECT_DIR$/docutranslate/converter/converter_mineru.py" afterDir="false" />
+      <change beforePath="$PROJECT_DIR$/.idea/workspace.xml" beforeDir="false" afterPath="$PROJECT_DIR$/.idea/workspace.xml" afterDir="false" />
+      <change beforePath="$PROJECT_DIR$/README.md" beforeDir="false" afterPath="$PROJECT_DIR$/README.md" afterDir="false" />
+      <change beforePath="$PROJECT_DIR$/docutranslate/app.py" beforeDir="false" afterPath="$PROJECT_DIR$/docutranslate/app.py" afterDir="false" />
+      <change beforePath="$PROJECT_DIR$/docutranslate/static/index.html" beforeDir="false" afterPath="$PROJECT_DIR$/docutranslate/static/index.html" afterDir="false" />
+      <change beforePath="$PROJECT_DIR$/docutranslate/translater.py" beforeDir="false" afterPath="$PROJECT_DIR$/docutranslate/translater.py" afterDir="false" />
+      <change beforePath="$PROJECT_DIR$/pyproject.toml" beforeDir="false" afterPath="$PROJECT_DIR$/pyproject.toml" afterDir="false" />
     </list>
     <option name="SHOW_DIALOG" value="false" />
     <option name="HIGHLIGHT_CONFLICTS" value="true" />
@@ -35,59 +44,62 @@
     <option name="hideEmptyMiddlePackages" value="true" />
     <option name="showLibraryContents" value="true" />
   </component>
-  <component name="PropertiesComponent">{
+  <component name="PropertiesComponent"><![CDATA[{
-  &quot;keyToString&quot;: {
+  "keyToString": {
-    &quot;DefaultHtmlFileTemplate&quot;: &quot;HTML File&quot;,
+    "DefaultHtmlFileTemplate": "HTML File",
-    &quot;JavaScript 调试.output.html (1).executor&quot;: &quot;Run&quot;,
+    "JavaScript 调试.output.html (1).executor": "Run",
-    &quot;JavaScript 调试.output.html.executor&quot;: &quot;Run&quot;,
+    "JavaScript 调试.output.html.executor": "Run",
-    &quot;JavaScript 调试.regex.md_中文.html.executor&quot;: &quot;Run&quot;,
+    "JavaScript 调试.regex.md_中文.html.executor": "Run",
-    &quot;JavaScript 调试.regex_中文.html.executor&quot;: &quot;Run&quot;,
+    "JavaScript 调试.regex_中文.html.executor": "Run",
-    &quot;JavaScript 调试.test.html.executor&quot;: &quot;Run&quot;,
+    "JavaScript 调试.test.html.executor": "Run",
-    &quot;JavaScript 调试.test2.html.executor&quot;: &quot;Run&quot;,
+    "JavaScript 调试.test2.html.executor": "Run",
-    &quot;JavaScript 调试.test2_英文.html.executor&quot;: &quot;Run&quot;,
+    "JavaScript 调试.test2_英文.html.executor": "Run",
-    &quot;JavaScript 调试.test4-1_中文.html.executor&quot;: &quot;Run&quot;,
+    "JavaScript 调试.test4-1_中文.html.executor": "Run",
-    &quot;JavaScript 调试.互联网认证授权机制.html.executor&quot;: &quot;Run&quot;,
+    "JavaScript 调试.互联网认证授权机制.html.executor": "Run",
-    &quot;JavaScript 调试.互联网认证授权机制_英文.html.executor&quot;: &quot;Run&quot;,
+    "JavaScript 调试.互联网认证授权机制_英文.html.executor": "Run",
-    &quot;JavaScript 调试.毕业论文_英文.html.executor&quot;: &quot;Run&quot;,
+    "JavaScript 调试.毕业论文_英文.html.executor": "Run",
-    &quot;ModuleVcsDetector.initialDetectionPerformed&quot;: &quot;true&quot;,
+    "ModuleVcsDetector.initialDetectionPerformed": "true",
-    &quot;Python 测试.Python 测试 (markdown_mask.py 内).executor&quot;: &quot;Run&quot;,
+    "Python 测试.Python 测试 (markdown_mask.py 内).executor": "Run",
-    &quot;Python 测试.markdown_mask.Test.test_basic_link_masking 的 Python 测试.executor&quot;: &quot;Run&quot;,
+    "Python 测试.markdown_mask.Test.test_basic_link_masking 的 Python 测试.executor": "Run",
-    &quot;Python 测试.pytest (test_html.py 内).executor&quot;: &quot;Run&quot;,
+    "Python 测试.pytest (test_html.py 内).executor": "Run",
-    &quot;Python.2test2 (1).executor&quot;: &quot;Run&quot;,
+    "Python.1test.executor": "Run",
-    &quot;Python.PDFtranslater (1).executor&quot;: &quot;Run&quot;,
+    "Python.2test2 (1).executor": "Run",
-    &quot;Python.PDFtranslater (2).executor&quot;: &quot;Run&quot;,
+    "Python.PDFtranslater (1).executor": "Run",
-    &quot;Python.agent.executor&quot;: &quot;Debug&quot;,
+    "Python.PDFtranslater (2).executor": "Run",
-    &quot;Python.agent_utils.executor&quot;: &quot;Run&quot;,
+    "Python.agent.executor": "Debug",
-    &quot;Python.app (1).executor&quot;: &quot;Run&quot;,
+    "Python.agent_utils.executor": "Run",
-    &quot;Python.app.executor&quot;: &quot;Run&quot;,
+    "Python.app (1).executor": "Run",
-    &quot;Python.app2.executor&quot;: &quot;Run&quot;,
+    "Python.app.executor": "Run",
-    &quot;Python.app_test (1).executor&quot;: &quot;Run&quot;,
+    "Python.app2.executor": "Run",
-    &quot;Python.convert.executor&quot;: &quot;Run&quot;,
+    "Python.app_test (1).executor": "Run",
-    &quot;Python.markdown_splitter.executor&quot;: &quot;Debug&quot;,
+    "Python.convert.executor": "Run",
-    &quot;Python.markdown_utils.executor&quot;: &quot;Run&quot;,
+    "Python.converter_docling.executor": "Run",
-    &quot;Python.test.executor&quot;: &quot;Run&quot;,
+    "Python.converter_mineru.executor": "Run",
-    &quot;Python.test1.executor&quot;: &quot;Run&quot;,
+    "Python.markdown_splitter.executor": "Debug",
-    &quot;Python.test2.executor&quot;: &quot;Run&quot;,
+    "Python.markdown_utils.executor": "Run",
-    &quot;Python.test3.executor&quot;: &quot;Run&quot;,
+    "Python.test.executor": "Run",
-    &quot;Python.test4.executor&quot;: &quot;Run&quot;,
+    "Python.test1.executor": "Run",
-    &quot;Python.testhtml.executor&quot;: &quot;Run&quot;,
+    "Python.test2.executor": "Run",
-    &quot;Python.translater.executor&quot;: &quot;Run&quot;,
+    "Python.test3.executor": "Run",
-    &quot;Python.切分测试.executor&quot;: &quot;Run&quot;,
+    "Python.test4.executor": "Run",
-    &quot;RunOnceActivity.ShowReadmeOnStart&quot;: &quot;true&quot;,
+    "Python.testhtml.executor": "Run",
-    &quot;RunOnceActivity.TerminalTabsStorage.copyFrom.TerminalArrangementManager&quot;: &quot;true&quot;,
+    "Python.translater.executor": "Run",
-    &quot;RunOnceActivity.git.unshallow&quot;: &quot;true&quot;,
+    "Python.切分测试.executor": "Run",
-    &quot;git-widget-placeholder&quot;: &quot;main&quot;,
+    "RunOnceActivity.ShowReadmeOnStart": "true",
-    &quot;last_opened_file_path&quot;: &quot;C:/Users/jxgm/Desktop/FileTranslate/dist/DocuTranslate&quot;,
+    "RunOnceActivity.TerminalTabsStorage.copyFrom.TerminalArrangementManager": "true",
-    &quot;list.type.of.created.stylesheet&quot;: &quot;CSS&quot;,
+    "RunOnceActivity.git.unshallow": "true",
-    &quot;node.js.detected.package.eslint&quot;: &quot;true&quot;,
+    "git-widget-placeholder": "main",
-    &quot;node.js.detected.package.tslint&quot;: &quot;true&quot;,
+    "last_opened_file_path": "C:/Users/jxgm/Desktop/FileTranslate/dist/DocuTranslate",
-    &quot;node.js.selected.package.eslint&quot;: &quot;(autodetect)&quot;,
+    "list.type.of.created.stylesheet": "CSS",
-    &quot;node.js.selected.package.tslint&quot;: &quot;(autodetect)&quot;,
+    "node.js.detected.package.eslint": "true",
-    &quot;nodejs_package_manager_path&quot;: &quot;npm&quot;,
+    "node.js.detected.package.tslint": "true",
-    &quot;settings.editor.selected.configurable&quot;: &quot;preferences.pluginManager&quot;,
+    "node.js.selected.package.eslint": "(autodetect)",
-    &quot;vue.rearranger.settings.migration&quot;: &quot;true&quot;
+    "node.js.selected.package.tslint": "(autodetect)",
+    "nodejs_package_manager_path": "npm",
+    "settings.editor.selected.configurable": "preferences.pluginManager",
+    "vue.rearranger.settings.migration": "true"
   }
-}</component>
+}]]></component>
   <component name="RecentsManager">
     <key name="CopyFile.RECENT_KEYS">
       <recent name="C:\Users\jxgm\Desktop\FileTranslate\dist\DocuTranslate" />
@@ -267,7 +279,27 @@
       <option name="OPTIONS" value="" />
       <method v="2" />
     </configuration>
-    <configuration name="test.html" type="JavascriptDebugType" temporary="true" nameIsGenerated="true" uri="http://localhost:63342/filetranslate/tests/test.html" useBuiltInWebServerPort="true">
+    <configuration name="1test" type="PythonConfigurationType" factoryName="Python" temporary="true" nameIsGenerated="true">
+      <module name="FileTranslate" />
+      <option name="ENV_FILES" value="" />
+      <option name="INTERPRETER_OPTIONS" value="" />
+      <option name="PARENT_ENVS" value="true" />
+      <envs>
+        <env name="PYTHONUNBUFFERED" value="1" />
+      </envs>
+      <option name="SDK_HOME" value="" />
+      <option name="WORKING_DIRECTORY" value="$PROJECT_DIR$/tests" />
+      <option name="IS_MODULE_SDK" value="true" />
+      <option name="ADD_CONTENT_ROOTS" value="true" />
+      <option name="ADD_SOURCE_ROOTS" value="true" />
+      <EXTENSION ID="PythonCoverageRunConfigurationExtension" runner="coverage.py" />
+      <option name="SCRIPT_NAME" value="$PROJECT_DIR$/tests/1test.py" />
+      <option name="PARAMETERS" value="" />
+      <option name="SHOW_COMMAND_LINE" value="false" />
+      <option name="EMULATE_TERMINAL" value="false" />
+      <option name="MODULE_MODE" value="false" />
+      <option name="REDIRECT_INPUT" value="false" />
+      <option name="INPUT_FILE" value="" />
       <method v="2" />
     </configuration>
     <configuration name="2test2 (1)" type="PythonConfigurationType" factoryName="Python" temporary="true" nameIsGenerated="true">
@@ -293,29 +325,6 @@
       <option name="INPUT_FILE" value="" />
       <method v="2" />
     </configuration>
-    <configuration name="3testhtml" type="PythonConfigurationType" factoryName="Python" temporary="true" nameIsGenerated="true">
-      <module name="FileTranslate" />
-      <option name="ENV_FILES" value="" />
-      <option name="INTERPRETER_OPTIONS" value="" />
-      <option name="PARENT_ENVS" value="true" />
-      <envs>
-        <env name="PYTHONUNBUFFERED" value="1" />
-      </envs>
-      <option name="SDK_HOME" value="" />
-      <option name="WORKING_DIRECTORY" value="$PROJECT_DIR$/tests" />
-      <option name="IS_MODULE_SDK" value="true" />
-      <option name="ADD_CONTENT_ROOTS" value="true" />
-      <option name="ADD_SOURCE_ROOTS" value="true" />
-      <EXTENSION ID="PythonCoverageRunConfigurationExtension" runner="coverage.py" />
-      <option name="SCRIPT_NAME" value="C:\Users\jxgm\Desktop\FileTranslate\tests\3testhtml.py" />
-      <option name="PARAMETERS" value="" />
-      <option name="SHOW_COMMAND_LINE" value="false" />
-      <option name="EMULATE_TERMINAL" value="false" />
-      <option name="MODULE_MODE" value="false" />
-      <option name="REDIRECT_INPUT" value="false" />
-      <option name="INPUT_FILE" value="" />
-      <method v="2" />
-    </configuration>
     <configuration default="true" type="PythonConfigurationType" factoryName="Python">
       <module name="filetranslate" />
