@gridstorm/pdf-plugin-text 0.1.2

package/README.md

@@ -0,0 +1,32 @@
+# @gridstorm/pdf-plugin-text
+
+PDF text extraction and full-text search.
+
+## Install
+
+```bash
+npm install @gridstorm/pdf-plugin-text
+```
+
+## Usage
+
+```typescript
+import { TextPlugin } from '@gridstorm/pdf-plugin-text';
+
+const pdf = createPDFEngine({ plugins: [TextPlugin()] });
+```
+
+## Features
+
+- **Full text extraction**
+- **Search with highlighting**
+- **Page and document-level search**
+- **Regex support**
+
+## Documentation
+
+[Full Documentation](https://grid-data-analytics-explorer.vercel.app/) | [GitHub](https://github.com/007krcs/grid-data)
+
+## License
+
+MIT

package/dist/index.cjs

@@ -0,0 +1,407 @@
+'use strict';
+
+// src/text-extractor.ts
+var DEFAULT_WORD_SPACING = 3;
+var DEFAULT_LINE_SPACING = 5;
+var TextExtractor = class {
+  constructor(config = {}) {
+    this.wordSpacingThreshold = config.wordSpacingThreshold ?? DEFAULT_WORD_SPACING;
+    this.lineSpacingThreshold = config.lineSpacingThreshold ?? DEFAULT_LINE_SPACING;
+  }
+  /** Extract text content from raw text items.
+   *  In Phase 2 this will accept pdf.js TextItem[]; for now it works with PdfCharInfo[]. */
+  extract(chars, _page) {
+    if (chars.length === 0) {
+      return { chars, words: [], lines: [] };
+    }
+    const words = this.segmentWords(chars);
+    const lines = this.segmentLines(words, chars);
+    return { chars, words, lines };
+  }
+  /** Build PdfCharInfo from a simple text string (for testing/placeholder). */
+  buildCharsFromString(text, startX, startY, fontSize, fontName = "Helvetica") {
+    const chars = [];
+    let x = startX;
+    const charWidth = fontSize * 0.6;
+    for (let i = 0; i < text.length; i++) {
+      const char = text[i];
+      const rect = [x, startY, x + charWidth, startY + fontSize];
+      chars.push({
+        char,
+        rect,
+        fontName,
+        fontSize,
+        transform: [fontSize, 0, 0, fontSize, x, startY]
+      });
+      x += charWidth;
+    }
+    return chars;
+  }
+  /** Segment characters into words based on spacing. */
+  segmentWords(chars) {
+    if (chars.length === 0) return [];
+    const words = [];
+    let wordStart = 0;
+    let wordChars = [chars[0]];
+    for (let i = 1; i < chars.length; i++) {
+      const prev = chars[i - 1];
+      const curr = chars[i];
+      const gap = curr.rect[0] - prev.rect[2];
+      const isSpace = curr.char === " " || prev.char === " ";
+      const isNewLine = Math.abs(curr.rect[1] - prev.rect[1]) > this.lineSpacingThreshold;
+      const isWordBreak = gap > this.wordSpacingThreshold || isSpace || isNewLine;
+      if (isWordBreak) {
+        const text = wordChars.map((c) => c.char).join("").trim();
+        if (text.length > 0) {
+          words.push({
+            text,
+            rect: this.boundingRect(wordChars),
+            charIndices: [wordStart, wordStart + wordChars.length - 1]
+          });
+        }
+        if (curr.char !== " ") {
+          wordStart = i;
+          wordChars = [curr];
+        } else {
+          wordStart = i + 1;
+          wordChars = [];
+        }
+      } else {
+        wordChars.push(curr);
+      }
+    }
+    if (wordChars.length > 0) {
+      const text = wordChars.map((c) => c.char).join("").trim();
+      if (text.length > 0) {
+        words.push({
+          text,
+          rect: this.boundingRect(wordChars),
+          charIndices: [wordStart, wordStart + wordChars.length - 1]
+        });
+      }
+    }
+    return words;
+  }
+  /** Segment words into lines based on vertical position. */
+  segmentLines(words, _chars) {
+    if (words.length === 0) return [];
+    const lines = [];
+    let lineStart = 0;
+    let lineWords = [words[0]];
+    let lineY = words[0].rect[1];
+    for (let i = 1; i < words.length; i++) {
+      const word = words[i];
+      const yDiff = Math.abs(word.rect[1] - lineY);
+      if (yDiff > this.lineSpacingThreshold) {
+        lines.push(this.createLine(lineWords, lineStart));
+        lineStart = i;
+        lineWords = [word];
+        lineY = word.rect[1];
+      } else {
+        lineWords.push(word);
+      }
+    }
+    if (lineWords.length > 0) {
+      lines.push(this.createLine(lineWords, lineStart));
+    }
+    return lines;
+  }
+  createLine(words, startIndex) {
+    const text = words.map((w) => w.text).join(" ");
+    const rects = words.map((w) => w.rect);
+    const rect = [
+      Math.min(...rects.map((r) => r[0])),
+      Math.min(...rects.map((r) => r[1])),
+      Math.max(...rects.map((r) => r[2])),
+      Math.max(...rects.map((r) => r[3]))
+    ];
+    return {
+      text,
+      rect,
+      wordIndices: [startIndex, startIndex + words.length - 1]
+    };
+  }
+  /** Compute bounding rect for a set of characters. */
+  boundingRect(chars) {
+    let minX = Infinity, minY = Infinity, maxX = -Infinity, maxY = -Infinity;
+    for (const c of chars) {
+      if (c.rect[0] < minX) minX = c.rect[0];
+      if (c.rect[1] < minY) minY = c.rect[1];
+      if (c.rect[2] > maxX) maxX = c.rect[2];
+      if (c.rect[3] > maxY) maxY = c.rect[3];
+    }
+    return [minX, minY, maxX, maxY];
+  }
+};
+
+// src/search-engine.ts
+var SearchEngine = class {
+  constructor() {
+    this.textContents = /* @__PURE__ */ new Map();
+    this.lastResult = {
+      matches: [],
+      totalCount: 0,
+      activeIndex: -1
+    };
+  }
+  /** Set the text content for a page. */
+  setPageTextContent(pageIndex, textContent) {
+    this.textContents.set(pageIndex, textContent);
+  }
+  /** Clear text content for a page. */
+  clearPageTextContent(pageIndex) {
+    this.textContents.delete(pageIndex);
+  }
+  /** Clear all cached text content. */
+  clearAll() {
+    this.textContents.clear();
+    this.lastResult = { matches: [], totalCount: 0, activeIndex: -1 };
+  }
+  /** Search for a query across all loaded pages. */
+  search(query, options = {}) {
+    if (!query) {
+      this.lastResult = { matches: [], totalCount: 0, activeIndex: -1 };
+      return this.lastResult;
+    }
+    const matches = [];
+    const sortedPages = [...this.textContents.entries()].sort(
+      ([a], [b]) => a - b
+    );
+    for (const [pageIndex, textContent] of sortedPages) {
+      const pageMatches = this.searchPage(
+        pageIndex,
+        textContent,
+        query,
+        options
+      );
+      matches.push(...pageMatches);
+    }
+    this.lastResult = {
+      matches,
+      totalCount: matches.length,
+      activeIndex: matches.length > 0 ? 0 : -1
+    };
+    return this.lastResult;
+  }
+  /** Navigate to the next search match. */
+  nextMatch() {
+    if (this.lastResult.totalCount === 0) return null;
+    this.lastResult.activeIndex = (this.lastResult.activeIndex + 1) % this.lastResult.totalCount;
+    return this.lastResult.matches[this.lastResult.activeIndex] ?? null;
+  }
+  /** Navigate to the previous search match. */
+  prevMatch() {
+    if (this.lastResult.totalCount === 0) return null;
+    this.lastResult.activeIndex = (this.lastResult.activeIndex - 1 + this.lastResult.totalCount) % this.lastResult.totalCount;
+    return this.lastResult.matches[this.lastResult.activeIndex] ?? null;
+  }
+  /** Get the current active match. */
+  getActiveMatch() {
+    if (this.lastResult.activeIndex < 0 || this.lastResult.activeIndex >= this.lastResult.totalCount) {
+      return null;
+    }
+    return this.lastResult.matches[this.lastResult.activeIndex] ?? null;
+  }
+  /** Get the last search result. */
+  getLastResult() {
+    return this.lastResult;
+  }
+  searchPage(pageIndex, textContent, query, options) {
+    const matches = [];
+    const fullText = textContent.lines.map((l) => l.text).join("\n");
+    let pattern;
+    try {
+      if (options.regex) {
+        const flags = options.caseSensitive ? "g" : "gi";
+        pattern = new RegExp(query, flags);
+      } else {
+        const escaped = query.replace(/[.*+?^${}()|[\]\\]/g, "\\$&");
+        const flags = options.caseSensitive ? "g" : "gi";
+        const word = options.wholeWord ? `\\b${escaped}\\b` : escaped;
+        pattern = new RegExp(word, flags);
+      }
+    } catch {
+      return matches;
+    }
+    let match;
+    while ((match = pattern.exec(fullText)) !== null) {
+      const matchText = match[0];
+      const matchStart = match.index;
+      const matchEnd = matchStart + matchText.length;
+      const rects = this.findMatchRects(textContent, matchStart, matchEnd);
+      matches.push({
+        pageIndex,
+        charStart: matchStart,
+        charEnd: matchEnd,
+        rects,
+        text: matchText
+      });
+      if (matchText.length === 0) {
+        pattern.lastIndex++;
+      }
+    }
+    return matches;
+  }
+  /** Find bounding rects for a text range. */
+  findMatchRects(textContent, _start, _end) {
+    const rects = [];
+    let charOffset = 0;
+    for (const line of textContent.lines) {
+      const lineEnd = charOffset + line.text.length;
+      if (lineEnd > _start && charOffset < _end) {
+        for (let wi = line.wordIndices[0]; wi <= line.wordIndices[1]; wi++) {
+          const word = textContent.words[wi];
+          if (word) {
+            rects.push(word.rect);
+          }
+        }
+      }
+      charOffset = lineEnd + 1;
+    }
+    return rects.length > 0 ? rects : [[0, 0, 0, 0]];
+  }
+};
+
+// src/text-plugin.ts
+var INITIAL_STATE = {
+  searchQuery: "",
+  searchOptions: {},
+  searchResult: null,
+  activeMatch: null
+};
+function createTextPlugin() {
+  return {
+    id: "text",
+    name: "Text Extraction & Search",
+    version: "0.1.0",
+    install(context) {
+      const extractor = new TextExtractor();
+      const searchEngine = new SearchEngine();
+      context.registerState("text", { ...INITIAL_STATE });
+      const unsubExtract = context.commandBus.registerHandler(
+        "text:extract",
+        (payload) => {
+          const state = context.store.getState();
+          const page = state.pages[payload.pageIndex];
+          if (!page) return;
+          const chars = extractor.buildCharsFromString(
+            `Sample text for page ${payload.pageIndex + 1}`,
+            72,
+            // 1 inch margin
+            72,
+            12
+          );
+          const textContent = extractor.extract(chars, page);
+          context.store.setState((prev) => {
+            const pages = [...prev.pages];
