Add image index and pdf parse (#3956)

* feat: think tag parse

* feat: parse think tag test

* feat: pdf parse ux

* feat: doc2x parse

* perf: rewrite training mode setting

* feat: image parse queue

* perf: image index

* feat: image parse process

* feat: add init sh

* fix: ts

packages/web/i18n/zh-CN/dataset.json

@@ -3,11 +3,16 @@
   "add_file": "添加文件",
   "api_file": "API 文件库",
   "api_url": "接口地址",
+  "auto_indexes": "自动生成补充索引",
+  "auto_indexes_tips": "通过大模型进行额外索引生成，提高语义丰富度，提高检索的精度。",
   "chunk_max_tokens": "分块上限",
   "close_auto_sync": "确认关闭自动同步功能？",
   "collection.Create update time": "创建/更新时间",
   "collection.Training type": "训练模式",
+  "collection.training_type": "处理模式",
   "collection_data_count": "数据量",
+  "collection_metadata_custom_pdf_parse": "PDF增强解析",
+  "collection_metadata_image_parse": "图片标注",
   "collection_not_support_retraining": "该集合类型不支持重新调整参数",
   "collection_not_support_sync": "该集合不支持同步",
   "collection_sync": "立即同步",
@@ -22,12 +27,21 @@
   "custom_data_process_params_desc": "自定义设置数据处理规则",
   "data.ideal_chunk_length": "理想分块长度",
   "data_amount": "{{dataAmount}} 组数据, {{indexAmount}} 组索引",
+  "data_index_custom": "自定义索引",
+  "data_index_default": "默认索引",
+  "data_index_image": "图片索引",
+  "data_index_num": "索引 {{index}}",
+  "data_index_question": "推测问题索引",
+  "data_index_summary": "摘要索引",
   "data_process_params": "处理参数",
   "data_process_setting": "数据处理配置",
   "dataset.Unsupported operation": "操作不支持",
   "dataset.no_collections": "暂无数据集",
   "dataset.no_tags": "暂无标签",
+  "default_params": "默认",
+  "default_params_desc": "使用系统默认的参数和规则",
   "edit_dataset_config": "编辑知识库配置",
+  "enhanced_indexes": "索引增强",
   "error.collectionNotFound": "集合找不到了~",
   "external_file": "外部文件库",
   "external_file_dataset_desc": "可以从外部文件库导入文件构建知识库，文件不会进行二次存储",
@@ -38,19 +52,38 @@
   "feishu_dataset": "飞书知识库",
   "feishu_dataset_config": "配置飞书知识库",
   "feishu_dataset_desc": "可通过配置飞书文档权限，使用飞书文档构建知识库，文档不会进行二次存储",
+  "file_list": "文件列表",
   "file_model_function_tip": "用于增强索引和 QA 生成",
   "filename": "文件名",
   "folder_dataset": "文件夹",
   "ideal_chunk_length": "理想分块长度",
   "ideal_chunk_length_tips": "按结束符号进行分段，并将多个分段组成一个分块，该值决定了分块的预估大小，如果会有上下浮动。",
+  "image_auto_parse": "图片自动索引",
+  "image_auto_parse_tips": "调用 VLM 自动标注文档里的图片，并生成额外的检索索引",
   "import.Auto mode Estimated Price Tips": "需调用文本理解模型，需要消耗较多AI 积分：{{price}} 积分/1K tokens",
   "import.Embedding Estimated Price Tips": "仅使用索引模型，消耗少量 AI 积分：{{price}} 积分/1K tokens",
+  "import_confirm": "确认上传",
+  "import_data_preview": "数据预览",
+  "import_data_process_setting": "数据处理方式设置",
+  "import_file_parse_setting": "文件解析设置",
+  "import_model_config": "模型选择",
+  "import_param_setting": "参数设置",
+  "import_select_file": "选择文件",
   "is_open_schedule": "启用定时同步",
+  "keep_image": "保留图片",
   "move.hint": "移动后，所选知识库/文件夹将继承新文件夹的权限设置，原先的权限设置失效。",
   "open_auto_sync": "开启定时同步后，系统将会每天不定时尝试同步集合，集合同步期间，会出现无法搜索到该集合数据现象。",
+  "params_setting": "参数设置",
+  "pdf_enhance_parse": "PDF增强解析",
+  "pdf_enhance_parse_price": "{{price}}积分/页",
+  "pdf_enhance_parse_tips": "解析 PDF 文件时，调用 PDF 识别模型进行识别，可以将其转换成 Markdown 并保留文档中的图片，同时也可以对扫描件进行识别。",
   "permission.des.manage": "可管理整个知识库数据和信息",
   "permission.des.read": "可查看知识库内容",
   "permission.des.write": "可增加和变更知识库内容",
+  "preview_chunk": "分块预览",
+  "preview_chunk_empty": "无法读取该文件内容",
+  "preview_chunk_intro": "最多展示 10 个分块",
+  "preview_chunk_not_selected": "点击左侧文件后进行预览",
   "rebuild_embedding_start_tip": "切换索引模型任务已开始",
   "rebuilding_index_count": "重建中索引数量：{{count}}",
   "request_headers": "请求头参数，会自动补充 Bearer",
@@ -72,8 +105,10 @@
   "tag.tags": "标签",
   "tag.total_tags": "共{{total}}个标签",
   "the_knowledge_base_has_indexes_that_are_being_trained_or_being_rebuilt": "知识库有训练中或正在重建的索引",
+  "total_num_files": "共 {{total}} 个文件",
   "training_mode": "处理方式",
   "vector_model_max_tokens_tip": "每个分块数据，最大长度为 3000 tokens",
+  "vllm_model": "图片理解模型",
   "website_dataset": "Web 站点同步",
   "website_dataset_desc": "Web 站点同步允许你直接使用一个网页链接构建知识库",
   "yuque_dataset": "语雀知识库",