@undefineds.co/xpod 0.3.58 → 0.3.61

package/dist/document/Chunker.jsonld

@@ -0,0 +1,110 @@
+{
+  "@context": [
+    "https://linkedsoftwaredependencies.org/bundles/npm/@undefineds.co/xpod/^0.0.0/components/context.jsonld"
+  ],
+  "@id": "npmd:@undefineds.co/xpod",
+  "components": [
+    {
+      "@id": "undefineds:dist/document/Chunker.jsonld#TextChunk",
+      "@type": "AbstractClass",
+      "requireElement": "TextChunk",
+      "comment": "文本块",
+      "parameters": [],
+      "memberFields": [
+        {
+          "@id": "undefineds:dist/document/Chunker.jsonld#TextChunk__member_id",
+          "memberFieldName": "id"
+        },
+        {
+          "@id": "undefineds:dist/document/Chunker.jsonld#TextChunk__member_level",
+          "memberFieldName": "level"
+        },
+        {
+          "@id": "undefineds:dist/document/Chunker.jsonld#TextChunk__member_heading",
+          "memberFieldName": "heading"
+        },
+        {
+          "@id": "undefineds:dist/document/Chunker.jsonld#TextChunk__member_content",
+          "memberFieldName": "content"
+        },
+        {
+          "@id": "undefineds:dist/document/Chunker.jsonld#TextChunk__member_path",
+          "memberFieldName": "path"
+        },
+        {
+          "@id": "undefineds:dist/document/Chunker.jsonld#TextChunk__member_startOffset",
+          "memberFieldName": "startOffset"
+        },
+        {
+          "@id": "undefineds:dist/document/Chunker.jsonld#TextChunk__member_endOffset",
+          "memberFieldName": "endOffset"
+        },
+        {
+          "@id": "undefineds:dist/document/Chunker.jsonld#TextChunk__member_parentId",
+          "memberFieldName": "parentId"
+        },
+        {
+          "@id": "undefineds:dist/document/Chunker.jsonld#TextChunk__member_children",
+          "memberFieldName": "children"
+        }
+      ],
+      "constructorArguments": []
+    },
+    {
+      "@id": "undefineds:dist/document/Chunker.jsonld#Chunker",
+      "@type": "AbstractClass",
+      "requireElement": "Chunker",
+      "comment": "分块器接口",
+      "parameters": [],
+      "memberFields": [
+        {
+          "@id": "undefineds:dist/document/Chunker.jsonld#Chunker__member_chunk",
+          "memberFieldName": "chunk"
+        },
+        {
+          "@id": "undefineds:dist/document/Chunker.jsonld#Chunker__member_flatten",
+          "memberFieldName": "flatten"
+        }
+      ],
+      "constructorArguments": []
+    },
+    {
+      "@id": "undefineds:dist/document/Chunker.jsonld#ChunkMetadata",
+      "@type": "AbstractClass",
+      "requireElement": "ChunkMetadata",
+      "comment": "块元数据（用于存储到 .meta 文件）",
+      "parameters": [],
+      "memberFields": [
+        {
+          "@id": "undefineds:dist/document/Chunker.jsonld#ChunkMetadata__member_id",
+          "memberFieldName": "id"
+        },
+        {
+          "@id": "undefineds:dist/document/Chunker.jsonld#ChunkMetadata__member_level",
+          "memberFieldName": "level"
+        },
+        {
+          "@id": "undefineds:dist/document/Chunker.jsonld#ChunkMetadata__member_heading",
+          "memberFieldName": "heading"
+        },
+        {
+          "@id": "undefineds:dist/document/Chunker.jsonld#ChunkMetadata__member_startOffset",
+          "memberFieldName": "startOffset"
+        },
+        {
+          "@id": "undefineds:dist/document/Chunker.jsonld#ChunkMetadata__member_endOffset",
+          "memberFieldName": "endOffset"
+        },
+        {
+          "@id": "undefineds:dist/document/Chunker.jsonld#ChunkMetadata__member_parentId",
+          "memberFieldName": "parentId"
+        },
+        {
+          "@id": "undefineds:dist/document/Chunker.jsonld#ChunkMetadata__member_vectorId",
+          "memberFieldName": "vectorId"
+        }
+      ],
+      "constructorArguments": []
+    }
+  ]
+}