       <option name="ENV_FILES" value="" />
@@ -362,6 +371,52 @@
       <option name="INPUT_FILE" value="" />
       <method v="2" />
     </configuration>
+    <configuration name="converter_docling" type="PythonConfigurationType" factoryName="Python" temporary="true" nameIsGenerated="true">
+      <module name="FileTranslate" />
+      <option name="ENV_FILES" value="" />
+      <option name="INTERPRETER_OPTIONS" value="" />
+      <option name="PARENT_ENVS" value="true" />
+      <envs>
+        <env name="PYTHONUNBUFFERED" value="1" />
+      </envs>
+      <option name="SDK_HOME" value="" />
+      <option name="WORKING_DIRECTORY" value="$PROJECT_DIR$/docutranslate/converter" />
+      <option name="IS_MODULE_SDK" value="true" />
+      <option name="ADD_CONTENT_ROOTS" value="true" />
+      <option name="ADD_SOURCE_ROOTS" value="true" />
+      <EXTENSION ID="PythonCoverageRunConfigurationExtension" runner="coverage.py" />
+      <option name="SCRIPT_NAME" value="$PROJECT_DIR$/docutranslate/converter/converter_docling.py" />
+      <option name="PARAMETERS" value="" />
+      <option name="SHOW_COMMAND_LINE" value="false" />
+      <option name="EMULATE_TERMINAL" value="false" />
+      <option name="MODULE_MODE" value="false" />
+      <option name="REDIRECT_INPUT" value="false" />
+      <option name="INPUT_FILE" value="" />
+      <method v="2" />
+    </configuration>
+    <configuration name="converter_mineru" type="PythonConfigurationType" factoryName="Python" temporary="true" nameIsGenerated="true">
+      <module name="FileTranslate" />
+      <option name="ENV_FILES" value="" />
+      <option name="INTERPRETER_OPTIONS" value="" />
+      <option name="PARENT_ENVS" value="true" />
+      <envs>
+        <env name="PYTHONUNBUFFERED" value="1" />
+      </envs>
+      <option name="SDK_HOME" value="" />
+      <option name="WORKING_DIRECTORY" value="$PROJECT_DIR$/docutranslate/converter" />
+      <option name="IS_MODULE_SDK" value="true" />
+      <option name="ADD_CONTENT_ROOTS" value="true" />
+      <option name="ADD_SOURCE_ROOTS" value="true" />
+      <EXTENSION ID="PythonCoverageRunConfigurationExtension" runner="coverage.py" />
+      <option name="SCRIPT_NAME" value="$PROJECT_DIR$/docutranslate/converter/converter_mineru.py" />
+      <option name="PARAMETERS" value="" />
+      <option name="SHOW_COMMAND_LINE" value="false" />
+      <option name="EMULATE_TERMINAL" value="false" />
+      <option name="MODULE_MODE" value="false" />
+      <option name="REDIRECT_INPUT" value="false" />
+      <option name="INPUT_FILE" value="" />
+      <method v="2" />
+    </configuration>
     <configuration default="true" type="Python.FlaskServer">
       <module name="filetranslate" />
       <option name="ENV_FILES" value="" />
@@ -459,31 +514,13 @@
       <option name="USE_PATTERN" value="false" />
       <method v="2" />
     </configuration>
-    <configuration name="pytest (3testhtml.py 内)" type="tests" factoryName="py.test" temporary="true" nameIsGenerated="true">
-      <module name="FileTranslate" />
-      <option name="ENV_FILES" value="" />
-      <option name="INTERPRETER_OPTIONS" value="" />
-      <option name="PARENT_ENVS" value="true" />
-      <option name="SDK_HOME" value="" />
-      <option name="WORKING_DIRECTORY" value="$PROJECT_DIR$/tests" />
-      <option name="IS_MODULE_SDK" value="true" />
-      <option name="ADD_CONTENT_ROOTS" value="true" />
-      <option name="ADD_SOURCE_ROOTS" value="true" />
-      <EXTENSION ID="PythonCoverageRunConfigurationExtension" runner="coverage.py" />
-      <option name="_new_keywords" value="&quot;&quot;" />
-      <option name="_new_parameters" value="&quot;&quot;" />
-      <option name="_new_additionalArguments" value="&quot;&quot;" />
-      <option name="_new_target" value="&quot;$PROJECT_DIR$/tests/3testhtml.py&quot;" />
-      <option name="_new_targetType" value="&quot;PATH&quot;" />
-      <method v="2" />
-    </configuration>
     <recent_temporary>
       <list>
         <item itemvalue="Python.app_test (1)" />
+        <item itemvalue="Python.1test" />
+        <item itemvalue="Python.converter_docling" />
+        <item itemvalue="Python.converter_mineru" />
         <item itemvalue="Python.2test2 (1)" />
-        <item itemvalue="Python.3testhtml" />
-        <item itemvalue="JavaScript 调试.test.html" />
-        <item itemvalue="Python 测试.pytest (3testhtml.py 内)" />
       </list>
     </recent_temporary>
   </component>
@@ -568,14 +605,26 @@
       <workItem from="1747628254543" duration="7347000" />
       <workItem from="1747635705571" duration="391000" />
       <workItem from="1747650908714" duration="121000" />
+      <workItem from="1747731199599" duration="4260000" />
     </task>
     <servers />
   </component>
   <component name="TypeScriptGeneratedFilesManager">
     <option name="version" value="3" />
   </component>
+  <component name="Vcs.Log.Tabs.Properties">
+    <option name="TAB_STATES">
+      <map>
+        <entry key="MAIN">
+          <value>
+            <State />
+          </value>
+        </entry>
+      </map>
+    </option>
+  </component>
   <component name="com.intellij.coverage.CoverageDataManagerImpl">
-    <SUITE FILE_PATH="coverage/filetranslate$app_test__1_.coverage" NAME="app_test (1) 覆盖结果" MODIFIED="1747634037187" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/tests" />
+    <SUITE FILE_PATH="coverage/filetranslate$app_test__1_.coverage" NAME="app_test (1) 覆盖结果" MODIFIED="1747733748258" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/tests" />
     <SUITE FILE_PATH="coverage/filetranslate$test.coverage" NAME="test 覆盖结果" MODIFIED="1747472297913" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/tests" />
     <SUITE FILE_PATH="coverage/filetranslate$convert.coverage" NAME="convert 覆盖结果" MODIFIED="1746963490689" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/docutranslate/utils" />
     <SUITE FILE_PATH="coverage/PDFtranslate$PDFtranslater__1_.coverage" NAME="PDFtranslater (1) 覆盖结果" MODIFIED="1746633258205" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/pdftranslate_packages" />
@@ -583,14 +632,17 @@
     <SUITE FILE_PATH="coverage/PDFtranslate$agent_utils.coverage" NAME="agent_utils 覆盖结果" MODIFIED="1746617703678" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/pdftranslate_packages/utils" />
     <SUITE FILE_PATH="coverage/filetranslate$app2.coverage" NAME="app2 覆盖结果" MODIFIED="1747108180309" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/docutranslate" />
     <SUITE FILE_PATH="coverage/filetranslate$app.coverage" NAME="app 覆盖结果" MODIFIED="1747448464521" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/docutranslate" />
+    <SUITE FILE_PATH="coverage/filetranslate$converter_mineru.coverage" NAME="converter_mineru 覆盖结果" MODIFIED="1747726229881" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/docutranslate/converter" />
     <SUITE FILE_PATH="coverage/PDFtranslate$markdown_splitter.coverage" NAME="markdown_splitter 覆盖结果" MODIFIED="1746599883603" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/pdftranslate_packages/utils" />
+    <SUITE FILE_PATH="coverage/filetranslate$converter_docling.coverage" NAME="converter_docling 覆盖结果" MODIFIED="1747726654277" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/docutranslate/converter" />
+    <SUITE FILE_PATH="coverage/filetranslate$1test.coverage" NAME="1test 覆盖结果" MODIFIED="1747732504752" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/tests" />
     <SUITE FILE_PATH="coverage/filetranslate$test3.coverage" NAME="test3 覆盖结果" MODIFIED="1746884110572" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/tests" />
     <SUITE FILE_PATH="coverage/filetranslate$app__1_.coverage" NAME="app (1) 覆盖结果" MODIFIED="1747136094477" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/tests" />
     <SUITE FILE_PATH="coverage/filetranslate$markdown_splitter.coverage" NAME="markdown_splitter 覆盖结果" MODIFIED="1746805063874" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/docutranslate/utils" />
     <SUITE FILE_PATH="coverage/filetranslate$agent.coverage" NAME="agent 覆盖结果" MODIFIED="1746805293987" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/docutranslate/Agents" />
     <SUITE FILE_PATH="coverage/filetranslate$PDFtranslater__2_.coverage" NAME="PDFtranslater (2) 覆盖结果" MODIFIED="1746679546680" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/filetranslate_packages" />
     <SUITE FILE_PATH="coverage/PDFtranslate$test.coverage" NAME="test 覆盖结果" MODIFIED="1746629433597" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/tests" />
-    <SUITE FILE_PATH="coverage/filetranslate$2test2__1_.coverage" NAME="2test2 (1) 覆盖结果" MODIFIED="1747579915531" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/tests" />
+    <SUITE FILE_PATH="coverage/filetranslate$2test2__1_.coverage" NAME="2test2 (1) 覆盖结果" MODIFIED="1747722801777" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/tests" />
     <SUITE FILE_PATH="coverage/filetranslate$agent_utils.coverage" NAME="agent_utils 覆盖结果" MODIFIED="1746708534311" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/docutranslate/utils" />
     <SUITE FILE_PATH="coverage/filetranslate$.coverage" NAME="切分测试 覆盖结果" MODIFIED="1747187128847" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/tests" />
     <SUITE FILE_PATH="coverage/filetranslate$test1.coverage" NAME="test1 覆盖结果" MODIFIED="1746936018440" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/tests" />
@@ -601,6 +653,7 @@
     <SUITE FILE_PATH="coverage/PDFtranslate$markdown_utils.coverage" NAME="markdown_utils 覆盖结果" MODIFIED="1746598797872" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/pdftranslate_packages/utils" />
     <SUITE FILE_PATH="coverage/filetranslate$test2.coverage" NAME="test2 覆盖结果" MODIFIED="1747553897731" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/tests" />
     <SUITE FILE_PATH="coverage/filetranslate$translater.coverage" NAME="translater 覆盖结果" MODIFIED="1746843159560" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/docutranslate" />
+    <SUITE FILE_PATH="coverage/filetranslate$docling_utils.coverage" NAME="docling_utils 覆盖结果" MODIFIED="1747710836730" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/docutranslate/utils" />
     <SUITE FILE_PATH="coverage/PDFtranslate$.coverage" NAME=" 覆盖结果" MODIFIED="1746588350286" SOURCE_PROVIDER="com.intellij.coverage.DefaultCoverageFileProvider" RUNNER="coverage.py" COVERAGE_BY_TEST_ENABLED="false" COVERAGE_TRACING_ENABLED="false" WORKING_DIRECTORY="$PROJECT_DIR$/decorator" />
   </component>
 </project>

