@semiont/content 0.5.4 → 0.5.6

package/README.md

@@ -64,14 +64,7 @@ console.log(duplicate.checksum === stored.checksum); // true
 ## Documentation
 
 - [API Reference](./docs/API.md) - Complete API documentation
-- [Architecture](./docs/ARCHITECTURE.md) - Design principles
-- [Patterns](./docs/PATTERNS.md) - Usage patterns and best practices
-
-## Examples
-
-- [Basic Example](./examples/basic.ts) - Storage and retrieval
-- [Deduplication](./examples/deduplication.ts) - Content addressing benefits
-- [Binary Content](./examples/binary.ts) - Images and documents
+- [Architecture](./docs/architecture.md) - Design principles
 
 ## Storage Architecture
 

package/dist/index.d.ts

@@ -1,9 +1,247 @@
+import { SemiontProject } from '@semiont/core/node';
+import { Logger, PdfCoordinate } from '@semiont/core';
+
 /**
- * @semiont/content
+ * WorkingTreeStore - Manages files in the project working tree
  *
- * Working tree storage for project resources.
+ * Unlike the old content-addressed RepresentationStore, this store treats
+ * the working tree (project root) as the source of truth for file content.
+ * Resources are identified by their file:// URI, which is stable across
+ * content changes and moves (tracked by events).
+ *
+ * Two write paths:
+ * - store(content, storageUri): Write bytes to disk (API/GUI/AI path).
+ *   Used when the file does not yet exist and the caller provides content.
+ * - register(storageUri, expectedChecksum?): Read an existing file and
+ *   return its metadata (CLI path). The file is already on disk; we just
+ *   verify and record it. If expectedChecksum is provided, throws on mismatch.
+ *
+ * Storage layout:
+ *   {projectRoot}/{path-from-uri}
+ *
+ * For example, storageUri "file://docs/overview.md" resolves to
+ *   {projectRoot}/docs/overview.md
+ */
+
+/**
+ * Result of store() or register()
+ */
+interface StoredResource {
+    storageUri: string;
+    checksum: string;
+    byteSize: number;
+    created: string;
+}
+/**
+ * Manages files in the project working tree
+ */
+declare class WorkingTreeStore {
+    private projectRoot;
+    private gitSync;
+    private logger?;
+    constructor(project: SemiontProject, logger?: Logger);
+    private shouldRunGit;
+    /**
+     * Write content to disk at the location indicated by storageUri.
+     *
+     * API/GUI/AI path: caller provides bytes; file may not yet exist.
+     *
+     * @param content - Raw bytes to write
+     * @param storageUri - file:// URI (e.g. "file://docs/overview.md")
+     * @returns Stored resource metadata
+     */
+    store(content: Buffer, storageUri: string, options?: {
+        noGit?: boolean;
+    }): Promise<StoredResource>;
+    /**
+     * Read an existing file and return its metadata.
+     *
+     * CLI path: the file is already on disk. We read it to compute the checksum.
+     * If expectedChecksum is provided, throws ChecksumMismatchError on mismatch.
+     *
+     * @param storageUri - file:// URI (e.g. "file://docs/overview.md")
+     * @param expectedChecksum - Optional SHA-256 to verify against
+     * @returns Stored resource metadata
+     * @throws ChecksumMismatchError if expectedChecksum is provided and does not match
+     * @throws Error if file does not exist
+     */
+    register(storageUri: string, expectedChecksum?: string, options?: {
+        noGit?: boolean;
+    }): Promise<StoredResource>;
+    /**
+     * Read file content by URI.
+     *
+     * @param storageUri - file:// URI
+     * @returns Raw bytes
+     */
+    retrieve(storageUri: string): Promise<Buffer>;
+    /**
+     * Move a file from one URI to another.
+     *
+     * If .git/ exists in the project root and noGit is not set, runs `git mv`.
+     * Otherwise (no .git/ or noGit: true), runs fs.rename.
+     *
+     * @param fromUri - Current file:// URI
+     * @param toUri - New file:// URI
+     * @param options.noGit - Skip git mv even if .git/ is present
+     */
+    move(fromUri: string, toUri: string, options?: {
+        noGit?: boolean;
+    }): Promise<void>;
+    /**
+     * Remove a file from the working tree.
+     *
+     * If .git/ exists and noGit is not set:
+     *   - keepFile false (default): runs `git rm` (removes from index and disk)
+     *   - keepFile true: runs `git rm --cached` (removes from index only, file stays on disk)
+     * If no .git/ or noGit: true:
+     *   - keepFile false: runs fs.unlink
+     *   - keepFile true: no-op on filesystem
+     *
+     * @param storageUri - file:// URI
+     * @param options.noGit - Skip git rm even if .git/ is present
+     * @param options.keepFile - Remove from git index only; leave file on disk
+     */
+    remove(storageUri: string, options?: {
+        noGit?: boolean;
+        keepFile?: boolean;
+    }): Promise<void>;
+    /**
+     * Convert a file:// URI to an absolute filesystem path.
+     *
+     * "file://docs/overview.md" → "{projectRoot}/docs/overview.md"
+     *
+     * @param storageUri - file:// URI
+     * @returns Absolute path
+     */
+    resolveUri(storageUri: string): string;
+}
+/**
+ * Thrown when a registered file's checksum does not match the expected value.
+ * This indicates the file on disk differs from what was recorded (e.g. modified
+ * after staging, or wrong file path provided).
+ */
+declare class ChecksumMismatchError extends Error {
+    readonly storageUri: string;
+    readonly expected: string;
+    readonly actual: string;
+    constructor(storageUri: string, expected: string, actual: string);
+}
+
+/**
+ * MIME Type to File Extension Mapping
+ *
+ * Maps common MIME types to their standard file extensions.
+ * Used by RepresentationStore to save files with proper extensions.
+ */
+/**
+ * Get file extension for a MIME type
+ *
+ * @param mediaType - MIME type (e.g., "text/markdown")
+ * @returns File extension with leading dot (e.g., ".md") or ".dat" if unknown
+ *
+ * @example
+ * getExtensionForMimeType('text/markdown') // => '.md'
+ * getExtensionForMimeType('image/png')     // => '.png'
+ * getExtensionForMimeType('unknown/type')  // => '.dat'
+ */
+declare function getExtensionForMimeType(mediaType: string): string;
+/**
+ * Derive a file:// storage URI from a resource name and MIME type.
+ *
+ * @example
+ * deriveStorageUri("My Document", "text/markdown") // => "file://my-document.md"
+ */
+declare function deriveStorageUri(name: string, format: string): string;
+/**
+ * Check if a MIME type has a known extension mapping
+ *
+ * @param mediaType - MIME type to check
+ * @returns true if extension is known, false if would fallback to .dat
+ */
+declare function hasKnownExtension(mediaType: string): boolean;
+
+/**
+ * Checksum utilities for content verification
+ */
+/**
+ * Calculate SHA-256 checksum of content
+ * @param content The content to hash
+ * @returns Hex-encoded SHA-256 hash
+ */
+declare function calculateChecksum(content: string | Buffer): string;
+/**
+ * Verify content against a checksum
+ * @param content The content to verify
+ * @param checksum The expected checksum
+ * @returns True if content matches checksum
+ */
+declare function verifyChecksum(content: string | Buffer, checksum: string): boolean;
+
+/**
+ * PDF Text Layer Types
+ *
+ * Represents the extracted text layer from a native PDF, including per-run (word-level)
+ * geometry in PDF point coordinates originating from the bottom-left of the page
+ * (Y increases upward). The Y-flip to canvas pixels happens downstream in the
+ * browser; the server has no canvas.
+ *
+ * `PdfCoordinate` — the geometry `locate()` emits — lives in `@semiont/core`
+ * alongside the viewrect FragmentSelector codec.
+ */
+/**
+ * A single text item (one text run, roughly a word) from the PDF text layer.
+ * Character offsets refer to positions in `PdfTextLayer.text`.
+ */
+interface PdfTextItem {
+    start: number;
+    end: number;
+    page: number;
+    x: number;
+    y: number;
+    width: number;
+    height: number;
+}
+/** Page dimensions in PDF points */
+interface PdfPageInfo {
+    pageNumber: number;
+    widthPt: number;
+    heightPt: number;
+}
+/**
+ * The full extracted text layer for a PDF.
+ * `text` is the reading-order concatenation across all pages.
+ * Each `item` is one text run carrying its character range into `text` plus PDF-point geometry.
+ */
+interface PdfTextLayer {
+    pages: PdfPageInfo[];
+    text: string;
+    items: PdfTextItem[];
+}
+
+/**
+ * PDF Text Layer Extraction
+ *
+ * Extracts positioned text from native, non-scanned PDFs using pdfjs-dist.
+ * Returns null for scanned/image-only PDFs (no text items).
+ *
+ * Coordinates are in PDF point space, originating from the bottom-left.
+ * The Y-flip to canvas pixels happens downstream.
+ */
+
+declare function extractPdfTextLayer(bytes: Uint8Array | Buffer): Promise<PdfTextLayer | null>;
+
+/**
+ * Locates bounding rectangles for a span of text in a PdfTextLayer
+ * (single-line or multi-line).
+ *
+ * Finds all overlapping items [start, end), groups them by page and line,
+ * and records one bounding rectangle per line as a PdfCoordinate.
+ *
+ * Returns array of PdfCoordinate, one per line of text covered by the span.
+ * Returns empty array if no items overlap the span.
  */
-export { WorkingTreeStore, ChecksumMismatchError, type StoredResource, } from './working-tree-store';
-export { getExtensionForMimeType, hasKnownExtension, deriveStorageUri, } from './mime-extensions';
-export { calculateChecksum, verifyChecksum } from './checksum';
-//# sourceMappingURL=index.d.ts.map
+declare function locate(layer: PdfTextLayer, start: number, end: number): PdfCoordinate[];
+
+export { ChecksumMismatchError, WorkingTreeStore, calculateChecksum, deriveStorageUri, extractPdfTextLayer, getExtensionForMimeType, hasKnownExtension, locate, verifyChecksum };
+export type { PdfPageInfo, PdfTextItem, PdfTextLayer, StoredResource };

package/dist/index.js

@@ -190,6 +190,9 @@ The file on disk differs from the recorded checksum. Has it been modified since
     this.actual = actual;
     this.name = "ChecksumMismatchError";
   }
+  storageUri;
+  expected;
+  actual;
 };
 
 // src/mime-extensions.ts
@@ -282,13 +285,113 @@ function hasKnownExtension(mediaType) {
   const normalized = mediaType.toLowerCase().split(";")[0].trim();
   return normalized in MIME_TO_EXTENSION;
 }
+
+// src/extract-pdf-text-layer.ts
+import * as pdfjs from "pdfjs-dist/legacy/build/pdf.mjs";
+async function extractPdfTextLayer(bytes) {
+  const doc = await pdfjs.getDocument({ data: bytes }).promise;
+  try {
+    const pages = [];
+    const items = [];
+    let text = "";
+    let hasAnyTextItems = false;
+    for (let pageNum = 1; pageNum <= doc.numPages; pageNum++) {
+      const page = await doc.getPage(pageNum);
+      const viewport = page.getViewport({ scale: 1 });
+      const content = await page.getTextContent();
+      pages.push({
+        pageNumber: pageNum,
+        widthPt: viewport.width,
+        heightPt: viewport.height
+      });
+      for (const item of content.items) {
+        if (!("str" in item)) continue;
+        if (item.str.trim()) {
+          hasAnyTextItems = true;
+          const start = text.length;
+          text += item.str;
+          const end = text.length;
+          const [, , , , x, y] = item.transform;
+          items.push({
+            start,
+            end,
+            page: pageNum,
+            x,
+            y,
+            width: item.width,
+            height: item.height
+          });
+          text += item.hasEOL ? "\n" : " ";
+        } else if (item.hasEOL) {
+          text += "\n";
+        }
+      }
+      text += "\n";
+    }
+    if (!hasAnyTextItems) return null;
+    return { pages, text, items };
+  } finally {
+    await doc.destroy();
+  }
+}
+
+// src/locate.ts
+var SAME_LINE_THRESHOLD_PT = 2;
+function locate(layer, start, end) {
+  const overlap = layer.items.filter(
+    (item) => item.start < end && item.end > start
+  );
+  if (overlap.length === 0) return [];
+  const pages = groupItemsByPage(overlap);
+  const rects = [];
+  for (const [page, pageItems] of pages) {
+    const lines = groupItemsByLine(pageItems, SAME_LINE_THRESHOLD_PT);
+    for (const lineItems of lines) {
+      const x = Math.min(...lineItems.map((i) => i.x));
+      const right = Math.max(...lineItems.map((i) => i.x + i.width));
+      const y = Math.min(...lineItems.map((i) => i.y));
+      const top = Math.max(...lineItems.map((i) => i.y + i.height));
+      rects.push({ page, x, y, width: right - x, height: top - y });
+    }
+  }
+  return rects;
+}
+function groupItemsByPage(items) {
+  const map = /* @__PURE__ */ new Map();
+  for (const item of items) {
+    const existing = map.get(item.page);
+    if (existing) {
+      existing.push(item);
+    } else {
+      map.set(item.page, [item]);
+    }
+  }
+  return map;
+}
+function groupItemsByLine(items, sameLineThreshold) {
+  const sorted = [...items].sort((a, b) => b.y - a.y || a.x - b.x);
+  const lines = [];
+  let currentLine = [];
+  for (const item of sorted) {
+    if (currentLine.length === 0 || Math.abs(item.y - currentLine[0].y) <= sameLineThreshold) {
+      currentLine.push(item);
+    } else {
+      lines.push(currentLine);
+      currentLine = [item];
+    }
+  }
+  if (currentLine.length > 0) lines.push(currentLine);
+  return lines;
+}
 export {
   ChecksumMismatchError,
   WorkingTreeStore,
   calculateChecksum,
   deriveStorageUri,
+  extractPdfTextLayer,
   getExtensionForMimeType,
   hasKnownExtension,
+  locate,
   verifyChecksum
 };
 //# sourceMappingURL=index.js.map