package/dist/document/{DocumentParser.d.ts → DocumentReader.d.ts}

@@ -1,12 +1,12 @@
 /**
- * DocumentParser - 文档解析接口
+ * DocumentReader - 文档读取接口
  *
  * 将各种格式的文档（PDF, Office, HTML 等）转换为 Markdown
  */
 /**
- * 解析后的文档
+ * 读取后的文档
  */
-export interface ParsedDocument {
+export interface ReadDocument {
     /** Markdown 格式的内容 */
     markdown: string;
     /** 文档元数据 */
@@ -21,21 +21,29 @@ export interface ParsedDocument {
         contentType?: string;
         /** 字数统计 */
         wordCount?: number;
+        /** reader provider，例如 paddleocr */
+        readerProvider?: string;
+        /** reader model，例如 pp-ocrv6 */
+        readerModel?: string;
+        /** 已读取页段 */
+        pageRange?: string;
+        /** 已读取页数 */
+        pageCount?: number;
     };
     /** 原始 URL */
     rawUrl: string;
 }
 /**
- * 文档解析器接口
+ * 文档读取器接口
  */
-export interface DocumentParser {
+export interface DocumentReader {
     /**
-     * 解析文档为 Markdown
+     * 读取文档为 Markdown
      * @param url 文档 URL
-     * @param options 解析选项
-     * @returns 解析后的文档
+     * @param options 读取选项
+     * @returns 读取后的文档
      */
-    parse(url: string, options?: ParseOptions): Promise<ParsedDocument>;
+    read(url: string, options?: ReadOptions): Promise<ReadDocument>;
     /**
      * 检查是否支持该 URL
      * @param url 文档 URL
@@ -44,9 +52,9 @@ export interface DocumentParser {
     supports(url: string): boolean;
 }
 /**
- * 解析选项
+ * 读取选项
  */
-export interface ParseOptions {
+export interface ReadOptions {
     /** 访问令牌（用于需要认证的资源） */
     accessToken?: string;
     /** 超时时间（毫秒） */
@@ -55,4 +63,8 @@ export interface ParseOptions {
     includeImages?: boolean;
     /** 是否包含链接 */
     includeLinks?: boolean;
+    /** 页码范围，例如 1-20 或 1,3,5-8 */
+    pageRange?: string;
+    /** Agent/系统读取目的，用于 provider adapter 记录与调度 */
+    expectedUse?: string;
 }

package/dist/document/{DocumentParser.js → DocumentReader.js}

@@ -1,8 +1,8 @@
 "use strict";
 /**
- * DocumentParser - 文档解析接口
+ * DocumentReader - 文档读取接口
  *
  * 将各种格式的文档（PDF, Office, HTML 等）转换为 Markdown
  */
 Object.defineProperty(exports, "__esModule", { value: true });
-//# sourceMappingURL=DocumentParser.js.map
+//# sourceMappingURL=DocumentReader.js.map

package/dist/document/DocumentReader.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"DocumentReader.js","sourceRoot":"","sources":["../../src/document/DocumentReader.ts"],"names":[],"mappings":";AAAA;;;;GAIG","sourcesContent":["/**\n * DocumentReader - 文档读取接口\n *\n * 将各种格式的文档（PDF, Office, HTML 等）转换为 Markdown\n */\n\n/**\n * 读取后的文档\n */\nexport interface ReadDocument {\n  /** Markdown 格式的内容 */\n  markdown: string;\n\n  /** 文档元数据 */\n  metadata: {\n    /** 文档标题 */\n    title?: string;\n    /** 文档描述 */\n    description?: string;\n    /** 原始 URL */\n    url: string;\n    /** 内容类型 */\n    contentType?: string;\n    /** 字数统计 */\n    wordCount?: number;\n    /** reader provider，例如 paddleocr */\n    readerProvider?: string;\n    /** reader model，例如 pp-ocrv6 */\n    readerModel?: string;\n    /** 已读取页段 */\n    pageRange?: string;\n    /** 已读取页数 */\n    pageCount?: number;\n  };\n\n  /** 原始 URL */\n  rawUrl: string;\n}\n\n/**\n * 文档读取器接口\n */\nexport interface DocumentReader {\n  /**\n   * 读取文档为 Markdown\n   * @param url 文档 URL\n   * @param options 读取选项\n   * @returns 读取后的文档\n   */\n  read(url: string, options?: ReadOptions): Promise<ReadDocument>;\n\n  /**\n   * 检查是否支持该 URL\n   * @param url 文档 URL\n   * @returns 是否支持\n   */\n  supports(url: string): boolean;\n}\n\n/**\n * 读取选项\n */\nexport interface ReadOptions {\n  /** 访问令牌（用于需要认证的资源） */\n  accessToken?: string;\n  /** 超时时间（毫秒） */\n  timeout?: number;\n  /** 是否包含图片描述 */\n  includeImages?: boolean;\n  /** 是否包含链接 */\n  includeLinks?: boolean;\n  /** 页码范围，例如 1-20 或 1,3,5-8 */\n  pageRange?: string;\n  /** Agent/系统读取目的，用于 provider adapter 记录与调度 */\n  expectedUse?: string;\n}\n"]}