README.md

@@ -4,12 +4,15 @@
 
 [![image](https://img.shields.io/badge/github-DocuTranslate-blue)](https://github.com/xunbu/docutranslate)
 
-文件翻译工具，借助[docling](https://github.com/docling-project/docling)与大语言模型实现多种格式文件的翻译
+文件翻译工具，借助[docling](https://github.com/docling-project/docling)、[minerU](https://mineru.net/)与大语言模型实现多种格式文件的翻译
+
+> QQ交流群：1047781902
 
 # 整合包
 
 对于只使用基本翻译功能的用户，可以在[github releases](https://github.com/xunbu/docutranslate/releases)
 上下载最新的整合包，该整合包点击即用，您所需的只是获取某个ai平台的api-key。
+以及可以在mineru申请token进行pdf识别【可选】
 
 # 安装
 
@@ -34,7 +37,16 @@
 
 # 前置条件
 
-## huggingface换源
+本翻译工具的翻译流程总体如下：
+
+1. 使用文本转换引擎将文档转换成markdown（有docling（本地）、minerU（联网）两种引擎）
+2. 使用大语言模型翻译markdown文本（需要申请api-key或本地部署）
+
+## 使用docling引擎注意事项
+
+使用docling将文档转换为markdown时，需要下载模型到本地（也可以提前下载，见FAQ），因此可能会遇到一些网络问题
+
+### huggingface换源
 
 > 不能科学上网的友友注意了
 
@@ -43,12 +55,12 @@
 - 第一次读取非markdown文本
 - 第一次使用公式识别或代码识别功能
 
-### 方法1
+#### 方法1
 
 设置电脑的环境变量(记得设置后重启IDE)  
 `HF_ENDPOINT=https://hf-mirror.com`
 
-### 方法2
+#### 方法2
 
 在代码开头设置环境变量
 
@@ -60,6 +72,13 @@ os.environ['HF_ENDPOINT'] = 'https://hf-mirror.com'
 ###其余代码写在下方
 ```
 
+## 使用minerU引擎注意事项
+
+使用minerU将文档转换为markdown时，需要在minerU平台申请token
+
+1. 打开[minerU官网](https://mineru.net/apiManage/docs)申请token
+2. 申请成功后，在[API Token管理界面](https://mineru.net/apiManage/token)创建API Token
+
 ## 获取大模型平台的baseurl、key、model-id
 
 由于需要使用大语言模型进行markdown调整与翻译，所以需要预先获取模型的baseurl、key、model-id  
@@ -90,7 +109,12 @@ from docutranslate.translater import FileTranslater
 
 translater = FileTranslater(base_url="<baseurl>",
                             key="<key>",
-                            model_id="<model-id>")
+                            model_id="<model-id>",
+                            convert_engin="docling"  # 默认使用docling
+                            # convert_engin="mineru",# 使用mineru
+                            # mineru_token="<申请的mineru_token>"#使用mineru时必填
+                            )
+
 # 不开启公式、代码识别（默认输出为markdown文件）
 translater.translate_file("<文件路径>", to_lang="中文")
 
@@ -141,12 +165,14 @@ translater.read_file("<文件路径>").save_as_markdown()
 from docutranslate import FileTranslater
 
 translater = FileTranslater(base_url="<baseurl>",  # 默认的模型baseurl
-                            key="<key>",  # 默认的模型api-key
+                            key="<api-key>",  # 默认的大语言模型平台api-key
                             model_id="<model-id>",  # 默认的模型id
                             chunksize=2000,  # markdown分块长度（单位byte），分块越大效果越好（也越慢），不建议超过8000
                             max_concurrent=20,  # 并发数，受到ai平台并发量限制，如果文章很长建议适当加大到20以上
-                            docling_artifact=None,  # 使用提前下载好的docling模型
                             timeout=2000,  # 调用api的超时时间
+                            docling_artifact=None,  # 使用提前下载好的docling模型
+                            convert_engin="mineru",  # 可选docling或minerU
+                            mineru_token="<mineru-token>",  # minerU的token，使用minerU时必填
                             tips=True  # 开场提示
                             )
 
@@ -206,7 +232,8 @@ from docutranslate.utils.docling_utils import get_docling_artifacts
 print(get_docling_artifacts())  # 会显示模型下载文件夹，通常在`C:\Users\<user>\.cache\docling\models`
 ```
 
