From 5a726019289a5e27e75019cb99860dd37b7d5592 Mon Sep 17 00:00:00 2001 From: xunbu Date: Tue, 4 Nov 2025 20:18:57 +0800 Subject: [PATCH] =?UTF-8?q?=E4=BF=AE=E5=A4=8Dmineru=E9=83=A8=E7=BD=B2?= =?UTF-8?q?=E7=9A=84backend=E9=80=89=E9=A1=B9?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- docutranslate/app.py | 2 +- .../converter/x2md/converter_mineru_deploy.py | 2 +- docutranslate/static/i18nData.json | 206 +++++++++--------- docutranslate/static/index.html | 2 +- 4 files changed, 109 insertions(+), 103 deletions(-) diff --git a/docutranslate/app.py b/docutranslate/app.py index d89c0eb..e3ec5ca 100644 --- a/docutranslate/app.py +++ b/docutranslate/app.py @@ -429,7 +429,7 @@ class MarkdownWorkflowParams(BaseWorkflowParams): "http://127.0.0.1:8000", description="[仅当 convert_engine='mineru_deploy'] 本地部署的 MinerU 服务地址。", ) - mineru_deploy_backend: Literal["pipeline", "vlm"] = Field( + mineru_deploy_backend: Literal["pipeline", "transformers", "mlx-engine", "vllm-engine", "vllm-async-engine"] = Field( "pipeline", description="[仅当 convert_engine='mineru_deploy'] 本地部署的 MinerU 服务使用的后端。", ) diff --git a/docutranslate/converter/x2md/converter_mineru_deploy.py b/docutranslate/converter/x2md/converter_mineru_deploy.py index bf8f981..ae04aea 100644 --- a/docutranslate/converter/x2md/converter_mineru_deploy.py +++ b/docutranslate/converter/x2md/converter_mineru_deploy.py @@ -18,7 +18,7 @@ class ConverterMineruDeployConfig(X2MarkdownConverterConfig): base_url: str = "http://127.0.0.1:8000" output_dir: str = "./output" # 覆盖默认值 ./output # lang_list: list[str] | None = None - backend: Literal["pipeline", "vlm"] = "pipeline" + backend: Literal["pipeline", "transformers", "mlx-engine", "vllm-engine", "vllm-async-engine"] = "pipeline" # parse_method: str = "auto" formula_enable: bool = True # table_enable: bool = True diff --git a/docutranslate/static/i18nData.json b/docutranslate/static/i18nData.json index 0a79abc..4695b47 100644 --- a/docutranslate/static/i18nData.json +++ b/docutranslate/static/i18nData.json @@ -46,7 +46,7 @@ "jsonSettingsTitleText": "JSON路径配置", "jsonPathLabel": "需要翻译的JSON路径", "jsonPathPlaceholder": "每行一个路径, 例如:\n$.name\n$.*", - "jsonPathHelp": "采用jsonpath-ng的路径选择语法，每一行表示一个json路径。\n 将翻译路径匹配对象内的所有字符串", + "jsonPathHelp": "采用jsonpath-ng的路径选择语法，每一行表示一个json路径。\n 将翻译路径匹配对象内的所有字符串", "parsingSettingsTitleText": "解析配置", "parsingEngineLabel": "解析引擎", "parsingEngineHelp": "如果上传的文件本身是.md格式，此项可不选。", @@ -60,7 +60,10 @@ "mineruDeployBaseUrlPlaceholder": "例如: http://127.0.0.1:8000", "mineruDeployBackendLabel": "后端类型", "mineruDeployBackendPipeline": "Pipeline", - "mineruDeployBackendVlm": "VLM", + "mineruDeployBackendTransformers": "Transformers", + "mineruDeployBackendMlx": "MLX Engine", + "mineruDeployBackendVllm": "VLLM Engine", + "mineruDeployBackendVllmAsync": "VLLM Async Engine", "mineruDeployStartPageLabel": "起始页面", "mineruDeployEndPageLabel": "结束页面", "mineruDeployFormulaEnableLabel": "启用公式解析", @@ -105,7 +108,7 @@ "glossaryGenConfigCustom": "自定义", "importConfigBtn": "导入配置", "exportConfigBtn": "导出配置", - "githubInfo": "GitHub主页(欢迎star❤):
\n https://github.com/xunbu/docutranslate", + "githubInfo": "GitHub主页(欢迎star❤):
\n https://github.com/xunbu/docutranslate", "qqGroupInfo": "交流QQ群: 1047781902", "taskListTitle": "任务列表", "newTaskBtn": "新建任务", @@ -134,7 +137,7 @@ "closeBtn": "关闭", "downloadBtn": "下载", "tutorialModalTitle": "使用教程", - "tutorialModalBody": "