package/dist/index.js.map

@@ -1 +1 @@
-{"version":3,"sources":["../src/working-tree-store.ts","../src/checksum.ts","../src/mime-extensions.ts"],"sourcesContent":["/**\n * WorkingTreeStore - Manages files in the project working tree\n *\n * Unlike the old content-addressed RepresentationStore, this store treats\n * the working tree (project root) as the source of truth for file content.\n * Resources are identified by their file:// URI, which is stable across\n * content changes and moves (tracked by events).\n *\n * Two write paths:\n * - store(content, storageUri): Write bytes to disk (API/GUI/AI path).\n *   Used when the file does not yet exist and the caller provides content.\n * - register(storageUri, expectedChecksum?): Read an existing file and\n *   return its metadata (CLI path). The file is already on disk; we just\n *   verify and record it. If expectedChecksum is provided, throws on mismatch.\n *\n * Storage layout:\n *   {projectRoot}/{path-from-uri}\n *\n * For example, storageUri \"file://docs/overview.md\" resolves to\n *   {projectRoot}/docs/overview.md\n */\n\nimport { promises as fs } from 'fs';\nimport { execFileSync } from 'child_process';\nimport path from 'path';\nimport type { SemiontProject } from '@semiont/core/node';\nimport type { Logger } from '@semiont/core';\nimport { calculateChecksum, verifyChecksum } from './checksum';\n\n/**\n * Result of store() or register()\n */\nexport interface StoredResource {\n  storageUri: string;    // file:// URI (e.g. \"file://docs/overview.md\")\n  checksum: string;      // SHA-256 hex of content\n  byteSize: number;      // Size in bytes\n  created: string;       // ISO 8601 timestamp\n}\n\n/**\n * Manages files in the project working tree\n */\nexport class WorkingTreeStore {\n  private projectRoot: string;\n  private gitSync: boolean;\n  private logger?: Logger;\n\n  constructor(project: SemiontProject, logger?: Logger) {\n    this.projectRoot = project.root;\n    this.gitSync = project.gitSync;\n    this.logger = logger;\n  }\n\n  private shouldRunGit(noGit?: boolean): boolean {\n    return this.gitSync && !noGit;\n  }\n\n  /**\n   * Write content to disk at the location indicated by storageUri.\n   *\n   * API/GUI/AI path: caller provides bytes; file may not yet exist.\n   *\n   * @param content - Raw bytes to write\n   * @param storageUri - file:// URI (e.g. \"file://docs/overview.md\")\n   * @returns Stored resource metadata\n   */\n  async store(content: Buffer, storageUri: string, options?: { noGit?: boolean }): Promise<StoredResource> {\n    const filePath = this.resolveUri(storageUri);\n    const checksum = calculateChecksum(content);\n\n    this.logger?.debug('Storing resource', { storageUri, byteSize: content.length });\n\n    await fs.mkdir(path.dirname(filePath), { recursive: true });\n    await fs.writeFile(filePath, content);\n\n    if (this.shouldRunGit(options?.noGit)) {\n      execFileSync('git', ['add', filePath], { cwd: this.projectRoot });\n    }\n\n    this.logger?.info('Resource stored', { storageUri, checksum, byteSize: content.length });\n\n    return {\n      storageUri,\n      checksum,\n      byteSize: content.length,\n      created: new Date().toISOString(),\n    };\n  }\n\n  /**\n   * Read an existing file and return its metadata.\n   *\n   * CLI path: the file is already on disk. We read it to compute the checksum.\n   * If expectedChecksum is provided, throws ChecksumMismatchError on mismatch.\n   *\n   * @param storageUri - file:// URI (e.g. \"file://docs/overview.md\")\n   * @param expectedChecksum - Optional SHA-256 to verify against\n   * @returns Stored resource metadata\n   * @throws ChecksumMismatchError if expectedChecksum is provided and does not match\n   * @throws Error if file does not exist\n   */\n  async register(storageUri: string, expectedChecksum?: string, options?: { noGit?: boolean }): Promise<StoredResource> {\n    const filePath = this.resolveUri(storageUri);\n\n    this.logger?.debug('Registering resource', { storageUri });\n\n    const content = await fs.readFile(filePath);\n    const checksum = calculateChecksum(content);\n\n    if (expectedChecksum !== undefined && !verifyChecksum(content, expectedChecksum)) {\n      throw new ChecksumMismatchError(storageUri, expectedChecksum, checksum);\n    }\n\n    if (this.shouldRunGit(options?.noGit)) {\n      execFileSync('git', ['add', filePath], { cwd: this.projectRoot });\n    }\n\n    this.logger?.info('Resource registered', { storageUri, checksum, byteSize: content.length });\n\n    return {\n      storageUri,\n      checksum,\n      byteSize: content.length,\n      created: new Date().toISOString(),\n    };\n  }\n\n  /**\n   * Read file content by URI.\n   *\n   * @param storageUri - file:// URI\n   * @returns Raw bytes\n   */\n  async retrieve(storageUri: string): Promise<Buffer> {\n    const filePath = this.resolveUri(storageUri);\n    try {\n      return await fs.readFile(filePath);\n    } catch (error: any) {\n      if (error.code === 'ENOENT') {\n        throw new Error(`Resource not found: ${storageUri}`);\n      }\n      throw error;\n    }\n  }\n\n  /**\n   * Move a file from one URI to another.\n   *\n   * If .git/ exists in the project root and noGit is not set, runs `git mv`.\n   * Otherwise (no .git/ or noGit: true), runs fs.rename.\n   *\n   * @param fromUri - Current file:// URI\n   * @param toUri - New file:// URI\n   * @param options.noGit - Skip git mv even if .git/ is present\n   */\n  async move(fromUri: string, toUri: string, options?: { noGit?: boolean }): Promise<void> {\n    const fromPath = this.resolveUri(fromUri);\n    const toPath = this.resolveUri(toUri);\n\n    this.logger?.debug('Moving resource', { fromUri, toUri });\n\n    await fs.mkdir(path.dirname(toPath), { recursive: true });\n\n    if (this.shouldRunGit(options?.noGit)) {\n      // git mv handles both the filesystem rename and the index update\n      execFileSync('git', ['mv', fromPath, toPath], { cwd: this.projectRoot });\n    } else {\n      await fs.rename(fromPath, toPath);\n    }\n\n    this.logger?.info('Resource moved', { fromUri, toUri });\n  }\n\n  /**\n   * Remove a file from the working tree.\n   *\n   * If .git/ exists and noGit is not set:\n   *   - keepFile false (default): runs `git rm` (removes from index and disk)\n   *   - keepFile true: runs `git rm --cached` (removes from index only, file stays on disk)\n   * If no .git/ or noGit: true:\n   *   - keepFile false: runs fs.unlink\n   *   - keepFile true: no-op on filesystem\n   *\n   * @param storageUri - file:// URI\n   * @param options.noGit - Skip git rm even if .git/ is present\n   * @param options.keepFile - Remove from git index only; leave file on disk\n   */\n  async remove(storageUri: string, options?: { noGit?: boolean; keepFile?: boolean }): Promise<void> {\n    const filePath = this.resolveUri(storageUri);\n    const keepFile = options?.keepFile ?? false;\n\n    this.logger?.debug('Removing resource', { storageUri, keepFile });\n\n    const useGit = this.shouldRunGit(options?.noGit);\n\n    if (useGit) {\n      const gitArgs = keepFile\n        ? ['rm', '--cached', filePath]\n        : ['rm', filePath];\n      execFileSync('git', gitArgs, { cwd: this.projectRoot });\n      this.logger?.info('Resource removed', { storageUri, keepFile, git: true });\n      return;\n    }\n\n    if (keepFile) {\n      this.logger?.info('Resource removed from index (file kept on disk)', { storageUri });\n      return;\n    }\n\n    try {\n      await fs.unlink(filePath);\n      this.logger?.info('Resource removed', { storageUri });\n    } catch (error: any) {\n      if (error.code === 'ENOENT') {\n        this.logger?.warn('Resource file already absent', { storageUri });\n        return;\n      }\n      throw error;\n    }\n  }\n\n  /**\n   * Convert a file:// URI to an absolute filesystem path.\n   *\n   * \"file://docs/overview.md\" → \"{projectRoot}/docs/overview.md\"\n   *\n   * @param storageUri - file:// URI\n   * @returns Absolute path\n   */\n  resolveUri(storageUri: string): string {\n    if (!storageUri.startsWith('file://')) {\n      throw new Error(`Invalid storage URI (must start with file://): ${storageUri}`);\n    }\n    const relativePath = storageUri.slice('file://'.length);\n    return path.join(this.projectRoot, relativePath);\n  }\n}\n\n/**\n * Thrown when a registered file's checksum does not match the expected value.\n * This indicates the file on disk differs from what was recorded (e.g. modified\n * after staging, or wrong file path provided).\n */\nexport class ChecksumMismatchError extends Error {\n  constructor(\n    readonly storageUri: string,\n    readonly expected: string,\n    readonly actual: string,\n  ) {\n    super(\n      `Checksum mismatch for ${storageUri}: expected ${expected.slice(0, 8)}... but got ${actual.slice(0, 8)}...\\n` +\n      `The file on disk differs from the recorded checksum. Has it been modified since staging?