+            const existing = pages[payload.pageIndex];
+            if (existing) {
+              pages[payload.pageIndex] = { ...existing, textContent };
+            }
+            return { ...prev, pages };
+          });
+          searchEngine.setPageTextContent(payload.pageIndex, textContent);
+          context.eventBus.emit("text:extracted", {
+            pageIndex: payload.pageIndex,
+            textContent
+          });
+        }
+      );
+      const unsubSearch = context.commandBus.registerHandler(
+        "text:search",
+        (payload) => {
+          const result = searchEngine.search(payload.query, payload.options);
+          context.setState("text", (prev) => ({
+            ...prev,
+            searchQuery: payload.query,
+            searchOptions: payload.options ?? {},
+            searchResult: result,
+            activeMatch: searchEngine.getActiveMatch()
+          }));
+          context.eventBus.emit("search:found", {
+            query: payload.query,
+            matches: result.matches,
+            total: result.totalCount
+          });
+        }
+      );
+      const unsubSearchNext = context.commandBus.registerHandler(
+        "text:searchNext",
+        () => {
+          const match = searchEngine.nextMatch();
+          context.setState("text", (prev) => ({
+            ...prev,
+            activeMatch: match,
+            searchResult: searchEngine.getLastResult()
+          }));
+          if (match) {
+            context.api.goToPage(match.pageIndex);
+          }
+        }
+      );
+      const unsubSearchPrev = context.commandBus.registerHandler(
+        "text:searchPrev",
+        () => {
+          const match = searchEngine.prevMatch();
+          context.setState("text", (prev) => ({
+            ...prev,
+            activeMatch: match,
+            searchResult: searchEngine.getLastResult()
+          }));
+          if (match) {
+            context.api.goToPage(match.pageIndex);
+          }
+        }
+      );
+      const pluginApi = {
+        extractPageText(pageIndex) {
+          context.commandBus.dispatch("text:extract", { pageIndex });
+        },
+        extractAllText() {
+          const state = context.store.getState();
+          for (let i = 0; i < state.pages.length; i++) {
+            context.commandBus.dispatch("text:extract", { pageIndex: i });
+          }
+        },
+        search(query, options) {
+          context.commandBus.dispatch("text:search", { query, options });
+          return searchEngine.getLastResult();
+        },
+        searchNext() {
+          context.commandBus.dispatch("text:searchNext", {});
+          return searchEngine.getActiveMatch();
+        },
+        searchPrev() {
+          context.commandBus.dispatch("text:searchPrev", {});
+          return searchEngine.getActiveMatch();
+        },
+        clearSearch() {
+          searchEngine.clearAll();
+          context.setState("text", () => ({
+            ...INITIAL_STATE
+          }));
+        },
+        getExtractor() {
+          return extractor;
+        },
+        getSearchEngine() {
+          return searchEngine;
+        }
+      };
+      context._pluginApi = pluginApi;
+      return () => {
+        unsubExtract();
+        unsubSearch();
+        unsubSearchNext();
+        unsubSearchPrev();
+        searchEngine.clearAll();
+      };
+    }
+  };
+}
+
+exports.SearchEngine = SearchEngine;
+exports.TextExtractor = TextExtractor;
+exports.createTextPlugin = createTextPlugin;
+//# sourceMappingURL=index.cjs.map
+//# sourceMappingURL=index.cjs.map

package/dist/index.cjs.map

@@ -0,0 +1 @@
+{"version":3,"sources":["../src/text-extractor.ts","../src/search-engine.ts","../src/text-plugin.ts"],"names":[],"mappings":";;;AAsBA,IAAM,oBAAA,GAAuB,CAAA;AAC7B,IAAM,oBAAA,GAAuB,CAAA;AAGtB,IAAM,gBAAN,MAAoB;AAAA,EAIzB,WAAA,CAAY,MAAA,GAA8B,EAAC,EAAG;AAC5C,IAAA,IAAA,CAAK,oBAAA,GAAuB,OAAO,oBAAA,IAAwB,oBAAA;AAC3D,IAAA,IAAA,CAAK,oBAAA,GAAuB,OAAO,oBAAA,IAAwB,oBAAA;AAAA,EAC7D;AAAA;AAAA;AAAA,EAIA,OAAA,CAAQ,OAAsB,KAAA,EAAqC;AACjE,IAAA,IAAI,KAAA,CAAM,WAAW,CAAA,EAAG;AACtB,MAAA,OAAO,EAAE,KAAA,EAAO,KAAA,EAAO,EAAC,EAAG,KAAA,EAAO,EAAC,EAAE;AAAA,IACvC;AAEA,IAAA,MAAM,KAAA,GAAQ,IAAA,CAAK,YAAA,CAAa,KAAK,CAAA;AACrC,IAAA,MAAM,KAAA,GAAQ,IAAA,CAAK,YAAA,CAAa,KAAA,EAAO,KAAK,CAAA;AAE5C,IAAA,OAAO,EAAE,KAAA,EAAO,KAAA,EAAO,KAAA,EAAM;AAAA,EAC/B;AAAA;AAAA,EAGA,qBACE,IAAA,EACA,MAAA,EACA,MAAA,EACA,QAAA,EACA,WAAW,WAAA,EACI;AACf,IAAA,MAAM,QAAuB,EAAC;AAC9B,IAAA,IAAI,CAAA,GAAI,MAAA;AACR,IAAA,MAAM,YAAY,QAAA,GAAW,GAAA;AAE7B,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,IAAA,CAAK,QAAQ,CAAA,EAAA,EAAK;AACpC,MAAA,MAAM,IAAA,GAAO,KAAK,CAAC,CAAA;AACnB,MAAA,MAAM,OAAgB,CAAC,CAAA,EAAG,QAAQ,CAAA,GAAI,SAAA,EAAW,SAAS,QAAQ,CAAA;AAElE,MAAA,KAAA,CAAM,IAAA,CAAK;AAAA,QACT,IAAA;AAAA,QACA,IAAA;AAAA,QACA,QAAA;AAAA,QACA,QAAA;AAAA,QACA,WAAW,CAAC,QAAA,EAAU,GAAG,CAAA,EAAG,QAAA,EAAU,GAAG,MAAM;AAAA,OAChD,CAAA;AAED,MAAA,CAAA,IAAK,SAAA;AAAA,IACP;AAEA,IAAA,OAAO,KAAA;AAAA,EACT;AAAA;AAAA,EAGQ,aAAa,KAAA,EAAqC;AACxD,IAAA,IAAI,KAAA,CAAM,MAAA,KAAW,CAAA,EAAG,OAAO,EAAC;AAEhC,IAAA,MAAM,QAAuB,EAAC;AAC9B,IAAA,IAAI,SAAA,GAAY,CAAA;AAChB,IAAA,IAAI,SAAA,GAA2B,CAAC,KAAA,CAAM,CAAC,CAAE,CAAA;AAEzC,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,KAAA,CAAM,QAAQ,CAAA,EAAA,EAAK;AACrC,MAAA,MAAM,IAAA,GAAO,KAAA,CAAM,CAAA,GAAI,CAAC,CAAA;AACxB,MAAA,MAAM,IAAA,GAAO,MAAM,CAAC,CAAA;AAEpB,MAAA,MAAM,MAAM,IAAA,CAAK,IAAA,CAAK,CAAC,CAAA,GAAI,IAAA,CAAK,KAAK,CAAC,CAAA;AACtC,MAAA,MAAM,OAAA,GAAU,IAAA,CAAK,IAAA,KAAS,GAAA,IAAO,KAAK,IAAA,KAAS,GAAA;AACnD,MAAA,MAAM,SAAA,GAAY,IAAA,CAAK,GAAA,CAAI,IAAA,CAAK,IAAA,CAAK,CAAC,CAAA,GAAI,IAAA,CAAK,IAAA,CAAK,CAAC,CAAC,CAAA,GAAI,IAAA,CAAK,oBAAA;AAC/D,MAAA,MAAM,WAAA,GAAc,GAAA,GAAM,IAAA,CAAK,oBAAA,IAAwB,OAAA,IAAW,SAAA;AAElE,MAAA,IAAI,WAAA,EAAa;AAEf,QAAA,MAAM,IAAA,GAAO,SAAA,CAAU,GAAA,CAAI,CAAC,CAAA,KAAM,CAAA,CAAE,IAAI,CAAA,CAAE,IAAA,CAAK,EAAE,CAAA,CAAE,IAAA,EAAK;AACxD,QAAA,IAAI,IAAA,CAAK,SAAS,CAAA,EAAG;AACnB,UAAA,KAAA,CAAM,IAAA,CAAK;AAAA,YACT,IAAA;AAAA,YACA,IAAA,EAAM,IAAA,CAAK,YAAA,CAAa,SAAS,CAAA;AAAA,YACjC,aAAa,CAAC,SAAA,EAAW,SAAA,GAAY,SAAA,CAAU,SAAS,CAAC;AAAA,WAC1D,CAAA;AAAA,QACH;AAGA,QAAA,IAAI,IAAA,CAAK,SAAS,GAAA,EAAK;AACrB,UAAA,SAAA,GAAY,CAAA;AACZ,UAAA,SAAA,GAAY,CAAC,IAAI,CAAA;AAAA,QACnB,CAAA,MAAO;AACL,UAAA,SAAA,GAAY,CAAA,GAAI,CAAA;AAChB,UAAA,SAAA,GAAY,EAAC;AAAA,QACf;AAAA,MACF,CAAA,MAAO;AACL,QAAA,SAAA,CAAU,KAAK,IAAI,CAAA;AAAA,MACrB;AAAA,IACF;AAGA,IAAA,IAAI,SAAA,CAAU,SAAS,CAAA,EAAG;AACxB,MAAA,MAAM,IAAA,GAAO,SAAA,CAAU,GAAA,CAAI,CAAC,CAAA,KAAM,CAAA,CAAE,IAAI,CAAA,CAAE,IAAA,CAAK,EAAE,CAAA,CAAE,IAAA,EAAK;AACxD,MAAA,IAAI,IAAA,CAAK,SAAS,CAAA,EAAG;AACnB,QAAA,KAAA,CAAM,IAAA,CAAK;AAAA,UACT,IAAA;AAAA,UACA,IAAA,EAAM,IAAA,CAAK,YAAA,CAAa,SAAS,CAAA;AAAA,UACjC,aAAa,CAAC,SAAA,EAAW,SAAA,GAAY,SAAA,CAAU,SAAS,CAAC;AAAA,SAC1D,CAAA;AAAA,MACH;AAAA,IACF;AAEA,IAAA,OAAO,KAAA;AAAA,EACT;AAAA;AAAA,EAGQ,YAAA,CAAa,OAAsB,MAAA,EAAsC;AAC/E,IAAA,IAAI,KAAA,CAAM,MAAA,KAAW,CAAA,EAAG,OAAO,EAAC;AAEhC,IAAA,MAAM,QAAuB,EAAC;AAC9B,IAAA,IAAI,SAAA,GAAY,CAAA;AAChB,IAAA,IAAI,SAAA,GAA2B,CAAC,KAAA,CAAM,CAAC,CAAE,CAAA;AACzC,IAAA,IAAI,KAAA,GAAQ,KAAA,CAAM,CAAC,CAAA,CAAG,KAAK,CAAC,CAAA;AAE5B,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,KAAA,CAAM,QAAQ,CAAA,EAAA,EAAK;AACrC,MAAA,MAAM,IAAA,GAAO,MAAM,CAAC,CAAA;AACpB,MAAA,MAAM,QAAQ,IAAA,CAAK,GAAA,CAAI,KAAK,IAAA,CAAK,CAAC,IAAI,KAAK,CAAA;AAE3C,MAAA,IAAI,KAAA,GAAQ,KAAK,oBAAA,EAAsB;AAErC,QAAA,KAAA,CAAM,IAAA,CAAK,IAAA,CAAK,UAAA,CAAW,SAAA,EAAW,SAAS,CAAC,CAAA;AAChD,QAAA,SAAA,GAAY,CAAA;AACZ,QAAA,SAAA,GAAY,CAAC,IAAI,CAAA;AACjB,QAAA,KAAA,GAAQ,IAAA,CAAK,KAAK,CAAC,CAAA;AAAA,MACrB,CAAA,MAAO;AACL,QAAA,SAAA,CAAU,KAAK,IAAI,CAAA;AAAA,MACrB;AAAA,IACF;AAGA,IAAA,IAAI,SAAA,CAAU,SAAS,CAAA,EAAG;AACxB,MAAA,KAAA,CAAM,IAAA,CAAK,IAAA,CAAK,UAAA,CAAW,SAAA,EAAW,SAAS,CAAC,CAAA;AAAA,IAClD;AAEA,IAAA,OAAO,KAAA;AAAA,EACT;AAAA,EAEQ,UAAA,CAAW,OAAsB,UAAA,EAAiC;AACxE,IAAA,MAAM,IAAA,GAAO,MAAM,GAAA,CAAI,CAAC,MAAM,CAAA,CAAE,IAAI,CAAA,CAAE,IAAA,CAAK,GAAG,CAAA;AAC9C,IAAA,MAAM,QAAQ,KAAA,CAAM,GAAA,CAAI,CAAC,CAAA,KAAM,EAAE,IAAI,CAAA;AACrC,IAAA,MAAM,IAAA,GAAgB;AAAA,MACpB,IAAA,CAAK,GAAA,CAAI,GAAG,KAAA,CAAM,GAAA,CAAI,CAAC,CAAA,KAAM,CAAA,CAAE,CAAC,CAAC,CAAC,CAAA;AAAA,MAClC,IAAA,CAAK,GAAA,CAAI,GAAG,KAAA,CAAM,GAAA,CAAI,CAAC,CAAA,KAAM,CAAA,CAAE,CAAC,CAAC,CAAC,CAAA;AAAA,MAClC,IAAA,CAAK,GAAA,CAAI,GAAG,KAAA,CAAM,GAAA,CAAI,CAAC,CAAA,KAAM,CAAA,CAAE,CAAC,CAAC,CAAC,CAAA;AAAA,MAClC,IAAA,CAAK,GAAA,CAAI,GAAG,KAAA,CAAM,GAAA,CAAI,CAAC,CAAA,KAAM,CAAA,CAAE,CAAC,CAAC,CAAC;AAAA,KACpC;AACA,IAAA,OAAO;AAAA,MACL,IAAA;AAAA,MACA,IAAA;AAAA,MACA,aAAa,CAAC,UAAA,EAAY,UAAA,GAAa,KAAA,CAAM,SAAS,CAAC;AAAA,KACzD;AAAA,EACF;AAAA;AAAA,EAGQ,aAAa,KAAA,EAA+B;AAClD,IAAA,IAAI,OAAO,QAAA,EACT,IAAA,GAAO,QAAA,EACP,IAAA,GAAO,WACP,IAAA,GAAO,CAAA,QAAA;AAET,IAAA,KAAA,MAAW,KAAK,KAAA,EAAO;AACrB,MAAA,IAAI,CAAA,CAAE,KAAK,CAAC,CAAA,GAAI,MAAM,IAAA,GAAO,CAAA,CAAE,KAAK,CAAC,CAAA;AACrC,MAAA,IAAI,CAAA,CAAE,KAAK,CAAC,CAAA,GAAI,MAAM,IAAA,GAAO,CAAA,CAAE,KAAK,CAAC,CAAA;AACrC,MAAA,IAAI,CAAA,CAAE,KAAK,CAAC,CAAA,GAAI,MAAM,IAAA,GAAO,CAAA,CAAE,KAAK,CAAC,CAAA;AACrC,MAAA,IAAI,CAAA,CAAE,KAAK,CAAC,CAAA,GAAI,MAAM,IAAA,GAAO,CAAA,CAAE,KAAK,CAAC,CAAA;AAAA,IACvC;AAEA,IAAA,OAAO,CAAC,IAAA,EAAM,IAAA,EAAM,IAAA,EAAM,IAAI,CAAA;AAAA,EAChC;AACF;;;AC7JO,IAAM,eAAN,MAAmB;AAAA,EAAnB,WAAA,GAAA;AACL,IAAA,IAAA,CAAQ,YAAA,uBAAmB,GAAA,EAA4B;AACvD,IAAA,IAAA,CAAQ,UAAA,GAA2B;AAAA,MACjC,SAAS,EAAC;AAAA,MACV,UAAA,EAAY,CAAA;AAAA,MACZ,WAAA,EAAa;AAAA,KACf;AAAA,EAAA;AAAA;AAAA,EAGA,kBAAA,CAAmB,WAAmB,WAAA,EAAmC;AACvE,IAAA,IAAA,CAAK,YAAA,CAAa,GAAA,CAAI,SAAA,EAAW,WAAW,CAAA;AAAA,EAC9C;AAAA;AAAA,EAGA,qBAAqB,SAAA,EAAyB;AAC5C,IAAA,IAAA,CAAK,YAAA,CAAa,OAAO,SAAS,CAAA;AAAA,EACpC;AAAA;AAAA,EAGA,QAAA,GAAiB;AACf,IAAA,IAAA,CAAK,aAAa,KAAA,EAAM;AACxB,IAAA,IAAA,CAAK,UAAA,GAAa,EAAE,OAAA,EAAS,IAAI,UAAA,EAAY,CAAA,EAAG,aAAa,EAAA,EAAG;AAAA,EAClE;AAAA;AAAA,EAGA,MAAA,CAAO,KAAA,EAAe,OAAA,GAAyB,EAAC,EAAiB;AAC/D,IAAA,IAAI,CAAC,KAAA,EAAO;AACV,MAAA,IAAA,CAAK,UAAA,GAAa,EAAE,OAAA,EAAS,IAAI,UAAA,EAAY,CAAA,EAAG,aAAa,EAAA,EAAG;AAChE,MAAA,OAAO,IAAA,CAAK,UAAA;AAAA,IACd;AAEA,IAAA,MAAM,UAAyB,EAAC;AAGhC,IAAA,MAAM,cAAc,CAAC,GAAG,KAAK,YAAA,CAAa,OAAA,EAAS,CAAA,CAAE,IAAA;AAAA,MACnD,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,MAAM,CAAA,GAAI;AAAA,KACpB;AAEA,IAAA,KAAA,MAAW,CAAC,SAAA,EAAW,WAAW,CAAA,IAAK,WAAA,EAAa;AAClD,MAAA,MAAM,cAAc,IAAA,CAAK,UAAA;AAAA,QACvB,SAAA;AAAA,QACA,WAAA;AAAA,QACA,KAAA;AAAA,QACA;AAAA,OACF;AACA,MAAA,OAAA,CAAQ,IAAA,CAAK,GAAG,WAAW,CAAA;AAAA,IAC7B;AAEA,IAAA,IAAA,CAAK,UAAA,GAAa;AAAA,MAChB,OAAA;AAAA,MACA,YAAY,OAAA,CAAQ,MAAA;AAAA,MACpB,WAAA,EAAa,OAAA,CAAQ,MAAA,GAAS,CAAA,GAAI,CAAA,GAAI