视频教程可以在B站搜索 docutranslate 获取。

\n

欢迎使用 DocuTranslate！请按照以下步骤完成文档翻译：

\n
    \n
  1. \n 第一步：选择工作流\n

    \n 在左侧配置面板的顶部，首先选择最适合您文件类型的处理流程。\n

    \n 提示:\n 默认已开启“自动选择工作流”。您只需上传文件，系统会自动为您匹配合适的工作流，简化操作。\n
    \n

    \n
      \n
    • 转Markdown再翻译: 适用于翻译PDF、markdown、图片等文件。这是最通用和强大的模式。
      • \n
    • 纯文本翻译: 用于翻译 .txt 纯文本文件。
      • \n
    • EPUB翻译: 用于翻译 .epub 电子书文件。
      • \n
    • DOCX翻译: 用于翻译 .docx Word文档。
      • \n
    • XLSX翻译: 用于翻译 .xlsx 或 .csv 电子表格文件。
      • \n
    • SRT字幕翻译: 用于翻译 .srt 字幕文件。
      • \n
    • ASS字幕翻译: 用于翻译 .ass 特效字幕文件。
      • \n
    • JSON翻译: 用于翻译 .json 文件中的特定字段。
      • \n
    • HTML翻译: 用于翻译 .html 网页文件。
      • \n
    \n
    2. \n
  2. \n 第二步：配置参数\n

    \n 选择工作流后，下方会显示相关的配置选项。请依次完成设置（所有配置都会自动保存在您的浏览器中）：

    \n

    A. 工作流特定选项 (根据您第一步的选择出现):

    \n
      \n
    • 如果选择“转Markdown再翻译”，请配置 解析配置：\n
        \n
      • 解析引擎:\n 选择一个引擎将您的文件（如PDF）转换为适合翻译的Markdown格式。如果您的文件已经是Markdown格式，则无需选择。\n
        • \n
      • Mineru Token: 如果您选择 minerU 引擎，需要在此处填入您的Token。\n
        • \n
      \n
      • \n
    • 如果选择“纯文本/DOCX/XLSX/SRT/ASS/EPUB/HTML”，请配置其 翻译选项：\n
        \n
      • 插入模式: 定义翻译结果如何放入文档。您可以选择直接“替换”原文，或是在原文之后“附加”，或是在原文之前“前置”。\n
        • \n
      • 分隔符: 当选择“附加”或“前置”模式时，此项用于在原文和译文之间插入分隔符（例如ASS格式常用 \\N，EPUB格式常用 <br /> 作为换行分隔符）。\n
        • \n
      \n
      • \n
    • 如果选择“JSON翻译”，请配置 JSON路径：\n
        \n
      • 需要翻译的JSON路径: 每行输入一个 JSONPath\n 表达式，将翻译所有与路径匹配的对象中的字符串。例如:$.*(翻译全部字符串)，$..description(翻译所有键为description的值)。\n
        • \n
      \n
      • \n
    \n

    B. 通用选项 (适用于所有工作流):

    \n
      \n
    • 翻译模型:\n
        \n
      • 选择平台/API 地址/API Key/模型ID: 配置您希望使用的AI翻译服务。模型能力指令遵循越强，出错、漏翻的概率越低。\n
        • \n
      • 跳过翻译: 勾选此项后，将只执行文档解析和格式转换，不调用AI进行翻译。\n
        • \n
      \n
      • \n
    • 翻译配置:\n
        \n
      • 目标语言: 指定翻译的目标语言。
        • \n
      • 自定义Prompt: 可选，添加额外指令，如“人名保持原文不翻译”。
        • \n
      • 思考模式: 针对部分支持混合推理的模型进行设置，建议选择“禁用(推荐)”。\n
        • \n
      • 分块大小/并发数等: 高级参数，用于调整性能和API请求行为，通常保持默认即可。\n
        • \n
      \n
      • \n
    • 术语表:\n
        \n
      • 上传术语表 (可选): 上传CSV文件（需包含'src'和'dst'列）来保证特定术语翻译的统一性和准确性。\n
        • \n
      • 自动生成术语表: 启用后，程序会先从原文中提取术语并生成一个术语表，然后再进行翻译。\n
        • \n
      \n
      • \n
    \n
    3. \n
  3. \n 第三步：上传文件\n

    在右侧的任务列表中，点击或拖拽您的文档到文件上传区域。

    \n
    4. \n
  4. \n 第四步：开始翻译\n

    文件选择成功后，点击任务卡片右下角的 开始翻译 按钮。系统将开始处理任务，您可以在日志区域查看实时进度。\n

    \n
    5. \n
  5. \n 第五步：查看与下载\n

    翻译完成后，任务卡片下方会出现操作按钮：

    \n
      \n
    • 预览: \n 在右侧滑出的面板中进行原文和译文的对照预览。\n
      • \n
    • 下载: \n 下载包括 PDF, DOCX, Markdown 等多种格式的译文。\n
      • \n
    • 附件: \n 如果翻译过程中生成了附加文件（如自动生成的术语表），可在此处下载。\n
      • \n
    \n
    6. \n
\n
\n 重要提示: \n 所有配置都会自动保存在您的浏览器本地，方便下次使用。您也可以使用新增的“导出配置”和“导入配置”按钮来备份和恢复您的设置。\n
", + "tutorialModalBody": "

视频教程可以在B站搜索 docutranslate 获取。

欢迎使用 DocuTranslate！请按照以下步骤完成文档翻译：

  1. 第一步：选择工作流

    在左侧配置面板的顶部，首先选择最适合您文件类型的处理流程。

    提示: 默认已开启“自动选择工作流”。您只需上传文件，系统会自动为您匹配合适的工作流，简化操作。

    • 转Markdown再翻译: 适用于翻译PDF、markdown、图片等文件。这是最通用和强大的模式。
    • 纯文本翻译: 用于翻译 .txt 纯文本文件。
    • EPUB翻译: 用于翻译 .epub 电子书文件。
    • DOCX翻译: 用于翻译 .docx Word文档。
    • XLSX翻译: 用于翻译 .xlsx 或 .csv 电子表格文件。
    • SRT字幕翻译: 用于翻译 .srt 字幕文件。
    • ASS字幕翻译: 用于翻译 .ass 特效字幕文件。
    • JSON翻译: 用于翻译 .json 文件中的特定字段。
    • HTML翻译: 用于翻译 .html 网页文件。
  2. 第二步：配置参数