package/dist/document/DocumentReader.jsonld

@@ -0,0 +1,82 @@
+{
+  "@context": [
+    "https://linkedsoftwaredependencies.org/bundles/npm/@undefineds.co/xpod/^0.0.0/components/context.jsonld"
+  ],
+  "@id": "npmd:@undefineds.co/xpod",
+  "components": [
+    {
+      "@id": "undefineds:dist/document/DocumentReader.jsonld#ReadDocument",
+      "@type": "AbstractClass",
+      "requireElement": "ReadDocument",
+      "comment": "读取后的文档",
+      "parameters": [],
+      "memberFields": [
+        {
+          "@id": "undefineds:dist/document/DocumentReader.jsonld#ReadDocument__member_markdown",
+          "memberFieldName": "markdown"
+        },
+        {
+          "@id": "undefineds:dist/document/DocumentReader.jsonld#ReadDocument__member_metadata",
+          "memberFieldName": "metadata"
+        },
+        {
+          "@id": "undefineds:dist/document/DocumentReader.jsonld#ReadDocument__member_rawUrl",
+          "memberFieldName": "rawUrl"
+        }
+      ],
+      "constructorArguments": []
+    },
+    {
+      "@id": "undefineds:dist/document/DocumentReader.jsonld#DocumentReader",
+      "@type": "AbstractClass",
+      "requireElement": "DocumentReader",
+      "comment": "文档读取器接口",
+      "parameters": [],
+      "memberFields": [
+        {
+          "@id": "undefineds:dist/document/DocumentReader.jsonld#DocumentReader__member_read",
+          "memberFieldName": "read"
+        },
+        {
+          "@id": "undefineds:dist/document/DocumentReader.jsonld#DocumentReader__member_supports",
+          "memberFieldName": "supports"
+        }
+      ],
+      "constructorArguments": []
+    },
+    {
+      "@id": "undefineds:dist/document/DocumentReader.jsonld#ReadOptions",
+      "@type": "AbstractClass",
+      "requireElement": "ReadOptions",
+      "comment": "读取选项",
+      "parameters": [],
+      "memberFields": [
+        {
+          "@id": "undefineds:dist/document/DocumentReader.jsonld#ReadOptions__member_accessToken",
+          "memberFieldName": "accessToken"
+        },
+        {
+          "@id": "undefineds:dist/document/DocumentReader.jsonld#ReadOptions__member_timeout",
+          "memberFieldName": "timeout"
+        },
+        {
+          "@id": "undefineds:dist/document/DocumentReader.jsonld#ReadOptions__member_includeImages",
+          "memberFieldName": "includeImages"
+        },
+        {
+          "@id": "undefineds:dist/document/DocumentReader.jsonld#ReadOptions__member_includeLinks",
+          "memberFieldName": "includeLinks"
+        },
+        {
+          "@id": "undefineds:dist/document/DocumentReader.jsonld#ReadOptions__member_pageRange",
+          "memberFieldName": "pageRange"
+        },
+        {
+          "@id": "undefineds:dist/document/DocumentReader.jsonld#ReadOptions__member_expectedUse",
+          "memberFieldName": "expectedUse"
+        }
+      ],
+      "constructorArguments": []
+    }
+  ]
+}

package/dist/document/HeadingChunker.jsonld

@@ -0,0 +1,41 @@
+{
+  "@context": [
+    "https://linkedsoftwaredependencies.org/bundles/npm/@undefineds.co/xpod/^0.0.0/components/context.jsonld"
+  ],
+  "@id": "npmd:@undefineds.co/xpod",
+  "components": [
+    {
+      "@id": "undefineds:dist/document/HeadingChunker.jsonld#HeadingChunker",
+      "@type": "Class",
+      "requireElement": "HeadingChunker",
+      "extends": [
+        "undefineds:dist/document/Chunker.jsonld#Chunker"
+      ],
+      "comment": "按标题层级分块的实现",
+      "parameters": [],
+      "memberFields": [
+        {
+          "@id": "undefineds:dist/document/HeadingChunker.jsonld#HeadingChunker__member_chunk",
+          "memberFieldName": "chunk"
+        },
+        {
+          "@id": "undefineds:dist/document/HeadingChunker.jsonld#HeadingChunker__member_flatten",
+          "memberFieldName": "flatten"
+        },
+        {
+          "@id": "undefineds:dist/document/HeadingChunker.jsonld#HeadingChunker__member_updateEndOffsets",
+          "memberFieldName": "updateEndOffsets"
+        },
+        {
+          "@id": "undefineds:dist/document/HeadingChunker.jsonld#HeadingChunker__member_findNextSiblingOrHigher",
+          "memberFieldName": "findNextSiblingOrHigher"
+        },
+        {
+          "@id": "undefineds:dist/document/HeadingChunker.jsonld#HeadingChunker__member_generateChunkId",
+          "memberFieldName": "generateChunkId"
+        }
+      ],
+      "constructorArguments": []
+    }
+  ]
+}