-> 创建FileTranslater时携带模型文件夹即可
+> 将模型文件夹命名为docling_artifact放置在项目下
+> 或创建FileTranslater时docling_artifact参数设置为文件夹位置
 
 ```python
 from docutranslate import FileTranslater

docutranslate/app.py

@@ -11,7 +11,7 @@ from fastapi import FastAPI, File, Form, UploadFile, Request, HTTPException
 from fastapi.responses import HTMLResponse, JSONResponse, StreamingResponse,FileResponse
 from fastapi.templating import Jinja2Templates
 from fastapi.staticfiles import StaticFiles
-from docutranslate import FileTranslater
+from docutranslate import FileTranslater # Assuming FileTranslater is in docutranslate module
 from docutranslate.logger import translater_logger
 from docutranslate.utils.resource_utils import resource_path
 
@@ -19,11 +19,10 @@ app = FastAPI()
 
 STATIC_DIR=resource_path("static")
 
-# print(f"__file__:{Path(__file__).resolve()}")
 app.mount("/static",StaticFiles(directory=STATIC_DIR), name="static")
 
 # --- 全局配置 ---
-log_queue: Optional[asyncio.Queue] = None  # Will be initialized in startup_event
+log_queue: Optional[asyncio.Queue] = None
 current_state: Dict[str, Any] = {
     "is_processing": False,
     "status_message": "空闲",
@@ -36,9 +35,9 @@ current_state: Dict[str, Any] = {
     "task_end_time": 0,
     "current_task_ref": None,
 }
-templates = Jinja2Templates(directory=".")
+templates = Jinja2Templates(directory=".") # Not strictly used if index.html is served as FileResponse
-MAX_LOG_HISTORY = 200  # Max items for the persistent log_history list
+MAX_LOG_HISTORY = 200
-log_history: List[str] = []  # Keeps a longer history, not directly for "unread"
+log_history: List[str] = []
 
 
 # --- 日志处理器 ---
@@ -51,25 +50,20 @@ class QueueAndHistoryHandler(logging.Handler):
 
     def emit(self, record: logging.LogRecord):
         log_entry = self.format(record)
-
+        print(log_entry) # Keep console log for server visibility
-        # Add to the persistent history (capped)
         self.history_list.append(log_entry)
         if len(self.history_list) > self.max_history:
             del self.history_list[:len(self.history_list) - self.max_history]
 
-        # Add to the "unread" queue for frontend consumption
-        try:
-            # Ensure self.queue is not None (it's initialized at startup)
         if self.queue is not None:
+            try:
                 main_loop = getattr(app.state, "main_event_loop", None)
                 if main_loop and main_loop.is_running():
                     main_loop.call_soon_threadsafe(self.queue.put_nowait, log_entry)
                 else:
-                    self.queue.put_nowait(log_entry)  # Fallback
+                    self.queue.put_nowait(log_entry)
-            else:
-                print(f"CRITICAL: Log queue not initialized. Log: {log_entry}")
             except asyncio.QueueFull:
-            print(f"Log queue is full. Log dropped: {log_entry}")  # Or handle differently
+                print(f"Log queue is full. Log dropped: {log_entry}")
             except Exception as e:
                 print(f"Error putting log to queue: {e}. Log: {log_entry}")
 
@@ -79,7 +73,7 @@ class QueueAndHistoryHandler(logging.Handler):
 async def startup_event():
     global log_queue
     app.state.main_event_loop = asyncio.get_running_loop()
-    log_queue = asyncio.Queue()  # Initialize the global log_queue
+    log_queue = asyncio.Queue()
 
     for handler in translater_logger.handlers[:]:
         translater_logger.removeHandler(handler)
@@ -93,7 +87,7 @@ async def startup_event():
     translater_logger.setLevel(logging.INFO)
 
     log_history.clear()
-    while not log_queue.empty():  # Clear queue just in case
+    while not log_queue.empty():
         try:
             log_queue.get_nowait()
         except asyncio.QueueEmpty:
@@ -112,6 +106,7 @@ async def _perform_translation(params: Dict[str, Any], file_contents: bytes, ori
     try:
         translater_logger.info(f"使用 Base URL: {params['base_url']}, Model: {params['model_id']}")
         translater_logger.info(f"文件大小: {len(file_contents)} 字节。目标语言: {params['to_lang']}")
+        translater_logger.info(f"使用转换引擎: {params['convert_engin']}")
         translater_logger.info(
             f"选项 - 公式: {params['formula_ocr']}, 代码: {params['code_ocr']}, 修正: {params['refine_markdown']}")
 
@@ -119,7 +114,9 @@ async def _perform_translation(params: Dict[str, Any], file_contents: bytes, ori
             base_url=params['base_url'],
             key=params['apikey'],
             model_id=params['model_id'],
-            tips=False
+            convert_engin=params['convert_engin'],
+            mineru_token=params['mineru_token'],
+            tips=False  # Assuming tips are not needed for server-side processing
         )
         await ft.translate_bytes_async(
             name=original_filename,
@@ -152,7 +149,7 @@ async def _perform_translation(params: Dict[str, Any], file_contents: bytes, ori
         translater_logger.info(f"翻译任务 '{original_filename}' 已被取消 (用时 {duration:.2f} 秒).")
         current_state.update({
             "status_message": f"翻译任务已取消（若有转换任务仍会后台进行） (用时 {duration:.2f} 秒).",
-            "error_flag": False,  # Cancellation is not an error in this context
+            "error_flag": False,
             "download_ready": False,
             "markdown_content": None,
             "html_content": None,
@@ -180,11 +177,25 @@ async def _perform_translation(params: Dict[str, Any], file_contents: bytes, ori
 # --- API Endpoints ---
 @app.get("/", response_class=HTMLResponse)
 async def main_page(request: Request):
-    return FileResponse(STATIC_DIR/"index.html")
+    # Serve index.html from the static directory or root project directory
+    # Assuming index.html is at the same level as app.py or in STATIC_DIR
+    # For simplicity, if index.html is at root:
+    # return FileResponse(Path(__file__).parent / "index.html")
+    # If using Jinja2Templates and index.html is in "templates" folder:
+    # return templates.TemplateResponse("index.html", {"request": request})
+    # Using FileResponse for index.html directly:
+    index_path = Path("index.html") # Adjust if index.html is elsewhere
+    if not index_path.exists():
+         # Fallback to static dir if not in root
+        index_path = STATIC_DIR / "index.html"
+        if not index_path.exists():
+            raise HTTPException(status_code=404, detail="index.html not found")
+    return FileResponse(index_path)
 
 
 @app.post("/translate")
 async def handle_translate(
+        request: Request, # Added request for potential future use, not strictly needed now
         base_url: str = Form(...),
         apikey: str = Form(...),
         model_id: str = Form(...),
@@ -192,6 +203,8 @@ async def handle_translate(
         formula_ocr: bool = Form(False),
         code_ocr: bool = Form(False),
         refine_markdown: bool = Form(False),
+        convert_engin: str = Form(...), # New parameter
+        mineru_token: Optional[str] = Form(None), # New parameter
         file: UploadFile = File(...)
 ):
     global current_state, log_queue, log_history
@@ -209,6 +222,12 @@ async def handle_translate(
             content={"task_started": False, "message": "没有选择文件或文件无效。"}
         )
 
+    if convert_engin == "mineru" and (not mineru_token or not mineru_token.strip()):
+        return JSONResponse(
+            status_code=400,
+            content={"task_started": False, "message": "使用 Mineru 引擎时必须提供有效的 Mineru Token。"}
+        )
+
     current_state["is_processing"] = True
     original_filename_for_init = file.filename or "uploaded_file"
 
@@ -224,26 +243,22 @@ async def handle_translate(
         "current_task_ref": None,
     })
 
-    # Clear logs for the new task
     log_history.clear()
-    if log_queue:  # Ensure log_queue is initialized
+    if log_queue:
         while not log_queue.empty():
             try:
                 log_queue.get_nowait()
             except asyncio.QueueEmpty:
                 break
 
-    # Add initial log entry for the new task
-    # We create a LogRecord manually to ensure it goes through the formatter and handler
     initial_log_msg = f"收到新的翻译请求: {original_filename_for_init}"
     if translater_logger.handlers and isinstance(translater_logger.handlers[0], QueueAndHistoryHandler):
-        # Use the existing handler to format and queue/store the log
         record = logging.LogRecord(
             name=translater_logger.name, level=logging.INFO, pathname="", lineno=0,
             msg=initial_log_msg, args=(), exc_info=None, func=""
         )
-        translater_logger.handlers[0].emit(record)  # This will add to both queue and history
+        translater_logger.handlers[0].emit(record)
-    else:  # Fallback if handler setup is unusual
+    else:
         translater_logger.info(initial_log_msg)
 
     try:
@@ -255,6 +270,8 @@ async def handle_translate(
             "base_url": base_url, "apikey": apikey, "model_id": model_id,
             "to_lang": to_lang, "formula_ocr": formula_ocr,
             "code_ocr": code_ocr, "refine_markdown": refine_markdown,
+            "convert_engin": convert_engin, # Pass to task
+            "mineru_token": mineru_token,   # Pass to task
         }
 
         loop = asyncio.get_running_loop()
@@ -332,18 +349,17 @@ async def get_status():
 
 
 @app.get("/get-logs")
-async def get_logs_from_queue():  # Renamed for clarity, though path is the same
+async def get_logs_from_queue():
     global log_queue
     new_logs = []
-    if log_queue:  # Ensure log_queue is initialized
+    if log_queue:
         while not log_queue.empty():
             try:
-                log_entry = log_queue.get_nowait()  # Consume from queue
+                log_entry = log_queue.get_nowait()
                 new_logs.append(log_entry)
-                log_queue.task_done()  # Important for queue management if using join() elsewhere
+                log_queue.task_done()
             except asyncio.QueueEmpty:
                 break
-                # No total_count, as the frontend just appends what it receives
     return JSONResponse(content={"logs": new_logs})
 
 
@@ -384,10 +400,11 @@ async def download_html(filename_with_ext: str):
 
 
 def run_app():
-    print("正在启动 DocuTranslate")
+    print("正在启动 DocuTranslate WebUI")
     print("请访问 http://127.0.0.1:8010")
     uvicorn.run(app, host="127.0.0.1", port=8010, workers=1)
 
 
 if __name__ == "__main__":
+
     run_app()

docutranslate/converter/__init__.py

@@ -0,0 +1,3 @@
+from .converter import Document,Converter
+from .converter_mineru import ConverterMineru
+from .converter_docling import ConverterDocling

docutranslate/converter/converter.py

@@ -0,0 +1,25 @@
+from typing import Protocol
+from pathlib import Path
+
+
+class Document:
+    def __init__(self,path:Path|str=None,filename:str=None,filebytes:bytes=None):
+        if path is None and (filename is None or filebytes is None):
+            raise Exception("Document的路径或filename、filebytes不能同时为空")
+        self.filebytes = filebytes
+        self.filename = filename
+        self.path = path
+        if path:
+            if isinstance(path,str):
+                path=Path(path)
+            self.path=path
+            self.filename=path.name
+            self.filebytes=path.read_bytes()
+
+class Converter(Protocol):
+    #转换为markdown
+    def convert(self,document:Document)->str:
+        ...
+
+    async def convert_async(self,document:Document)->str:
+        ...

docutranslate/converter/converter_docling.py

@@ -0,0 +1,80 @@
+import os
+import time
+from io import BytesIO