;AAAA,KACxC;AAEA,IAAA,OAAO,IAAA,CAAK,UAAA;AAAA,EACd;AAAA;AAAA,EAGA,SAAA,GAAgC;AAC9B,IAAA,IAAI,IAAA,CAAK,UAAA,CAAW,UAAA,KAAe,CAAA,EAAG,OAAO,IAAA;AAE7C,IAAA,IAAA,CAAK,WAAW,WAAA,GAAA,CACb,IAAA,CAAK,WAAW,WAAA,GAAc,CAAA,IAAK,KAAK,UAAA,CAAW,UAAA;AACtD,IAAA,OAAO,KAAK,UAAA,CAAW,OAAA,CAAQ,IAAA,CAAK,UAAA,CAAW,WAAW,CAAA,IAAK,IAAA;AAAA,EACjE;AAAA;AAAA,EAGA,SAAA,GAAgC;AAC9B,IAAA,IAAI,IAAA,CAAK,UAAA,CAAW,UAAA,KAAe,CAAA,EAAG,OAAO,IAAA;AAE7C,IAAA,IAAA,CAAK,UAAA,CAAW,WAAA,GAAA,CACb,IAAA,CAAK,UAAA,CAAW,WAAA,GAAc,IAAI,IAAA,CAAK,UAAA,CAAW,UAAA,IACnD,IAAA,CAAK,UAAA,CAAW,UAAA;AAClB,IAAA,OAAO,KAAK,UAAA,CAAW,OAAA,CAAQ,IAAA,CAAK,UAAA,CAAW,WAAW,CAAA,IAAK,IAAA;AAAA,EACjE;AAAA;AAAA,EAGA,cAAA,GAAqC;AACnC,IAAA,IACE,IAAA,CAAK,WAAW,WAAA,GAAc,CAAA,IAC9B,KAAK,UAAA,CAAW,WAAA,IAAe,IAAA,CAAK,UAAA,CAAW,UAAA,EAC/C;AACA,MAAA,OAAO,IAAA;AAAA,IACT;AACA,IAAA,OAAO,KAAK,UAAA,CAAW,OAAA,CAAQ,IAAA,CAAK,UAAA,CAAW,WAAW,CAAA,IAAK,IAAA;AAAA,EACjE;AAAA;AAAA,EAGA,aAAA,GAA8B;AAC5B,IAAA,OAAO,IAAA,CAAK,UAAA;AAAA,EACd;AAAA,EAEQ,UAAA,CACN,SAAA,EACA,WAAA,EACA,KAAA,EACA,OAAA,EACe;AACf,IAAA,MAAM,UAAyB,EAAC;AAGhC,IAAA,MAAM,QAAA,GAAW,WAAA,CAAY,KAAA,CAAM,GAAA,CAAI,CAAC,MAAM,CAAA,CAAE,IAAI,CAAA,CAAE,IAAA,CAAK,IAAI,CAAA;AAE/D,IAAA,IAAI,OAAA;AACJ,IAAA,IAAI;AACF,MAAA,IAAI,QAAQ,KAAA,EAAO;AACjB,QAAA,MAAM,KAAA,GAAQ,OAAA,CAAQ,aAAA,GAAgB,GAAA,GAAM,IAAA;AAC5C,QAAA,OAAA,GAAU,IAAI,MAAA,CAAO,KAAA,EAAO,KAAK,CAAA;AAAA,MACnC,CAAA,MAAO;AACL,QAAA,MAAM,OAAA,GAAU,KAAA,CAAM,OAAA,CAAQ,qBAAA,EAAuB,MAAM,CAAA;AAC3D,QAAA,MAAM,KAAA,GAAQ,OAAA,CAAQ,aAAA,GAAgB,GAAA,GAAM,IAAA;AAC5C,QAAA,MAAM,IAAA,GAAO,OAAA,CAAQ,SAAA,GAAY,CAAA,GAAA,EAAM,OAAO,CAAA,GAAA,CAAA,GAAQ,OAAA;AACtD,QAAA,OAAA,GAAU,IAAI,MAAA,CAAO,IAAA,EAAM,KAAK,CAAA;AAAA,MAClC;AAAA,IACF,CAAA,CAAA,MAAQ;AACN,MAAA,OAAO,OAAA;AAAA,IACT;AAEA,IAAA,IAAI,KAAA;AACJ,IAAA,OAAA,CAAQ,KAAA,GAAQ,OAAA,CAAQ,IAAA,CAAK,QAAQ,OAAO,IAAA,EAAM;AAChD,MAAA,MAAM,SAAA,GAAY,MAAM,CAAC,CAAA;AACzB,MAAA,MAAM,aAAa,KAAA,CAAM,KAAA;AACzB,MAAA,MAAM,QAAA,GAAW,aAAa,SAAA,CAAU,MAAA;AAGxC,MAAA,MAAM,KAAA,GAAQ,IAAA,CAAK,cAAA,CAAe,WAAA,EAAa,YAAY,QAAQ,CAAA;AAEnE,MAAA,OAAA,CAAQ,IAAA,CAAK;AAAA,QACX,SAAA;AAAA,QACA,SAAA,EAAW,UAAA;AAAA,QACX,OAAA,EAAS,QAAA;AAAA,QACT,KAAA;AAAA,QACA,IAAA,EAAM;AAAA,OACP,CAAA;AAGD,MAAA,IAAI,SAAA,CAAU,WAAW,CAAA,EAAG;AAC1B,QAAA,OAAA,CAAQ,SAAA,EAAA;AAAA,MACV;AAAA,IACF;AAEA,IAAA,OAAO,OAAA;AAAA,EACT;AAAA;AAAA,EAGQ,cAAA,CACN,WAAA,EACA,MAAA,EACA,IAAA,EACW;AAGX,IAAA,MAAM,QAAmB,EAAC;AAG1B,IAAA,IAAI,UAAA,GAAa,CAAA;AACjB,IAAA,KAAA,MAAW,IAAA,IAAQ,YAAY,KAAA,EAAO;AACpC,MAAA,MAAM,OAAA,GAAU,UAAA,GAAa,IAAA,CAAK,IAAA,CAAK,MAAA;AAEvC,MAAA,IAAI,OAAA,GAAU,MAAA,IAAU,UAAA,GAAa,IAAA,EAAM;AAEzC,QAAA,KAAA,IAAS,EAAA,GAAK,IAAA,CAAK,WAAA,CAAY,CAAC,CAAA,EAAG,MAAM,IAAA,CAAK,WAAA,CAAY,CAAC,CAAA,EAAG,EAAA,EAAA,EAAM;AAClE,UAAA,MAAM,IAAA,GAAO,WAAA,CAAY,KAAA,CAAM,EAAE,CAAA;AACjC,UAAA,IAAI,IAAA,EAAM;AACR,YAAA,KAAA,CAAM,IAAA,CAAK,KAAK,IAAI,CAAA;AAAA,UACtB;AAAA,QACF;AAAA,MACF;AAEA,MAAA,UAAA,GAAa,OAAA,GAAU,CAAA;AAAA,IACzB;AAEA,IAAA,OAAO,KAAA,CAAM,MAAA,GAAS,CAAA,GAAI,KAAA,GAAQ,CAAC,CAAC,CAAA,EAAG,CAAA,EAAG,CAAA,EAAG,CAAC,CAAC,CAAA;AAAA,EACjD;AACF;;;AC1KA,IAAM,aAAA,GAAiC;AAAA,EACrC,WAAA,EAAa,EAAA;AAAA,EACb,eAAe,EAAC;AAAA,EAChB,YAAA,EAAc,IAAA;AAAA,EACd,WAAA,EAAa;AACf,CAAA;AAGO,SAAS,gBAAA,GAA8B;AAC5C,EAAA,OAAO;AAAA,IACL,EAAA,EAAI,MAAA;AAAA,IACJ,IAAA,EAAM,0BAAA;AAAA,IACN,OAAA,EAAS,OAAA;AAAA,IAET,QAAQ,OAAA,EAA8C;AACpD,MAAA,MAAM,SAAA,GAAY,IAAI,aAAA,EAAc;AACpC,MAAA,MAAM,YAAA,GAAe,IAAI,YAAA,EAAa;AAGtC,MAAA,OAAA,CAAQ,aAAA,CAA+B,MAAA,EAAQ,EAAE,GAAG,eAAe,CAAA;AAInE,MAAA,MAAM,YAAA,GAAe,QAAQ,UAAA,CAAW,eAAA;AAAA,QACtC,cAAA;AAAA,QACA,CAAC,OAAA,KAAmC;AAClC,UAAA,MAAM,KAAA,GAAQ,OAAA,CAAQ,KAAA,CAAM,QAAA,EAAS;AACrC,UAAA,MAAM,IAAA,GAAO,KAAA,CAAM,KAAA,CAAM,OAAA,CAAQ,SAAS,CAAA;AAC1C,UAAA,IAAI,CAAC,IAAA,EAAM;AAGX,UAAA,MAAM,QAAQ,SAAA,CAAU,oBAAA;AAAA,YACtB,CAAA,qBAAA,EAAwB,OAAA,CAAQ,SAAA,GAAY,CAAC,CAAA,CAAA;AAAA,YAC7C,EAAA;AAAA;AAAA,YACA,EAAA;AAAA,YACA;AAAA,WACF;AAEA,UAAA,MAAM,WAAA,GAAc,SAAA,CAAU,OAAA,CAAQ,KAAA,EAAO,IAAI,CAAA;AAGjD,UAAA,OAAA,CAAQ,KAAA,CAAM,QAAA,CAAS,CAAC,IAAA,KAAS;AAC/B,YAAA,MAAM,KAAA,GAAQ,CAAC,GAAG,IAAA,CAAK,KAAK,CAAA;AAC5B,YAAA,MAAM,QAAA,GAAW,KAAA,CAAM,OAAA,CAAQ,SAAS,CAAA;AACxC,YAAA,IAAI,QAAA,EAAU;AACZ,cAAA,KAAA,CAAM,QAAQ,SAAS,CAAA,GAAI,EAAE,GAAG,UAAU,WAAA,EAAY;AAAA,YACxD;AACA,YAAA,OAAO,EAAE,GAAG,IAAA,EAAM,KAAA,EAAM;AAAA,UAC1B,CAAC,CAAA;AAGD,UAAA,YAAA,CAAa,kBAAA,CAAmB,OAAA,CAAQ,SAAA,EAAW,WAAW,CAAA;AAE9D,UAAA,OAAA,CAAQ,QAAA,CAAS,KAAK,gBAAA,EAAkB;AAAA,YACtC,WAAW,OAAA,CAAQ,SAAA;AAAA,YACnB;AAAA,WACD,CAAA;AAAA,QACH;AAAA,OACF;AAEA,MAAA,MAAM,WAAA,GAAc,QAAQ,UAAA,CAAW,eAAA;AAAA,QACrC,aAAA;AAAA,QACA,CAAC,OAAA,KAAwD;AACvD,UAAA,MAAM,SAAS,YAAA,CAAa,MAAA,CAAO,OAAA,CAAQ,KAAA,EAAO,QAAQ,OAAO,CAAA;AAEjE,UAAA,OAAA,CAAQ,QAAA,CAA0B,MAAA,EAAQ,CAAC,IAAA,MAAU;AAAA,YACnD,GAAG,IAAA;AAAA,YACH,aAAa,OAAA,CAAQ,KAAA;AAAA,YACrB,aAAA,EAAe,OAAA,CAAQ,OAAA,IAAW,EAAC;AAAA,YACnC,YAAA,EAAc,MAAA;AAAA,YACd,WAAA,EAAa,aAAa,cAAA;AAAe,WAC3C,CAAE,CAAA;AAEF,UAAA,OAAA,CAAQ,QAAA,CAAS,KAAK,cAAA,EAAgB;AAAA,YACpC,OAAO,OAAA,CAAQ,KAAA;AAAA,YACf,SAAS,MAAA,CAAO,OAAA;AAAA,YAChB,OAAO,MAAA,CAAO;AAAA,WACf,CAAA;AAAA,QACH;AAAA,OACF;AAEA,MAAA,MAAM,eAAA,GAAkB,QAAQ,UAAA,CAAW,eAAA;AAAA,QACzC,iBAAA;AAAA,QACA,MAAM;AACJ,UAAA,MAAM,KAAA,GAAQ,aAAa,SAAA,EAAU;AACrC,UAAA,OAAA,CAAQ,QAAA,CAA0B,MAAA,EAAQ,CAAC,IAAA,MAAU;AAAA,YACnD,GAAG,IAAA;AAAA,YACH,WAAA,EAAa,KAAA;AAAA,YACb,YAAA,EAAc,aAAa,aAAA;AAAc,WAC3C,CAAE,CAAA;AAEF,UAAA,IAAI,KAAA,EAAO;AAET,YAAA,OAAA,CAAQ,GAAA,CAAI,QAAA,CAAS,KAAA,CAAM,SAAS,CAAA;AAAA,UACtC;AAAA,QACF;AAAA,OACF;AAEA,MAAA,MAAM,eAAA,GAAkB,QAAQ,UAAA,CAAW,eAAA;AAAA,QACzC,iBAAA;AAAA,QACA,MAAM;AACJ,UAAA,MAAM,KAAA,GAAQ,aAAa,SAAA,EAAU;AACrC,UAAA,OAAA,CAAQ,QAAA,CAA0B,MAAA,EAAQ,CAAC,IAAA,MAAU;AAAA,YACnD,GAAG,IAAA;AAAA,YACH,WAAA,EAAa,KAAA;AAAA,YACb,YAAA,EAAc,aAAa,aAAA;AAAc,WAC3C,CAAE,CAAA;AAEF,UAAA,IAAI,KAAA,EAAO;AACT,YAAA,OAAA,CAAQ,GAAA,CAAI,QAAA,CAAS,KAAA,CAAM,SAAS,CAAA;AAAA,UACtC;AAAA,QACF;AAAA,OACF;AAIA,MAAA,MAAM,SAAA,GAA2B;AAAA,QAC/B,gBAAgB,SAAA,EAAmB;AACjC,UAAA,OAAA,CAAQ,UAAA,CAAW,QAAA,CAAS,cAAA,EAAgB,EAAE,WAAW,CAAA;AAAA,QAC3D,CAAA;AAAA,QACA,cAAA,GAAiB;AACf,UAAA,MAAM,KAAA,GAAQ,OAAA,CAAQ,KAAA,CAAM,QAAA,EAAS;AACrC,UAAA,KAAA,IAAS,IAAI,CAAA,EAAG,CAAA,GAAI,KAAA,CAAM,KAAA,CAAM,QAAQ,CAAA,EAAA,EAAK;AAC3C,YAAA,OAAA,CAAQ,WAAW,QAAA,CAAS,cAAA,EAAgB,EAAE,SAAA,EAAW,GAAG,CAAA;AAAA,UAC9D;AAAA,QACF,CAAA;AAAA,QACA,MAAA,CAAO,OAAe,OAAA,EAAyB;AAC7C,UAAA,OAAA,CAAQ,WAAW,QAAA,CAAS,aAAA,EAAe,EAAE,KAAA,EAAO,SAAS,CAAA;AAC7D,UAAA,OAAO,aAAa,aAAA,EAAc;AAAA,QACpC,CAAA;AAAA,QACA,UAAA,GAAa;AACX,UAAA,OAAA,CAAQ,UAAA,CAAW,QAAA,CAAS,iBAAA,EAAmB,EAAE,CAAA;AACjD,UAAA,OAAO,aAAa,cAAA,EAAe;AAAA,QACrC,CAAA;AAAA,QACA,UAAA,GAAa;AACX,UAAA,OAAA,CAAQ,UAAA,CAAW,QAAA,CAAS,iBAAA,EAAmB,EAAE,CAAA;AACjD,UAAA,OAAO,aAAa,cAAA,EAAe;AAAA,QACrC,CAAA;AAAA,QACA,WAAA,GAAc;AACZ,UAAA,YAAA,CAAa,QAAA,EAAS;AACtB,UAAA,OAAA,CAAQ,QAAA,CAA0B,QAAQ,OAAO;AAAA,YAC/C,GAAG;AAAA,WACL,CAAE,CAAA;AAAA,QACJ,CAAA;AAAA,QACA,YAAA,GAAe;AACb,UAAA,OAAO,SAAA;AAAA,QACT,CAAA;AAAA,QACA,eAAA,GAAkB;AAChB,UAAA,OAAO,YAAA;AAAA,QACT;AAAA,OACF;AAGA,MAAC,QAAgB,UAAA,GAAa,SAAA;AAE9B,MAAA,OAAO,MAAM;AACX,QAAA,YAAA,EAAa;AACb,QAAA,WAAA,EAAY;AACZ,QAAA,eAAA,EAAgB;AAChB,QAAA,eAAA,EAAgB;AAChB,QAAA,YAAA,CAAa,QAAA,EAAS;AAAA,MACxB,CAAA;AAAA,IACF;AAAA,GACF;AACF","file":"index.cjs","sourcesContent":["// ─── Text Extractor ───\n//\n// Extracts text content from PDF pages with character, word, and line bounding boxes.\n// Phase 1 implementation uses synthetic extraction; Phase 2 will integrate pdf.js getTextContent().\n\nimport type {\n  PdfTextContent,\n  PdfCharInfo,\n  PdfWordInfo,\n  PdfLineInfo,\n  PdfRect,\n  PdfPageState,\n} from '@gridstorm/pdf-core';\n\n/** Configuration for text extraction. */\nexport interface TextExtractorConfig {\n  /** Threshold for word break detection (in PDF points). */\n  wordSpacingThreshold?: number;\n  /** Threshold for line break detection (in PDF points). */\n  lineSpacingThreshold?: number;\n}\n\nconst DEFAULT_WORD_SPACING = 3;\nconst DEFAULT_LINE_SPACING = 5;\n\n/** Extracts structured text content from PDF text items. */\nexport class TextExtractor {\n  private wordSpacingThreshold: number;\n  private lineSpacingThreshold: number;\n\n  constructor(config: TextExtractorConfig = {}) {\n    this.wordSpacingThreshold = config.wordSpacingThreshold ?? DEFAULT_WORD_SPACING;\n    this.lineSpacingThreshold = config.lineSpacingThreshold ?? DEFAULT_LINE_SPACING;\n  }\n\n  /** Extract text content from raw text items.\n   *  In Phase 2 this will accept pdf.js TextItem[]; for now it works with PdfCharInfo[]. */\n  extract(chars: PdfCharInfo[], _page: PdfPageState): PdfTextContent {\n    if (chars.length === 0) {\n      return { chars, words: [], lines: [] };\n    }\n\n    const words = this.segmentWords(chars);\n    const lines = this.segmentLines(words, chars);\n\n    return { chars, words, lines };\n  }\n\n  /** Build PdfCharInfo from a simple text string (for testing/placeholder). */\n  buildCharsFromString(\n    text: string,\n    startX: number,\n    startY: number,\n    fontSize: number,\n    fontName = 'Helvetica',\n  ): PdfCharInfo[] {\n    const chars: PdfCharInfo[] = [];\n    let x = startX;\n    const charWidth = fontSize * 0.6; // Approximate monospace width\n\n    for (let i = 0; i < text.length; i++) {\n      const char = text[i]!;\n      const rect: PdfRect = [x, startY, x + charWidth, startY + fontSize];\n\n      chars.push({\n        char,\n        rect,\n        fontName,\n        fontSize,\n        transform: [fontSize, 0, 0, fontSize, x, startY],\n      });\n\n      x += charWidth;\n    }\n\n    return chars;\n  }\n\n  /** Segment characters into words based on spacing. */\n  private segmentWords(chars: PdfCharInfo[]): PdfWordInfo[] {\n    if (chars.length === 0) return [];\n\n    const words: PdfWordInfo[] = [];\n    let wordStart = 0;\n    let wordChars: PdfCharInfo[] = [chars[0]!];\n\n    for (let i = 1; i < chars.length; i++) {\n      const prev = chars[i - 1]!;\n      const curr = chars[i]!;\n\n      const gap = curr.rect[0] - prev.rect[2]; // x1 of current - x2 of previous\n      const isSpace = curr.char === ' ' || prev.char === ' ';\n      const isNewLine = Math.abs(curr.rect[1] - prev.rect[1]) > this.lineSpacingThreshold;\n      const isWordBreak = gap > this.wordSpacingThreshold || isSpace || isNewLine;\n\n      if (isWordBreak) {\n        // Complete current word (skip if only spaces)\n        const text = wordChars.map((c) => c.char).join('').trim();\n        if (text.length > 0) {\n          words.push({\n            text,\n            rect: this.boundingRect(wordChars),\n            charIndices: [wordStart, wordStart + wordChars.length - 1],\n          });\n        }\n\n        // Skip space characters\n        if (curr.char !