package/dist/document/{JinaDocumentParser.d.ts → JinaReader.d.ts}

@@ -1,13 +1,13 @@
 /**
- * JinaDocumentParser - 基于 JINA Reader API 的文档解析器
+ * JinaReader - 基于 JINA Reader API 的文档解析器
  *
  * 使用 JINA Reader API (https://r.jina.ai/) 将各种格式的文档转换为 Markdown
  * 支持：PDF, Office 文档, HTML, 图片等
  *
  * @see https://jina.ai/reader/
  */
-import type { DocumentParser, ParsedDocument, ParseOptions } from './DocumentParser';
-export interface JinaDocumentParserOptions {
+import type { DocumentReader, ReadDocument, ReadOptions } from './DocumentReader';
+export interface JinaReaderOptions {
     /** JINA API Key */
     apiKey: string;
     /** 自定义 API 端点（默认 https://r.jina.ai） */
@@ -18,16 +18,16 @@ export interface JinaDocumentParserOptions {
 /**
  * JINA Reader API 文档解析器
  */
-export declare class JinaDocumentParser implements DocumentParser {
+export declare class JinaReader implements DocumentReader {
     protected readonly logger: import("global-logger-factory").Logger<unknown>;
     private readonly apiKey;
     private readonly baseUrl;
     private readonly defaultTimeout;
-    constructor(options: JinaDocumentParserOptions);
+    constructor(options: JinaReaderOptions);
     /**
-     * 解析文档为 Markdown
+     * 读取文档为 Markdown
      */
-    parse(url: string, options?: ParseOptions): Promise<ParsedDocument>;
+    read(url: string, options?: ReadOptions): Promise<ReadDocument>;
     /**
      * 检查是否支持该 URL
      * JINA Reader 支持大多数 URL

package/dist/document/{JinaDocumentParser.js → JinaReader.js}

@@ -1,6 +1,6 @@
 "use strict";
 /**
- * JinaDocumentParser - 基于 JINA Reader API 的文档解析器
+ * JinaReader - 基于 JINA Reader API 的文档解析器
  *
  * 使用 JINA Reader API (https://r.jina.ai/) 将各种格式的文档转换为 Markdown
  * 支持：PDF, Office 文档, HTML, 图片等
@@ -8,12 +8,12 @@
  * @see https://jina.ai/reader/
  */
 Object.defineProperty(exports, "__esModule", { value: true });
-exports.JinaDocumentParser = void 0;
+exports.JinaReader = void 0;
 const global_logger_factory_1 = require("global-logger-factory");
 /**
  * JINA Reader API 文档解析器
  */
-class JinaDocumentParser {
+class JinaReader {
     constructor(options) {
         this.logger = (0, global_logger_factory_1.getLoggerFor)(this);
         this.apiKey = options.apiKey;
@@ -21,11 +21,11 @@ class JinaDocumentParser {
         this.defaultTimeout = options.defaultTimeout ?? 30000;
     }
     /**
-     * 解析文档为 Markdown
+     * 读取文档为 Markdown
      */
-    async parse(url, options) {
+    async read(url, options) {
         const timeout = options?.timeout ?? this.defaultTimeout;
-        this.logger.debug(`Parsing document: ${url}`);
+        this.logger.debug(`Reading document: ${url}`);
         try {
             // 构建 JINA Reader URL
             const readerUrl = `${this.baseUrl}/${encodeURIComponent(url)}`;
@@ -53,12 +53,12 @@ class JinaDocumentParser {
             clearTimeout(timeoutId);
             if (!response.ok) {
                 const errorText = await response.text();
-                throw new Error(`JINA parse failed: ${response.status} ${errorText}`);
+                throw new Error(`JINA read failed: ${response.status} ${errorText}`);
             }
             const markdown = await response.text();
             // 提取元数据（从 Markdown 头部）
             const metadata = this.extractMetadata(markdown, url);
-            this.logger.info(`Parsed document: ${url}, ${markdown.length} chars`);
+            this.logger.info(`Read document: ${url}, ${markdown.length} chars`);
             return {
                 markdown,
                 metadata,
@@ -67,9 +67,9 @@ class JinaDocumentParser {
         }
         catch (error) {
             if (error instanceof Error && error.name === 'AbortError') {
-                throw new Error(`Document parsing timed out after ${timeout}ms: ${url}`);
+                throw new Error(`Document reading timed out after ${timeout}ms: ${url}`);
             }
-            this.logger.error(`Failed to parse document ${url}: ${error}`);
+            this.logger.error(`Failed to read document ${url}: ${error}`);
             throw error;
         }
     }
@@ -125,5 +125,5 @@ class JinaDocumentParser {
         return metadata;
     }
 }
-exports.JinaDocumentParser = JinaDocumentParser;
-//# sourceMappingURL=JinaDocumentParser.js.map
+exports.JinaReader = JinaReader;
+//# sourceMappingURL=JinaReader.js.map