+from pathlib import Path
+
+from docling.datamodel.base_models import InputFormat
+from docling.datamodel.document import DocumentStream
+from docling.datamodel.pipeline_options import PdfPipelineOptions
+from docling.datamodel.settings import settings
+from docling.document_converter import DocumentConverter, PdfFormatOption
+from docling_core.types.doc import ImageRefMode
+from huggingface_hub.errors import LocalEntryNotFoundError
+
+from docutranslate.logger import translater_logger
+
+from docutranslate.converter import Converter, Document
+
+import asyncio
+
+IMAGE_RESOLUTION_SCALE = 4
+
+
+def file2markdown_embed_images(file_path: Path | str | DocumentStream, formula=False, code=False,
+                               artifacts_path: Path | str | None = None) -> str:
+    pipeline_options = PdfPipelineOptions(artifacts_path=artifacts_path)
+    pipeline_options.do_ocr = False
+    pipeline_options.images_scale = IMAGE_RESOLUTION_SCALE
+    pipeline_options.generate_picture_images = True
+    # pipeline_options.table_structure_options.mode = TableFormerMode.FAST
+    pipeline_options.table_structure_options.do_cell_matching = False
+    if formula:
+        pipeline_options.do_formula_enrichment = True
+    if code:
+        pipeline_options.do_code_enrichment = True
+    # pipeline_options.accelerator_options= AcceleratorOptions(
+    #     num_threads=4, device=AcceleratorDevice.AUTO
+    # )
+    # 打印时间
+    settings.debug.profile_pipeline_timings = True
+    converter = DocumentConverter(format_options={
+        InputFormat.PDF: PdfFormatOption(pipeline_options=pipeline_options)
+
+    })
+    try:
+        conversion_result = converter.convert(file_path)
+        result = conversion_result.document.export_to_markdown(image_mode=ImageRefMode.EMBEDDED)
+    except LocalEntryNotFoundError:
+        translater_logger.info(f"无法连接huggingface，正在尝试换源")
+        os.environ['HF_ENDPOINT'] = 'https://hf-mirror.com'
+        conversion_result = converter.convert(file_path)
+        result = conversion_result.document.export_to_markdown(image_mode=ImageRefMode.EMBEDDED)
+        # translater_logger.info(f"docling转换耗时: {conversion_result.timings["pipeline_total"].times}")
+    return result
+
+
+class ConverterDocling(Converter):
+    def __init__(self, code=True, formula=True, artifact=None):
+        self.code = code
+        self.formula = formula
+        self.artifact = artifact
+
+    def convert(self, document):
+        assert isinstance(document.filename, str)
+        translater_logger.info(f"正在将文档转换为markdown")
+        time1 = time.time()
+        document_stream = DocumentStream(name=document.filename, stream=BytesIO(document.filebytes))
+        result = file2markdown_embed_images(document_stream, formula=self.formula, code=self.code,
+                                            artifacts_path=self.artifact)
+        translater_logger.info(f"已转换为markdown，耗时{time.time() - time1}秒")
+        return result
+
+    async def convert_async(self, document: Document) -> str:
+        return await asyncio.to_thread(
+            self.convert,
+            document
+        )
+
+
+if __name__ == '__main__':
+    pass

docutranslate/converter/converter_mineru.py

@@ -0,0 +1,220 @@
+import asyncio
+import base64
+import io
+import mimetypes
+import os
+import re
+import time
+import zipfile
+import httpx
+from docutranslate.converter import Converter, Document
+from docutranslate.logger import translater_logger
+
+URL = 'https://mineru.net/api/v4/file-urls/batch'
+
+client=httpx.Client(trust_env=False)
+
+#TODO: 提供更详细的logger
+class ConverterMineru(Converter):
+    def __init__(self, token: str, formula=True):
+        self.mineru_token = token.strip()
+        self.client_async = httpx.AsyncClient()
+        self.formula = formula
+
+    def _get_header(self):
+        return {
+            'Content-Type': 'application/json',
+            "Authorization": f"Bearer {self.mineru_token}"
+        }
+
+    def _get_upload_data(self, document: Document):
+        return {
+            "enable_formula": self.formula,
+            "language": "auto",
+            "enable_table": True,
+            "files": [
+                {"name": f"{document.filename}", "is_ocr": True}
+            ]
+        }
+
+    def upload(self, document: Document):
+        # 获取上传链接
+        response = client.post(URL, headers=self._get_header(), json=self._get_upload_data(document))
+        response.raise_for_status()
+        result = response.json()
+        # print('response success. result:{}'.format(result))
+        if result["code"] == 0:
+            batch_id = result["data"]["batch_id"]
+            urls = result["data"]["file_urls"]
+            # print('batch_id:{},urls:{}'.format(batch_id, urls))
+            # 获取
+            res_upload = client.put(urls[0], content=document.filebytes)
+            res_upload.raise_for_status()
+            # print(f"{urls[0]} upload success")
+            return batch_id
+        else:
+            raise Exception('apply upload url failed,reason:{}'.format(result.msg))
+
+    def get_file_url(self, batch_id: str) -> str:
+        while True:
+            url = f'https://mineru.net/api/v4/extract-results/batch/{batch_id}'
+            header = self._get_header()
+            res = client.get(url, headers=header)
+            res.raise_for_status()
+            fileinfo = res.json()["data"]["extract_result"][0]
+            if fileinfo["state"] == "done":
+                fileurl = fileinfo["full_zip_url"]
+                return fileurl
+            else:
+                time.sleep(3)
+
+    def convert(self, document: Document) -> str:
+        translater_logger.info(f"正在将文档转换为markdown")
+        time1=time.time()
+        batch_id = self.upload(document)
+        file_url = self.get_file_url(batch_id)
+        result=get_md_from_zip_url_with_inline_images(zip_url=file_url)
+        translater_logger.info(f"已转换为markdown，耗时{time.time()-time1}秒")
+        return result
+
+    # TODO: 实现细粒度更高的协程
+    async def convert_async(self, document: Document) -> str:
+        # 待优化
+        return await asyncio.to_thread(
+            self.convert,
+            document
+        )
+
+
+def get_md_from_zip_url_with_inline_images(
+        zip_url: str,
+        filename_in_zip: str = "full.md",
+        encoding: str = "utf-8"
+) -> str | None:
+    """
+    从给定的ZIP文件URL中下载并提取指定文件的内容，
+    并将Markdown文件中的相对路径图片转换为内联Base64图片。
+
+    Args:
+        zip_url (str): ZIP文件的下载链接。
+        filename_in_zip (str): ZIP压缩包内目标Markdown文件的名称（包括路径）。
+                               默认为 "full.md"。
+        encoding (str): 目标文件的预期编码。默认为 "utf-8"。
+
+    Returns:
+        str | None: 如果成功，返回处理后的Markdown文本内容；否则返回 None。
+    """
+    try:
+        print(f"正在从 {zip_url} 下载ZIP文件 (使用 httpx.get)...")
+        response = client.get(zip_url, timeout=60.0)  # 增加超时
+        response.raise_for_status()
+        print("ZIP文件下载完成。")
+
+        zip_file_bytes = io.BytesIO(response.content)
+
+        print(f"正在尝试打开内存中的ZIP存档...")
+        with zipfile.ZipFile(zip_file_bytes, 'r') as archive:
+            print(f"ZIP存档已打开。正在查找文件 '{filename_in_zip}'...")
+
+            if filename_in_zip not in archive.namelist():
+                print(f"错误: 文件 '{filename_in_zip}' 在ZIP压缩包中未找到。")
+                print(f"压缩包中的可用文件列表: {archive.namelist()}")
+                return None
+
+            md_content_bytes = archive.read(filename_in_zip)
+            print(f"文件 '{filename_in_zip}' 已找到并读取。")
+            md_content_text = md_content_bytes.decode(encoding)
+            print(f"文件内容已使用 '{encoding}' 编码成功解码。")
+
+            # --- 新增：处理图片 ---
+            print("开始处理Markdown中的图片...")
+            # 获取Markdown文件在ZIP包内的基本目录，用于解析相对图片路径
+            # 例如，如果 filename_in_zip 是 "docs/guide/full.md", base_md_path_in_zip 是 "docs/guide"
+            # 如果 filename_in_zip 是 "full.md", base_md_path_in_zip 是 ""
+            base_md_path_in_zip = os.path.dirname(filename_in_zip)
+
+            def replace_image_with_base64(match):
+                alt_text = match.group(1)
+                original_image_path = match.group(2)
+
+                # 检查是否是外部链接或已经是data URI
+                if original_image_path.startswith(('http://', 'https://', 'data:')):
+                    print(f"  跳过外部或已内联图片: {original_image_path}")
+                    return match.group(0)  # 返回原始匹配
+
+                # 构建图片在ZIP文件中的绝对路径
+                # os.path.join 会正确处理 base_md_path_in_zip 为空字符串的情况
+                image_path_in_zip = os.path.join(base_md_path_in_zip, original_image_path)
+                # zipfile 使用正斜杠，并且路径是相对于zip根目录的，os.path.normpath确保路径格式正确
+                image_path_in_zip = os.path.normpath(image_path_in_zip).replace(os.sep, '/')
+
+                # 确保路径不是以 './' 开头，如果filename_in_zip在根目录且图片路径也是相对的
+                if image_path_in_zip.startswith('./'):
+                    image_path_in_zip = image_path_in_zip[2:]
+
+                # print(f"  尝试内联图片: '{original_image_path}' (解析为ZIP内路径: '{image_path_in_zip}')")
+
+                try:
+                    image_bytes = archive.read(image_path_in_zip)
+
+                    # 猜测MIME类型
+                    mime_type, _ = mimetypes.guess_type(image_path_in_zip)
+                    if not mime_type:
+                        # 备用：根据扩展名手动判断一些常见类型
+                        ext = os.path.splitext(image_path_in_zip)[1].lower()
+                        if ext == '.png':
+                            mime_type = 'image/png'
+                        elif ext in ['.jpg', '.jpeg']:
+                            mime_type = 'image/jpeg'
+                        elif ext == '.gif':
+                            mime_type = 'image/gif'
+                        elif ext == '.svg':
+                            mime_type = 'image/svg+xml'
+                        elif ext == '.webp':
+                            mime_type = 'image/webp'
+                        else:
+                            print(f"    警告: 无法确定图片 '{image_path_in_zip}' 的MIME类型。跳过内联。")
+                            return match.group(0)  # 返回原始匹配
+
+                    base64_encoded_data = base64.b64encode(image_bytes).decode('utf-8')
+                    new_image_tag = f"![{alt_text}](data:{mime_type};base64,{base64_encoded_data})"
+                    # print(f"    成功内联图片: {original_image_path} -> data:{mime_type[:20]}...")
+                    return new_image_tag
+                except KeyError:
+                    print(f"    警告: 图片 '{image_path_in_zip}' 在ZIP压缩包中未找到。原始链接将被保留。")
+                    return match.group(0)  # 图片不在zip中，返回原始匹配
+                except Exception as e_img:
+                    print(f"    错误: 处理图片 '{image_path_in_zip}' 时发生错误: {e_img}。原始链接将被保留。")
+                    return match.group(0)
+
+            # 正则表达式查找Markdown图片: ![alt text](path/to/image.ext)
+            # 修改了正则表达式，使其不贪婪地匹配alt文本和路径
+            image_regex = r"!\[(.*?)\]\((.*?)\)"
+            modified_md_content = re.sub(image_regex, replace_image_with_base64, md_content_text)
+
+            print("图片处理完成。")
+            return modified_md_content
+
+    except httpx.HTTPStatusError as e:
+        print(f"HTTP 错误 (httpx): {e.response.status_code} - {e.request.url}")
+        print(f"响应内容: {e.response.text[:200]}...")
+        return None
+    except httpx.RequestError as e:
+        print(f"下载ZIP文件时发生错误 (httpx): {e}")
+        return None
+    except zipfile.BadZipFile:
+        print("错误: 下载的文件不是一个有效的ZIP压缩文件或已损坏。")
+        return None
+    except UnicodeDecodeError:
+        print(f"错误: 无法使用 '{encoding}' 编码解码文件 '{filename_in_zip}' 的内容。")
+        print("请尝试其他编码，如 'gbk', 'latin1' 等，或确认文件本身的编码。")
+        return None
+    except Exception as e:
+        print(f"发生未知错误: {e}")
+        import traceback
+        traceback.print_exc()  # 打印完整的堆栈跟踪，便于调试
+        return None
+
+
+if __name__ == '__main__':
+    pass

docutranslate/static/index.html

@@ -28,11 +28,11 @@
         }
 
         .error-message {
-            color: #d32f2f;
+            color: #d32f2f; /* Pico invalid color */
         }
 
         .success-message {
-            color: #2e7d32;
+            color: #2e7d32; /* Pico valid color */
         }
 
         .form-group {
@@ -65,9 +65,15 @@
         .checkbox-group {
             display: flex;
             flex-wrap: wrap;
+            gap: 1rem; /* Added gap for better spacing */
             margin-bottom: 1rem;
         }
 
+        .checkbox-group label { /* Ensure checkboxes are aligned */
+            margin-right: 10px;
+        }
+
+
         #resultArea {
             margin-top: 1.5rem;
             padding-top: 1rem;
@@ -116,7 +122,6 @@
             display: none;
         }
 
-        /* Styles for drag and drop area */
         #fileDropArea {
             border: 2px dashed #ccc;
             padding: 20px;
@@ -126,16 +131,16 @@
         }
 
         #fileDropArea.drag-over {
-            border-color: #1095c1; /* Pico primary color (定量替换 var(--pico-primary-focus)) */
+            border-color: #1095c1;
-            background-color: #e7f5fa; /* Pico primary background (定量替换 var(--pico-primary-background)) */
+            background-color: #e7f5fa;
         }
 
         #fileDropArea.file-selected {
-            border-color: #2e7d32; /* Pico success color (定量替换 var(--pico-form-element-valid-border-color, #2e7d32)) */
+            border-color: #2e7d32;
-            background-color: #e8f5e9; /* Light green (定量替换 var(--pico-form-element-valid-background-color, #e8f5e9)) */
+            background-color: #e8f5e9;
         }
 
-        #fileDropArea p { /* General style for <p> inside drop area */
+        #fileDropArea p {
             margin: 0.5rem 0;
             color: #555;
         }
@@ -149,19 +154,18 @@
         #fileNameDisplay.has-file {
             font-style: normal;
             font-weight: bold;
-            color: #1a531d; /* Darker green or success color (定量替换 var(--pico-form-element-valid-border-color, #1a531d)) */
+            color: #1a531d;
         }
 
-        #fileDropArea.input-error {
+        #fileDropArea.input-error, input.input-error, select.input-error { /* Extended to input/select */
-            border-color: #d32f2f !important; /* (定量替换 var(--pico-form-element-invalid-border-color, #d32f2f)) */
+            border-color: #d32f2f !important;
         }
 
         #fileNameDisplay.input-error-text {
-            color: #d32f2f !important; /* (定量替换 var(--pico-form-element-invalid-border-color, #d32f2f)) */
+            color: #d32f2f !important;
             font-weight: bold;
         }
 
-
         @media (max-width: 768px) {
             .form-grid {
                 grid-template-columns: 1fr;
@@ -176,7 +180,6 @@
     </h1>
     <form id="translateForm">
 
-        <!-- Modified File Input Area -->
         <div class="form-group">
             <label for="file">文档选择</label>
             <div id="fileDropArea">
@@ -206,17 +209,34 @@
                 </select>
             </div>
             <div class="form-group">
-                <label>高级选项</label>
+                <label>选项</label>
                 <div class="checkbox-group">
-                    <label for="formula_ocr"><input type="checkbox" id="formula_ocr" name="formula_ocr">公式识别</label>
+                    <label for="formula_ocr"><input type="checkbox" id="formula_ocr" name="formula_ocr" role="switch">公式识别</label>
-                    <label for="code_ocr"><input type="checkbox" id="code_ocr" name="code_ocr">代码识别</label>
+                    <label for="code_ocr"><input type="checkbox" id="code_ocr" name="code_ocr"
-                    <label for="refine_markdown"><input type="checkbox" id="refine_markdown"
+                                                 role="switch">代码识别</label>
-                                                        name="refine_markdown">修正文本（耗时）</label>
+                    <label for="refine_markdown"><input type="checkbox" id="refine_markdown" name="refine_markdown"
+                                                        role="switch">修正文本（耗时）</label>
                 </div>
             </div>
         </div>
         <details>
-            <summary>API 配置</summary>
+            <summary>文档转换引擎配置</summary>
+            <div class="form-group">
+                <label for="convert_engin">转换引擎</label>
+                <select id="convert_engin" name="convert_engin">
+                    <option value="mineru" selected>Mineru</option>
+                    <option value="docling">Docling</option>
+                </select>
+            </div>
+            <div class="form-group hidden" id="mineruTokenGroup">
+                <label for="mineru_token">Mineru Token</label>
+                <input type="password" id="mineru_token" name="mineru_token" placeholder="使用 Mineru 引擎时必须填写">
+            </div>
+        </details>
+
+
+        <details>
+            <summary>翻译API配置</summary>
             <div class="form-grid">
                 <div class="form-group">
                     <label for="platform_select">AI 平台</label>
@@ -225,8 +245,7 @@
                         <option value="https://api.openai.com/v1">OpenAI</option>
                         <option value="https://open.bigmodel.cn/api/paas/v4">智谱AI</option>
                         <option value="https://api.deepseek.com/v1">DeepSeek</option>
-                        <option value="https://dashscope.aliyuncs.com/compatible-mode/v1">阿里云百炼
+                        <option value="https://dashscope.aliyuncs.com/compatible-mode/v1">阿里云百炼</option>
-                        </option>
                         <option value="https://www.dmxapi.cn/v1">DMXAPI</option>
                         <option value="https://openrouter.ai/api/v1">OpenRouter</option>
                         <option value="https://ark.cn-beijing.volces.com/api/v3">火山引擎</option>
@@ -235,14 +254,12 @@
                 </div>
                 <div class="form-group hidden" id="baseUrlGroup">
                     <label for="base_url">API 地址 (Base URL)</label>
-                    <input type="text" id="base_url" name="base_url"
+                    <input type="text" id="base_url" name="base_url" placeholder="https://api.openai.com/v1">
-                           placeholder="https://api.openai.com/v1">
                 </div>
             </div>
             <div class="form-group">
                 <label for="apikey">API 密钥</label>
-                <input type="password" id="apikey" name="apikey" placeholder="平台对应的API Key"
+                <input type="password" id="apikey" name="apikey" placeholder="平台对应的API Key" required>
-                       required>
             </div>
             <div class="form-group">
                 <label for="model_id">模型 ID</label>
@@ -267,7 +284,7 @@
 </main>
 <div id="previewModal" class="modal">
     <div class="modal-content">
-        <span id="closeModalBtn" style="cursor:pointer; float:right;">×</span>
+        <span id="closeModalBtn" style="cursor:pointer; float:right; font-size: 1.5rem; line-height: 1;">×</span>
         <h3>HTML 预览</h3>
         <iframe id="previewFrame"></iframe>
         <div class="button-group">
@@ -277,7 +294,6 @@
     </div>
 </div>
 <iframe id="printFrame" style="display:none;"></iframe>
-
 <script>
     const platformSelect = document.getElementById('platform_select');
     const baseUrlGroup = document.getElementById('baseUrlGroup');
@@ -288,6 +304,11 @@
     const formulaCheckbox = document.getElementById('formula_ocr');
     const codeCheckbox = document.getElementById('code_ocr');
     const refineCheckbox = document.getElementById('refine_markdown');
+
+    const convertEnginSelect = document.getElementById('convert_engin');
+    const mineruTokenGroup = document.getElementById('mineruTokenGroup');
+    const mineruTokenInput = document.getElementById('mineru_token');
+
     const form = document.getElementById('translateForm');
     const submitButton = document.getElementById('submitButton');
     const logArea = document.getElementById('logArea');
@@ -311,7 +332,6 @@
 
     let logPollIntervalId = null;
     let statusPollIntervalId = null;
-    // let lastLogCount = 0; // No longer needed for fetching logs
     let isTranslating = false;
 
     function saveToStorage(key, value) {
@@ -347,7 +367,34 @@
         saveToStorage('translator_last_platform', selectedPlatformValue);
     }
 
-    loadSettings();
+    function updateConvertEnginUI() {
+        const selectedEngin = convertEnginSelect.value;
+        if (selectedEngin === 'mineru') {
+            mineruTokenGroup.classList.remove('hidden');
+            mineruTokenInput.required = true;
+            mineruTokenInput.value = getFromStorage('translator_mineru_token');
+        } else {
+            mineruTokenGroup.classList.add('hidden');
+            mineruTokenInput.required = false;
+            // Optionally clear if not needed: mineruTokenInput.value = '';
+        }
+        saveToStorage('translator_convert_engin', selectedEngin);
+    }
+
+    function loadSettings() {
+        platformSelect.value = getFromStorage('translator_last_platform', 'custom');
+        updatePlatformUI();
+
+        convertEnginSelect.value = getFromStorage('translator_convert_engin', 'mineru');
+        updateConvertEnginUI(); // Must be after setting convertEnginSelect.value
+
+        toLangSelect.value = getFromStorage('translator_to_lang', '中文');
+        formulaCheckbox.checked = getFromStorage('translator_formula_ocr') === 'true';
+        codeCheckbox.checked = getFromStorage('translator_code_ocr') === 'true';
+        refineCheckbox.checked = getFromStorage('translator_refine_markdown') === 'true';
+    }
+
+    loadSettings(); // Initial load
 
     platformSelect.addEventListener('change', updatePlatformUI);
     apikeyInput.addEventListener('input', (e) => saveToStorage(`translator_platform_${platformSelect.value}_apikey`, e.target.value));
@@ -355,10 +402,14 @@
     baseUrlInput.addEventListener('input', (e) => {
         if (platformSelect.value === 'custom') saveToStorage('translator_platform_custom_base_url', e.target.value);
     });
+
+    convertEnginSelect.addEventListener('change', updateConvertEnginUI);
+    mineruTokenInput.addEventListener('input', (e) => saveToStorage('translator_mineru_token', e.target.value));
+
     toLangSelect.addEventListener('change', e => saveToStorage('translator_to_lang', e.target.value));
-    formulaCheckbox.addEventListener('change', e => saveToStorage('translator_formula_ocr', e.target.checked));
+    formulaCheckbox.addEventListener('change', e => saveToStorage('translator_formula_ocr', e.target.checked.toString()));
-    codeCheckbox.addEventListener('change', e => saveToStorage('translator_code_ocr', e.target.checked));
+    codeCheckbox.addEventListener('change', e => saveToStorage('translator_code_ocr', e.target.checked.toString()));
-    refineCheckbox.addEventListener('change', e => saveToStorage('translator_refine_markdown', e.target.checked));
+    refineCheckbox.addEventListener('change', e => saveToStorage('translator_refine_markdown', e.target.checked.toString()));
 