`\n    );\n    this.name = 'ChecksumMismatchError';\n  }\n}\n","/**\n * Checksum utilities for content verification\n */\n\nimport { createHash } from 'crypto';\n\n/**\n * Calculate SHA-256 checksum of content\n * @param content The content to hash\n * @returns Hex-encoded SHA-256 hash\n */\nexport function calculateChecksum(content: string | Buffer): string {\n  const hash = createHash('sha256');\n  hash.update(content);\n  return hash.digest('hex');\n}\n\n/**\n * Verify content against a checksum\n * @param content The content to verify\n * @param checksum The expected checksum\n * @returns True if content matches checksum\n */\nexport function verifyChecksum(content: string | Buffer, checksum: string): boolean {\n  return calculateChecksum(content) === checksum;\n}\n","/**\n * MIME Type to File Extension Mapping\n *\n * Maps common MIME types to their standard file extensions.\n * Used by RepresentationStore to save files with proper extensions.\n */\n\n/**\n * Comprehensive MIME type to extension mapping\n */\nconst MIME_TO_EXTENSION: Record<string, string> = {\n  // Text formats\n  'text/plain': '.txt',\n  'text/markdown': '.md',\n  'text/html': '.html',\n  'text/css': '.css',\n  'text/csv': '.csv',\n  'text/xml': '.xml',\n\n  // Application formats - structured data\n  'application/json': '.json',\n  'application/xml': '.xml',\n  'application/yaml': '.yaml',\n  'application/x-yaml': '.yaml',\n\n  // Application formats - documents\n  'application/pdf': '.pdf',\n  'application/msword': '.doc',\n  'application/vnd.openxmlformats-officedocument.wordprocessingml.document': '.docx',\n  'application/vnd.ms-excel': '.xls',\n  'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet': '.xlsx',\n  'application/vnd.ms-powerpoint': '.ppt',\n  'application/vnd.openxmlformats-officedocument.presentationml.presentation': '.pptx',\n\n  // Application formats - archives\n  'application/zip': '.zip',\n  'application/gzip': '.gz',\n  'application/x-tar': '.tar',\n  'application/x-7z-compressed': '.7z',\n\n  // Application formats - executables/binaries\n  'application/octet-stream': '.bin',\n  'application/wasm': '.wasm',\n\n  // Image formats\n  'image/png': '.png',\n  'image/jpeg': '.jpg',\n  'image/gif': '.gif',\n  'image/webp': '.webp',\n  'image/svg+xml': '.svg',\n  'image/bmp': '.bmp',\n  'image/tiff': '.tiff',\n  'image/x-icon': '.ico',\n\n  // Audio formats\n  'audio/mpeg': '.mp3',\n  'audio/wav': '.wav',\n  'audio/ogg': '.ogg',\n  'audio/webm': '.webm',\n  'audio/aac': '.aac',\n  'audio/flac': '.flac',\n\n  // Video formats\n  'video/mp4': '.mp4',\n  'video/mpeg': '.mpeg',\n  'video/webm': '.webm',\n  'video/ogg': '.ogv',\n  'video/quicktime': '.mov',\n  'video/x-msvideo': '.avi',\n\n  // Programming languages\n  'text/javascript': '.js',\n  'application/javascript': '.js',\n  'text/x-typescript': '.ts',\n  'application/typescript': '.ts',\n  'text/x-python': '.py',\n  'text/x-java': '.java',\n  'text/x-c': '.c',\n  'text/x-c++': '.cpp',\n  'text/x-csharp': '.cs',\n  'text/x-go': '.go',\n  'text/x-rust': '.rs',\n  'text/x-ruby': '.rb',\n  'text/x-php': '.php',\n  'text/x-swift': '.swift',\n  'text/x-kotlin': '.kt',\n  'text/x-shell': '.sh',\n\n  // Font formats\n  'font/woff': '.woff',\n  'font/woff2': '.woff2',\n  'font/ttf': '.ttf',\n  'font/otf': '.otf',\n};\n\n/**\n * Get file extension for a MIME type\n *\n * @param mediaType - MIME type (e.g., \"text/markdown\")\n * @returns File extension with leading dot (e.g., \".md\") or \".dat\" if unknown\n *\n * @example\n * getExtensionForMimeType('text/markdown') // => '.md'\n * getExtensionForMimeType('image/png')     // => '.png'\n * getExtensionForMimeType('unknown/type')  // => '.dat'\n */\nexport function getExtensionForMimeType(mediaType: string): string {\n  // Normalize MIME type (lowercase, remove parameters)\n  const normalized = mediaType.toLowerCase().split(';')[0]!.trim();\n\n  // Look up in mapping\n  const extension = MIME_TO_EXTENSION[normalized];\n\n  // Return mapped extension or fallback to .dat\n  return extension || '.dat';\n}\n\n/**\n * Derive a file:// storage URI from a resource name and MIME type.\n *\n * @example\n * deriveStorageUri(\"My Document\", \"text/markdown\") // => \"file://my-document.md\"\n */\nexport function deriveStorageUri(name: string, format: string): string {\n  const slug = name\n    .toLowerCase()\n    .replace(/[^a-z0-9]+/g, '-')\n    .replace(/^-|-$/g, '');\n  const ext = getExtensionForMimeType(format);\n  return `file://${slug}${ext}`;\n}\n\n/**\n * Check if a MIME type has a known extension mapping\n *\n * @param mediaType - MIME type to check\n * @returns true if extension is known, false if would fallback to .dat\n */\nexport function hasKnownExtension(mediaType: string): boolean {\n  const normalized = mediaType.toLowerCase().split(';')[0]!.trim();\n  return normalized in MIME_TO_EXTENSION;\n}\n"],"mappings":";AAsBA,SAAS,YAAY,UAAU;AAC/B,SAAS,oBAAoB;AAC7B,OAAO,UAAU;;;ACpBjB,SAAS,kBAAkB;AAOpB,SAAS,kBAAkB,SAAkC;AAClE,QAAM,OAAO,WAAW,QAAQ;AAChC,OAAK,OAAO,OAAO;AACnB,SAAO,KAAK,OAAO,KAAK;AAC1B;AAQO,SAAS,eAAe,SAA0B,UAA2B;AAClF,SAAO,kBAAkB,OAAO,MAAM;AACxC;;;ADiBO,IAAM,mBAAN,MAAuB;AAAA,EACpB;AAAA,EACA;AAAA,EACA;AAAA,EAER,YAAY,SAAyB,QAAiB;AACpD,SAAK,cAAc,QAAQ;AAC3B,SAAK,UAAU,QAAQ;AACvB,SAAK,SAAS;AAAA,EAChB;AAAA,EAEQ,aAAa,OAA0B;AAC7C,WAAO,KAAK,WAAW,CAAC;AAAA,EAC1B;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAWA,MAAM,MAAM,SAAiB,YAAoB,SAAwD;AACvG,UAAM,WAAW,KAAK,WAAW,UAAU;AAC3C,UAAM,WAAW,kBAAkB,OAAO;AAE1C,SAAK,QAAQ,MAAM,oBAAoB,EAAE,YAAY,UAAU,QAAQ,OAAO,CAAC;AAE/E,UAAM,GAAG,MAAM,KAAK,QAAQ,QAAQ,GAAG,EAAE,WAAW,KAAK,CAAC;AAC1D,UAAM,GAAG,UAAU,UAAU,OAAO;AAEpC,QAAI,KAAK,aAAa,SAAS,KAAK,GAAG;AACrC,mBAAa,OAAO,CAAC,OAAO,QAAQ,GAAG,EAAE,KAAK,KAAK,YAAY,CAAC;AAAA,IAClE;AAEA,SAAK,QAAQ,KAAK,mBAAmB,EAAE,YAAY,UAAU,UAAU,QAAQ,OAAO,CAAC;AAEvF,WAAO;AAAA,MACL;AAAA,MACA;AAAA,MACA,UAAU,QAAQ;AAAA,MAClB,UAAS,oBAAI,KAAK,GAAE,YAAY;AAAA,IAClC;AAAA,EACF;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAcA,MAAM,SAAS,YAAoB,kBAA2B,SAAwD;AACpH,UAAM,WAAW,KAAK,WAAW,UAAU;AAE3C,SAAK,QAAQ,MAAM,wBAAwB,EAAE,WAAW,CAAC;AAEzD,UAAM,UAAU,MAAM,GAAG,SAAS,QAAQ;AAC1C,UAAM,WAAW,kBAAkB,OAAO;AAE1C,QAAI,qBAAqB,UAAa,CAAC,eAAe,SAAS,gBAAgB,GAAG;AAChF,YAAM,IAAI,sBAAsB,YAAY,kBAAkB,QAAQ;AAAA,IACxE;AAEA,QAAI,KAAK,aAAa,SAAS,KAAK,GAAG;AACrC,mBAAa,OAAO,CAAC,OAAO,QAAQ,GAAG,EAAE,KAAK,KAAK,YAAY,CAAC;AAAA,IAClE;AAEA,SAAK,QAAQ,KAAK,uBAAuB,EAAE,YAAY,UAAU,UAAU,QAAQ,OAAO,CAAC;AAE3F,WAAO;AAAA,MACL;AAAA,MACA;AAAA,MACA,UAAU,QAAQ;AAAA,MAClB,UAAS,oBAAI,KAAK,GAAE,YAAY;AAAA,IAClC;AAAA,EACF;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAQA,MAAM,SAAS,YAAqC;AAClD,UAAM,WAAW,KAAK,WAAW,UAAU;AAC3C,QAAI;AACF,aAAO,MAAM,GAAG,SAAS,QAAQ;AAAA,IACnC,SAAS,OAAY;AACnB,UAAI,MAAM,SAAS,UAAU;AAC3B,cAAM,IAAI,MAAM,uBAAuB,UAAU,EAAE;AAAA,MACrD;AACA,YAAM;AAAA,IACR;AAAA,EACF;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAYA,MAAM,KAAK,SAAiB,OAAe,SAA8C;AACvF,UAAM,WAAW,KAAK,WAAW,OAAO;AACxC,UAAM,SAAS,KAAK,WAAW,KAAK;AAEpC,SAAK,QAAQ,MAAM,mBAAmB,EAAE,SAAS,MAAM,CAAC;AAExD,UAAM,GAAG,MAAM,KAAK,QAAQ,MAAM,GAAG,EAAE,WAAW,KAAK,CAAC;AAExD,QAAI,KAAK,aAAa,SAAS,KAAK,GAAG;AAErC,mBAAa,OAAO,CAAC,MAAM,UAAU,MAAM,GAAG,EAAE,KAAK,KAAK,YAAY,CAAC;AAAA,IACzE,OAAO;AACL,YAAM,GAAG,OAAO,UAAU,MAAM;AAAA,IAClC;AAEA,SAAK,QAAQ,KAAK,kBAAkB,EAAE,SAAS,MAAM,CAAC;AAAA,EACxD;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAgBA,MAAM,OAAO,YAAoB,SAAkE;AACjG,UAAM,WAAW,KAAK,WAAW,UAAU;AAC3C,UAAM,WAAW,SAAS,YAAY;AAEtC,SAAK,QAAQ,MAAM,qBAAqB,EAAE,YAAY,SAAS,CAAC;AAEhE,UAAM,SAAS,KAAK,aAAa,SAAS,KAAK;AAE/C,QAAI,QAAQ;AACV,YAAM,UAAU,WACZ,CAAC,MAAM,YAAY,QAAQ,IAC3B,CAAC,MAAM,QAAQ;AACnB,mBAAa,OAAO,SAAS,EAAE,KAAK,KAAK,YAAY,CAAC;AACtD,WAAK,QAAQ,KAAK,oBAAoB,EAAE,YAAY,UAAU,KAAK,KAAK,CAAC;AACzE;AAAA,IACF;AAEA,QAAI,UAAU;AACZ,WAAK,QAAQ,KAAK,mDAAmD,EAAE,WAAW,CAAC;AACnF;AAAA,IACF;AAEA,QAAI;AACF,YAAM,GAAG,OAAO,QAAQ;AACxB,WAAK,QAAQ,KAAK,oBAAoB,EAAE,WAAW,CAAC;AAAA,IACtD,SAAS,OAAY;AACnB,UAAI,MAAM,SAAS,UAAU;AAC3B,aAAK,QAAQ,KAAK,gCAAgC,EAAE,WAAW,CAAC;AAChE;AAAA,MACF;AACA,YAAM;AAAA,IACR;AAAA,EACF;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAUA,WAAW,YAA4B;AACrC,QAAI,CAAC,WAAW,WAAW,SAAS,GAAG;AACrC,YAAM,IAAI,MAAM,kDAAkD,UAAU,EAAE;AAAA,IAChF;AACA,UAAM,eAAe,WAAW,MAAM,UAAU,MAAM;AACtD,WAAO,KAAK,KAAK,KAAK,aAAa,YAAY;AAAA,EACjD;AACF;AAOO,IAAM,wBAAN,cAAoC,MAAM;AAAA,EAC/C,YACW,YACA,UACA,QACT;AACA;AAAA,MACE,yBAAyB,UAAU,cAAc,SAAS,MAAM,GAAG,CAAC,CAAC,eAAe,OAAO,MAAM,GAAG,CAAC,CAAC;AAAA;AAAA,IAExG;AAPS;AACA;AACA;AAMT,SAAK,OAAO;AAAA,EACd;AACF;;;AErPA,IAAM,oBAA4C;AAAA;AAAA,EAEhD,cAAc;AAAA,EACd,iBAAiB;AAAA,EACjB,aAAa;AAAA,EACb,YAAY;AAAA,EACZ,YAAY;AAAA,EACZ,YAAY;AAAA;AAAA,EAGZ,oBAAoB;AAAA,EACpB,mBAAmB;AAAA,EACnB,oBAAoB;AAAA,EACpB,sBAAsB;AAAA;AAAA,EAGtB,mBAAmB;AAAA,EACnB,sBAAsB;AAAA,EACtB,2EAA2E;AAAA,EAC3E,4BAA4B;AAAA,EAC5B,qEAAqE;AAAA,EACrE,iCAAiC;AAAA,EACjC,6EAA6E;AAAA;AAAA,EAG7E,mBAAmB;AAAA,EACnB,oBAAoB;AAAA,EACpB,qBAAqB;AAAA,EACrB,+BAA+B;AAAA;AAAA,EAG/B,4BAA4B;AAAA,EAC5B,oBAAoB;AAAA;AAAA,EAGpB,aAAa;AAAA,EACb,cAAc;AAAA,EACd,aAAa;AAAA,EACb,cAAc;AAAA,EACd,iBAAiB;AAAA,EACjB,aAAa;AAAA,EACb,cAAc;AAAA,EACd,gBAAgB;AAAA;AAAA,EAGhB,cAAc;AAAA,EACd,aAAa;AAAA,EACb,aAAa;AAAA,EACb,cAAc;AAAA,EACd,aAAa;AAAA,EACb,cAAc;AAAA;AAAA,EAGd,aAAa;AAAA,EACb,cAAc;AAAA,EACd,cAAc;AAAA,EACd,aAAa;AAAA,EACb,mBAAmB;AAAA,EACnB,mBAAmB;AAAA;AAAA,EAGnB,mBAAmB;AAAA,EACnB,0BAA0B;AAAA,EAC1B,qBAAqB;AAAA,EACrB,0BAA0B;AAAA,EAC1B,iBAAiB;AAAA,EACjB,eAAe;AAAA,EACf,YAAY;AAAA,EACZ,cAAc;AAAA,EACd,iBAAiB;AAAA,EACjB,aAAa;AAAA,EACb,eAAe;AAAA,EACf,eAAe;AAAA,EACf,cAAc;AAAA,EACd,gBAAgB;AAAA,EAChB,iBAAiB;AAAA,EACjB,gBAAgB;AAAA;AAAA,EAGhB,aAAa;AAAA,EACb,cAAc;AAAA,EACd,YAAY;AAAA,EACZ,YAAY;AACd;AAaO,SAAS,wBAAwB,WAA2B;AAEjE,QAAM,aAAa,UAAU,YAAY,EAAE,MAAM,GAAG,EAAE,CAAC,EAAG,KAAK;AAG/D,QAAM,YAAY,kBAAkB,UAAU;AAG9C,SAAO,aAAa;AACtB;AAQO,SAAS,iBAAiB,MAAc,QAAwB;AACrE,QAAM,OAAO,KACV,YAAY,EACZ,QAAQ,eAAe,GAAG,EAC1B,QAAQ,UAAU,EAAE;AACvB,QAAM,MAAM,wBAAwB,MAAM;AAC1C,SAAO,UAAU,IAAI,GAAG,GAAG;AAC7B;AAQO,SAAS,kBAAkB,WAA4B;AAC5D,QAAM,aAAa,UAAU,YAAY,EAAE,MAAM,GAAG,EAAE,CAAC,EAAG,KAAK;AAC/D,SAAO,cAAc;AACvB;","names":[]}