package/dist/document/JinaReader.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"JinaReader.js","sourceRoot":"","sources":["../../src/document/JinaReader.ts"],"names":[],"mappings":";AAAA;;;;;;;GAOG;;;AAEH,iEAAqD;AAYrD;;GAEG;AACH,MAAa,UAAU;IAOrB,YAAmB,OAA0B;QAN1B,WAAM,GAAG,IAAA,oCAAY,EAAC,IAAI,CAAC,CAAC;QAO7C,IAAI,CAAC,MAAM,GAAG,OAAO,CAAC,MAAM,CAAC;QAC7B,IAAI,CAAC,OAAO,GAAG,OAAO,CAAC,OAAO,IAAI,mBAAmB,CAAC;QACtD,IAAI,CAAC,cAAc,GAAG,OAAO,CAAC,cAAc,IAAI,KAAK,CAAC;IACxD,CAAC;IAED;;OAEG;IACI,KAAK,CAAC,IAAI,CAAC,GAAW,EAAE,OAAqB;QAClD,MAAM,OAAO,GAAG,OAAO,EAAE,OAAO,IAAI,IAAI,CAAC,cAAc,CAAC;QAExD,IAAI,CAAC,MAAM,CAAC,KAAK,CAAC,qBAAqB,GAAG,EAAE,CAAC,CAAC;QAE9C,IAAI,CAAC;YACH,qBAAqB;YACrB,MAAM,SAAS,GAAG,GAAG,IAAI,CAAC,OAAO,IAAI,kBAAkB,CAAC,GAAG,CAAC,EAAE,CAAC;YAE/D,4BAA4B;YAC5B,MAAM,UAAU,GAAG,IAAI,eAAe,EAAE,CAAC;YACzC,MAAM,SAAS,GAAG,UAAU,CAAC,GAAG,EAAE,CAAC,UAAU,CAAC,KAAK,EAAE,EAAE,OAAO,CAAC,CAAC;YAEhE,MAAM,OAAO,GAA2B;gBACtC,aAAa,EAAE,UAAU,IAAI,CAAC,MAAM,EAAE;gBACtC,MAAM,EAAE,eAAe;gBACvB,iBAAiB,EAAE,UAAU;aAC9B,CAAC;YAEF,aAAa;YACb,IAAI,OAAO,EAAE,aAAa,EAAE,CAAC;gBAC3B,OAAO,CAAC,uBAAuB,CAAC,GAAG,MAAM,CAAC;YAC5C,CAAC;YAED,WAAW;YACX,IAAI,OAAO,EAAE,YAAY,KAAK,KAAK,EAAE,CAAC;gBACpC,OAAO,CAAC,sBAAsB,CAAC,GAAG,MAAM,CAAC;YAC3C,CAAC;YAED,MAAM,QAAQ,GAAG,MAAM,KAAK,CAAC,SAAS,EAAE;gBACtC,MAAM,EAAE,KAAK;gBACb,OAAO;gBACP,MAAM,EAAE,UAAU,CAAC,MAAM;aAC1B,CAAC,CAAC;YAEH,YAAY,CAAC,SAAS,CAAC,CAAC;YAExB,IAAI,CAAC,QAAQ,CAAC,EAAE,EAAE,CAAC;gBACjB,MAAM,SAAS,GAAG,MAAM,QAAQ,CAAC,IAAI,EAAE,CAAC;gBACxC,MAAM,IAAI,KAAK,CAAC,qBAAqB,QAAQ,CAAC,MAAM,IAAI,SAAS,EAAE,CAAC,CAAC;YACvE,CAAC;YAED,MAAM,QAAQ,GAAG,MAAM,QAAQ,CAAC,IAAI,EAAE,CAAC;YAEvC,uBAAuB;YACvB,MAAM,QAAQ,GAAG,IAAI,CAAC,eAAe,CAAC,QAAQ,EAAE,GAAG,CAAC,CAAC;YAErD,IAAI,CAAC,MAAM,CAAC,IAAI,CAAC,kBAAkB,GAAG,KAAK,QAAQ,CAAC,MAAM,QAAQ,CAAC,CAAC;YAEpE,OAAO;gBACL,QAAQ;gBACR,QAAQ;gBACR,MAAM,EAAE,GAAG;aACZ,CAAC;QACJ,CAAC;QAAC,OAAO,KAAK,EAAE,CAAC;YACf,IAAI,KAAK,YAAY,KAAK,IAAI,KAAK,CAAC,IAAI,KAAK,YAAY,EAAE,CAAC;gBAC1D,MAAM,IAAI,KAAK,CAAC,oCAAoC,OAAO,OAAO,GAAG,EAAE,CAAC,CAAC;YAC3E,CAAC;YACD,IAAI,CAAC,MAAM,CAAC,KAAK,CAAC,2BAA2B,GAAG,KAAK,KAAK,EAAE,CAAC,CAAC;YAC9D,MAAM,KAAK,CAAC;QACd,CAAC;IACH,CAAC;IAED;;;OAGG;IACI,QAAQ,CAAC,GAAW;QACzB,IAAI,CAAC;YACH,MAAM,MAAM,GAAG,IAAI,GAAG,CAAC,GAAG,CAAC,CAAC;YAC5B,kBAAkB;YAClB,OAAO,MAAM,CAAC,QAAQ,KAAK,OAAO,IAAI,MAAM,CAAC,QAAQ,KAAK,QAAQ,CAAC;QACrE,CAAC;QAAC,MAAM,CAAC;YACP,OAAO,KAAK,CAAC;QACf,CAAC;IACH,CAAC;IAED;;OAEG;IACK,eAAe,CACrB,QAAgB,EAChB,GAAW;QAEX,MAAM,QAAQ,GAA6B;YACzC,GAAG;SACJ,CAAC;QAEF,2BAA2B;QAC3B,MAAM,gBAAgB,GAAG,QAAQ,CAAC,KAAK,CAAC,uBAAuB,CAAC,CAAC;QACjE,IAAI,gBAAgB,EAAE,CAAC;YACrB,MAAM,WAAW,GAAG,gBAAgB,CAAC,CAAC,CAAC,CAAC;YAExC,WAAW;YACX,MAAM,UAAU,GAAG,WAAW,CAAC,KAAK,CAAC,kBAAkB,CAAC,CAAC;YACzD,IAAI,UAAU,EAAE,CAAC;gBACf,QAAQ,CAAC,KAAK,GAAG,UAAU,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC,cAAc,EAAE,EAAE,CAAC,CAAC;YAC7D,CAAC;YAED,iBAAiB;YACjB,MAAM,SAAS,GAAG,WAAW,CAAC,KAAK,CAAC,wBAAwB,CAAC,CAAC;YAC9D,IAAI,SAAS,EAAE,CAAC;gBACd,QAAQ,CAAC,WAAW,GAAG,SAAS,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC,cAAc,EAAE,EAAE,CAAC,CAAC;YAClE,CAAC;QACH,CAAC;QAED,2CAA2C;QAC3C,IAAI,CAAC,QAAQ,CAAC,KAAK,EAAE,CAAC;YACpB,MAAM,OAAO,GAAG,QAAQ,CAAC,KAAK,CAAC,aAAa,CAAC,CAAC;YAC9C,IAAI,OAAO,EAAE,CAAC;gBACZ,QAAQ,CAAC,KAAK,GAAG,OAAO,CAAC,CAAC,CAAC,CAAC;YAC9B,CAAC;QACH,CAAC;QAED,aAAa;QACb,MAAM,WAAW,GAAG,QAAQ;aACzB,OAAO,CAAC,iBAAiB,EAAE,EAAE,CAAC,CAAC,QAAQ;aACvC,OAAO,CAAC,eAAe,EAAE,EAAE,CAAC,CAAC,iBAAiB;aAC9C,IAAI,EAAE,CAAC;QACV,QAAQ,CAAC,SAAS,GAAG,WAAW,CAAC,KAAK,CAAC,KAAK,CAAC,CAAC,MAAM,CAAC,OAAO,CAAC,CAAC,MAAM,CAAC;QAErE,OAAO,QAAQ,CAAC;IAClB,CAAC;CACF;AA3ID,gCA2IC","sourcesContent":["/**\n * JinaReader - 基于 JINA Reader API 的文档解析器\n *\n * 使用 JINA Reader API (https://r.jina.ai/) 将各种格式的文档转换为 Markdown\n * 支持：PDF, Office 文档, HTML, 图片等\n *\n * @see https://jina.ai/reader/\n */\n\nimport { getLoggerFor } from 'global-logger-factory';\nimport type { DocumentReader, ReadDocument, ReadOptions } from './DocumentReader';\n\nexport interface JinaReaderOptions {\n  /** JINA API Key */\n  apiKey: string;\n  /** 自定义 API 端点（默认 https://r.jina.ai） */\n  baseUrl?: string;\n  /** 默认超时时间（毫秒） */\n  defaultTimeout?: number;\n}\n\n/**\n * JINA Reader API 文档解析器\n */\nexport class JinaReader implements DocumentReader {\n  protected readonly logger = getLoggerFor(this);\n\n  private readonly apiKey: string;\n  private readonly baseUrl: string;\n  private readonly defaultTimeout: number;\n\n  public constructor(options: JinaReaderOptions) {\n    this.apiKey = options.apiKey;\n    this.baseUrl = options.baseUrl ?? 'https://r.jina.ai';\n    this.defaultTimeout = options.defaultTimeout ?? 30000;\n  }\n\n  /**\n   * 读取文档为 Markdown\n   */\n  public async read(url: string, options?: ReadOptions): Promise<ReadDocument> {\n    const timeout = options?.timeout ?? this.defaultTimeout;\n\n    this.logger.debug(`Reading document: ${url}`);\n\n    try {\n      // 构建 JINA Reader URL\n      const readerUrl = `${this.baseUrl}/${encodeURIComponent(url)}`;\n\n      // 创建 AbortController 用于超时控制\n      const controller = new AbortController();\n      const timeoutId = setTimeout(() => controller.abort(), timeout);\n\n      const headers: Record<string, string> = {\n        Authorization: `Bearer ${this.apiKey}`,\n        Accept: 'text/markdown',\n        'X-Return-Format': 'markdown',\n      };\n\n      // 如果需要包含图片描述\n      if (options?.includeImages) {\n        headers['X-With-Images-Summary'] = 'true';\n      }\n\n      // 如果需要包含链接\n      if (options?.includeLinks !