    选择工作流后，下方会显示相关的配置选项。请依次完成设置（所有配置都会自动保存在您的浏览器中）：

    A. 工作流特定选项 (根据您第一步的选择出现):

    • 如果选择“转Markdown再翻译”，请配置 解析配置：
      • 解析引擎: 选择一个引擎将您的文件（如PDF）转换为适合翻译的Markdown格式。如果您的文件已经是Markdown格式，则无需选择。
      • Mineru Token: 如果您选择 minerU 引擎，需要在此处填入您的Token。
    • 如果选择“纯文本/DOCX/XLSX/SRT/ASS/EPUB/HTML”，请配置其 翻译选项：
      • 插入模式: 定义翻译结果如何放入文档。您可以选择直接“替换”原文，或是在原文之后“附加”，或是在原文之前“前置”。
      • 分隔符: 当选择“附加”或“前置”模式时，此项用于在原文和译文之间插入分隔符（例如ASS格式常用 \\N，EPUB格式常用 <br /> 作为换行分隔符）。
    • 如果选择“JSON翻译”，请配置 JSON路径：
      • 需要翻译的JSON路径: 每行输入一个 JSONPath 表达式，将翻译所有与路径匹配的对象中的字符串。例如:$.*(翻译全部字符串)，$..description(翻译所有键为description的值)。

    B. 通用选项 (适用于所有工作流):

    • 翻译模型:
      • 选择平台/API 地址/API Key/模型ID: 配置您希望使用的AI翻译服务。模型能力指令遵循越强，出错、漏翻的概率越低。
      • 跳过翻译: 勾选此项后，将只执行文档解析和格式转换，不调用AI进行翻译。
    • 翻译配置:
      • 目标语言: 指定翻译的目标语言。
      • 自定义Prompt: 可选，添加额外指令，如“人名保持原文不翻译”。
      • 思考模式: 针对部分支持混合推理的模型进行设置，建议选择“禁用(推荐)”。
      • 分块大小/并发数等: 高级参数，用于调整性能和API请求行为，通常保持默认即可。
    • 术语表:
      • 上传术语表 (可选): 上传CSV文件（需包含'src'和'dst'列）来保证特定术语翻译的统一性和准确性。
      • 自动生成术语表: 启用后，程序会先从原文中提取术语并生成一个术语表，然后再进行翻译。
  3. 第三步：上传文件