+{"version":3,"sources":["../src/working-tree-store.ts","../src/checksum.ts","../src/mime-extensions.ts","../src/extract-pdf-text-layer.ts","../src/locate.ts"],"sourcesContent":["/**\n * WorkingTreeStore - Manages files in the project working tree\n *\n * Unlike the old content-addressed RepresentationStore, this store treats\n * the working tree (project root) as the source of truth for file content.\n * Resources are identified by their file:// URI, which is stable across\n * content changes and moves (tracked by events).\n *\n * Two write paths:\n * - store(content, storageUri): Write bytes to disk (API/GUI/AI path).\n *   Used when the file does not yet exist and the caller provides content.\n * - register(storageUri, expectedChecksum?): Read an existing file and\n *   return its metadata (CLI path). The file is already on disk; we just\n *   verify and record it. If expectedChecksum is provided, throws on mismatch.\n *\n * Storage layout:\n *   {projectRoot}/{path-from-uri}\n *\n * For example, storageUri \"file://docs/overview.md\" resolves to\n *   {projectRoot}/docs/overview.md\n */\n\nimport { promises as fs } from 'fs';\nimport { execFileSync } from 'child_process';\nimport path from 'path';\nimport type { SemiontProject } from '@semiont/core/node';\nimport type { Logger } from '@semiont/core';\nimport { calculateChecksum, verifyChecksum } from './checksum';\n\n/**\n * Result of store() or register()\n */\nexport interface StoredResource {\n  storageUri: string;    // file:// URI (e.g. \"file://docs/overview.md\")\n  checksum: string;      // SHA-256 hex of content\n  byteSize: number;      // Size in bytes\n  created: string;       // ISO 8601 timestamp\n}\n\n/**\n * Manages files in the project working tree\n */\nexport class WorkingTreeStore {\n  private projectRoot: string;\n  private gitSync: boolean;\n  private logger?: Logger;\n\n  constructor(project: SemiontProject, logger?: Logger) {\n    this.projectRoot = project.root;\n    this.gitSync = project.gitSync;\n    this.logger = logger;\n  }\n\n  private shouldRunGit(noGit?: boolean): boolean {\n    return this.gitSync && !noGit;\n  }\n\n  /**\n   * Write content to disk at the location indicated by storageUri.\n   *\n   * API/GUI/AI path: caller provides bytes; file may not yet exist.\n   *\n   * @param content - Raw bytes to write\n   * @param storageUri - file:// URI (e.g. \"file://docs/overview.md\")\n   * @returns Stored resource metadata\n   */\n  async store(content: Buffer, storageUri: string, options?: { noGit?: boolean }): Promise<StoredResource> {\n    const filePath = this.resolveUri(storageUri);\n    const checksum = calculateChecksum(content);\n\n    this.logger?.debug('Storing resource', { storageUri, byteSize: content.length });\n\n    await fs.mkdir(path.dirname(filePath), { recursive: true });\n    await fs.writeFile(filePath, content);\n\n    if (this.shouldRunGit(options?.noGit)) {\n      execFileSync('git', ['add', filePath], { cwd: this.projectRoot });\n    }\n\n    this.logger?.info('Resource stored', { storageUri, checksum, byteSize: content.length });\n\n    return {\n      storageUri,\n      checksum,\n      byteSize: content.length,\n      created: new Date().toISOString(),\n    };\n  }\n\n  /**\n   * Read an existing file and return its metadata.\n   *\n   * CLI path: the file is already on disk. We read it to compute the checksum.\n   * If expectedChecksum is provided, throws ChecksumMismatchError on mismatch.\n   *\n   * @param storageUri - file:// URI (e.g. \"file://docs/overview.md\")\n   * @param expectedChecksum - Optional SHA-256 to verify against\n   * @returns Stored resource metadata\n   * @throws ChecksumMismatchError if expectedChecksum is provided and does not match\n   * @throws Error if file does not exist\n   */\n  async register(storageUri: string, expectedChecksum?: string, options?: { noGit?: boolean }): Promise<StoredResource> {\n    const filePath = this.resolveUri(storageUri);\n\n    this.logger?.debug('Registering resource', { storageUri });\n\n    const content = await fs.readFile(filePath);\n    const checksum = calculateChecksum(content);\n\n    if (expectedChecksum !== undefined && !verifyChecksum(content, expectedChecksum)) {\n      throw new ChecksumMismatchError(storageUri, expectedChecksum, checksum);\n    }\n\n    if (this.shouldRunGit(options?.noGit)) {\n      execFileSync('git', ['add', filePath], { cwd: this.projectRoot });\n    }\n\n    this.logger?.info('Resource registered', { storageUri, checksum, byteSize: content.length });\n\n    return {\n      storageUri,\n      checksum,\n      byteSize: content.length,\n      created: new Date().toISOString(),\n    };\n  }\n\n  /**\n   * Read file content by URI.\n   *\n   * @param storageUri - file:// URI\n   * @returns Raw bytes\n   */\n  async retrieve(storageUri: string): Promise<Buffer> {\n    const filePath = this.resolveUri(storageUri);\n    try {\n      return await fs.readFile(filePath);\n    } catch (error: any) {\n      if (error.code === 'ENOENT') {\n        throw new Error(`Resource not found: ${storageUri}`);\n      }\n      throw error;\n    }\n  }\n\n  /**\n   * Move a file from one URI to another.\n   *\n   * If .git/ exists in the project root and noGit is not set, runs `git mv`.\n   * Otherwise (no .git/ or noGit: true), runs fs.rename.\n   *\n   * @param fromUri - Current file:// URI\n   * @param toUri - New file:// URI\n   * @param options.noGit - Skip git mv even if .git/ is present\n   */\n  async move(fromUri: string, toUri: string, options?: { noGit?: boolean }): Promise<void> {\n    const fromPath = this.resolveUri(fromUri);\n    const toPath = this.resolveUri(toUri);\n\n    this.logger?.debug('Moving resource', { fromUri, toUri });\n\n    await fs.mkdir(path.dirname(toPath), { recursive: true });\n\n    if (this.shouldRunGit(options?.noGit)) {\n      // git mv handles both the filesystem rename and the index update\n      execFileSync('git', ['mv', fromPath, toPath], { cwd: this.projectRoot });\n    } else {\n      await fs.rename(fromPath, toPath);\n    }\n\n    this.logger?.info('Resource moved', { fromUri, toUri });\n  }\n\n  /**\n   * Remove a file from the working tree.\n   *\n   * If .git/ exists and noGit is not set:\n   *   - keepFile false (default): runs `git rm` (removes from index and disk)\n   *   - keepFile true: runs `git rm --cached` (removes from index only, file stays on disk)\n   * If no .git/ or noGit: true:\n   *   - keepFile false: runs fs.unlink\n   *   - keepFile true: no-op on filesystem\n   *\n   * @param storageUri - file:// URI\n   * @param options.noGit - Skip git rm even if .git/ is present\n   * @param options.keepFile - Remove from git index only; leave file on disk\n   */\n  async remove(storageUri: string, options?: { noGit?: boolean; keepFile?: boolean }): Promise<void> {\n    const filePath = this.resolveUri(storageUri);\n    const keepFile = options?.keepFile ?? false;\n\n    this.logger?.debug('Removing resource', { storageUri, keepFile });\n\n    const useGit = this.shouldRunGit(options?.noGit);\n\n    if (useGit) {\n      const gitArgs = keepFile\n        ? ['rm', '--cached', filePath]\n        : ['rm', filePath];\n      execFileSync('git', gitArgs, { cwd: this.projectRoot });\n      this.logger?.info('Resource removed', { storageUri, keepFile, git: true });\n      return;\n    }\n\n    if (keepFile) {\n      this.logger?.info('Resource removed from index (file kept on disk)', { storageUri });\n      return;\n    }\n\n    try {\n      await fs.unlink(filePath);\n      this.logger?.info('Resource removed', { storageUri });\n    } catch (error: any) {\n      if (error.code === 'ENOENT') {\n        this.logger?.warn('Resource file already absent', { storageUri });\n        return;\n      }\n      throw error;\n    }\n  }\n\n  /**\n   * Convert a file:// URI to an absolute filesystem path.\n   *\n   * \"file://docs/overview.md\" → \"{projectRoot}/docs/overview.md\"\n   *\n   * @param storageUri - file:// URI\n   * @returns Absolute path\n   */\n  resolveUri(storageUri: string): string {\n    if (!storageUri.startsWith('file://')) {\n      throw new Error(`Invalid storage URI (must start with file://): ${storageUri}`);\n    }\n    const relativePath = storageUri.slice('file://'.length);\n    return path.join(this.projectRoot, relativePath);\n  }\n}\n\n/**\n * Thrown when a registered file's checksum does not match the expected value.\n * This indicates the file on disk differs from what was recorded (e.g. modified\n * after staging, or wrong file path provided).\n */\nexport class ChecksumMismatchError extends Error {\n  constructor(\n    readonly storageUri: string,\n    readonly expected: string,\n    readonly actual: string,\n  ) {\n    super(\n      `Checksum mismatch for ${storageUri}: expected ${expected.slice(0, 8)}... but got ${actual.slice(0, 8)}...\\n` +\n      `The file on disk differs from the recorded checksum. Has it been modified since staging?