== false) {\n        headers['X-With-Links-Summary'] = 'true';\n      }\n\n      const response = await fetch(readerUrl, {\n        method: 'GET',\n        headers,\n        signal: controller.signal,\n      });\n\n      clearTimeout(timeoutId);\n\n      if (!response.ok) {\n        const errorText = await response.text();\n        throw new Error(`JINA read failed: ${response.status} ${errorText}`);\n      }\n\n      const markdown = await response.text();\n\n      // 提取元数据（从 Markdown 头部）\n      const metadata = this.extractMetadata(markdown, url);\n\n      this.logger.info(`Read document: ${url}, ${markdown.length} chars`);\n\n      return {\n        markdown,\n        metadata,\n        rawUrl: url,\n      };\n    } catch (error) {\n      if (error instanceof Error && error.name === 'AbortError') {\n        throw new Error(`Document reading timed out after ${timeout}ms: ${url}`);\n      }\n      this.logger.error(`Failed to read document ${url}: ${error}`);\n      throw error;\n    }\n  }\n\n  /**\n   * 检查是否支持该 URL\n   * JINA Reader 支持大多数 URL\n   */\n  public supports(url: string): boolean {\n    try {\n      const parsed = new URL(url);\n      // 支持 http 和 https\n      return parsed.protocol === 'http:' || parsed.protocol === 'https:';\n    } catch {\n      return false;\n    }\n  }\n\n  /**\n   * 从 Markdown 内容中提取元数据\n   */\n  private extractMetadata(\n    markdown: string,\n    url: string,\n  ): ReadDocument['metadata'] {\n    const metadata: ReadDocument['metadata'] = {\n      url,\n    };\n\n    // 尝试从 YAML front matter 提取\n    const frontMatterMatch = markdown.match(/^---\\n([\\s\\S]*?)\\n---/);\n    if (frontMatterMatch) {\n      const frontMatter = frontMatterMatch[1];\n\n      // 提取 title\n      const titleMatch = frontMatter.match(/^title:\\s*(.+)$/m);\n      if (titleMatch) {\n        metadata.title = titleMatch[1].replace(/^[\"']|[\"']$/g, '');\n      }\n\n      // 提取 description\n      const descMatch = frontMatter.match(/^description:\\s*(.+)$/m);\n      if (descMatch) {\n        metadata.description = descMatch[1].replace(/^[\"']|[\"']$/g, '');\n      }\n    }\n\n    // 如果没有从 front matter 获取 title，尝试从第一个 H1 获取\n    if (!metadata.title) {\n      const h1Match = markdown.match(/^#\\s+(.+)$/m);\n      if (h1Match) {\n        metadata.title = h1Match[1];\n      }\n    }\n\n    // 计算字数（简单估算）\n    const textContent = markdown\n      .replace(/```[\\s\\S]*?```/g, '') // 移除代码块\n      .replace(/[#*_`\\[\\]()]/g, '') // 移除 Markdown 标记\n      .trim();\n    metadata.wordCount = textContent.split(/\\s+/).filter(Boolean).length;\n\n    return metadata;\n  }\n}\n"]}