    在右侧的任务列表中，点击或拖拽您的文档到文件上传区域。

  4. 第四步：开始翻译

    文件选择成功后，点击任务卡片右下角的 开始翻译 按钮。系统将开始处理任务，您可以在日志区域查看实时进度。

  5. 第五步：查看与下载

    翻译完成后，任务卡片下方会出现操作按钮：

    • 预览: 在右侧滑出的面板中进行原文和译文的对照预览。
    • 下载: 下载包括 PDF, DOCX, Markdown 等多种格式的译文。
    • 附件: 如果翻译过程中生成了附加文件（如自动生成的术语表），可在此处下载。
重要提示: 所有配置都会自动保存在您的浏览器本地，方便下次使用。您也可以使用新增的“导出配置”和“导入配置”按钮来备份和恢复您的设置。
", "tutorialUnderstandBtn": "我明白了", "contributorsModalTitle": "感谢贡献", "contributorsPara1": "DocuTranslate是一个开源项目！大家的需求与使用是项目进步的动力。", @@ -149,46 +152,46 @@ "glossaryTableDestination": "译文 (dst)", "init_i18n_failed_alert": "加载界面翻译资源失败，请检查网络连接或联系管理员。", "init_failed_alert": "初始化失败，无法连接到后端服务。请检查服务是否运行或刷新页面。", - "apiHrefInfo302ai": "👈 通过此链接注册可享1美金免费额度", - "glossaryEmpty": "术语表为空。", + "status_selectFileFirst": "请先选择文件", + "status_fillRequired": "请填写所有必填项", + "btn_initializing": "初始化中...", + "status_encodingAndSubmitting": "文件编码和提交中...", + "status_requestOk": "请求成功，任务已开始", + "btn_cancelTranslation": "取消翻译", + "status_requestFail": "请求失败", + "status_initFail": "初始化失败", + "status_cancelling": "取消中...", + "status_cancelSent": "已发送取消请求", + "status_cancelFail": "取消失败", + "status_gettingStatus": "获取状态中...", + "btn_reTranslate": "重新翻译", + "status_updateError": "状态更新失败", + "preview_loading": "加载预览中...", + "preview_cantReadOriginal": "无法读取原文文件内容", + "preview_cantPreviewType": "无法预览此文件类型", + "preview_noOriginalCache": "无原文文件缓存，无法预览", + "preview_loadFailed": "预览加载失败", + "pdf_preparing": "PDF生成中，请稍候...", + "pdf_print_failed": "调用打印功能失败，请尝试手动右键打印", + "pdf_fetch_failed": "获取预览内容失败，无法生成PDF", + "preview_bilingual": "双语预览", + "preview_translatedOnly": "仅译文预览", + "admin_tasklist_failed": "管理员模式：加载任务列表失败", + "configImportSuccess": "配置导入成功！", + "configImportError": "配置导入失败，请检查文件格式是否正确。", + "apiHrefInfo302ai": "👈 通过此链接注册可享1美元免费额度", + "glossaryEmpty": "术语表为空", "engineOptionIdentity": "已经是Markdown格式", "engineOptionMineru": "Mineru（推荐）", "engineOptionDocling": "Docling", - "engineOptionMineruDeploy": "Mineru部署服务", - "preview_loading": "正在加载预览...", - "preview_cantReadOriginal": "无法读取原始文件内容。", - "preview_cantPreviewType": "无法预览此文件类型", - "preview_noOriginalCache": "无原始文件缓存可供预览。", - "preview_loadFailed": "预览加载失败。", - "pdf_preparing": "正在准备PDF，请稍候...", - "pdf_print_failed": "调用打印功能失败。请检查浏览器设置或手动保存为PDF。", - "pdf_fetch_failed": "获取译文内容失败，无法生成PDF。", - "preview_bilingual": "双语对照预览", - "preview_translatedOnly": "仅译文预览", - "admin_tasklist_failed": "管理员模式：获取任务列表失败。", - "status_selectFileFirst": "请先选择文件！", - "status_fillRequired": "请填写所有必填项！", - "btn_initializing": "初始化中...", - "status_encodingAndSubmitting": "正在编码并提交任务...", - "status_requestOk": "请求成功，任务已开始。", - "btn_cancelTranslation": "取消翻译", - "status_requestFail": "请求失败", - "status_initFail": "任务初始化失败", - "status_cancelling": "取消中...", - "status_cancelSent": "已发送取消请求。", - "status_cancelFail": "取消失败", - "status_gettingStatus": "正在获取状态...", - "btn_reTranslate": "重新翻译", - "status_updateError": "状态更新出错", - "configImportSuccess": "配置导入成功！设置已应用。", - "configImportError": "导入失败！文件格式错误或内容不合法。" + "engineOptionMineruDeploy": "Mineru部署服务" }, "en": { "pageTitle": "DocuTranslate - Interactive Document Translation", "tutorialBtn": "Tutorial", - "projectContributeBtn": "Contribute", + "projectContributeBtn": "Project Contribution", "workflowTitle": "Select Workflow", - "workflowOptionMarkdown": "To Markdown then Translate (.pdf/.md/.png, etc.)", + "workflowOptionMarkdown": "Convert to Markdown then Translate (.pdf/.md/.png, etc.)", "workflowOptionTxt": "Plain Text Translation (.txt)", "workflowOptionEpub": "EPUB Translation (.epub)", "workflowOptionDocx": "DOCX Translation (.docx)", @@ -197,20 +200,20 @@ "workflowOptionAss": "ASS Subtitle Translation (.ass)", "workflowOptionJson": "JSON Translation (.json)", "workflowOptionHtml": "HTML Translation (.html)", - "autoWorkflowLabel": "Auto-select workflow", + "autoWorkflowLabel": "Auto-select Workflow", "txtSettingsTitleText": "TXT Translation Options", "insertModeLabel": "Insert Mode", - "insertModeReplace": "Replace Original", - "insertModeAppend": "Append to Original", - "insertModePrepend": "Prepend to Original", + "insertModeReplace": "Replace Original (Replace)", + "insertModeAppend": "Append to Original (Append)", + "insertModePrepend": "Prepend to Original (Prepend)", "insertModeHelpTxt": "Choose how to insert the translated text.", "separatorLabel": "Separator", "separatorPlaceholderSimple": "e.g., \\n---\\n", - "separatorHelp": "Characters to separate original and translated text in append/prepend mode. \\n for new line.", + "separatorHelp": "Character used to separate original and translated text in append or prepend mode. \\n represents a newline.", "docxSettingsTitleText": "DOCX Translation Options", "insertModeHelpDocx": "Choose how to insert the translated text.", "separatorPlaceholderStructured": "e.g., ---", - "separatorHelpDocx": "In append mode, translation starts in a new paragraph. This adds extra content between paragraphs. \\n for new lines inside.", + "separatorHelpDocx": "In append mode, the translation will start a new paragraph. This is used to add extra content between the original and translated paragraphs. \\n can be used for newlines within the separator.", "xlsxSettingsTitleText": "XLSX Translation Options", "insertModeHelpXlsx": "Choose how to insert the translated text into cells.", "xlsxTranslateRegionsLabel": "Translation Regions (Optional)", @@ -219,32 +222,35 @@ "insertModeHelpSrt": "Choose how to insert the translated text.", "epubSettingsTitleText": "EPUB Translation Options", "insertModeHelpEpub": "Choose how to insert the translated text.", - "separatorHelpEpub": "In append mode, translation starts in a new block. This adds extra content between blocks. \\n for new lines inside.", + "separatorHelpEpub": "In append mode, the translation will start a new block. This is used to add extra content between the original and translated blocks. \\n can be used for newlines within the separator.", "htmlSettingsTitleText": "HTML Translation Options", "insertModeHelpHtml": "Choose how to insert the translated text.", - "separatorHelpHtml": "In append mode, translation starts in a new block. This adds extra content between blocks. \\n for new lines inside.", + "separatorHelpHtml": "In append mode, the translation will start a new block. This is used to add extra content between the original and translated blocks. \\n can be used for newlines within the separator.", "assSettingsTitleText": "ASS Translation Options", "insertModeHelpAss": "Choose how to insert the translated text.", - "separatorPlaceholderAss": "e.g., \\N (newline)", - "separatorHelpAss": "Characters to separate original and translated text in append/prepend mode. \\N is the newline for ASS format.", + "separatorPlaceholderAss": "e.g., \\N (newline character)", + "separatorHelpAss": "Character used to separate original and translated text in append or prepend mode. \\N is the newline character for the ASS format.", "jsonSettingsTitleText": "JSON Path Configuration", "jsonPathLabel": "JSON Paths to Translate", "jsonPathPlaceholder": "One path per line, e.g.:\n$.name\n$.