`\n    );\n    this.name = 'ChecksumMismatchError';\n  }\n}\n","/**\n * Checksum utilities for content verification\n */\n\nimport { createHash } from 'crypto';\n\n/**\n * Calculate SHA-256 checksum of content\n * @param content The content to hash\n * @returns Hex-encoded SHA-256 hash\n */\nexport function calculateChecksum(content: string | Buffer): string {\n  const hash = createHash('sha256');\n  hash.update(content);\n  return hash.digest('hex');\n}\n\n/**\n * Verify content against a checksum\n * @param content The content to verify\n * @param checksum The expected checksum\n * @returns True if content matches checksum\n */\nexport function verifyChecksum(content: string | Buffer, checksum: string): boolean {\n  return calculateChecksum(content) === checksum;\n}\n","/**\n * MIME Type to File Extension Mapping\n *\n * Maps common MIME types to their standard file extensions.\n * Used by RepresentationStore to save files with proper extensions.\n */\n\n/**\n * Comprehensive MIME type to extension mapping\n */\nconst MIME_TO_EXTENSION: Record<string, string> = {\n  // Text formats\n  'text/plain': '.txt',\n  'text/markdown': '.md',\n  'text/html': '.html',\n  'text/css': '.css',\n  'text/csv': '.csv',\n  'text/xml': '.xml',\n\n  // Application formats - structured data\n  'application/json': '.json',\n  'application/xml': '.xml',\n  'application/yaml': '.yaml',\n  'application/x-yaml': '.yaml',\n\n  // Application formats - documents\n  'application/pdf': '.pdf',\n  'application/msword': '.doc',\n  'application/vnd.openxmlformats-officedocument.wordprocessingml.document': '.docx',\n  'application/vnd.ms-excel': '.xls',\n  'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet': '.xlsx',\n  'application/vnd.ms-powerpoint': '.ppt',\n  'application/vnd.openxmlformats-officedocument.presentationml.presentation': '.pptx',\n\n  // Application formats - archives\n  'application/zip': '.zip',\n  'application/gzip': '.gz',\n  'application/x-tar': '.tar',\n  'application/x-7z-compressed': '.7z',\n\n  // Application formats - executables/binaries\n  'application/octet-stream': '.bin',\n  'application/wasm': '.wasm',\n\n  // Image formats\n  'image/png': '.png',\n  'image/jpeg': '.jpg',\n  'image/gif': '.gif',\n  'image/webp': '.webp',\n  'image/svg+xml': '.svg',\n  'image/bmp': '.bmp',\n  'image/tiff': '.tiff',\n  'image/x-icon': '.ico',\n\n  // Audio formats\n  'audio/mpeg': '.mp3',\n  'audio/wav': '.wav',\n  'audio/ogg': '.ogg',\n  'audio/webm': '.webm',\n  'audio/aac': '.aac',\n  'audio/flac': '.flac',\n\n  // Video formats\n  'video/mp4': '.mp4',\n  'video/mpeg': '.mpeg',\n  'video/webm': '.webm',\n  'video/ogg': '.ogv',\n  'video/quicktime': '.mov',\n  'video/x-msvideo': '.avi',\n\n  // Programming languages\n  'text/javascript': '.js',\n  'application/javascript': '.js',\n  'text/x-typescript': '.ts',\n  'application/typescript': '.ts',\n  'text/x-python': '.py',\n  'text/x-java': '.java',\n  'text/x-c': '.c',\n  'text/x-c++': '.cpp',\n  'text/x-csharp': '.cs',\n  'text/x-go': '.go',\n  'text/x-rust': '.rs',\n  'text/x-ruby': '.rb',\n  'text/x-php': '.php',\n  'text/x-swift': '.swift',\n  'text/x-kotlin': '.kt',\n  'text/x-shell': '.sh',\n\n  // Font formats\n  'font/woff': '.woff',\n  'font/woff2': '.woff2',\n  'font/ttf': '.ttf',\n  'font/otf': '.otf',\n};\n\n/**\n * Get file extension for a MIME type\n *\n * @param mediaType - MIME type (e.g., \"text/markdown\")\n * @returns File extension with leading dot (e.g., \".md\") or \".dat\" if unknown\n *\n * @example\n * getExtensionForMimeType('text/markdown') // => '.md'\n * getExtensionForMimeType('image/png')     // => '.png'\n * getExtensionForMimeType('unknown/type')  // => '.dat'\n */\nexport function getExtensionForMimeType(mediaType: string): string {\n  // Normalize MIME type (lowercase, remove parameters)\n  const normalized = mediaType.toLowerCase().split(';')[0]!.trim();\n\n  // Look up in mapping\n  const extension = MIME_TO_EXTENSION[normalized];\n\n  // Return mapped extension or fallback to .dat\n  return extension || '.dat';\n}\n\n/**\n * Derive a file:// storage URI from a resource name and MIME type.\n *\n * @example\n * deriveStorageUri(\"My Document\", \"text/markdown\") // => \"file://my-document.md\"\n */\nexport function deriveStorageUri(name: string, format: string): string {\n  const slug = name\n    .toLowerCase()\n    .replace(/[^a-z0-9]+/g, '-')\n    .replace(/^-|-$/g, '');\n  const ext = getExtensionForMimeType(format);\n  return `file://${slug}${ext}`;\n}\n\n/**\n * Check if a MIME type has a known extension mapping\n *\n * @param mediaType - MIME type to check\n * @returns true if extension is known, false if would fallback to .dat\n */\nexport function hasKnownExtension(mediaType: string): boolean {\n  const normalized = mediaType.toLowerCase().split(';')[0]!.trim();\n  return normalized in MIME_TO_EXTENSION;\n}\n","/**\n * PDF Text Layer Extraction\n *\n * Extracts positioned text from native, non-scanned PDFs using pdfjs-dist.\n * Returns null for scanned/image-only PDFs (no text items).\n *\n * Coordinates are in PDF point space, originating from the bottom-left.\n * The Y-flip to canvas pixels happens downstream.\n */\n\nimport * as pdfjs from 'pdfjs-dist/legacy/build/pdf.mjs';\nimport type { PdfTextLayer, PdfPageInfo, PdfTextItem } from './pdf-text-layer';\n\nexport async function extractPdfTextLayer(\n    bytes: Uint8Array | Buffer\n): Promise<PdfTextLayer | null> {\n    // pdf.js v5 removed the isEvalSupported option; this path only calls\n    // getTextContent (no rendering / no PDF functions).\n    const doc = await pdfjs.getDocument({ data: bytes }).promise;\n\n    try {\n        const pages: PdfPageInfo[] = [];\n        const items: PdfTextItem[] = [];\n        let text = '';\n        let hasAnyTextItems = false;\n\n        for (let pageNum = 1; pageNum <= doc.numPages; pageNum++) {\n            const page = await doc.getPage(pageNum);\n            const viewport = page.getViewport({ scale: 1.0 });\n            const content = await page.getTextContent();  // all text items on the page\n\n            pages.push({\n                pageNumber: pageNum,\n                widthPt: viewport.width,\n                heightPt: viewport.height,\n            });\n\n            for (const item of content.items) {\n                if (!('str' in item)) continue;  // skip marked-content items (no text)\n\n                if (item.str.trim()) {\n                    hasAnyTextItems = true;\n                    const start = text.length;\n                    text += item.str;\n                    const end = text.length;  // range covers only this run's own chars\n\n                    const [, , , , x, y] = item.transform as number[];\n\n                    items.push({\n                        start,\n                        end,\n                        page: pageNum,\n                        x,\n                        y,\n                        width: item.width,\n                        height: item.height,\n                    });\n\n                    // Separator AFTER recording the run, so its [start, end) never\n                    // includes it. pdf.js flags the last run on a line with hasEOL —\n                    // newline there, space between words otherwise, so reading-order\n                    // lines don't glue (e.g. \"textsecond\").\n                    text += item.hasEOL ? '\\n' : ' ';\n                } else if (item.hasEOL) {\n                    // Standalone end-of-line marker (empty/whitespace str): keep the\n                    // line break without letting whitespace-only runs add stray spaces.\n                    text += '\\n';\n                }\n            }\n\n            text += '\\n';  // page break\n        }\n\n        if (!hasAnyTextItems) return null;\n\n        return { pages, text, items };\n    } finally {\n        // Release the pdf.js document — Phase 2 runs this in a long-lived worker pool.\n        await doc.destroy();\n    }\n}\n","import type { PdfCoordinate } from '@semiont/core';\nimport type { PdfTextLayer, PdfTextItem } from './pdf-text-layer';\n\n/**\n * Items whose baseline Y is within this many PDF points are treated as being on\n * the same line. Tuned for ~12pt body text; revisit for documents with large or\n * variable font sizes (Phase 4 / #738).\n */\nconst SAME_LINE_THRESHOLD_PT = 2;\n\n/**\n * Locates bounding rectangles for a span of text in a PdfTextLayer\n * (single-line or multi-line).\n * \n * Finds all overlapping items [start, end), groups them by page and line,\n * and records one bounding rectangle per line as a PdfCoordinate.\n * \n * Returns array of PdfCoordinate, one per line of text covered by the span.\n * Returns empty array if no items overlap the span.