     [closeModalButton, closePreviewBtn].forEach(elem => elem.addEventListener('click', () => modal.style.display = 'none'));
     window.addEventListener('click', (event) => {
@@ -374,9 +425,7 @@
         }
     });
 
-    fileDropArea.addEventListener('click', () => {
+    fileDropArea.addEventListener('click', () => fileInput.click());
-        fileInput.click();
-    });
 
     fileInput.addEventListener('change', () => {
         if (fileInput.files.length > 0) {
@@ -422,7 +471,6 @@
     fileDropArea.addEventListener('drop', (e) => {
         const dt = e.dataTransfer;
         const files = dt.files;
-
         if (files.length > 0) {
             fileInput.files = files;
             const event = new Event('change', {bubbles: true});
@@ -432,8 +480,7 @@
 
     async function pollLogs() {
         try {
-            // const response = await fetch(`/get-logs?since=${lastLogCount}`); // OLD
+            const response = await fetch('/get-logs');
-            const response = await fetch('/get-logs'); // NEW: No 'since' parameter
             if (!response.ok) {
                 console.warn(`Log polling failed: ${response.status}`);
                 return;
@@ -444,9 +491,8 @@
                     const escapedLog = log.replace(/&/g, "&").replace(/</g, "<").replace(/>/g, ">");
                     logArea.innerHTML += escapedLog + "<br>";
                 });
-                logArea.scrollTop = logArea.scrollHeight; // Scroll to bottom
+                logArea.scrollTop = logArea.scrollHeight;
             }
-            // lastLogCount = data.total_count; // OLD: No longer tracking count this way
         } catch (error) {
             console.warn("Error polling logs:", error);
         }
@@ -561,7 +607,7 @@
             } else {
                 submitButton.textContent = '取消翻译';
                 submitButton.classList.remove('primary');
-                submitButton.classList.add('secondary');
+                submitButton.classList.add('secondary', 'contrast'); // Using contrast for cancel
                 isTranslating = true;
                 submitButton.disabled = false;
                 submitButton.removeAttribute('aria-busy');
@@ -576,10 +622,9 @@
 
     function startPolling() {
         stopPolling();
-        // lastLogCount = 0; // No longer needed
+        logArea.innerHTML = '';
-        logArea.innerHTML = ''; // Clear log area for new task
+        pollLogs();
-        pollLogs(); // Initial poll for logs
+        pollStatus();
-        pollStatus(); // Initial poll for status
         logPollIntervalId = setInterval(pollLogs, 2000);
         statusPollIntervalId = setInterval(pollStatus, 1500);
     }
@@ -589,16 +634,7 @@
         if (statusPollIntervalId) clearInterval(statusPollIntervalId);
         logPollIntervalId = null;
         statusPollIntervalId = null;
-        setTimeout(pollLogs, 500);
+        setTimeout(pollLogs, 500); // One last poll for logs
-    }
-
-    function loadSettings() {
-        platformSelect.value = getFromStorage('translator_last_platform', 'custom');
-        updatePlatformUI();
-        toLangSelect.value = getFromStorage('translator_to_lang', '中文');
-        formulaCheckbox.checked = getFromStorage('translator_formula_ocr') === 'true';
-        codeCheckbox.checked = getFromStorage('translator_code_ocr') === 'true';
-        refineCheckbox.checked = getFromStorage('translator_refine_markdown') === 'true';
     }
 
     async function cancelTranslation() {
@@ -609,13 +645,13 @@
         try {
             const response = await fetch('/cancel-translate', {method: 'POST'});
             const result = await response.json();
-
             if (response.ok && result.cancelled) {
                 statusMsg.textContent = result.message || '取消请求已发送。';
-                statusMsg.className = '';
+                statusMsg.className = ''; // Clear error class
             } else {
                 statusMsg.textContent = result.message || '取消失败。';
                 statusMsg.className = 'error-message';
+                // Re-enable button if cancellation failed to register server-side
                 submitButton.disabled = false;
                 submitButton.textContent = '取消翻译';
                 submitButton.removeAttribute('aria-busy');
@@ -628,6 +664,7 @@
             submitButton.textContent = '取消翻译';
             submitButton.removeAttribute('aria-busy');
         }
+        // Status poller will eventually update the button state correctly
     }
 
     form.addEventListener('submit', async function (event) {
@@ -638,6 +675,10 @@
             return;
         }
 
+        // Clear previous input errors
+        [fileDropArea, mineruTokenInput].forEach(el => el.classList.remove('input-error'));
+        fileNameDisplay.classList.remove('input-error-text');
+
         if (fileInput.files.length === 0) {
             statusMsg.textContent = '请选择一个文件进行翻译。';
             statusMsg.className = 'error-message';
@@ -648,14 +689,18 @@
             setTimeout(() => {
                 fileDropArea.classList.remove('input-error');
                 fileNameDisplay.classList.remove('input-error-text');
-                if (fileNameDisplay.textContent === '请选择文件！') {
+                if (fileNameDisplay.textContent === '请选择文件！') fileNameDisplay.textContent = '未选择文件';
-                    fileNameDisplay.textContent = '未选择文件';
+                if (fileInput.files.length === 0) fileDropPrompt.classList.remove('hidden');
-                }
+            }, 3000);
-                if (fileInput.files.length === 0) {
+            return;
-                    fileDropPrompt.classList.remove('hidden');
         }
 
-            }, 3000);
+        if (convertEnginSelect.value === 'mineru' && !mineruTokenInput.value.trim()) {
+            statusMsg.textContent = '使用 Mineru 引擎时，必须填写 Mineru Token。';
+            statusMsg.className = 'error-message';
+            mineruTokenInput.classList.add('input-error');
+            mineruTokenInput.focus();
+            setTimeout(() => mineruTokenInput.classList.remove('input-error'), 3000);
             return;
         }
 
@@ -667,9 +712,10 @@
         statusMsg.textContent = '正在提交任务...';
         statusMsg.className = '';
         downloadBtns.style.display = 'none';
-        // lastLogCount = 0; // No longer needed
 
         const formData = new FormData(form);
+        // FormData automatically includes convert_engin and mineru_token due to 'name' attributes
+
         try {
             const response = await fetch('/translate', {method: 'POST', body: formData});
             const result = await response.json();
@@ -678,8 +724,9 @@
                 statusMsg.className = '';
                 submitButton.textContent = '取消翻译';
                 submitButton.classList.remove('primary');
-                submitButton.classList.add('secondary');
+                submitButton.classList.add('secondary', 'contrast');
                 isTranslating = true;
+                submitButton.disabled = false; // Enable cancel button
                 submitButton.removeAttribute('aria-busy');
                 startPolling();
             } else {

docutranslate/translater.py

@@ -1,24 +1,25 @@
 import asyncio
-from io import BytesIO
 from pathlib import Path
 from typing import Literal
-
 import markdown2
-from docling.datamodel.document import DocumentStream
-
 from docutranslate.agents import Agent, AgentArgs
 from docutranslate.agents import MDRefineAgent, MDTranslateAgent
-from docutranslate.utils.convert import file2markdown_embed_images
+from docutranslate.converter import Document, ConverterDocling, ConverterMineru
-from docutranslate.utils.markdown_splitter import split_markdown_text,join_markdown_texts
+from docutranslate.utils.markdown_splitter import split_markdown_text, join_markdown_texts
 from docutranslate.utils.markdown_utils import uris2placeholder, placeholder2_uris, MaskDict
 from docutranslate.logger import translater_logger
 
 
-
 class FileTranslater:
-    def __init__(self, file_path: Path | str | None = None, chunksize: int = 2000, base_url="", key=None,
+    def __init__(self, file_path: Path | str | None = None, chunksize: int = 2000,
-                 model_id="", temperature=0.7, max_concurrent=20, docling_artifact: Path | str | None = None,
+                 base_url="", key=None, model_id="", temperature=0.7,
-                 timeout=2000, tips=True):
+                 max_concurrent=20, timeout=2000,
+                 convert_engin: Literal["docling", "mineru"] = "docling",
+                 docling_artifact: Path | str | None = None,
+                 mineru_token: str = None,
+                 tips=True):
+        self.convert_engin = convert_engin
+        self.mineru_token = mineru_token.strip() if mineru_token is not None else None
         if isinstance(file_path, str):
             file_path = Path(file_path)
         self.file_path: Path = file_path
@@ -33,10 +34,10 @@ class FileTranslater:
         self.temperature = temperature
         self.docling_artifact = docling_artifact
         if docling_artifact is None:
-            artifact_path=Path("./artifact")
+            artifact_path = Path("./docling_artifact")
             if artifact_path.is_dir():
-                translater_logger.info("检测到artifact文件夹")
+                translater_logger.info("检测到docling_artifact文件夹")
-                self.docling_artifact=artifact_path
+                self.docling_artifact = artifact_path
         self.timeout = timeout
         if tips:
             print("""
@@ -66,7 +67,7 @@ class FileTranslater:
         translater_logger.info(f"markdown分为{len(chunks)}块")
         return chunks
 
-    def default_agent_params(self) -> AgentArgs:
+    def _default_agent_params(self) -> AgentArgs:
         result: AgentArgs = {
             "baseurl": self.base_url,
             "key": self.key,
@@ -77,16 +78,48 @@ class FileTranslater:
         }
         return result
 
+    def _convert2markdown(self, document: Document, formula: bool, code: bool, artifact: Path = None) -> str:
+        translater_logger.info(f"正在使用{self.convert_engin}转换文件为markdown")
+        if self.convert_engin == "docling":
+            if artifact is None:
+                artifact = self.docling_artifact
+            mdconverter = ConverterDocling(formula=formula, code=code, artifact=artifact)
+            result = mdconverter.convert(document)
+        else:
+            if self.mineru_token is None:
+                raise Exception("mineru_token未配置")
+            if code:
+                translater_logger.info("mineru暂不支持code识别")
+            mdconverter = ConverterMineru(token=self.mineru_token, formula=formula)
+            result = mdconverter.convert(document)
+        return result
+
+    async def _convert2markdown_async(self, document: Document, formula: bool, code: bool,
+                                      artifact: Path = None) -> str:
+        if self.convert_engin == "docling":
+            if artifact is None:
+                artifact = self.docling_artifact
+            mdconverter = ConverterDocling(formula=formula, code=code, artifact=artifact)
+            result = await mdconverter.convert_async(document)
+        else:
+            if self.mineru_token is None:
+                raise Exception("mineru_token未配置")
+            if code:
+                translater_logger.info("mineru暂不支持code识别")
+            mdconverter = ConverterMineru(token=self.mineru_token, formula=formula)
+            result = await mdconverter.convert_async(document)