== ' ') {\n          wordStart = i;\n          wordChars = [curr];\n        } else {\n          wordStart = i + 1;\n          wordChars = [];\n        }\n      } else {\n        wordChars.push(curr);\n      }\n    }\n\n    // Final word\n    if (wordChars.length > 0) {\n      const text = wordChars.map((c) => c.char).join('').trim();\n      if (text.length > 0) {\n        words.push({\n          text,\n          rect: this.boundingRect(wordChars),\n          charIndices: [wordStart, wordStart + wordChars.length - 1],\n        });\n      }\n    }\n\n    return words;\n  }\n\n  /** Segment words into lines based on vertical position. */\n  private segmentLines(words: PdfWordInfo[], _chars: PdfCharInfo[]): PdfLineInfo[] {\n    if (words.length === 0) return [];\n\n    const lines: PdfLineInfo[] = [];\n    let lineStart = 0;\n    let lineWords: PdfWordInfo[] = [words[0]!];\n    let lineY = words[0]!.rect[1]; // y1 of first word\n\n    for (let i = 1; i < words.length; i++) {\n      const word = words[i]!;\n      const yDiff = Math.abs(word.rect[1] - lineY);\n\n      if (yDiff > this.lineSpacingThreshold) {\n        // New line\n        lines.push(this.createLine(lineWords, lineStart));\n        lineStart = i;\n        lineWords = [word];\n        lineY = word.rect[1];\n      } else {\n        lineWords.push(word);\n      }\n    }\n\n    // Final line\n    if (lineWords.length > 0) {\n      lines.push(this.createLine(lineWords, lineStart));\n    }\n\n    return lines;\n  }\n\n  private createLine(words: PdfWordInfo[], startIndex: number): PdfLineInfo {\n    const text = words.map((w) => w.text).join(' ');\n    const rects = words.map((w) => w.rect);\n    const rect: PdfRect = [\n      Math.min(...rects.map((r) => r[0])),\n      Math.min(...rects.map((r) => r[1])),\n      Math.max(...rects.map((r) => r[2])),\n      Math.max(...rects.map((r) => r[3])),\n    ];\n    return {\n      text,\n      rect,\n      wordIndices: [startIndex, startIndex + words.length - 1],\n    };\n  }\n\n  /** Compute bounding rect for a set of characters. */\n  private boundingRect(chars: PdfCharInfo[]): PdfRect {\n    let minX = Infinity,\n      minY = Infinity,\n      maxX = -Infinity,\n      maxY = -Infinity;\n\n    for (const c of chars) {\n      if (c.rect[0] < minX) minX = c.rect[0];\n      if (c.rect[1] < minY) minY = c.rect[1];\n      if (c.rect[2] > maxX) maxX = c.rect[2];\n      if (c.rect[3] > maxY) maxY = c.rect[3];\n    }\n\n    return [minX, minY, maxX, maxY];\n  }\n}\n","// ─── Search Engine ───\n//\n// Full-text search across PDF pages with match navigation.\n\nimport type { PdfTextContent, PdfRect } from '@gridstorm/pdf-core';\n\n/** Search options. */\nexport interface SearchOptions {\n  /** Case-sensitive search. */\n  caseSensitive?: boolean;\n  /** Match whole words only. */\n  wholeWord?: boolean;\n  /** Use regex pattern. */\n  regex?: boolean;\n}\n\n/** A single search match. */\nexport interface SearchMatch {\n  /** Page index where the match was found. */\n  pageIndex: number;\n  /** Start character index within the page text. */\n  charStart: number;\n  /** End character index (exclusive). */\n  charEnd: number;\n  /** Bounding rectangles covering the match (may span multiple lines). */\n  rects: PdfRect[];\n  /** The matched text. */\n  text: string;\n}\n\n/** Search result set. */\nexport interface SearchResult {\n  /** All matches across all pages. */\n  matches: SearchMatch[];\n  /** Total match count. */\n  totalCount: number;\n  /** Currently active match index (-1 if none). */\n  activeIndex: number;\n}\n\n/** Search engine for finding text in PDF pages. */\nexport class SearchEngine {\n  private textContents = new Map<number, PdfTextContent>();\n  private lastResult: SearchResult = {\n    matches: [],\n    totalCount: 0,\n    activeIndex: -1,\n  };\n\n  /** Set the text content for a page. */\n  setPageTextContent(pageIndex: number, textContent: PdfTextContent): void {\n    this.textContents.set(pageIndex, textContent);\n  }\n\n  /** Clear text content for a page. */\n  clearPageTextContent(pageIndex: number): void {\n    this.textContents.delete(pageIndex);\n  }\n\n  /** Clear all cached text content. */\n  clearAll(): void {\n    this.textContents.clear();\n    this.lastResult = { matches: [], totalCount: 0, activeIndex: -1 };\n  }\n\n  /** Search for a query across all loaded pages. */\n  search(query: string, options: SearchOptions = {}): SearchResult {\n    if (!query) {\n      this.lastResult = { matches: [], totalCount: 0, activeIndex: -1 };\n      return this.lastResult;\n    }\n\n    const matches: SearchMatch[] = [];\n\n    // Sort pages by index for consistent ordering\n    const sortedPages = [...this.textContents.entries()].sort(\n      ([a], [b]) => a - b,\n    );\n\n    for (const [pageIndex, textContent] of sortedPages) {\n      const pageMatches = this.searchPage(\n        pageIndex,\n        textContent,\n        query,\n        options,\n      );\n      matches.push(...pageMatches);\n    }\n\n    this.lastResult = {\n      matches,\n      totalCount: matches.length,\n      activeIndex: matches.length > 0 ? 0 : -1,\n    };\n\n    return this.lastResult;\n  }\n\n  /** Navigate to the next search match. */\n  nextMatch(): SearchMatch | null {\n    if (this.lastResult.totalCount === 0) return null;\n\n    this.lastResult.activeIndex =\n      (this.lastResult.activeIndex + 1) % this.lastResult.totalCount;\n    return this.lastResult.matches[this.lastResult.activeIndex] ?? null;\n  }\n\n  /** Navigate to the previous search match. */\n  prevMatch(): SearchMatch | null {\n    if (this.lastResult.totalCount === 0) return null;\n\n    this.lastResult.activeIndex =\n      (this.lastResult.activeIndex - 1 + this.lastResult.totalCount) %\n      this.lastResult.totalCount;\n    return this.lastResult.matches[this.lastResult.activeIndex] ?? null;\n  }\n\n  /** Get the current active match. */\n  getActiveMatch(): SearchMatch | null {\n    if (\n      this.lastResult.activeIndex < 0 ||\n      this.lastResult.activeIndex >= this.lastResult.totalCount\n    ) {\n      return null;\n    }\n    return this.lastResult.matches[this.lastResult.activeIndex] ?? null;\n  }\n\n  /** Get the last search result. */\n  getLastResult(): SearchResult {\n    return this.lastResult;\n  }\n\n  private searchPage(\n    pageIndex: number,\n    textContent: PdfTextContent,\n    query: string,\n    options: SearchOptions,\n  ): SearchMatch[] {\n    const matches: SearchMatch[] = [];\n\n    // Build full page text from lines\n    const fullText = textContent.lines.map((l) => l.text).join('\\n');\n\n    let pattern: RegExp;\n    try {\n      if (options.regex) {\n        const flags = options.caseSensitive ? 'g' : 'gi';\n        pattern = new RegExp(query, flags);\n      } else {\n        const escaped = query.replace(/[.*+?^${}()|[\\]\\\\]/g, '\\\\$&');\n        const flags = options.caseSensitive ? 'g' : 'gi';\n        const word = options.wholeWord ? `\\\\b${escaped}\\\\b` : escaped;\n        pattern = new RegExp(word, flags);\n      }\n    } catch {\n      return matches;\n    }\n\n    let match: RegExpExecArray | null;\n    while ((match = pattern.exec(fullText)) !== null) {\n      const matchText = match[0];\n      const matchStart = match.index;\n      const matchEnd = matchStart + matchText.length;\n\n      // Find covering word rects\n      const rects = this.findMatchRects(textContent, matchStart, matchEnd);\n\n      matches.push({\n        pageIndex,\n        charStart: matchStart,\n        charEnd: matchEnd,\n        rects,\n        text: matchText,\n      });\n\n      // Avoid infinite loop for zero-length matches\n      if (matchText.length === 0) {\n        pattern.lastIndex++;\n      }\n    }\n\n    return matches;\n  }\n\n  /** Find bounding rects for a text range. */\n  private findMatchRects(\n    textContent: PdfTextContent,\n    _start: number,\n    _end: number,\n  ): PdfRect[] {\n    // In Phase 1 we use word-level rects; Phase 2 will use char-level precision.\n    // For now, return rects for all words that overlap the match range.\n    const rects: PdfRect[] = [];\n\n    // Simple approach: iterate lines and match by character offset\n    let charOffset = 0;\n    for (const line of textContent.lines) {\n      const lineEnd = charOffset + line.text.length;\n\n      if (lineEnd > _start && charOffset < _end) {\n        // This line overlaps the match — use word-level rects within the line\n        for (let wi = line.wordIndices[0]; wi <= line.wordIndices[1]; wi++) {\n          const word = textContent.words[wi];\n          if (word) {\n            rects.push(word.rect);\n          }\n        }\n      }\n\n      charOffset = lineEnd + 1; // +1 for \\n\n    }\n\n    return rects.length > 0 ? rects : [[0, 0, 0, 0]];\n  }\n}\n","// ─── Text Plugin ───\n//\n// GridStorm PDF plugin for text extraction and search.