\n */\nexport function locate(\n    layer: PdfTextLayer,\n    start: number,\n    end: number\n): PdfCoordinate[] {\n    const overlap: PdfTextItem[] = layer.items.filter(\n        item => item.start < end && item.end > start\n    );\n    if (overlap.length === 0) return [];\n\n    const pages: Map<number, PdfTextItem[]> = groupItemsByPage(overlap);\n    const rects: PdfCoordinate[] = [];\n\n    // for each page, group items into lines and compute one rectangle per line\n    for (const [page, pageItems] of pages) {\n        const lines = groupItemsByLine(pageItems, SAME_LINE_THRESHOLD_PT);\n        // Compute one bounding rectangle per line and add it to rects\n        for (const lineItems of lines) {\n            const x = Math.min(...lineItems.map(i => i.x));\n            const right = Math.max(...lineItems.map(i => i.x + i.width));\n            const y = Math.min(...lineItems.map(i => i.y));\n            const top = Math.max(...lineItems.map(i => i.y + i.height));\n            rects.push({page, x, y, width: right - x, height: top - y});\n        }\n    }\n    return rects;\n}\n\nfunction groupItemsByPage(items: PdfTextItem[]): Map<number, PdfTextItem[]> {\n    const map = new Map<number, PdfTextItem[]>();\n    for (const item of items) {\n        const existing = map.get(item.page);\n        if (existing) {\n            existing.push(item);\n        } else {\n            map.set(item.page, [item]);\n        }\n    }\n    return map;\n}\n\n\n/**\n * Sorts text items into lines when their y coordinates are\n * within `sameLineThreshold` points of each other.\n * Sorted top-to-bottom (descending y in PDF space), then left-to-right.\n * \n * Returns 2D array: \n * Outer array = list of lines\n * Inner array = list of items on that line\n*/\nfunction groupItemsByLine(items: PdfTextItem[], sameLineThreshold: number): PdfTextItem[][] {\n    // Sort top-to-bottom by y; if y is equal, sort left-to-right by x\n    const sorted = [...items].sort((a, b) => b.y - a.y || a.x - b.x);\n    const lines: PdfTextItem[][] = [];\n    let currentLine: PdfTextItem[] = [];\n\n    for (const item of sorted) {\n        if (currentLine.length === 0 || Math.abs(item.y - currentLine[0].y) <= sameLineThreshold) {\n            currentLine.push(item);\n        } else {\n            lines.push(currentLine);\n            currentLine = [item];\n        }\n    }\n    if (currentLine.length > 0) lines.push(currentLine);\n    return lines;\n}\n"],"mappings":";AAsBA,SAAS,YAAY,UAAU;AAC/B,SAAS,oBAAoB;AAC7B,OAAO,UAAU;;;ACpBjB,SAAS,kBAAkB;AAOpB,SAAS,kBAAkB,SAAkC;AAClE,QAAM,OAAO,WAAW,QAAQ;AAChC,OAAK,OAAO,OAAO;AACnB,SAAO,KAAK,OAAO,KAAK;AAC1B;AAQO,SAAS,eAAe,SAA0B,UAA2B;AAClF,SAAO,kBAAkB,OAAO,MAAM;AACxC;;;ADiBO,IAAM,mBAAN,MAAuB;AAAA,EACpB;AAAA,EACA;AAAA,EACA;AAAA,EAER,YAAY,SAAyB,QAAiB;AACpD,SAAK,cAAc,QAAQ;AAC3B,SAAK,UAAU,QAAQ;AACvB,SAAK,SAAS;AAAA,EAChB;AAAA,EAEQ,aAAa,OAA0B;AAC7C,WAAO,KAAK,WAAW,CAAC;AAAA,EAC1B;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAWA,MAAM,MAAM,SAAiB,YAAoB,SAAwD;AACvG,UAAM,WAAW,KAAK,WAAW,UAAU;AAC3C,UAAM,WAAW,kBAAkB,OAAO;AAE1C,SAAK,QAAQ,MAAM,oBAAoB,EAAE,YAAY,UAAU,QAAQ,OAAO,CAAC;AAE/E,UAAM,GAAG,MAAM,KAAK,QAAQ,QAAQ,GAAG,EAAE,WAAW,KAAK,CAAC;AAC1D,UAAM,GAAG,UAAU,UAAU,OAAO;AAEpC,QAAI,KAAK,aAAa,SAAS,KAAK,GAAG;AACrC,mBAAa,OAAO,CAAC,OAAO,QAAQ,GAAG,EAAE,KAAK,KAAK,YAAY,CAAC;AAAA,IAClE;AAEA,SAAK,QAAQ,KAAK,mBAAmB,EAAE,YAAY,UAAU,UAAU,QAAQ,OAAO,CAAC;AAEvF,WAAO;AAAA,MACL;AAAA,MACA;AAAA,MACA,UAAU,QAAQ;AAAA,MAClB,UAAS,oBAAI,KAAK,GAAE,YAAY;AAAA,IAClC;AAAA,EACF;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAcA,MAAM,SAAS,YAAoB,kBAA2B,SAAwD;AACpH,UAAM,WAAW,KAAK,WAAW,UAAU;AAE3C,SAAK,QAAQ,MAAM,wBAAwB,EAAE,WAAW,CAAC;AAEzD,UAAM,UAAU,MAAM,GAAG,SAAS,QAAQ;AAC1C,UAAM,WAAW,kBAAkB,OAAO;AAE1C,QAAI,qBAAqB,UAAa,CAAC,eAAe,SAAS,gBAAgB,GAAG;AAChF,YAAM,IAAI,sBAAsB,YAAY,kBAAkB,QAAQ;AAAA,IACxE;AAEA,QAAI,KAAK,aAAa,SAAS,KAAK,GAAG;AACrC,mBAAa,OAAO,CAAC,OAAO,QAAQ,GAAG,EAAE,KAAK,KAAK,YAAY,CAAC;AAAA,IAClE;AAEA,SAAK,QAAQ,KAAK,uBAAuB,EAAE,YAAY,UAAU,UAAU,QAAQ,OAAO,CAAC;AAE3F,WAAO;AAAA,MACL;AAAA,MACA;AAAA,MACA,UAAU,QAAQ;AAAA,MAClB,UAAS,oBAAI,KAAK,GAAE,YAAY;AAAA,IAClC;AAAA,EACF;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAQA,MAAM,SAAS,YAAqC;AAClD,UAAM,WAAW,KAAK,WAAW,UAAU;AAC3C,QAAI;AACF,aAAO,MAAM,GAAG,SAAS,QAAQ;AAAA,IACnC,SAAS,OAAY;AACnB,UAAI,MAAM,SAAS,UAAU;AAC3B,cAAM,IAAI,MAAM,uBAAuB,UAAU,EAAE;AAAA,MACrD;AACA,YAAM;AAAA,IACR;AAAA,EACF;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAYA,MAAM,KAAK,SAAiB,OAAe,SAA8C;AACvF,UAAM,WAAW,KAAK,WAAW,OAAO;AACxC,UAAM,SAAS,KAAK,WAAW,KAAK;AAEpC,SAAK,QAAQ,MAAM,mBAAmB,EAAE,SAAS,MAAM,CAAC;AAExD,UAAM,GAAG,MAAM,KAAK,QAAQ,MAAM,GAAG,EAAE,WAAW,KAAK,CAAC;AAExD,QAAI,KAAK,aAAa,SAAS,KAAK,GAAG;AAErC,mBAAa,OAAO,CAAC,MAAM,UAAU,MAAM,GAAG,EAAE,KAAK,KAAK,YAAY,CAAC;AAAA,IACzE,OAAO;AACL,YAAM,GAAG,OAAO,UAAU,MAAM;AAAA,IAClC;AAEA,SAAK,QAAQ,KAAK,kBAAkB,EAAE,SAAS,MAAM,CAAC;AAAA,EACxD;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAgBA,MAAM,OAAO,YAAoB,SAAkE;AACjG,UAAM,WAAW,KAAK,WAAW,UAAU;AAC3C,UAAM,WAAW,SAAS,YAAY;AAEtC,SAAK,QAAQ,MAAM,qBAAqB,EAAE,YAAY,SAAS,CAAC;AAEhE,UAAM,SAAS,KAAK,aAAa,SAAS,KAAK;AAE/C,QAAI,QAAQ;AACV,YAAM,UAAU,WACZ,CAAC,MAAM,YAAY,QAAQ,IAC3B,CAAC,MAAM,QAAQ;AACnB,mBAAa,OAAO,SAAS,EAAE,KAAK,KAAK,YAAY,CAAC;AACtD,WAAK,QAAQ,KAAK,oBAAoB,EAAE,YAAY,UAAU,KAAK,KAAK,CAAC;AACzE;AAAA,IACF;AAEA,QAAI,UAAU;AACZ,WAAK,QAAQ,KAAK,mDAAmD,EAAE,WAAW,CAAC;AACnF;AAAA,IACF;AAEA,QAAI;AACF,YAAM,GAAG,OAAO,QAAQ;AACxB,WAAK,QAAQ,KAAK,oBAAoB,EAAE,WAAW,CAAC;AAAA,IACtD,SAAS,OAAY;AACnB,UAAI,MAAM,SAAS,UAAU;AAC3B,aAAK,QAAQ,KAAK,gCAAgC,EAAE,WAAW,CAAC;AAChE;AAAA,MACF;AACA,YAAM;AAAA,IACR;AAAA,EACF;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA;AAAA,EAUA,WAAW,YAA4B;AACrC,QAAI,CAAC,WAAW,WAAW,SAAS,GAAG;AACrC,YAAM,IAAI,MAAM,kDAAkD,UAAU,EAAE;AAAA,IAChF;AACA,UAAM,eAAe,WAAW,MAAM,UAAU,MAAM;AACtD,WAAO,KAAK,KAAK,KAAK,aAAa,YAAY;AAAA,EACjD;AACF;AAOO,IAAM,wBAAN,cAAoC,MAAM;AAAA,EAC/C,YACW,YACA,UACA,QACT;AACA;AAAA,MACE,yBAAyB,UAAU,cAAc,SAAS,MAAM,GAAG,CAAC,CAAC,eAAe,OAAO,MAAM,GAAG,CAAC,CAAC;AAAA;AAAA,IAExG;AAPS;AACA;AACA;AAMT,SAAK,OAAO;AAAA,EACd;AAAA,EATW;AAAA,EACA;AAAA,EACA;AAQb;;;AErPA,IAAM,oBAA4C;AAAA;AAAA,EAEhD,cAAc;AAAA,EACd,iBAAiB;AAAA,EACjB,aAAa;AAAA,EACb,YAAY;AAAA,EACZ,YAAY;AAAA,EACZ,YAAY;AAAA;AAAA,EAGZ,oBAAoB;AAAA,EACpB,mBAAmB;AAAA,EACnB,oBAAoB;AAAA,EACpB,sBAAsB;AAAA;AAAA,EAGtB,mBAAmB;AAAA,EACnB,sBAAsB;AAAA,EACtB,2EAA2E;AAAA,EAC3E,4BAA4B;AAAA,EAC5B,qEAAqE;AAAA,EACrE,iCAAiC;AAAA,EACjC,6EAA6E;AAAA;AAAA,EAG7E,mBAAmB;AAAA,EACnB,oBAAoB;AAAA,EACpB,qBAAqB;AAAA,EACrB,+BAA+B;AAAA;AAAA,EAG/B,4BAA4B;AAAA,EAC5B,oBAAoB;AAAA;AAAA,EAGpB,aAAa;AAAA,EACb,cAAc;AAAA,EACd,aAAa;AAAA,EACb,cAAc;AAAA,EACd,iBAAiB;AAAA,EACjB,aAAa;AAAA,EACb,cAAc;AAAA,EACd,gBAAgB;AAAA;AAAA,EAGhB,cAAc;AAAA,EACd,aAAa;AAAA,EACb,aAAa;AAAA,EACb,cAAc;AAAA,EACd,aAAa;AAAA,EACb,cAAc;AAAA;AAAA,EAGd,aAAa;AAAA,EACb,cAAc;AAAA,EACd,cAAc;AAAA,EACd,aAAa;AAAA,EACb,mBAAmB;AAAA,EACnB,mBAAmB;AAAA;AAAA,EAGnB,mBAAmB;AAAA,EACnB,0BAA0B;AAAA,EAC1B,qBAAqB;AAAA,EACrB,0BAA0B;AAAA,EAC1B,iBAAiB;AAAA,EACjB,eAAe;AAAA,EACf,YAAY;AAAA,EACZ,cAAc;AAAA,EACd,iBAAiB;AAAA,EACjB,aAAa;AAAA,EACb,eAAe;AAAA,EACf,eAAe;AAAA,EACf,cAAc;AAAA,EACd,gBAAgB;AAAA,EAChB,iBAAiB;AAAA,EACjB,gBAAgB;AAAA;AAAA,EAGhB,aAAa;AAAA,EACb,cAAc;AAAA,EACd,YAAY;AAAA,EACZ,YAAY;AACd;AAaO,SAAS,wBAAwB,WAA2B;AAEjE,QAAM,aAAa,UAAU,YAAY,EAAE,MAAM,GAAG,EAAE,CAAC,EAAG,KAAK;AAG/D,QAAM,YAAY,kBAAkB,UAAU;AAG9C,SAAO,aAAa;AACtB;AAQO,SAAS,iBAAiB,MAAc,QAAwB;AACrE,QAAM,OAAO,KACV,YAAY,EACZ,QAAQ,eAAe,GAAG,EAC1B,QAAQ,UAAU,EAAE;AACvB,QAAM,MAAM,wBAAwB,MAAM;AAC1C,SAAO,UAAU,IAAI,GAAG,GAAG;AAC7B;AAQO,SAAS,kBAAkB,WAA4B;AAC5D,QAAM,aAAa,UAAU,YAAY,EAAE,MAAM,GAAG,EAAE,CAAC,EAAG,KAAK;AAC/D,SAAO,cAAc;AACvB;;;ACnIA,YAAY,WAAW;AAGvB,eAAsB,oBAClB,OAC4B;AAG5B,QAAM,MAAM,MAAY,kBAAY,EAAE,MAAM,MAAM,CAAC,EAAE;AAErD,MAAI;AACA,UAAM,QAAuB,CAAC;AAC9B,UAAM,QAAuB,CAAC;AAC9B,QAAI,OAAO;AACX,QAAI,kBAAkB;AAEtB,aAAS,UAAU,GAAG,WAAW,IAAI,UAAU,WAAW;AACtD,YAAM,OAAO,MAAM,IAAI,QAAQ,OAAO;AACtC,YAAM,WAAW,KAAK,YAAY,EAAE,OAAO,EAAI,CAAC;AAChD,YAAM,UAAU,MAAM,KAAK,eAAe;AAE1C,YAAM,KAAK;AAAA,QACP,YAAY;AAAA,QACZ,SAAS,SAAS;AAAA,QAClB,UAAU,SAAS;AAAA,MACvB,CAAC;AAED,iBAAW,QAAQ,QAAQ,OAAO;AAC9B,YAAI,EAAE,SAAS,MAAO;AAEtB,YAAI,KAAK,IAAI,KAAK,GAAG;AACjB,4BAAkB;AAClB,gBAAM,QAAQ,KAAK;AACnB,kBAAQ,KAAK;AACb,gBAAM,MAAM,KAAK;AAEjB,gBAAM,CAAC,EAAE,EAAE,EAAE,EAAE,GAAG,CAAC,IAAI,KAAK;AAE5B,gBAAM,KAAK;AAAA,YACP;AAAA,YACA;AAAA,YACA,MAAM;AAAA,YACN;AAAA,YACA;AAAA,YACA,OAAO,KAAK;AAAA,YACZ,QAAQ,KAAK;AAAA,UACjB,CAAC;AAMD,kBAAQ,KAAK,SAAS,OAAO;AAAA,QACjC,WAAW,KAAK,QAAQ;AAGpB,kBAAQ;AAAA,QACZ;AAAA,MACJ;AAEA,cAAQ;AAAA,IACZ;AAEA,QAAI,CAAC,gBAAiB,QAAO;AAE7B,WAAO,EAAE,OAAO,MAAM,MAAM;AAAA,EAChC,UAAE;AAEE,UAAM,IAAI,QAAQ;AAAA,EACtB;AACJ;;;ACxEA,IAAM,yBAAyB;AAYxB,SAAS,OACZ,OACA,OACA,KACe;AACf,QAAM,UAAyB,MAAM,MAAM;AAAA,IACvC,UAAQ,KAAK,QAAQ,OAAO,KAAK,MAAM;AAAA,EAC3C;AACA,MAAI,QAAQ,WAAW,EAAG,QAAO,CAAC;AAElC,QAAM,QAAoC,iBAAiB,OAAO;AAClE,QAAM,QAAyB,CAAC;AAGhC,aAAW,CAAC,MAAM,SAAS,KAAK,OAAO;AACnC,UAAM,QAAQ,iBAAiB,WAAW,sBAAsB;AAEhE,eAAW,aAAa,OAAO;AAC3B,YAAM,IAAI,KAAK,IAAI,GAAG,UAAU,IAAI,OAAK,EAAE,CAAC,CAAC;AAC7C,YAAM,QAAQ,KAAK,IAAI,GAAG,UAAU,IAAI,OAAK,EAAE,IAAI,EAAE,KAAK,CAAC;AAC3D,YAAM,IAAI,KAAK,IAAI,GAAG,UAAU,IAAI,OAAK,EAAE,CAAC,CAAC;AAC7C,YAAM,MAAM,KAAK,IAAI,GAAG,UAAU,IAAI,OAAK,EAAE,IAAI,EAAE,MAAM,CAAC;AAC1D,YAAM,KAAK,EAAC,MAAM,GAAG,GAAG,OAAO,QAAQ,GAAG,QAAQ,MAAM,EAAC,CAAC;AAAA,IAC9D;AAAA,EACJ;AACA,SAAO;AACX;AAEA,SAAS,iBAAiB,OAAkD;AACxE,QAAM,MAAM,oBAAI,IAA2B;AAC3C,aAAW,QAAQ,OAAO;AACtB,UAAM,WAAW,IAAI,IAAI,KAAK,IAAI;AAClC,QAAI,UAAU;AACV,eAAS,KAAK,IAAI;AAAA,IACtB,OAAO;AACH,UAAI,IAAI,KAAK,MAAM,CAAC,IAAI,CAAC;AAAA,IAC7B;AAAA,EACJ;AACA,SAAO;AACX;AAYA,SAAS,iBAAiB,OAAsB,mBAA4C;AAExF,QAAM,SAAS,CAAC,GAAG,KAAK,EAAE,KAAK,CAAC,GAAG,MAAM,EAAE,IAAI,EAAE,KAAK,EAAE,IAAI,EAAE,CAAC;AAC/D,QAAM,QAAyB,CAAC;AAChC,MAAI,cAA6B,CAAC;AAElC,aAAW,QAAQ,QAAQ;AACvB,QAAI,YAAY,WAAW,KAAK,KAAK,IAAI,KAAK,IAAI,YAAY,CAAC,EAAE,CAAC,KAAK,mBAAmB;AACtF,kBAAY,KAAK,IAAI;AAAA,IACzB,OAAO;AACH,YAAM,KAAK,WAAW;AACtB,oBAAc,CAAC,IAAI;AAAA,IACvB;AAAA,EACJ;AACA,MAAI,YAAY,SAAS,EAAG,OAAM,KAAK,WAAW;AAClD,SAAO;AACX;","names":[]}