package/dist/document/JinaReader.jsonld

@@ -0,0 +1,133 @@
+{
+  "@context": [
+    "https://linkedsoftwaredependencies.org/bundles/npm/@undefineds.co/xpod/^0.0.0/components/context.jsonld"
+  ],
+  "@id": "npmd:@undefineds.co/xpod",
+  "components": [
+    {
+      "@id": "undefineds:dist/document/JinaReader.jsonld#JinaReader",
+      "@type": "Class",
+      "requireElement": "JinaReader",
+      "extends": [
+        "undefineds:dist/document/DocumentReader.jsonld#DocumentReader"
+      ],
+      "comment": "JINA Reader API 文档解析器",
+      "parameters": [
+        {
+          "@id": "undefineds:dist/document/JinaReader.jsonld#JinaReader_options_apiKey",
+          "range": "xsd:string",
+          "comment": "JINA API Key"
+        },
+        {
+          "@id": "undefineds:dist/document/JinaReader.jsonld#JinaReader_options_baseUrl",
+          "range": {
+            "@type": "ParameterRangeUnion",
+            "parameterRangeElements": [
+              "xsd:string",
+              {
+                "@type": "ParameterRangeUndefined"
+              }
+            ]
+          },
+          "comment": "自定义 API 端点（默认 https://r.jina.ai）"
+        },
+        {
+          "@id": "undefineds:dist/document/JinaReader.jsonld#JinaReader_options_defaultTimeout",
+          "range": {
+            "@type": "ParameterRangeUnion",
+            "parameterRangeElements": [
+              "xsd:number",
+              {
+                "@type": "ParameterRangeUndefined"
+              }
+            ]
+          },
+          "comment": "默认超时时间（毫秒）"
+        }
+      ],
+      "memberFields": [
+        {
+          "@id": "undefineds:dist/document/JinaReader.jsonld#JinaReader__member_logger",
+          "memberFieldName": "logger",
+          "range": {
+            "@type": "ParameterRangeWildcard"
+          }
+        },
+        {
+          "@id": "undefineds:dist/document/JinaReader.jsonld#JinaReader__member_apiKey",
+          "memberFieldName": "apiKey"
+        },
+        {
+          "@id": "undefineds:dist/document/JinaReader.jsonld#JinaReader__member_baseUrl",
+          "memberFieldName": "baseUrl"
+        },
+        {
+          "@id": "undefineds:dist/document/JinaReader.jsonld#JinaReader__member_defaultTimeout",
+          "memberFieldName": "defaultTimeout"
+        },
+        {
+          "@id": "undefineds:dist/document/JinaReader.jsonld#JinaReader__member_constructor",
+          "memberFieldName": "constructor"
+        },
+        {
+          "@id": "undefineds:dist/document/JinaReader.jsonld#JinaReader__member_read",
+          "memberFieldName": "read"
+        },
+        {
+          "@id": "undefineds:dist/document/JinaReader.jsonld#JinaReader__member_supports",
+          "memberFieldName": "supports"
+        },
+        {
+          "@id": "undefineds:dist/document/JinaReader.jsonld#JinaReader__member_extractMetadata",
+          "memberFieldName": "extractMetadata"
+        }
+      ],
+      "constructorArguments": [
+        {
+          "@id": "undefineds:dist/document/JinaReader.jsonld#JinaReader_options__constructorArgument",
+          "fields": [
+            {
+              "keyRaw": "apiKey",
+              "value": {
+                "@id": "undefineds:dist/document/JinaReader.jsonld#JinaReader_options_apiKey"
+              }
+            },
+            {
+              "keyRaw": "baseUrl",
+              "value": {
+                "@id": "undefineds:dist/document/JinaReader.jsonld#JinaReader_options_baseUrl"
+              }
+            },
+            {
+              "keyRaw": "defaultTimeout",
+              "value": {
+                "@id": "undefineds:dist/document/JinaReader.jsonld#JinaReader_options_defaultTimeout"
+              }
+            }
+          ]
+        }
+      ]
+    },
+    {
+      "@id": "undefineds:dist/document/JinaReader.jsonld#JinaReaderOptions",
+      "@type": "AbstractClass",
+      "requireElement": "JinaReaderOptions",
+      "parameters": [],
+      "memberFields": [
+        {
+          "@id": "undefineds:dist/document/JinaReader.jsonld#JinaReaderOptions__member_apiKey",
+          "memberFieldName": "apiKey"
+        },
+        {
+          "@id": "undefineds:dist/document/JinaReader.jsonld#JinaReaderOptions__member_baseUrl",
+          "memberFieldName": "baseUrl"
+        },
+        {
+          "@id": "undefineds:dist/document/JinaReader.jsonld#JinaReaderOptions__member_defaultTimeout",
+          "memberFieldName": "defaultTimeout"
+        }
+      ],
+      "constructorArguments": []
+    }
+  ]
+}