\n\nimport type {\n  PdfPlugin,\n  PdfPluginContext,\n  PdfPluginDisposer,\n} from '@gridstorm/pdf-core';\nimport { TextExtractor } from './text-extractor';\nimport { SearchEngine } from './search-engine';\nimport type { SearchOptions, SearchResult, SearchMatch } from './search-engine';\n\n/** Plugin state for text operations. */\nexport interface TextPluginState {\n  /** Current search query. */\n  searchQuery: string;\n  /** Search options. */\n  searchOptions: SearchOptions;\n  /** Last search result. */\n  searchResult: SearchResult | null;\n  /** Active search match. */\n  activeMatch: SearchMatch | null;\n}\n\n/** Public API exposed by the text plugin. */\nexport interface TextPluginApi {\n  /** Extract text from a specific page. */\n  extractPageText(pageIndex: number): void;\n  /** Extract text from all pages. */\n  extractAllText(): void;\n  /** Search for text across all pages. */\n  search(query: string, options?: SearchOptions): SearchResult;\n  /** Navigate to next search match. */\n  searchNext(): SearchMatch | null;\n  /** Navigate to previous search match. */\n  searchPrev(): SearchMatch | null;\n  /** Clear search results. */\n  clearSearch(): void;\n  /** Get the text extractor instance. */\n  getExtractor(): TextExtractor;\n  /** Get the search engine instance. */\n  getSearchEngine(): SearchEngine;\n}\n\nconst INITIAL_STATE: TextPluginState = {\n  searchQuery: '',\n  searchOptions: {},\n  searchResult: null,\n  activeMatch: null,\n};\n\n/** Create the text extraction and search plugin. */\nexport function createTextPlugin(): PdfPlugin {\n  return {\n    id: 'text',\n    name: 'Text Extraction & Search',\n    version: '0.1.0',\n\n    install(context: PdfPluginContext): PdfPluginDisposer {\n      const extractor = new TextExtractor();\n      const searchEngine = new SearchEngine();\n\n      // Register plugin state\n      context.registerState<TextPluginState>('text', { ...INITIAL_STATE });\n\n      // ─── Command Handlers ───\n\n      const unsubExtract = context.commandBus.registerHandler(\n        'text:extract',\n        (payload: { pageIndex: number }) => {\n          const state = context.store.getState();\n          const page = state.pages[payload.pageIndex];\n          if (!page) return;\n\n          // In Phase 1, use placeholder chars; Phase 2 will use pdf.js\n          const chars = extractor.buildCharsFromString(\n            `Sample text for page ${payload.pageIndex + 1}`,\n            72, // 1 inch margin\n            72,\n            12,\n          );\n\n          const textContent = extractor.extract(chars, page);\n\n          // Update page state with extracted text\n          context.store.setState((prev) => {\n            const pages = [...prev.pages];\n            const existing = pages[payload.pageIndex];\n            if (existing) {\n              pages[payload.pageIndex] = { ...existing, textContent };\n            }\n            return { ...prev, pages };\n          });\n\n          // Feed text to search engine\n          searchEngine.setPageTextContent(payload.pageIndex, textContent);\n\n          context.eventBus.emit('text:extracted', {\n            pageIndex: payload.pageIndex,\n            textContent,\n          });\n        },\n      );\n\n      const unsubSearch = context.commandBus.registerHandler(\n        'text:search',\n        (payload: { query: string; options?: SearchOptions }) => {\n          const result = searchEngine.search(payload.query, payload.options);\n\n          context.setState<TextPluginState>('text', (prev) => ({\n            ...prev,\n            searchQuery: payload.query,\n            searchOptions: payload.options ?? {},\n            searchResult: result,\n            activeMatch: searchEngine.getActiveMatch(),\n          }));\n\n          context.eventBus.emit('search:found', {\n            query: payload.query,\n            matches: result.matches,\n            total: result.totalCount,\n          });\n        },\n      );\n\n      const unsubSearchNext = context.commandBus.registerHandler(\n        'text:searchNext',\n        () => {\n          const match = searchEngine.nextMatch();\n          context.setState<TextPluginState>('text', (prev) => ({\n            ...prev,\n            activeMatch: match,\n            searchResult: searchEngine.getLastResult(),\n          }));\n\n          if (match) {\n            // Navigate to the match page\n            context.api.goToPage(match.pageIndex);\n          }\n        },\n      );\n\n      const unsubSearchPrev = context.commandBus.registerHandler(\n        'text:searchPrev',\n        () => {\n          const match = searchEngine.prevMatch();\n          context.setState<TextPluginState>('text', (prev) => ({\n            ...prev,\n            activeMatch: match,\n            searchResult: searchEngine.getLastResult(),\n          }));\n\n          if (match) {\n            context.api.goToPage(match.pageIndex);\n          }\n        },\n      );\n\n      // ─── Plugin API ───\n\n      const pluginApi: TextPluginApi = {\n        extractPageText(pageIndex: number) {\n          context.commandBus.dispatch('text:extract', { pageIndex });\n        },\n        extractAllText() {\n          const state = context.store.getState();\n          for (let i = 0; i < state.pages.length; i++) {\n            context.commandBus.dispatch('text:extract', { pageIndex: i });\n          }\n        },\n        search(query: string, options?: SearchOptions) {\n          context.commandBus.dispatch('text:search', { query, options });\n          return searchEngine.getLastResult();\n        },\n        searchNext() {\n          context.commandBus.dispatch('text:searchNext', {});\n          return searchEngine.getActiveMatch();\n        },\n        searchPrev() {\n          context.commandBus.dispatch('text:searchPrev', {});\n          return searchEngine.getActiveMatch();\n        },\n        clearSearch() {\n          searchEngine.clearAll();\n          context.setState<TextPluginState>('text', () => ({\n            ...INITIAL_STATE,\n          }));\n        },\n        getExtractor() {\n          return extractor;\n        },\n        getSearchEngine() {\n          return searchEngine;\n        },\n      };\n\n      // Expose API via plugin context (can be retrieved via api.getPluginApi('text'))\n      (context as any)._pluginApi = pluginApi;\n\n      return () => {\n        unsubExtract();\n        unsubSearch();\n        unsubSearchNext();\n        unsubSearchPrev();\n        searchEngine.clearAll();\n      };\n    },\n  };\n}\n"]}

package/dist/index.d.cts

@@ -0,0 +1,118 @@
+import { PdfCharInfo, PdfPageState, PdfTextContent, PdfRect, PdfPlugin } from '@gridstorm/pdf-core';
+
+/** Configuration for text extraction. */
+interface TextExtractorConfig {
+    /** Threshold for word break detection (in PDF points). */
+    wordSpacingThreshold?: number;
+    /** Threshold for line break detection (in PDF points). */
+    lineSpacingThreshold?: number;
+}
+/** Extracts structured text content from PDF text items. */
+declare class TextExtractor {
+    private wordSpacingThreshold;
+    private lineSpacingThreshold;
+    constructor(config?: TextExtractorConfig);
+    /** Extract text content from raw text items.
+     *  In Phase 2 this will accept pdf.js TextItem[]; for now it works with PdfCharInfo[]. */
+    extract(chars: PdfCharInfo[], _page: PdfPageState): PdfTextContent;
+    /** Build PdfCharInfo from a simple text string (for testing/placeholder). */
+    buildCharsFromString(text: string, startX: number, startY: number, fontSize: number, fontName?: string): PdfCharInfo[];
+    /** Segment characters into words based on spacing. */
+    private segmentWords;
+    /** Segment words into lines based on vertical position. */
+    private segmentLines;
+    private createLine;
+    /** Compute bounding rect for a set of characters. */
+    private boundingRect;
+}
+
+/** Search options. */
+interface SearchOptions {
+    /** Case-sensitive search. */
+    caseSensitive?: boolean;
+    /** Match whole words only. */
+    wholeWord?: boolean;
+    /** Use regex pattern. */
+    regex?: boolean;
+}
+/** A single search match. */
+interface SearchMatch {
+    /** Page index where the match was found. */
+    pageIndex: number;
+    /** Start character index within the page text. */
+    charStart: number;
+    /** End character index (exclusive). */
+    charEnd: number;
+    /** Bounding rectangles covering the match (may span multiple lines). */
+    rects: PdfRect[];
+    /** The matched text. */
+    text: string;
+}
+/** Search result set. */
+interface SearchResult {
+    /** All matches across all pages. */
+    matches: SearchMatch[];
+    /** Total match count. */
+    totalCount: number;
+    /** Currently active match index (-1 if none). */
+    activeIndex: number;
+}
+/** Search engine for finding text in PDF pages. */
+declare class SearchEngine {
+    private textContents;
+    private lastResult;
+    /** Set the text content for a page. */
+    setPageTextContent(pageIndex: number, textContent: PdfTextContent): void;
+    /** Clear text content for a page. */
+    clearPageTextContent(pageIndex: number): void;
+    /** Clear all cached text content. */
+    clearAll(): void;
+    /** Search for a query across all loaded pages. */
+    search(query: string, options?: SearchOptions): SearchResult;
+    /** Navigate to the next search match. */
+    nextMatch(): SearchMatch | null;
+    /** Navigate to the previous search match. */
+    prevMatch(): SearchMatch | null;
+    /** Get the current active match. */
+    getActiveMatch(): SearchMatch | null;
+    /** Get the last search result. */
+    getLastResult(): SearchResult;
+    private searchPage;
+    /** Find bounding rects for a text range. */
+    private findMatchRects;
+}
+
+/** Plugin state for text operations. */
+interface TextPluginState {
+    /** Current search query. */
+    searchQuery: string;
+    /** Search options. */
+    searchOptions: SearchOptions;
+    /** Last search result. */
+    searchResult: SearchResult | null;
+    /** Active search match. */
+    activeMatch: SearchMatch | null;
+}
+/** Public API exposed by the text plugin. */
+interface TextPluginApi {
+    /** Extract text from a specific page. */
+    extractPageText(pageIndex: number): void;
+    /** Extract text from all pages. */
+    extractAllText(): void;
+    /** Search for text across all pages. */
+    search(query: string, options?: SearchOptions): SearchResult;
+    /** Navigate to next search match. */
+    searchNext(): SearchMatch | null;
+    /** Navigate to previous search match. */
+    searchPrev(): SearchMatch | null;
+    /** Clear search results. */
+    clearSearch(): void;
+    /** Get the text extractor instance. */
+    getExtractor(): TextExtractor;
+    /** Get the search engine instance. */
+    getSearchEngine(): SearchEngine;
+}
+/** Create the text extraction and search plugin. */
+declare function createTextPlugin(): PdfPlugin;
+
+export { SearchEngine, type SearchMatch, type SearchOptions, type SearchResult, TextExtractor, type TextExtractorConfig, type TextPluginApi, type TextPluginState, createTextPlugin };

package/dist/index.d.ts

@@ -0,0 +1,118 @@
+import { PdfCharInfo, PdfPageState, PdfTextContent, PdfRect, PdfPlugin } from '@gridstorm/pdf-core';
+
+/** Configuration for text extraction. */
+interface TextExtractorConfig {
+    /** Threshold for word break detection (in PDF points). */
+    wordSpacingThreshold?: number;
+    /** Threshold for line break detection (in PDF points). */
+    lineSpacingThreshold?: number;
+}
+/** Extracts structured text content from PDF text items. */
+declare class TextExtractor {
+    private wordSpacingThreshold;
+    private lineSpacingThreshold;
+    constructor(config?: TextExtractorConfig);
+    /** Extract text content from raw text items.