*", - "jsonPathHelp": "Uses jsonpath-ng syntax. Each line is a JSON path. All strings within matching objects will be translated.", + "jsonPathHelp": "Uses jsonpath-ng syntax. Each line represents a JSON path.\n All strings within the matched objects will be translated.", "parsingSettingsTitleText": "Parsing Configuration", "parsingEngineLabel": "Parsing Engine", - "parsingEngineHelp": "Not required if your uploaded file is already in .md format.", + "parsingEngineHelp": "If the uploaded file is already in .md format, this option is not required.", "getMineruTokenTitle": "Get Mineru Token", "mineruTokenPlaceholder": "Required when using Mineru engine", "modelVersionLabel": "Mineru Model Version", "modelVersionVlm": "VLM", "modelVersionPipline": "Pipeline", - "modelVersionHelp": "Mineru VLM is a newer, internal test model.", + "modelVersionHelp": "mineru VLM is a newer model in closed beta.", "mineruDeployBaseUrlLabel": "Service Address (Base URL)", "mineruDeployBaseUrlPlaceholder": "e.g., http://127.0.0.1:8000", "mineruDeployBackendLabel": "Backend Type", "mineruDeployBackendPipeline": "Pipeline", - "mineruDeployBackendVlm": "VLM", + "mineruDeployBackendTransformers": "Transformers", + "mineruDeployBackendMlx": "MLX Engine", + "mineruDeployBackendVllm": "VLLM Engine", + "mineruDeployBackendVllmAsync": "VLLM Async Engine", "mineruDeployStartPageLabel": "Start Page", "mineruDeployEndPageLabel": "End Page", "mineruDeployFormulaEnableLabel": "Enable Formula Parsing", @@ -253,9 +259,9 @@ "aiSettingsTitleText": "Translation Model", "skipTranslationLabel": "Skip Translation", "platformLabel": "Select Platform", - "platformCustom": "Custom Endpoint", + "platformCustom": "Custom API", "baseUrlLabel": "API Address (Base URL)", - "baseUrlPlaceholder": "OpenAI-compatible address", + "baseUrlPlaceholder": "OpenAI-compatible Address", "getApiKeyTitle": "Get API Key", "apiKeyPlaceholder": "Please enter your API Key", "modelIdLabel": "Model ID", @@ -266,65 +272,65 @@ "targetLanguageCustom": "Other (Custom)", "customLangPlaceholder": "Enter target language, e.g., Italian", "thinkingModeLabel": "Thinking Mode", - "thinkingModeTooltip": "Sets whether the hybrid inference model should 'think'. Currently supports Zhipu's glm4.5 series, Volcengine's seed1.6 series, SiliconFlow platform, Google's Gemini series, and 302AI (partial). Disabling is recommended.", + "thinkingModeTooltip": "Sets whether mixed-inference models should 'think'. Currently supported by Zhipu's glm4.5 series, Volcengine's seed1.6 series, SiliconFlow, Google's Gemini series, and 302AI (partial). Disabling thinking is recommended.", "thinkingModeEnable": "Enable", "thinkingModeDisable": "Disable (Recommended)", "thinkingModeDefault": "Default", "customPromptLabel": "Custom Prompt", - "customPromptPlaceholder": "Optional, e.g., 'Keep names in their original language.'", + "customPromptPlaceholder": "Optional, e.g., 'Do not translate proper names.'", "chunkSizeLabel": "Chunk Size", "resetBtn": "Reset", "concurrentLabel": "Concurrency", "retryLabel": "Retries", "glossaryGenTitle": "Glossary", "glossaryLabel": "Glossary (Optional)", - "glossaryHelp": "Select one or more CSV files. Files must have 'src' and 'dst' columns for source and destination terms.", + "glossaryHelp": "Select one or more CSV files. Files must contain 'src' and 'dst' headers, representing source and destination text respectively.", "viewGlossaryBtn": "View Glossary", "clearGlossaryBtn": "Clear", - "glossaryGenEnableLabel": "Auto-generate glossary", + "glossaryGenEnableLabel": "Auto-generate Glossary", "glossaryCustomPromptLabel": "Custom Prompt", - "glossaryCustomPromptPlaceholder": "Prompt for glossary generation", + "glossaryCustomPromptPlaceholder": "Glossary generation prompt", "glossaryGenConfigLabel": "Glossary Generation Config", - "glossaryGenConfigSame": "Same as Translation", + "glossaryGenConfigSame": "Same as Translation Config", "glossaryGenConfigCustom": "Custom", "importConfigBtn": "Import Config", "exportConfigBtn": "Export Config", - "githubInfo": "GitHub Homepage (stars❤ are welcome):
\n https://github.com/xunbu/docutranslate", + "githubInfo": "GitHub (star welcome❤):
\n https://github.com/xunbu/docutranslate", "qqGroupInfo": "QQ Group: 1047781902", "taskListTitle": "Task List", "newTaskBtn": "New Task", "noTaskPlaceholder": "No tasks yet. Click 'New Task' to get started!", "taskCardIdLabel": "Task ID", - "taskCardIdPlaceholder": "Awaiting submission...", + "taskCardIdPlaceholder": "Waiting for submission...", "taskCardFileDrop": "Click or drag file here", - "taskCardFileSelected": "File selected", + "taskCardFileSelected": "File Selected", "taskCardFilenameLabel": "Filename: ", - "taskCardLogLabel": "Logs", - "copyLogsTooltip": "Copy logs", + "taskCardLogLabel": "Log", + "copyLogsTooltip": "Copy Logs", "taskCardStatusWaiting": "Waiting for file upload...", "taskCardPreviewBtn": "Preview", "taskCardDownloadBtn": "Download", "taskCardAttachmentBtn": "Attachments", "taskCardStartBtn": "Start Translation", - "downloadMdEmbedded": "Markdown (Embedded Img)", - "downloadMdZip": "Markdown ZIP", + "downloadMdEmbedded": "Markdown (Embedded Images)", + "downloadMdZip": "Markdown (Zip)", "downloadAss": "ASS", "previewTitle": "Preview", "previewBilingualBtn": "Bilingual", - "previewTranslatedOnlyBtn": "Translation Only", + "previewTranslatedOnlyBtn": "Translated Only", "syncScrollTooltip": "Sync Scrolling", "previewOriginal": "Original", - "previewTranslated": "Translation", + "previewTranslated": "Translated", "closeBtn": "Close", "downloadBtn": "Download", - "tutorialModalTitle": "User Guide", - "tutorialModalBody": "