+        return result
+
     def read_bytes(self, name: str, file: bytes, formula=True, code=True, save=False,
                    save_format: Literal["markdown", "html"] = "markdown", refine=False,
                    refine_agent: Agent | None = None):
-        ds = DocumentStream(name=name, stream=BytesIO(file))
+        document = Document(filename=name, filebytes=file)
         file_path = Path(name)
         # 如果是markdown，直接读取
         if file_path.suffix == ".md":
             self.markdown = file.decode()
         else:
-            self.markdown = file2markdown_embed_images(ds, formula, code, artifacts_path=self.docling_artifact)
+            self.markdown = self._convert2markdown(document, formula=formula, code=code, artifact=self.docling_artifact)
         if refine:
             self.refine_markdown_by_agent(refine_agent)
         if save:
@@ -96,6 +129,26 @@ class FileTranslater:
                 self.save_as_markdown(filename=f"{file_path.stem}.md")
         return self
 
+    async def read_bytes_async(self, name: str, file: bytes, formula=True, code=True, save=False,
+                               save_format: Literal["markdown", "html"] = "markdown", refine=False,
+                               refine_agent: Agent | None = None):
+        document = Document(filename=name, filebytes=file)
+        file_path = Path(name)
+        # 如果是markdown，直接读取
+        if file_path.suffix == ".md":
+            self.markdown = file.decode()
+        else:
+            self.markdown = await self._convert2markdown_async(document, formula=formula, code=code,
+                                                               artifact=self.docling_artifact)
+        if refine:
+            await self.refine_markdown_by_agent_async(refine_agent)
+        if save:
+            if save_format == "html":
+                self.save_as_html(filename=f"{file_path.stem}.html")
+            else:
+                self.save_as_markdown(filename=f"{file_path.stem}.md")
+        return self
+
     def read_file(self, file_path: Path | str | None = None, formula=True, code=True, save=False,
                   save_format: Literal["markdown", "html"] = "markdown", refine=False,
                   refine_agent: Agent | None = None):
@@ -112,7 +165,8 @@ class FileTranslater:
             with open(file_path, "r") as f:
                 self.markdown = f.read()
         else:
-            self.markdown = file2markdown_embed_images(file_path, formula, code, artifacts_path=self.docling_artifact)
+            document = Document(file_path)
+            self.markdown = self._convert2markdown(document, formula=formula, code=code, artifact=self.docling_artifact)
         if refine:
             self.refine_markdown_by_agent(refine_agent)
         if save:
@@ -122,14 +176,42 @@ class FileTranslater:
                 self.save_as_markdown(filename=f"{file_path.stem}.md")
         return self
 
+    async def read_file_async(self, file_path: Path | str | None = None, formula=True, code=True, save=False,
+                              save_format: Literal["markdown", "html"] = "markdown", refine=False,
+                              refine_agent: Agent | None = None):
+        if file_path is None:
+            if self.file_path is None:
+                translater_logger.debug("未设置文件路径")
+                raise Exception("未设置文件路径")
+            file_path = self.file_path
+        if isinstance(file_path, str):
+            file_path = Path(file_path)
+        translater_logger.info(f"读取文件：{file_path.name}")
+        # 如果是markdown，直接读取
+        if file_path.suffix == ".md":
+            with open(file_path, "r") as f:
+                self.markdown = f.read()
+        else:
+            document = Document(file_path)
+            self.markdown = await self._convert2markdown_async(document, formula=formula, code=code,
+                                                               artifact=self.docling_artifact)
+        if refine:
+            await self.refine_markdown_by_agent_async(refine_agent)
+        if save:
+            if save_format == "html":
+                self.save_as_html(filename=f"{file_path.stem}.html")
+            else:
+                self.save_as_markdown(filename=f"{file_path.stem}.md")
+        return self
+
     def refine_markdown_by_agent(self, refine_agent: Agent | None = None) -> str:
         translater_logger.info("正在修正markdown")
         self._mask_uris_in_markdown()
         chuncks = self._split_markdown_into_chunks()
         if refine_agent is None:
-            refine_agent = MDRefineAgent(**self.default_agent_params())
+            refine_agent = MDRefineAgent(**self._default_agent_params())
         result: list[str] = refine_agent.send_prompts(chuncks)
-        self.markdown=join_markdown_texts(result)
+        self.markdown = join_markdown_texts(result)
         self._unmask_uris_in_markdown()
         translater_logger.info("markdown已修正")
         return self.markdown
@@ -139,22 +221,21 @@ class FileTranslater:
         self._mask_uris_in_markdown()
         chuncks = self._split_markdown_into_chunks()
         if translate_agent is None:
-            translate_agent = MDTranslateAgent(to_lang=to_lang, **self.default_agent_params())
+            translate_agent = MDTranslateAgent(to_lang=to_lang, **self._default_agent_params())
         result: list[str] = translate_agent.send_prompts(chuncks)
-        self.markdown=join_markdown_texts(result)
+        self.markdown = join_markdown_texts(result)
         self._unmask_uris_in_markdown()
         translater_logger.info("翻译完成")
         return self.markdown
 
-
     async def refine_markdown_by_agent_async(self, refine_agent: Agent | None = None) -> str:
         translater_logger.info("正在修正markdown")
         self._mask_uris_in_markdown()
         chuncks = self._split_markdown_into_chunks()
         if refine_agent is None:
-            refine_agent = MDRefineAgent(**self.default_agent_params())
+            refine_agent = MDRefineAgent(**self._default_agent_params())
         result: list[str] = await refine_agent.send_prompts_async(chuncks)
-        self.markdown=join_markdown_texts(result)
+        self.markdown = join_markdown_texts(result)
         self._unmask_uris_in_markdown()
         translater_logger.info("markdown已修正")
         return self.markdown
@@ -164,9 +245,9 @@ class FileTranslater:
         self._mask_uris_in_markdown()
         chuncks = self._split_markdown_into_chunks()
         if translate_agent is None:
-            translate_agent = MDTranslateAgent(to_lang=to_lang, **self.default_agent_params())
+            translate_agent = MDTranslateAgent(to_lang=to_lang, **self._default_agent_params())
         result: list[str] = await translate_agent.send_prompts_async(chuncks)
-        self.markdown=join_markdown_texts(result)
+        self.markdown = join_markdown_texts(result)
         self._unmask_uris_in_markdown()
         translater_logger.info("翻译完成")
         return self.markdown
@@ -217,6 +298,7 @@ class FileTranslater:
 
     def export_to_html(self, title="title") -> str:
         markdowner = markdown2.Markdown(extras=['tables', 'fenced-code-blocks', 'mermaid', "code-friendly"])
+        # TODO:实现完全本地化css和js
         # language=html
         html = f"""<!DOCTYPE html>
 <html lang="en">
@@ -245,7 +327,7 @@ class FileTranslater:
     </script>
 </head>
 <body>
-{markdowner.convert(self.markdown.replace("\\","\\\\"))}
+{markdowner.convert(self.markdown.replace("\\", "\\\\"))}
 </body>
 <script type="module" defer>
   import mermaid from 'https://cdn.jsdelivr.net/npm/mermaid@9/dist/mermaid.esm.min.mjs';
@@ -273,7 +355,7 @@ class FileTranslater:
     def translate_file(self, file_path: Path | str | None = None, to_lang="中文", output_dir="./output",
                        formula=True,
                        code=True, output_format: Literal["markdown", "html"] = "markdown", refine=False,
-                       refine_agent: Agent | None = None, translate_agent: Agent | None = None,save=True):
+                       refine_agent: Agent | None = None, translate_agent: Agent | None = None, save=True):
         if file_path is None:
             assert self.file_path is not None, "未输入文件路径"
             file_path = self.file_path
@@ -291,10 +373,11 @@ class FileTranslater:
                 filename = f"{file_path.stem}_{to_lang}.html"
                 self.save_as_html(filename=filename, output_dir=output_dir)
         return self
+
     async def translate_file_async(self, file_path: Path | str | None = None, to_lang="中文", output_dir="./output",
                                    formula=True,
                                    code=True, output_format: Literal["markdown", "html"] = "markdown", refine=False,
-                       refine_agent: Agent | None = None, translate_agent: Agent | None = None,save=True):
+                                   refine_agent: Agent | None = None, translate_agent: Agent | None = None, save=True):
         if file_path is None:
             assert self.file_path is not None, "未输入文件路径"
             file_path = self.file_path
@@ -317,11 +400,12 @@ class FileTranslater:
                 filename = f"{file_path.stem}_{to_lang}.html"
                 self.save_as_html(filename=filename, output_dir=output_dir)
         return self
-    def translate_bytes(self, name:str,file: bytes, to_lang="中文", output_dir="./output",
+
+    def translate_bytes(self, name: str, file: bytes, to_lang="中文", output_dir="./output",
                         formula=True,
                         code=True, output_format: Literal["markdown", "html"] = "markdown", refine=False,
-                       refine_agent: Agent | None = None, translate_agent: Agent | None = None,save=True):
+                        refine_agent: Agent | None = None, translate_agent: Agent | None = None, save=True):
-        self.read_bytes(name=name,file=file, formula=formula, code=code)
+        self.read_bytes(name=name, file=file, formula=formula, code=code)
         if refine:
             self.refine_markdown_by_agent(refine_agent)
         self.translate_markdown_by_agent(translate_agent, to_lang=to_lang)
@@ -334,17 +418,12 @@ class FileTranslater:
                 self.save_as_html(filename=filename, output_dir=output_dir)
         return self
 
-    async def translate_bytes_async(self, name:str,file: bytes, to_lang="中文", output_dir="./output",
+    async def translate_bytes_async(self, name: str, file: bytes, to_lang="中文", output_dir="./output",
                                     formula=True,
                                     code=True, output_format: Literal["markdown", "html"] = "markdown", refine=False,
-                       refine_agent: Agent | None = None, translate_agent: Agent | None = None,save=True):
+                                    refine_agent: Agent | None = None, translate_agent: Agent | None = None, save=True):
-        await asyncio.to_thread(
+        await self.read_bytes_async(name=name, file=file, formula=formula, code=code)
-            self.read_bytes,
+
-            name=name,
-            file=file,
-            formula=formula,
-            code=code
-        )
         if refine:
             await self.refine_markdown_by_agent_async(refine_agent)
         await self.translate_markdown_by_agent_async(translate_agent, to_lang=to_lang)

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "docutranslate"
-version = "0.2.19"
+version = "0.2.20"
 description = "文件翻译工具"
 readme = "README.md"
 requires-python = ">=3.10"