+     *  In Phase 2 this will accept pdf.js TextItem[]; for now it works with PdfCharInfo[]. */
+    extract(chars: PdfCharInfo[], _page: PdfPageState): PdfTextContent;
+    /** Build PdfCharInfo from a simple text string (for testing/placeholder). */
+    buildCharsFromString(text: string, startX: number, startY: number, fontSize: number, fontName?: string): PdfCharInfo[];
+    /** Segment characters into words based on spacing. */
+    private segmentWords;
+    /** Segment words into lines based on vertical position. */
+    private segmentLines;
+    private createLine;
+    /** Compute bounding rect for a set of characters. */
+    private boundingRect;
+}
+
+/** Search options. */
+interface SearchOptions {
+    /** Case-sensitive search. */
+    caseSensitive?: boolean;
+    /** Match whole words only. */
+    wholeWord?: boolean;
+    /** Use regex pattern. */
+    regex?: boolean;
+}
+/** A single search match. */
+interface SearchMatch {
+    /** Page index where the match was found. */
+    pageIndex: number;
+    /** Start character index within the page text. */
+    charStart: number;
+    /** End character index (exclusive). */
+    charEnd: number;
+    /** Bounding rectangles covering the match (may span multiple lines). */
+    rects: PdfRect[];
+    /** The matched text. */
+    text: string;
+}
+/** Search result set. */
+interface SearchResult {
+    /** All matches across all pages. */
+    matches: SearchMatch[];
+    /** Total match count. */
+    totalCount: number;
+    /** Currently active match index (-1 if none). */
+    activeIndex: number;
+}
+/** Search engine for finding text in PDF pages. */
+declare class SearchEngine {
+    private textContents;
+    private lastResult;
+    /** Set the text content for a page. */
+    setPageTextContent(pageIndex: number, textContent: PdfTextContent): void;
+    /** Clear text content for a page. */
+    clearPageTextContent(pageIndex: number): void;
+    /** Clear all cached text content. */
+    clearAll(): void;
+    /** Search for a query across all loaded pages. */
+    search(query: string, options?: SearchOptions): SearchResult;
+    /** Navigate to the next search match. */
+    nextMatch(): SearchMatch | null;
+    /** Navigate to the previous search match. */
+    prevMatch(): SearchMatch | null;
+    /** Get the current active match. */
+    getActiveMatch(): SearchMatch | null;
+    /** Get the last search result. */
+    getLastResult(): SearchResult;
+    private searchPage;
+    /** Find bounding rects for a text range. */
+    private findMatchRects;
+}
+
+/** Plugin state for text operations. */
+interface TextPluginState {
+    /** Current search query. */
+    searchQuery: string;
+    /** Search options. */
+    searchOptions: SearchOptions;
+    /** Last search result. */
+    searchResult: SearchResult | null;
+    /** Active search match. */
+    activeMatch: SearchMatch | null;
+}
+/** Public API exposed by the text plugin. */
+interface TextPluginApi {
+    /** Extract text from a specific page. */
+    extractPageText(pageIndex: number): void;
+    /** Extract text from all pages. */
+    extractAllText(): void;
+    /** Search for text across all pages. */
+    search(query: string, options?: SearchOptions): SearchResult;
+    /** Navigate to next search match. */
+    searchNext(): SearchMatch | null;
+    /** Navigate to previous search match. */
+    searchPrev(): SearchMatch | null;
+    /** Clear search results. */
+    clearSearch(): void;
+    /** Get the text extractor instance. */
+    getExtractor(): TextExtractor;
+    /** Get the search engine instance. */
+    getSearchEngine(): SearchEngine;
+}
+/** Create the text extraction and search plugin. */
+declare function createTextPlugin(): PdfPlugin;
+
+export { SearchEngine, type SearchMatch, type SearchOptions, type SearchResult, TextExtractor, type TextExtractorConfig, type TextPluginApi, type TextPluginState, createTextPlugin };

package/dist/index.js

@@ -0,0 +1,403 @@
+// src/text-extractor.ts
+var DEFAULT_WORD_SPACING = 3;
+var DEFAULT_LINE_SPACING = 5;
+var TextExtractor = class {
+  constructor(config = {}) {
+    this.wordSpacingThreshold = config.wordSpacingThreshold ?? DEFAULT_WORD_SPACING;
+    this.lineSpacingThreshold = config.lineSpacingThreshold ?? DEFAULT_LINE_SPACING;
+  }
+  /** Extract text content from raw text items.
+   *  In Phase 2 this will accept pdf.js TextItem[]; for now it works with PdfCharInfo[]. */
+  extract(chars, _page) {
+    if (chars.length === 0) {
+      return { chars, words: [], lines: [] };
+    }
+    const words = this.segmentWords(chars);
+    const lines = this.segmentLines(words, chars);
+    return { chars, words, lines };
+  }
+  /** Build PdfCharInfo from a simple text string (for testing/placeholder). */
+  buildCharsFromString(text, startX, startY, fontSize, fontName = "Helvetica") {
+    const chars = [];
+    let x = startX;
+    const charWidth = fontSize * 0.6;
+    for (let i = 0; i < text.length; i++) {
+      const char = text[i];
+      const rect = [x, startY, x + charWidth, startY + fontSize];
+      chars.push({
+        char,
+        rect,
+        fontName,
+        fontSize,
+        transform: [fontSize, 0, 0, fontSize, x, startY]
+      });
+      x += charWidth;
+    }
+    return chars;
+  }
+  /** Segment characters into words based on spacing. */
+  segmentWords(chars) {
+    if (chars.length === 0) return [];
+    const words = [];
+    let wordStart = 0;
+    let wordChars = [chars[0]];
+    for (let i = 1; i < chars.length; i++) {
+      const prev = chars[i - 1];
+      const curr = chars[i];
+      const gap = curr.rect[0] - prev.rect[2];
+      const isSpace = curr.char === " " || prev.char === " ";
+      const isNewLine = Math.abs(curr.rect[1] - prev.rect[1]) > this.lineSpacingThreshold;
+      const isWordBreak = gap > this.wordSpacingThreshold || isSpace || isNewLine;
+      if (isWordBreak) {
+        const text = wordChars.map((c) => c.char).join("").trim();
+        if (text.length > 0) {
+          words.push({
+            text,
+            rect: this.boundingRect(wordChars),
+            charIndices: [wordStart, wordStart + wordChars.length - 1]
+          });
+        }
+        if (curr.char !== " ") {
+          wordStart = i;
+          wordChars = [curr];
+        } else {
+          wordStart = i + 1;
+          wordChars = [];
+        }
+      } else {
+        wordChars.push(curr);
+      }
+    }
+    if (wordChars.length > 0) {
+      const text = wordChars.map((c) => c.char).join("").trim();
+      if (text.length > 0) {
+        words.push({
+          text,
+          rect: this.boundingRect(wordChars),
+          charIndices: [wordStart, wordStart + wordChars.length - 1]
+        });
+      }
+    }
+    return words;
+  }
+  /** Segment words into lines based on vertical position. */
+  segmentLines(words, _chars) {
+    if (words.length === 0) return [];
+    const lines = [];
+    let lineStart = 0;
+    let lineWords = [words[0]];
+    let lineY = words[0].rect[1];
+    for (let i = 1; i < words.length; i++) {
+      const word = words[i];
+      const yDiff = Math.abs(word.rect[1] - lineY);
+      if (yDiff > this.lineSpacingThreshold) {
+        lines.push(this.createLine(lineWords, lineStart));
+        lineStart = i;
+        lineWords = [word];
+        lineY = word.rect[1];
+      } else {
+        lineWords.push(word);
+      }
+    }
+    if (lineWords.length > 0) {
+      lines.push(this.createLine(lineWords, lineStart));
+    }
+    return lines;
+  }
+  createLine(words, startIndex) {
+    const text = words.map((w) => w.text).join(" ");
+    const rects = words.map((w) => w.rect);
+    const rect = [
+      Math.min(...rects.map((r) => r[0])),
+      Math.min(...rects.map((r) => r[1])),
+      Math.max(...rects.map((r) => r[2])),
+      Math.max(...rects.map((r) => r[3]))
+    ];
+    return {
+      text,
+      rect,
+      wordIndices: [startIndex, startIndex + words.length - 1]
+    };
+  }
+  /** Compute bounding rect for a set of characters. */
+  boundingRect(chars) {
+    let minX = Infinity, minY = Infinity, maxX = -Infinity, maxY = -Infinity;
+    for (const c of chars) {
+      if (c.rect[0] < minX) minX = c.rect[0];
+      if (c.rect[1] < minY) minY = c.rect[1];
+      if (c.rect[2] > maxX) maxX = c.rect[2];
+      if (c.rect[3] > maxY) maxY = c.rect[3];
+    }
+    return [minX, minY, maxX, maxY];
+  }
+};
+
+// src/search-engine.ts
+var SearchEngine = class {
+  constructor() {
+    this.textContents = /* @__PURE__ */ new Map();
+    this.lastResult = {
+      matches: [],
+      totalCount: 0,
+      activeIndex: -1
+    };
+  }
+  /** Set the text content for a page. */
+  setPageTextContent(pageIndex, textContent) {
+    this.textContents.set(pageIndex, textContent);
+  }
+  /** Clear text content for a page. */
+  clearPageTextContent(pageIndex) {
+    this.textContents.delete(pageIndex);
+  }
+  /** Clear all cached text content. */
+  clearAll() {
+    this.textContents.clear();
+    this.lastResult = { matches: [], totalCount: 0, activeIndex: -1 };
+  }
+  /** Search for a query across all loaded pages. */
+  search(query, options = {}) {
+    if (!query) {
+      this.lastResult = { matches: [], totalCount: 0, activeIndex: -1 };
+      return this.lastResult;
+    }
+    const matches = [];
+    const sortedPages = [...this.textContents.entries()].sort(
+      ([a], [b]) => a - b
+    );
+    for (const [pageIndex, textContent] of sortedPages) {
+      const pageMatches = this.searchPage(