Video tutorials are available on Bilibili by searching for docutranslate.

Welcome to DocuTranslate! Follow these steps to translate your documents:

  1. Step 1: Select Workflow

    At the top of the left-side settings panel, choose the processing flow that best suits your file type.

    Tip: 'Auto-select workflow' is enabled by default. Just upload your file, and the system will automatically match the appropriate workflow for you, simplifying the process.
    • To Markdown then Translate: Suitable for translating PDFs, markdown files, images, etc. This is the most versatile and powerful mode.
    • Plain Text Translation: For translating .txt files.
    • EPUB Translation: For translating .epub e-books.
    • DOCX Translation: For translating .docx Word documents.
    • XLSX Translation: For translating .xlsx or .csv spreadsheets.
    • SRT Subtitle Translation: For translating .srt subtitle files.
    • ASS Subtitle Translation: For translating .ass advanced subtitle files.
    • JSON Translation: For translating specific fields within .json files.
    • HTML Translation: For translating .html web files.
  2. Step 2: Configure Parameters

    After selecting a workflow, relevant configuration options will appear below. Please complete the settings in order (all configurations are automatically saved in your browser):

    A. Workflow-Specific Options (appear based on your choice in Step 1):

    • If 'To Markdown then Translate' is selected, configure Parsing Configuration:
      • Parsing Engine: Choose an engine to convert your file (like a PDF) into a translation-friendly Markdown format. If your file is already Markdown, no selection is needed.
      • Mineru Token: If you select the minerU engine, you must enter your token here.
    • If 'Plain Text/DOCX/XLSX/SRT/ASS/EPUB/HTML' is selected, configure its Translation Options:
      • Insert Mode: Defines how the translation result is placed in the document. You can choose to 'Replace' the original, 'Append' after it, or 'Prepend' before it.
      • Separator: When 'Append' or 'Prepend' is selected, this is used to insert a separator between the original and translated text (e.g., \\N is common for ASS, <br /> for EPUB).
    • If 'JSON Translation' is selected, configure JSON Paths:
      • JSON Paths to Translate: Enter one JSONPath expression per line to translate all strings within matching objects. E.g., $.* (translates all strings), $..description (translates all values with the key description).

    B. General Options (apply to all workflows):

    • Translation Model:
      • Select Platform/API Address/API Key/Model ID: Configure the AI translation service you wish to use. The stronger the model's instruction-following capability, the lower the chance of errors or missed translations.
      • Skip Translation: If checked, only document parsing and format conversion will be performed, without calling an AI for translation.
    • Translation Configuration:
      • Target Language: Specify the language to translate to.
      • Custom Prompt: Optional, add extra instructions, like 'Keep personal names in their original language.'
      • Thinking Mode: For models that support hybrid inference; 'Disable (Recommended)' is suggested.
      • Chunk Size/Concurrency, etc.: Advanced parameters to tune performance and API request behavior. The defaults are usually fine.
    • Glossary:
      • Upload Glossary (Optional): Upload a CSV file (must have 'src' and 'dst' columns) to ensure consistency and accuracy for specific terms.
      • Auto-generate glossary: When enabled, the program first extracts terms from the original text to generate a glossary, then proceeds with the translation.
  3. Step 3: Upload File

    In the task list on the right, click or drag your document into the file upload area.