package/package.json

@@ -1,6 +1,9 @@
 {
   "name": "@semiont/content",
-  "version": "0.5.4",
+  "version": "0.5.6",
+  "engines": {
+    "node": ">=24.0.0"
+  },
   "type": "module",
   "description": "Content-addressed storage for resource representations",
   "main": "./dist/index.js",
@@ -16,20 +19,25 @@
     "README.md"
   ],
   "scripts": {
-    "build": "npm run typecheck && tsup && tsc -p tsconfig.build.json",
+    "build": "npm run typecheck && tsup && tsc -p tsconfig.build.json && rollup -c rollup.dts.config.mjs && rm -rf dist-types",
     "typecheck": "tsc --noEmit",
-    "clean": "rm -rf dist",
+    "clean": "rm -rf dist dist-types",
     "test": "vitest run",
     "test:watch": "vitest",
     "test:coverage": "vitest run --coverage"
   },
   "dependencies": {
-    "@semiont/core": "*"
+    "@semiont/core": "*",
+    "pdfjs-dist": "^5.7.284"
   },
   "devDependencies": {
-    "@vitest/coverage-v8": "^4.1.0",
+    "@vitest/coverage-v8": "^4.1.8",
+    "pdf-lib": "^1.17.1",
+    "rollup": "^4.61.0",
+    "rollup-plugin-dts": "^6.4.1",
     "tsup": "^8.0.1",
-    "typescript": "^6.0.2"
+    "typescript": "^6.0.2",
+    "vitest": "^4.1.8"
   },
   "keywords": [
     "content",

package/dist/checksum.d.ts

@@ -1,17 +0,0 @@
-/**
- * Checksum utilities for content verification
- */
-/**
- * Calculate SHA-256 checksum of content
- * @param content The content to hash
- * @returns Hex-encoded SHA-256 hash
- */
-export declare function calculateChecksum(content: string | Buffer): string;
-/**
- * Verify content against a checksum
- * @param content The content to verify
- * @param checksum The expected checksum
- * @returns True if content matches checksum
- */
-export declare function verifyChecksum(content: string | Buffer, checksum: string): boolean;
-//# sourceMappingURL=checksum.d.ts.map

package/dist/checksum.d.ts.map

@@ -1 +0,0 @@
-{"version":3,"file":"checksum.d.ts","sourceRoot":"","sources":["../src/checksum.ts"],"names":[],"mappings":"AAAA;;GAEG;AAIH;;;;GAIG;AACH,wBAAgB,iBAAiB,CAAC,OAAO,EAAE,MAAM,GAAG,MAAM,GAAG,MAAM,CAIlE;AAED;;;;;GAKG;AACH,wBAAgB,cAAc,CAAC,OAAO,EAAE,MAAM,GAAG,MAAM,EAAE,QAAQ,EAAE,MAAM,GAAG,OAAO,CAElF"}

package/dist/index.d.ts.map

@@ -1 +0,0 @@
-{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../src/index.ts"],"names":[],"mappings":"AAAA;;;;GAIG;AAGH,OAAO,EACL,gBAAgB,EAChB,qBAAqB,EACrB,KAAK,cAAc,GACpB,MAAM,sBAAsB,CAAC;AAG9B,OAAO,EACL,uBAAuB,EACvB,iBAAiB,EACjB,gBAAgB,GACjB,MAAM,mBAAmB,CAAC;AAG3B,OAAO,EACL,iBAAiB,EACjB,cAAc,EACf,MAAM,YAAY,CAAC"}

package/dist/mime-extensions.d.ts

@@ -1,33 +0,0 @@
-/**
- * MIME Type to File Extension Mapping
- *
- * Maps common MIME types to their standard file extensions.
- * Used by RepresentationStore to save files with proper extensions.
- */
-/**
- * Get file extension for a MIME type
- *
- * @param mediaType - MIME type (e.g., "text/markdown")
- * @returns File extension with leading dot (e.g., ".md") or ".dat" if unknown
- *
- * @example
- * getExtensionForMimeType('text/markdown') // => '.md'
- * getExtensionForMimeType('image/png')     // => '.png'
- * getExtensionForMimeType('unknown/type')  // => '.dat'
- */
-export declare function getExtensionForMimeType(mediaType: string): string;
-/**
- * Derive a file:// storage URI from a resource name and MIME type.
- *
- * @example
- * deriveStorageUri("My Document", "text/markdown") // => "file://my-document.md"
- */
-export declare function deriveStorageUri(name: string, format: string): string;
-/**
- * Check if a MIME type has a known extension mapping
- *
- * @param mediaType - MIME type to check
- * @returns true if extension is known, false if would fallback to .dat
- */
-export declare function hasKnownExtension(mediaType: string): boolean;
-//# sourceMappingURL=mime-extensions.d.ts.map

package/dist/mime-extensions.d.ts.map

@@ -1 +0,0 @@
-{"version":3,"file":"mime-extensions.d.ts","sourceRoot":"","sources":["../src/mime-extensions.ts"],"names":[],"mappings":"AAAA;;;;;GAKG;AA0FH;;;;;;;;;;GAUG;AACH,wBAAgB,uBAAuB,CAAC,SAAS,EAAE,MAAM,GAAG,MAAM,CASjE;AAED;;;;;GAKG;AACH,wBAAgB,gBAAgB,CAAC,IAAI,EAAE,MAAM,EAAE,MAAM,EAAE,MAAM,GAAG,MAAM,CAOrE;AAED;;;;;GAKG;AACH,wBAAgB,iBAAiB,CAAC,SAAS,EAAE,MAAM,GAAG,OAAO,CAG5D"}

package/dist/representation-store.d.ts

@@ -1,94 +0,0 @@
-/**
- * RepresentationStore - Content-addressed storage for byte-level resource representations
- *
- * Handles storage and retrieval of concrete byte-level renditions of resources.
- * Uses content-addressed storage where the checksum IS the filename.
- * Supports multiple storage backends (filesystem, S3, IPFS, etc.)
- *
- * Storage structure (filesystem):
- * basePath/representations/{mediaType}/{ab}/{cd}/rep-{checksum}{extension}
- *
- * Where:
- * - {mediaType} is base MIME type with "/" encoded as "~1" (e.g., "text~1markdown")
- * - {ab}/{cd} are first 4 hex digits of checksum for sharding
- * - {checksum} is the raw SHA-256 hex hash (e.g., "5aaa0b72abc123...")
- * - {extension} is derived from base MIME type (.md, .txt, .png, etc.)
- *
- * Example:
- * For content with checksum "5aaa0b72abc123..." and mediaType "text/markdown; charset=iso-8859-1":
- * - Storage path: basePath/representations/text~1markdown/5a/aa/rep-5aaa0b72abc123....md
- * - Stored mediaType: "text/markdown; charset=iso-8859-1" (full type with charset preserved)
- *
- * Character Encoding:
- * - Charset parameters in mediaType are preserved in metadata (e.g., "text/plain; charset=iso-8859-1")
- * - Storage path uses only base MIME type (strips charset for directory structure)
- * - Content stored as raw bytes - charset only affects decoding on retrieval
- *
- * This design provides:
- * - O(1) content retrieval by checksum + mediaType
- * - Automatic deduplication (identical content = same file)
- * - Idempotent storage operations
- * - Proper file extensions for filesystem browsing
- * - Faithful preservation of character encoding metadata
- */
-import type { SemiontProject } from '@semiont/core/node';
-import type { Logger } from '@semiont/core';
-/**
- * Metadata for a representation being stored
- */
-export interface RepresentationMetadata {
-    mediaType: string;
-    filename?: string;
-    encoding?: string;
-    language?: string;
-    rel?: 'original' | 'thumbnail' | 'preview' | 'optimized' | 'derived' | 'other';
-}
-/**
- * Complete representation information
- */
-export interface StoredRepresentation extends RepresentationMetadata {
-    '@id': string;
-    byteSize: number;
-    checksum: string;
-    created: string;
-}
-/**
- * Interface for representation storage backends
- */
-export interface RepresentationStore {
-    /**
-     * Store content and return representation metadata
-     *
-     * @param content - Raw bytes to store
-     * @param metadata - Representation metadata
-     * @returns Complete representation info with checksum
-     */
-    store(content: Buffer, metadata: RepresentationMetadata): Promise<StoredRepresentation>;
-    /**
-     * Retrieve content by checksum (content-addressed lookup)
-     *
-     * @param checksum - Content checksum as raw hex (e.g., "5aaa0b72...")
-     * @param mediaType - MIME type (e.g., "text/markdown")
-     * @returns Raw bytes
-     */
-    retrieve(checksum: string, mediaType: string): Promise<Buffer>;
-}
-/**
- * Filesystem implementation of RepresentationStore
- */
-export declare class FilesystemRepresentationStore implements RepresentationStore {
-    private basePath;
-    private logger?;
-    constructor(project: SemiontProject, logger?: Logger);
-    store(content: Buffer, metadata: RepresentationMetadata): Promise<StoredRepresentation>;
-    retrieve(checksum: string, mediaType: string): Promise<Buffer>;
-    /**
-     * Encode media type for filesystem path
-     * Replaces "/" with "~1" to avoid directory separators
-     *
-     * @param mediaType - MIME type (e.g., "text/markdown")
-     * @returns Encoded path segment (e.g., "text~1markdown")
-     */
-    private encodeMediaType;
-}
-//# sourceMappingURL=representation-store.d.ts.map