  4. Step 4: Start Translation

    Once the file is selected, click the Start Translation button on the task card. The system will begin processing, and you can watch the real-time progress in the log area.

  5. Step 5: View & Download

    After the translation is complete, action buttons will appear on the task card:

    • Preview: Compare the original and translated text side-by-side in a slide-out panel.
    • Download: Download the translation in various formats, including PDF, DOCX, and Markdown.
    • Attachments: If any additional files were generated (like an auto-generated glossary), download them here.
Important Note: All settings are automatically saved locally in your browser for your convenience. You can also use the new 'Export Config' and 'Import Config' buttons to back up and restore your settings.
", + "tutorialModalTitle": "Tutorial", + "tutorialModalBody": "

Video tutorials can be found by searching for docutranslate on Bilibili.

Welcome to DocuTranslate! Please follow these steps to translate your documents:

  1. Step 1: Select Workflow

    At the top of the left settings panel, first select the processing flow that best suits your file type.

    Tip: 'Auto-select Workflow' is enabled by default. Simply upload your file, and the system will automatically match the appropriate workflow for you, simplifying the process.

    • Convert to Markdown then Translate: Suitable for translating PDF, Markdown, images, etc. This is the most versatile and powerful mode.
    • Plain Text Translation: For translating .txt plain text files.
    • EPUB Translation: For translating .epub e-book files.
    • DOCX Translation: For translating .docx Word documents.
    • XLSX Translation: For translating .xlsx or .csv spreadsheet files.
    • SRT Subtitle Translation: For translating .srt subtitle files.
    • ASS Subtitle Translation: For translating .ass advanced subtitle files.
    • JSON Translation: For translating specific fields in .json files.
    • HTML Translation: For translating .html web page files.
  2. Step 2: Configure Parameters

    After selecting a workflow, the relevant configuration options will appear below. Please complete the settings in order (all configurations are automatically saved in your browser):

    A. Workflow-Specific Options (Appear based on your choice in Step 1):

    • If you selected 'Convert to Markdown then Translate', configure Parsing Configuration:
      • Parsing Engine: Choose an engine to convert your file (like a PDF) into a translation-friendly Markdown format. If your file is already in Markdown format, no selection is needed.
      • Mineru Token: If you choose the minerU engine, you need to enter your Token here.
    • If you selected 'Plain Text/DOCX/XLSX/SRT/ASS/EPUB/HTML', configure its Translation Options:
      • Insert Mode: Defines how the translation result is placed in the document. You can choose to directly 'Replace' the original, 'Append' after the original, or 'Prepend' before the original.
      • Separator: When 'Append' or 'Prepend' mode is selected, this is used to insert a separator between the original and translated text (e.g., \\N is often used in ASS format, <br /> in EPUB format as a newline separator).
    • If you selected 'JSON Translation', configure JSON Paths:
      • JSON Paths to Translate: Enter one JSONPath expression per line. It will translate all strings in objects matching the path. For example: $.* (translate all strings), $..description (translate all values with the key description).

    B. General Options (Apply to all workflows):

    • Translation Model:
      • Select Platform/API Address/API Key/Model ID: Configure the AI translation service you wish to use. The stronger the model's capabilities, the lower the probability of errors and missed translations.
      • Skip Translation: If checked, only document parsing and format conversion will be performed, without calling the AI for translation.
    • Translation Configuration:
      • Target Language: Specify the target language for the translation.
      • Custom Prompt: Optional, add extra instructions, such as 'Do not translate proper names.'
      • Thinking Mode: A setting for some models that support mixed-inference. It is recommended to select 'Disable (Recommended)'.
      • Chunk Size/Concurrency, etc.: Advanced parameters for adjusting performance and API request behavior. Usually, the default values are fine.
    • Glossary:
      • Upload Glossary (Optional): Upload a CSV file (must contain 'src' and 'dst' columns) to ensure consistency and accuracy of specific term translations.
      • Auto-generate Glossary: When enabled, the program will first extract terms from the original text to generate a glossary before proceeding with the translation.
  3. Step 3: Upload File

    In the task list on the right, click or drag your document into the file upload area.

  4. Step 4: Start Translation

    After the file is successfully selected, click the Start Translation button at the bottom right of the task card. The system will start processing the task, and you can view the real-time progress in the log area.

  5. Step 5: View and Download

    Once the translation is complete, action buttons will appear at the bottom of the task card:

    • Preview: Compare the original and translated text in a side-panel that slides out.
    • Download: Download the translated document in various formats, including PDF, DOCX, Markdown, etc.
    • Attachments: If any additional files were generated during the translation process (like an auto-generated glossary), they can be downloaded here.
Important Note: All configurations are automatically saved locally in your browser for future use. You can also use the new 'Export Config' and 'Import Config' buttons to back up and restore your settings.
", "tutorialUnderstandBtn": "I Understand", - "contributorsModalTitle": "Thank You for Contributing", + "contributorsModalTitle": "Thanks for Contributing", "contributorsPara1": "DocuTranslate is an open-source project! The community's needs and usage are the driving force behind its progress.", - "contributorsPara2": "A heartfelt thank you to everyone who has sponsored the project, submitted code, provided valuable suggestions, and starred the repository!", - "contributorsWelcome": "You are welcome to contribute in the following ways:", - "contributorsGithub": "GitHub Page", + "contributorsPara2": "Thank you to everyone who has funded the project, submitted code, provided valuable suggestions, and starred the project!", + "contributorsWelcome": "We welcome contributions in the following ways:", + "contributorsGithub": "GitHub Homepage", "contributorsPR": "Submit a Pull Request", "contributorsIssue": "Report an Issue", "contributorsQQ": "Or contact the author via QQ Group: 1047781902", @@ -333,38 +339,38 @@ "glossaryTableDestination": "Destination (dst)", "init_i18n_failed_alert": "Failed to load interface translations. Please check your network connection or contact an administrator.", "init_failed_alert": "Initialization failed, could not connect to the backend service. Please ensure the service is running and refresh the page.", - "apiHrefInfo302ai": "👈 Register through this link to enjoy a $1 free credit", - "glossaryEmpty": "The glossary is empty.", - "engineOptionIdentity": "Already Markdown", - "engineOptionMineru": "Mineru (Recommended)", - "engineOptionDocling": "Docling", - "engineOptionMineruDeploy": "Mineru Deploy Service", - "preview_loading": "Loading preview...", - "preview_cantReadOriginal": "Could not read original file content.", - "preview_cantPreviewType": "Cannot preview this file type", - "preview_noOriginalCache": "No original file cached for preview.", - "preview_loadFailed": "Failed to load preview.", - "pdf_preparing": "Preparing PDF, please wait...", - "pdf_print_failed": "Failed to invoke print function. Please check browser settings or save as PDF manually.", - "pdf_fetch_failed": "Failed to fetch translated content, cannot generate PDF.", - "preview_bilingual": "Bilingual Preview", - "preview_translatedOnly": "Translation Only Preview", - "admin_tasklist_failed": "Admin Mode: Failed to fetch task list.", - "status_selectFileFirst": "Please select a file first!", - "status_fillRequired": "Please fill in all required fields!", + "status_selectFileFirst": "Please select a file first", + "status_fillRequired": "Please fill in all required fields", "btn_initializing": "Initializing...", - "status_encodingAndSubmitting": "Encoding and submitting task...", - "status_requestOk": "Request successful, task has started.", - "btn_cancelTranslation": "Cancel", + "status_encodingAndSubmitting": "Encoding and submitting file...", + "status_requestOk": "Request successful, task has started", + "btn_cancelTranslation": "Cancel Translation", "status_requestFail": "Request failed", - "status_initFail": "Task initialization failed", + "status_initFail": "Initialization failed", "status_cancelling": "Cancelling...", - "status_cancelSent": "Cancellation request sent.", + "status_cancelSent": "Cancellation request sent", "status_cancelFail": "Cancellation failed", "status_gettingStatus": "Getting status...", "btn_reTranslate": "Re-translate", - "status_updateError": "Error updating status", - "configImportSuccess": "Configuration imported successfully! Settings have been applied.", - "configImportError": "Import failed! Invalid file format or content." + "status_updateError": "Status update failed", + "preview_loading": "Loading preview...", + "preview_cantReadOriginal": "Could not read original file content", + "preview_cantPreviewType": "Cannot preview this file type", + "preview_noOriginalCache": "No original file cached, cannot preview", + "preview_loadFailed": "Preview failed to load", + "pdf_preparing": "Generating PDF, please wait...", + "pdf_print_failed": "Failed to invoke print function. Please try right-clicking to print manually.", + "pdf_fetch_failed": "Failed to fetch preview content, cannot generate PDF.", + "preview_bilingual": "Bilingual Preview", + "preview_translatedOnly": "Translated Only Preview", + "admin_tasklist_failed": "Admin mode: Failed to load task list.", + "configImportSuccess": "Configuration imported successfully!", + "configImportError": "Configuration import failed. Please check if the file format is correct.", + "apiHrefInfo302ai": "👈 Register through this link to enjoy a $1 free credit", + "glossaryEmpty": "Glossary is empty", + "engineOptionIdentity": "Already in Markdown format", + "engineOptionMineru": "Mineru (Recommended)", + "engineOptionDocling": "Docling", + "engineOptionMineruDeploy": "Mineru Deploy Service" } } \ No newline at end of file diff --git a/docutranslate/static/index.html b/docutranslate/static/index.html index 535549e..a0d3870 100644 --- a/docutranslate/static/index.html +++ b/docutranslate/static/index.html @@ -1 +1 @@ - DocuTranslate - 交互式文档翻译

DocuTranslate

如果上传的文件本身是.md格式，此项可不选。

Base URL:

选择一个或多个CSV文件。文件需包含'src'和'dst'两列标题，分别代表原文和译文。

GitHub主页(欢迎star❤):
https://github.com/xunbu/docutranslate

交流QQ群: 1047781902

version:

任务列表

LOGO

当前没有任务，点击“新建任务”开始吧！

预览
原文
译文
\ No newline at end of file + DocuTranslate - 交互式文档翻译

DocuTranslate

如果上传的文件本身是.md格式，此项可不选。

Base URL:

选择一个或多个CSV文件。文件需包含'src'和'dst'两列标题，分别代表原文和译文。

GitHub主页(欢迎star❤):
https://github.com/xunbu/docutranslate

交流QQ群: 1047781902

version:

任务列表

LOGO

当前没有任务，点击“新建任务”开始吧！

预览
原文
译文
\ No newline at end of file