package/dist/representation-store.d.ts.map

@@ -1 +0,0 @@
-{"version":3,"file":"representation-store.d.ts","sourceRoot":"","sources":["../src/representation-store.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GAgCG;AAIH,OAAO,KAAK,EAAE,cAAc,EAAE,MAAM,oBAAoB,CAAC;AACzD,OAAO,KAAK,EAAE,MAAM,EAAE,MAAM,eAAe,CAAC;AAI5C;;GAEG;AACH,MAAM,WAAW,sBAAsB;IACrC,SAAS,EAAE,MAAM,CAAC;IAClB,QAAQ,CAAC,EAAE,MAAM,CAAC;IAClB,QAAQ,CAAC,EAAE,MAAM,CAAC;IAClB,QAAQ,CAAC,EAAE,MAAM,CAAC;IAClB,GAAG,CAAC,EAAE,UAAU,GAAG,WAAW,GAAG,SAAS,GAAG,WAAW,GAAG,SAAS,GAAG,OAAO,CAAC;CAChF;AAED;;GAEG;AACH,MAAM,WAAW,oBAAqB,SAAQ,sBAAsB;IAClE,KAAK,EAAE,MAAM,CAAC;IACd,QAAQ,EAAE,MAAM,CAAC;IACjB,QAAQ,EAAE,MAAM,CAAC;IACjB,OAAO,EAAE,MAAM,CAAC;CACjB;AAED;;GAEG;AACH,MAAM,WAAW,mBAAmB;IAClC;;;;;;OAMG;IACH,KAAK,CAAC,OAAO,EAAE,MAAM,EAAE,QAAQ,EAAE,sBAAsB,GAAG,OAAO,CAAC,oBAAoB,CAAC,CAAC;IAExF;;;;;;OAMG;IACH,QAAQ,CAAC,QAAQ,EAAE,MAAM,EAAE,SAAS,EAAE,MAAM,GAAG,OAAO,CAAC,MAAM,CAAC,CAAC;CAChE;AAED;;GAEG;AACH,qBAAa,6BAA8B,YAAW,mBAAmB;IACvE,OAAO,CAAC,QAAQ,CAAS;IACzB,OAAO,CAAC,MAAM,CAAC,CAAS;gBAEZ,OAAO,EAAE,cAAc,EAAE,MAAM,CAAC,EAAE,MAAM;IAK9C,KAAK,CAAC,OAAO,EAAE,MAAM,EAAE,QAAQ,EAAE,sBAAsB,GAAG,OAAO,CAAC,oBAAoB,CAAC;IAwDvF,QAAQ,CAAC,QAAQ,EAAE,MAAM,EAAE,SAAS,EAAE,MAAM,GAAG,OAAO,CAAC,MAAM,CAAC;IAyDpE;;;;;;OAMG;IACH,OAAO,CAAC,eAAe;CAGxB"}

package/dist/working-tree-store.d.ts

@@ -1,128 +0,0 @@
-/**
- * WorkingTreeStore - Manages files in the project working tree
- *
- * Unlike the old content-addressed RepresentationStore, this store treats
- * the working tree (project root) as the source of truth for file content.
- * Resources are identified by their file:// URI, which is stable across
- * content changes and moves (tracked by events).
- *
- * Two write paths:
- * - store(content, storageUri): Write bytes to disk (API/GUI/AI path).
- *   Used when the file does not yet exist and the caller provides content.
- * - register(storageUri, expectedChecksum?): Read an existing file and
- *   return its metadata (CLI path). The file is already on disk; we just
- *   verify and record it. If expectedChecksum is provided, throws on mismatch.
- *
- * Storage layout:
- *   {projectRoot}/{path-from-uri}
- *
- * For example, storageUri "file://docs/overview.md" resolves to
- *   {projectRoot}/docs/overview.md
- */
-import type { SemiontProject } from '@semiont/core/node';
-import type { Logger } from '@semiont/core';
-/**
- * Result of store() or register()
- */
-export interface StoredResource {
-    storageUri: string;
-    checksum: string;
-    byteSize: number;
-    created: string;
-}
-/**
- * Manages files in the project working tree
- */
-export declare class WorkingTreeStore {
-    private projectRoot;
-    private gitSync;
-    private logger?;
-    constructor(project: SemiontProject, logger?: Logger);
-    private shouldRunGit;
-    /**
-     * Write content to disk at the location indicated by storageUri.
-     *
-     * API/GUI/AI path: caller provides bytes; file may not yet exist.
-     *
-     * @param content - Raw bytes to write
-     * @param storageUri - file:// URI (e.g. "file://docs/overview.md")
-     * @returns Stored resource metadata
-     */
-    store(content: Buffer, storageUri: string, options?: {
-        noGit?: boolean;
-    }): Promise<StoredResource>;
-    /**
-     * Read an existing file and return its metadata.
-     *
-     * CLI path: the file is already on disk. We read it to compute the checksum.
-     * If expectedChecksum is provided, throws ChecksumMismatchError on mismatch.
-     *
-     * @param storageUri - file:// URI (e.g. "file://docs/overview.md")
-     * @param expectedChecksum - Optional SHA-256 to verify against
-     * @returns Stored resource metadata
-     * @throws ChecksumMismatchError if expectedChecksum is provided and does not match
-     * @throws Error if file does not exist
-     */
-    register(storageUri: string, expectedChecksum?: string, options?: {
-        noGit?: boolean;
-    }): Promise<StoredResource>;
-    /**
-     * Read file content by URI.
-     *
-     * @param storageUri - file:// URI
-     * @returns Raw bytes
-     */
-    retrieve(storageUri: string): Promise<Buffer>;
-    /**
-     * Move a file from one URI to another.
-     *
-     * If .git/ exists in the project root and noGit is not set, runs `git mv`.
-     * Otherwise (no .git/ or noGit: true), runs fs.rename.
-     *
-     * @param fromUri - Current file:// URI
-     * @param toUri - New file:// URI
-     * @param options.noGit - Skip git mv even if .git/ is present
-     */
-    move(fromUri: string, toUri: string, options?: {
-        noGit?: boolean;
-    }): Promise<void>;
-    /**
-     * Remove a file from the working tree.
-     *
-     * If .git/ exists and noGit is not set:
-     *   - keepFile false (default): runs `git rm` (removes from index and disk)
-     *   - keepFile true: runs `git rm --cached` (removes from index only, file stays on disk)
-     * If no .git/ or noGit: true:
-     *   - keepFile false: runs fs.unlink
-     *   - keepFile true: no-op on filesystem
-     *
-     * @param storageUri - file:// URI
-     * @param options.noGit - Skip git rm even if .git/ is present
-     * @param options.keepFile - Remove from git index only; leave file on disk
-     */
-    remove(storageUri: string, options?: {
-        noGit?: boolean;
-        keepFile?: boolean;
-    }): Promise<void>;
-    /**
-     * Convert a file:// URI to an absolute filesystem path.
-     *
-     * "file://docs/overview.md" → "{projectRoot}/docs/overview.md"
-     *
-     * @param storageUri - file:// URI
-     * @returns Absolute path
-     */
-    resolveUri(storageUri: string): string;
-}
-/**
- * Thrown when a registered file's checksum does not match the expected value.
- * This indicates the file on disk differs from what was recorded (e.g. modified
- * after staging, or wrong file path provided).
- */
-export declare class ChecksumMismatchError extends Error {
-    readonly storageUri: string;
-    readonly expected: string;
-    readonly actual: string;
-    constructor(storageUri: string, expected: string, actual: string);
-}
-//# sourceMappingURL=working-tree-store.d.ts.map

package/dist/working-tree-store.d.ts.map

@@ -1 +0,0 @@
-{"version":3,"file":"working-tree-store.d.ts","sourceRoot":"","sources":["../src/working-tree-store.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;GAoBG;AAKH,OAAO,KAAK,EAAE,cAAc,EAAE,MAAM,oBAAoB,CAAC;AACzD,OAAO,KAAK,EAAE,MAAM,EAAE,MAAM,eAAe,CAAC;AAG5C;;GAEG;AACH,MAAM,WAAW,cAAc;IAC7B,UAAU,EAAE,MAAM,CAAC;IACnB,QAAQ,EAAE,MAAM,CAAC;IACjB,QAAQ,EAAE,MAAM,CAAC;IACjB,OAAO,EAAE,MAAM,CAAC;CACjB;AAED;;GAEG;AACH,qBAAa,gBAAgB;IAC3B,OAAO,CAAC,WAAW,CAAS;IAC5B,OAAO,CAAC,OAAO,CAAU;IACzB,OAAO,CAAC,MAAM,CAAC,CAAS;gBAEZ,OAAO,EAAE,cAAc,EAAE,MAAM,CAAC,EAAE,MAAM;IAMpD,OAAO,CAAC,YAAY;IAIpB;;;;;;;;OAQG;IACG,KAAK,CAAC,OAAO,EAAE,MAAM,EAAE,UAAU,EAAE,MAAM,EAAE,OAAO,CAAC,EAAE;QAAE,KAAK,CAAC,EAAE,OAAO,CAAA;KAAE,GAAG,OAAO,CAAC,cAAc,CAAC;IAuBxG;;;;;;;;;;;OAWG;IACG,QAAQ,CAAC,UAAU,EAAE,MAAM,EAAE,gBAAgB,CAAC,EAAE,MAAM,EAAE,OAAO,CAAC,EAAE;QAAE,KAAK,CAAC,EAAE,OAAO,CAAA;KAAE,GAAG,OAAO,CAAC,cAAc,CAAC;IA0BrH;;;;;OAKG;IACG,QAAQ,CAAC,UAAU,EAAE,MAAM,GAAG,OAAO,CAAC,MAAM,CAAC;IAYnD;;;;;;;;;OASG;IACG,IAAI,CAAC,OAAO,EAAE,MAAM,EAAE,KAAK,EAAE,MAAM,EAAE,OAAO,CAAC,EAAE;QAAE,KAAK,CAAC,EAAE,OAAO,CAAA;KAAE,GAAG,OAAO,CAAC,IAAI,CAAC;IAkBxF;;;;;;;;;;;;;OAaG;IACG,MAAM,CAAC,UAAU,EAAE,MAAM,EAAE,OAAO,CAAC,EAAE;QAAE,KAAK,CAAC,EAAE,OAAO,CAAC;QAAC,QAAQ,CAAC,EAAE,OAAO,CAAA;KAAE,GAAG,OAAO,CAAC,IAAI,CAAC;IAkClG;;;;;;;OAOG;IACH,UAAU,CAAC,UAAU,EAAE,MAAM,GAAG,MAAM;CAOvC;AAED;;;;GAIG;AACH,qBAAa,qBAAsB,SAAQ,KAAK;IAE5C,QAAQ,CAAC,UAAU,EAAE,MAAM;IAC3B,QAAQ,CAAC,QAAQ,EAAE,MAAM;IACzB,QAAQ,CAAC,MAAM,EAAE,MAAM;gBAFd,UAAU,EAAE,MAAM,EAClB,QAAQ,EAAE,MAAM,EAChB,MAAM,EAAE,MAAM;CAQ1B"}