+        pageIndex,
+        textContent,
+        query,
+        options
+      );
+      matches.push(...pageMatches);
+    }
+    this.lastResult = {
+      matches,
+      totalCount: matches.length,
+      activeIndex: matches.length > 0 ? 0 : -1
+    };
+    return this.lastResult;
+  }
+  /** Navigate to the next search match. */
+  nextMatch() {
+    if (this.lastResult.totalCount === 0) return null;
+    this.lastResult.activeIndex = (this.lastResult.activeIndex + 1) % this.lastResult.totalCount;
+    return this.lastResult.matches[this.lastResult.activeIndex] ?? null;
+  }
+  /** Navigate to the previous search match. */
+  prevMatch() {
+    if (this.lastResult.totalCount === 0) return null;
+    this.lastResult.activeIndex = (this.lastResult.activeIndex - 1 + this.lastResult.totalCount) % this.lastResult.totalCount;
+    return this.lastResult.matches[this.lastResult.activeIndex] ?? null;
+  }
+  /** Get the current active match. */
+  getActiveMatch() {
+    if (this.lastResult.activeIndex < 0 || this.lastResult.activeIndex >= this.lastResult.totalCount) {
+      return null;
+    }
+    return this.lastResult.matches[this.lastResult.activeIndex] ?? null;
+  }
+  /** Get the last search result. */
+  getLastResult() {
+    return this.lastResult;
+  }
+  searchPage(pageIndex, textContent, query, options) {
+    const matches = [];
+    const fullText = textContent.lines.map((l) => l.text).join("\n");
+    let pattern;
+    try {
+      if (options.regex) {
+        const flags = options.caseSensitive ? "g" : "gi";
+        pattern = new RegExp(query, flags);
+      } else {
+        const escaped = query.replace(/[.*+?^${}()|[\]\\]/g, "\\$&");
+        const flags = options.caseSensitive ? "g" : "gi";
+        const word = options.wholeWord ? `\\b${escaped}\\b` : escaped;
+        pattern = new RegExp(word, flags);
+      }
+    } catch {
+      return matches;
+    }
+    let match;
+    while ((match = pattern.exec(fullText)) !== null) {
+      const matchText = match[0];
+      const matchStart = match.index;
+      const matchEnd = matchStart + matchText.length;
+      const rects = this.findMatchRects(textContent, matchStart, matchEnd);
+      matches.push({
+        pageIndex,
+        charStart: matchStart,
+        charEnd: matchEnd,
+        rects,
+        text: matchText
+      });
+      if (matchText.length === 0) {
+        pattern.lastIndex++;
+      }
+    }
+    return matches;
+  }
+  /** Find bounding rects for a text range. */
+  findMatchRects(textContent, _start, _end) {
+    const rects = [];
+    let charOffset = 0;
+    for (const line of textContent.lines) {
+      const lineEnd = charOffset + line.text.length;
+      if (lineEnd > _start && charOffset < _end) {
+        for (let wi = line.wordIndices[0]; wi <= line.wordIndices[1]; wi++) {
+          const word = textContent.words[wi];
+          if (word) {
+            rects.push(word.rect);
+          }
+        }
+      }
+      charOffset = lineEnd + 1;
+    }
+    return rects.length > 0 ? rects : [[0, 0, 0, 0]];
+  }
+};
+
+// src/text-plugin.ts
+var INITIAL_STATE = {
+  searchQuery: "",
+  searchOptions: {},
+  searchResult: null,
+  activeMatch: null
+};
+function createTextPlugin() {
+  return {
+    id: "text",
+    name: "Text Extraction & Search",
+    version: "0.1.0",
+    install(context) {
+      const extractor = new TextExtractor();
+      const searchEngine = new SearchEngine();
+      context.registerState("text", { ...INITIAL_STATE });
+      const unsubExtract = context.commandBus.registerHandler(
+        "text:extract",
+        (payload) => {
+          const state = context.store.getState();
+          const page = state.pages[payload.pageIndex];
+          if (!page) return;
+          const chars = extractor.buildCharsFromString(
+            `Sample text for page ${payload.pageIndex + 1}`,
+            72,
+            // 1 inch margin
+            72,
+            12
+          );
+          const textContent = extractor.extract(chars, page);
+          context.store.setState((prev) => {
+            const pages = [...prev.pages];
+            const existing = pages[payload.pageIndex];
+            if (existing) {
+              pages[payload.pageIndex] = { ...existing, textContent };
+            }
+            return { ...prev, pages };
+          });
+          searchEngine.setPageTextContent(payload.pageIndex, textContent);
+          context.eventBus.emit("text:extracted", {
+            pageIndex: payload.pageIndex,
+            textContent
+          });
+        }
+      );
+      const unsubSearch = context.commandBus.registerHandler(
+        "text:search",
+        (payload) => {
+          const result = searchEngine.search(payload.query, payload.options);
+          context.setState("text", (prev) => ({
+            ...prev,
+            searchQuery: payload.query,
+            searchOptions: payload.options ?? {},
+            searchResult: result,
+            activeMatch: searchEngine.getActiveMatch()
+          }));
+          context.eventBus.emit("search:found", {
+            query: payload.query,
+            matches: result.matches,
+            total: result.totalCount
+          });
+        }
+      );
+      const unsubSearchNext = context.commandBus.registerHandler(
+        "text:searchNext",
+        () => {
+          const match = searchEngine.nextMatch();
+          context.setState("text", (prev) => ({
+            ...prev,
+            activeMatch: match,
+            searchResult: searchEngine.getLastResult()
+          }));
+          if (match) {
+            context.api.goToPage(match.pageIndex);
+          }
+        }
+      );
+      const unsubSearchPrev = context.commandBus.registerHandler(
+        "text:searchPrev",
+        () => {
+          const match = searchEngine.prevMatch();
+          context.setState("text", (prev) => ({
+            ...prev,
+            activeMatch: match,
+            searchResult: searchEngine.getLastResult()
+          }));
+          if (match) {
+            context.api.goToPage(match.pageIndex);
+          }
+        }
+      );
+      const pluginApi = {
+        extractPageText(pageIndex) {
+          context.commandBus.dispatch("text:extract", { pageIndex });
+        },
+        extractAllText() {
+          const state = context.store.getState();
+          for (let i = 0; i < state.pages.length; i++) {
+            context.commandBus.dispatch("text:extract", { pageIndex: i });
+          }
+        },
+        search(query, options) {
+          context.commandBus.dispatch("text:search", { query, options });
+          return searchEngine.getLastResult();
+        },
+        searchNext() {
+          context.commandBus.dispatch("text:searchNext", {});
+          return searchEngine.getActiveMatch();
+        },
+        searchPrev() {
+          context.commandBus.dispatch("text:searchPrev", {});
+          return searchEngine.getActiveMatch();
+        },
+        clearSearch() {
+          searchEngine.clearAll();
+          context.setState("text", () => ({
+            ...INITIAL_STATE
+          }));
+        },
+        getExtractor() {
+          return extractor;
+        },
+        getSearchEngine() {
+          return searchEngine;
+        }
+      };
+      context._pluginApi = pluginApi;
+      return () => {
+        unsubExtract();
+        unsubSearch();
+        unsubSearchNext();
+        unsubSearchPrev();
+        searchEngine.clearAll();
+      };
+    }
+  };
+}
+
+export { SearchEngine, TextExtractor, createTextPlugin };
+//# sourceMappingURL=index.js.map
+//# sourceMappingURL=index.js.map

package/dist/index.js.map

@@ -0,0 +1 @@
+{"version":3,"sources":["../src/text-extractor.ts","../src/search-engine.ts","../src/text-plugin.ts"],"names":[],"mappings":";AAsBA,IAAM,oBAAA,GAAuB,CAAA;AAC7B,IAAM,oBAAA,GAAuB,CAAA;AAGtB,IAAM,gBAAN,MAAoB;AAAA,EAIzB,WAAA,CAAY,MAAA,GAA8B,EAAC,EAAG;AAC5C,IAAA,IAAA,CAAK,oBAAA,GAAuB,OAAO,oBAAA,IAAwB,oBAAA;AAC3D,IAAA,IAAA,CAAK,oBAAA,GAAuB,OAAO,oBAAA,IAAwB,oBAAA;AAAA,EAC7D;AAAA;AAAA;AAAA,EAIA,OAAA,CAAQ,OAAsB,KAAA,EAAqC;AACjE,IAAA,IAAI,KAAA,CAAM,WAAW,CAAA,EAAG;AACtB,MAAA,OAAO,EAAE,KAAA,EAAO,KAAA,EAAO,EAAC,EAAG,KAAA,EAAO,EAAC,EAAE;AAAA,IACvC;AAEA,IAAA,MAAM,KAAA,GAAQ,IAAA,CAAK,YAAA,CAAa,KAAK,CAAA;AACrC,IAAA,MAAM,KAAA,GAAQ,IAAA,CAAK,YAAA,CAAa,KAAA,EAAO,KAAK,CAAA;AAE5C,IAAA,OAAO,EAAE,KAAA,EAAO,KAAA,EAAO,KAAA,EAAM;AAAA,EAC/B;AAAA;AAAA,EAGA,qBACE,IAAA,EACA,MAAA,EACA,MAAA,EACA,QAAA,EACA,WAAW,WAAA,EACI;AACf,IAAA,MAAM,QAAuB,EAAC;AAC9B,IAAA,IAAI,CAAA,GAAI,MAAA;AACR,IAAA,MAAM,YAAY,QAAA,GAAW,GAAA;AAE7B,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,IAAA,CAAK,QAAQ,CAAA,EAAA,EAAK;AACpC,MAAA,MAAM,IAAA,GAAO,KAAK,CAAC,CAAA;AACnB,MAAA,MAAM,OAAgB,CAAC,CAAA,EAAG,QAAQ,CAAA,GAAI,SAAA,EAAW,SAAS,QAAQ,CAAA;AAElE,MAAA,KAAA,CAAM,IAAA,CAAK;AAAA,QACT,IAAA;AAAA,QACA,IAAA;AAAA,QACA,QAAA;AAAA,QACA,QAAA;AAAA,QACA,WAAW,CAAC,QAAA,EAAU,GAAG,CAAA,EAAG,QAAA,EAAU,GAAG,MAAM;AAAA,OAChD,CAAA;AAED,MAAA,CAAA,IAAK,SAAA;AAAA,IACP;AAEA,IAAA,OAAO,KAAA;AAAA,EACT;AAAA;AAAA,EAGQ,aAAa,KAAA,EAAqC;AACxD,IAAA,IAAI,KAAA,CAAM,MAAA,KAAW,CAAA,EAAG,OAAO,EAAC;AAEhC,IAAA,MAAM,QAAuB,EAAC;AAC9B,IAAA,IAAI,SAAA,GAAY,CAAA;AAChB,IAAA,IAAI,SAAA,GAA2B,CAAC,KAAA,CAAM,CAAC,CAAE,CAAA;AAEzC,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,KAAA,CAAM,QAAQ,CAAA,EAAA,EAAK;AACrC,MAAA,MAAM,IAAA,GAAO,KAAA,CAAM,CAAA,GAAI,CAAC,CAAA;AACxB,MAAA,MAAM,IAAA,GAAO,MAAM,CAAC,CAAA;AAEpB,MAAA,MAAM,MAAM,IAAA,CAAK,IAAA,CAAK,CAAC,CAAA,GAAI,IAAA,CAAK,KAAK,CAAC,CAAA;AACtC,MAAA,MAAM,OAAA,GAAU,IAAA,CAAK,IAAA,KAAS,GAAA,IAAO,KAAK,IAAA,KAAS,GAAA;AACnD,MAAA,MAAM,SAAA,GAAY,IAAA,CAAK,GAAA,CAAI,IAAA,CAAK,IAAA,CAAK,CAAC,CAAA,GAAI,IAAA,CAAK,IAAA,CAAK,CAAC,CAAC,CAAA,GAAI,IAAA,CAAK,oBAAA;AAC/D,MAAA,MAAM,WAAA,GAAc,GAAA,GAAM,IAAA,CAAK,oBAAA,IAAwB,OAAA,IAAW,SAAA;AAElE,MAAA,IAAI,WAAA,EAAa;AAEf,QAAA,MAAM,IAAA,GAAO,SAAA,CAAU,GAAA,CAAI,CAAC,CAAA,KAAM,CAAA,CAAE,IAAI,CAAA,CAAE,IAAA,CAAK,EAAE,CAAA,CAAE,IAAA,EAAK;AACxD,QAAA,IAAI,IAAA,CAAK,SAAS,CAAA,EAAG;AACnB,UAAA,KAAA,CAAM,IAAA,CAAK;AAAA,YACT,IAAA;AAAA,YACA,IAAA,EAAM,IAAA,CAAK,YAAA,CAAa,SAAS,CAAA;AAAA,YACjC,aAAa,CAAC,SAAA,EAAW,SAAA,GAAY,SAAA,CAAU,SAAS,CAAC;AAAA,WAC1D,CAAA;AAAA,QACH;AAGA,QAAA,IAAI,IAAA,CAAK,SAAS,GAAA,EAAK;AACrB,UAAA,SAAA,GAAY,CAAA;AACZ,UAAA,SAAA,GAAY,CAAC,IAAI,CAAA;AAAA,QACnB,CAAA,MAAO;AACL,UAAA,SAAA,GAAY,CAAA,GAAI,CAAA;AAChB,UAAA,SAAA,GAAY,EAAC;AAAA,QACf;AAAA,MACF,CAAA,MAAO;AACL,QAAA,SAAA,CAAU,KAAK,IAAI,CAAA;AAAA,MACrB;AAAA,IACF;AAGA,IAAA,IAAI,SAAA,CAAU,SAAS,CAAA,EAAG;AACxB,MAAA,MAAM,IAAA,GAAO,SAAA,CAAU,GAAA,CAAI,CAAC,CAAA,KAAM,CAAA,CAAE,IAAI,CAAA,CAAE,IAAA,CAAK,EAAE,CAAA,CAAE,IAAA,EAAK;AACxD,MAAA,IAAI,IAAA,CAAK,SAAS,CAAA,EAAG;AACnB,QAAA,KAAA,CAAM,IAAA,CAAK;AAAA,UACT,IAAA;AAAA,UACA,IAAA,EAAM,IAAA,CAAK,YAAA,CAAa,SAAS,CAAA;AAAA,UACjC,aAAa,CAAC,SAAA,EAAW,SAAA,GAAY,SAAA,CAAU,SAAS,CAAC;AAAA,SAC1D,CAAA;AAAA,MACH;AAAA,IACF;AAEA,IAAA,OAAO,KAAA;AAAA,EACT;AAAA;AAAA,EAGQ,YAAA,CAAa,OAAsB,MAAA,EAAsC;AAC/E,IAAA,IAAI,KAAA,CAAM,MAAA,KAAW,CAAA,EAAG,OAAO,EAAC;AAEhC,IAAA,MAAM,QAAuB,EAAC;AAC9B,IAAA,IAAI,SAAA,GAAY,CAAA;AAChB,IAAA,IAAI,SAAA,GAA2B,CAAC,KAAA,CAAM,CAAC,CAAE,CAAA;AACzC,IAAA,IAAI,KAAA,GAAQ,KAAA,CAAM,CAAC,CAAA,CAAG,KAAK,CAAC,CAAA;AAE5B,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,KAAA,CAAM,QAAQ,CAAA,EAAA,EAAK;AACrC,MAAA,MAAM,IAAA,GAAO,MAAM,CAAC,CAAA;AACpB,MAAA,MAAM,QAAQ,IAAA,CAAK,GAAA,CAAI,KAAK,IAAA,CAAK,CAAC,IAAI,KAAK,CAAA;AAE3C,MAAA,IAAI,KAAA,GAAQ,KAAK,oBAAA,EAAsB;AAErC,QAAA,KAAA,CAAM,IAAA,CAAK,IAAA,CAAK,UAAA,CAAW,SAAA,EAAW,SAAS,CAAC,CAAA;AAChD,QAAA,SAAA,GAAY,CAAA;AACZ,QAAA,SAAA,GAAY,CAAC,IAAI,CAAA;AACjB,QAAA,KAAA,GAAQ,IAAA,CAAK,KAAK,CAAC,CAAA;AAAA,MACrB,CAAA,MAAO;AACL,QAAA,SAAA,CAAU,KAAK,IAAI,CAAA;AAAA,MACrB;AAAA,IACF;AAGA,IAAA,IAAI,SAAA,CAAU,SAAS,CAAA,EAAG;AACxB,MAAA,KAAA,CAAM,IAAA,CAAK,IAAA,CAAK,UAAA,CAAW,SAAA,EAAW,SAAS,CAAC,CAAA;AAAA,IAClD;AAEA,IAAA,OAAO,KAAA;AAAA,EACT;AAAA,EAEQ,UAAA,CAAW,OAAsB,UAAA,EAAiC;AACxE,IAAA,MAAM,IAAA,GAAO,MAAM,GAAA,CAAI,CAAC,MAAM,CAAA,CAAE,IAAI,CAAA,CAAE,IAAA,CAAK,GAAG,CAAA;AAC9C,IAAA,MAAM,QAAQ,KAAA,CAAM,GAAA,CAAI,CAAC,CAAA,KAAM,EAAE,IAAI,CAAA;AACrC,IAAA,MAAM,IAAA,GAAgB;AAAA,MACpB,IAAA,CAAK,GAAA,CAAI,GAAG,KAAA,CAAM,GAAA,CAAI,CAAC,CAAA,KAAM,CAAA,CAAE,CAAC,CAAC,CAAC,CAAA;AAAA,MAClC,IAAA,CAAK,GAAA,CAAI,GAAG,KAAA,CAAM,GAAA,CAAI,CAAC,CAAA,KAAM,CAAA,CAAE,CAAC,CAAC,CAAC,CAAA;AAAA,MAClC,IAAA,CAAK,GAAA,CAAI,GAAG,KAAA,CAAM,GAAA,CAAI,CAAC,CAAA,KAAM,CAAA,CAAE,CAAC,CAAC,CAAC,CAAA;AAAA,MAClC,IAAA,CAAK,GAAA,CAAI,GAAG,KAAA,CAAM,GAAA,CAAI,CAAC,CAAA,KAAM,CAAA,CAAE,CAAC,CAAC,CAAC;AAAA,KACpC;AACA,IAAA,OAAO;AAAA,MACL,IAAA;AAAA,MACA,IAAA;AAAA,MACA,aAAa,CAAC,UAAA,EAAY,UAAA,GAAa,KAAA,CAAM,SAAS,CAAC;AAAA,KACzD;AAAA,EACF;AAAA;AAAA,EAGQ,aAAa,KAAA,EAA+B;AAClD,IAAA,IAAI,OAAO,QAAA,EACT,IAAA,GAAO,QAAA,EACP,IAAA,GAAO,WACP,IAAA,GAAO,CAAA,QAAA;AAET,IAAA,KAAA,MAAW,KAAK,KAAA,EAAO;AACrB,MAAA,IAAI,CAAA,CAAE,KAAK,CAAC,CAAA,GAAI,MAAM,IAAA,GAAO,CAAA,CAAE,KAAK,CAAC,CAAA;AACrC,MAAA,IAAI,CAAA,CAAE,KAAK,CAAC,CAAA,GAAI,MAAM,IAAA,GAAO,CAAA,CAAE,KAAK,CAAC,CAAA;AACrC,MAAA,IAAI,CAAA,CAAE,KAAK,CAAC,CAAA,GAAI,MAAM,IAAA,GAAO,CAAA,CAAE,KAAK,CAAC,CAAA;AACrC,MAAA,IAAI,CAAA,CAAE,KAAK,CAAC,CAAA,GAAI,MAAM,IAAA,GAAO,CAAA,CAAE,KAAK,CAAC,CAAA;AAAA,IACvC;AAEA,IAAA,OAAO,CAAC,IAAA,EAAM,IAAA,EAAM,IAAA,EAAM,IAAI,CAAA;AAAA,EAChC;AACF;;;AC7JO,IAAM,eAAN,MAAmB;AAAA,EAAnB,WAAA,GAAA;AACL,IAAA,IAAA,CAAQ,YAAA,uBAAmB,GAAA,EAA4B;AACvD,IAAA,IAAA,CAAQ,UAAA,GAA2B;AAAA,MACjC,SAAS,EAAC;AAAA,MACV,UAAA,EAAY,CAAA;AAAA,MACZ,WAAA,EAAa;AAAA,KACf;AAAA,EAAA;AAAA;AAAA,EAGA,kBAAA,CAAmB,WAAmB,WAAA,EAAmC;AACvE,IAAA,IAAA,CAAK,YAAA,CAAa,GAAA,CAAI,SAAA,EAAW,WAAW,CAAA;AAAA,EAC9C;AAAA;AAAA,EAGA,qBAAqB,SAAA,EAAyB;AAC5C,IAAA,IAAA,CAAK,YAAA,CAAa,OAAO,SAAS,CAAA;AAAA,EACpC;AAAA;AAAA,EAGA,QAAA,GAAiB;AACf,IAAA,IAAA,CAAK,aAAa,KAAA,EAAM;AACxB,IAAA,IAAA,CAAK,UAAA,GAAa,EAAE,OAAA,EAAS,IAAI,UAAA,EAAY,CAAA,EAAG,aAAa,EAAA,EAAG;AAAA,EAClE;AAAA;AAAA,EAGA,MAAA,CAAO,KAAA,EAAe,OAAA,GAAyB,EAAC,EAAiB;AAC/D,IAAA,IAAI,CAAC,KAAA,EAAO;AACV,MAAA,IAAA,CAAK,UAAA,GAAa,EAAE,OAAA,EAAS,IAAI,UAAA,EAAY,CAAA,EAAG,aAAa,EAAA,EAAG;AAChE,MAAA,OAAO,IAAA,CAAK,UAAA;AAAA,IACd;AAEA,IAAA,MAAM,UAAyB,EAAC;AAGhC,IAAA,MAAM,cAAc,CAAC,GAAG,KAAK,YAAA,CAAa,OAAA,EAAS,CAAA,CAAE,IAAA;AAAA,MACnD,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,MAAM,CAAA,GAAI;AAAA,KACpB;AAEA,IAAA,KAAA,MAAW,CAAC,SAAA,EAAW,WAAW,CAAA,IAAK,WAAA,EAAa;AAClD,MAAA,MAAM,cAAc,IAAA,CAAK,UAAA;AAAA,QACvB,SAAA;AAAA,QACA,WAAA;AAAA,QACA,KAAA;AAAA,QACA;AAAA,OACF;AACA,MAAA,OAAA,CAAQ,IAAA,CAAK,GAAG,WAAW,CAAA;AAAA,IAC7B;AAEA,IAAA,IAAA,CAAK,UAAA,GAAa;AAAA,MAChB,OAAA;AAAA,MACA,YAAY,OAAA,CAAQ,MAAA;AAAA,MACpB,WAAA,EAAa,OAAA,CAAQ,MAAA,GAAS,CAAA,GAAI,CAAA,GAAI;AAAA,KACxC;AAEA,IAAA,OAAO,IAAA,CAAK,UAAA;AAAA,EACd;AAAA;AAAA,EAGA,SAAA,GAAgC;AAC9B,IAAA,IAAI,IAAA,CAAK,UAAA,CAAW,UAAA,KAAe,CAAA,EAAG,OAAO,IAAA;AAE7C,IAAA,IAAA,CAAK,WAAW,WAAA,GAAA,CACb,IAAA,CAAK,WAAW,WAAA,GAAc,CAAA,IAAK,KAAK,UAAA,CAAW,UAAA;AACtD,IAAA,OAAO,KAAK,UAAA,CAAW,OAAA,CAAQ,IAAA,CAAK,UAAA,CAAW,WAAW,CAAA,IAAK,IAAA;AAAA,EACjE;AAAA;AAAA,EAGA,SAAA,GAAgC;AAC9B,IAAA,IAAI,IAAA,CAAK,UAAA,CAAW,UAAA,KAAe,CAAA,EAAG,OAAO,IAAA;AAE7C,IAAA,IAAA,CAAK,UAAA,CAAW,WAAA,GAAA,CACb,IAAA,CAAK,UAAA,CAAW,WAAA,GAAc,IAAI,IAAA,CAAK,UAAA,CAAW,UAAA,IACnD,IAAA,CAAK,UAAA,CAAW,UAAA;AAClB,IAAA,OAAO,KAAK,UAAA,CAAW,OAAA,CAAQ,IAAA,CAAK,UAAA,CAAW,WAAW,CAAA,IAAK,IAAA;AAAA,EACjE;AAAA;AAAA,EAGA,cAAA,GAAqC;AACnC,IAAA,IACE,IAAA,CAAK,WAAW,WAAA,GAAc,CAAA,IAC9B,KAAK,UAAA,CAAW,WAAA,IAAe,IAAA,CAAK,UAAA,CAAW,UAAA,EAC/C;AACA,MAAA,OAAO,IAAA;AAAA,IACT;AACA,IAAA,OAAO,KAAK,UAAA,CAAW,OAAA,CAAQ,IAAA,CAAK,UAAA,CAAW,WAAW,CAAA,IAAK,IAAA;AAAA,EACjE;AAAA;AAAA,EAGA,aAAA,GAA8B;AAC5B,IAAA,OAAO,IAAA,CAAK,UAAA;AAAA,EACd;AAAA,EAEQ,UAAA,CACN,SAAA,EACA,WAAA,EACA,KAAA,EACA,OAAA,EACe;AACf,IAAA,MAAM,UAAyB,EAAC;AAGhC,IAAA,MAAM,QAAA,GAAW,WAAA,CAAY,KAAA,CAAM,GAAA,CAAI,CAAC,MAAM,CAAA,CAAE,IAAI,CAAA,CAAE,IAAA,CAAK,IAAI,CAAA;AAE/D,IAAA,IAAI,OAAA;AACJ,IAAA,IAAI;AACF,MAAA,IAAI,QAAQ,KAAA,EAAO;AACjB,QAAA,MAAM,KAAA,GAAQ,OAAA,CAAQ,aAAA,GAAgB,GAAA,GAAM,IAAA;AAC5C,QAAA,OAAA,GAAU,IAAI,MAAA,CAAO,KAAA,EAAO,KAAK,CAAA;AAAA,MACnC,CAAA,MAAO;AACL,QAAA,MAAM,OAAA,GAAU,KAAA,CAAM,OAAA,CAAQ,qBAAA,EAAuB,MAAM,CAAA;AAC3D,QAAA,MAAM,KAAA,GAAQ,OAAA,CAAQ,aAAA,GAAgB,GAAA,GAAM,IAAA;AAC5C,QAAA,MAAM,IAAA,GAAO,OAAA,CAAQ,SAAA,GAAY,CAAA,GAAA,EAAM,OAAO,CAAA,GAAA,CAAA,GAAQ,OAAA;AACtD,QAAA,OAAA,GAAU,IAAI,MAAA,CAAO,IAAA,EAAM,KAAK,CAAA;AAAA,MAClC;AAAA,IACF,CAAA,CAAA,MAAQ;AACN,MAAA,OAAO,OAAA;AAAA,IACT;AAEA,IAAA,IAAI,KAAA;AACJ,IAAA,OAAA,CAAQ,KAAA,GAAQ,OAAA,CAAQ,IAAA,CAAK,QAAQ,OAAO,IAAA,EAAM;AAChD,MAAA,MAAM,SAAA,GAAY,MAAM,CAAC,CAAA;AACzB,MAAA,MAAM,aAAa,KAAA,CAAM,KAAA;AACzB,MAAA,MAAM,QAAA,GAAW,aAAa,SAAA,CAAU,MAAA;AAGxC,MAAA,MAAM,KAAA,GAAQ,IAAA,CAAK,cAAA,CAAe,WAAA,EAAa,YAAY,QAAQ,CAAA;AAEnE,MAAA,OAAA,CAAQ,IAAA,CAAK;AAAA,QACX,SAAA;AAAA,QACA,SAAA,EAAW,UAAA;AAAA,QACX,OAAA,EAAS,QAAA;AAAA,QACT,KAAA;AAAA,QACA,IAAA,EAAM;AAAA,OACP,CAAA;AAGD,MAAA,IAAI,SAAA,CAAU,WAAW,CAAA,EAAG;AAC1B,QAAA,OAAA,CAAQ,SAAA,EAAA;AAAA,MACV;AAAA,IACF;AAEA,IAAA,OAAO,OAAA;AAAA,EACT;AAAA;AAAA,EAGQ,cAAA,CACN,WAAA,EACA,MAAA,EACA,IAAA,EACW;AAGX,IAAA,MAAM,QAAmB,EAAC;AAG1B,IAAA,IAAI,UAAA,GAAa,CAAA;AACjB,IAAA,KAAA,MAAW,IAAA,IAAQ,YAAY,KAAA,EAAO;AACpC,MAAA,MAAM,OAAA,GAAU,UAAA,GAAa,IAAA,CAAK,IAAA,CAAK,MAAA;AAEvC,MAAA,IAAI,OAAA,GAAU,MAAA,IAAU,UAAA,GAAa,IAAA,EAAM;AAEzC,QAAA,KAAA,IAAS,EAAA,GAAK,IAAA,CAAK,WAAA,CAAY,CAAC,CAAA,EAAG,MAAM,IAAA,CAAK,WAAA,CAAY,CAAC,CAAA,EAAG,EAAA,EAAA,EAAM;AAClE,UAAA,MAAM,IAAA,GAAO,WAAA,CAAY,KAAA,CAAM,EAAE,CAAA;AACjC,UAAA,IAAI,IAAA,EAAM;AACR,YAAA,KAAA,CAAM,IAAA,CAAK,KAAK,IAAI,CAAA;AAAA,UACtB;AAAA,QACF;AAAA,MACF;AAEA,MAAA,UAAA,GAAa,OAAA,GAAU,CAAA;AAAA,IACzB;AAEA,IAAA,OAAO,KAAA,CAAM,MAAA,GAAS,CAAA,GAAI,KAAA,GAAQ,CAAC,CAAC,CAAA,EAAG,CAAA,EAAG,CAAA,EAAG,CAAC,CAAC,CAAA;AAAA,EACjD;AACF;;;AC1KA,IAAM,aAAA,GAAiC;AAAA,EACrC,WAAA,EAAa,EAAA;AAAA,EACb,eAAe,EAAC;AAAA,EAChB,YAAA,EAAc,IAAA;AAAA,EACd,WAAA,EAAa;AACf,CAAA;AAGO,SAAS,gBAAA,GAA8B;AAC5C,EAAA,OAAO;AAAA,IACL,EAAA,EAAI,MAAA;AAAA,IACJ,IAAA,EAAM,0BAAA;AAAA,IACN,OAAA,EAAS,OAAA;AAAA,IAET,QAAQ,OAAA,EAA8C;AACpD,MAAA,MAAM,SAAA,GAAY,IAAI,aAAA,EAAc;AACpC,MAAA,MAAM,YAAA,GAAe,IAAI,YAAA,EAAa;AAGtC,MAAA,OAAA,CAAQ,aAAA,CAA+B,MAAA,EAAQ,EAAE,GAAG,eAAe,CAAA;AAInE,MAAA,MAAM,YAAA,GAAe,QAAQ,UAAA,CAAW,eAAA;AAAA,QACtC,cAAA;AAAA,QACA,CAAC,OAAA,KAAmC;AAClC,UAAA,MAAM,KAAA,GAAQ,OAAA,CAAQ,KAAA,CAAM,QAAA,EAAS;AACrC,UAAA,MAAM,IAAA,GAAO,KAAA,CAAM,KAAA,CAAM,OAAA,CAAQ,SAAS,CAAA;AAC1C,UAAA,IAAI,CAAC,IAAA,EAAM;AAGX,UAAA,MAAM,QAAQ,SAAA,CAAU,oBAAA;AAAA,YACtB,CAAA,qBAAA,EAAwB,OAAA,CAAQ,SAAA,GAAY,CAAC,CAAA,CAAA;AAAA,YAC7C,EAAA;AAAA;AAAA,YACA,EAAA;AAAA,YACA;AAAA,WACF;AAEA,UAAA,MAAM,WAAA,GAAc,SAAA,CAAU,OAAA,CAAQ,KAAA,EAAO,IAAI,CAAA;AAGjD,UAAA,OAAA,CAAQ,KAAA,CAAM,QAAA,CAAS,CAAC,IAAA,KAAS;AAC/B,YAAA,MAAM,KAAA,GAAQ,CAAC,GAAG,IAAA,CAAK,KAAK,CAAA;AAC5B,YAAA,MAAM,QAAA,GAAW,KAAA,CAAM,OAAA,CAAQ,SAAS,CAAA;AACxC,YAAA,IAAI,QAAA,EAAU;AACZ,cAAA,KAAA,CAAM,QAAQ,SAAS,CAAA,GAAI,EAAE,GAAG,UAAU,WAAA,EAAY;AAAA,YACxD;AACA,YAAA,OAAO,EAAE,GAAG,IAAA,EAAM,KAAA,EAAM;AAAA,UAC1B,CAAC,CAAA;AAGD,UAAA,YAAA,CAAa,kBAAA,CAAmB,OAAA,CAAQ,SAAA,EAAW,WAAW,CAAA;AAE9D,UAAA,OAAA,CAAQ,QAAA,CAAS,KAAK,gBAAA,EAAkB;AAAA,YACtC,WAAW,OAAA,CAAQ,SAAA;AAAA,YACnB;AAAA,WACD,CAAA;AAAA,QACH;AAAA,OACF;AAEA,MAAA,MAAM,WAAA,GAAc,QAAQ,UAAA,CAAW,eAAA;AAAA,QACrC,aAAA;AAAA,QACA,CAAC,OAAA,KAAwD;AACvD,UAAA,MAAM,SAAS,YAAA,CAAa,MAAA,CAAO,OAAA,CAAQ,KAAA,EAAO,QAAQ,OAAO,CAAA;AAEjE,UAAA,OAAA,CAAQ,QAAA,CAA0B,MAAA,EAAQ,CAAC,IAAA,MAAU;AAAA,YACnD,GAAG,IAAA;AAAA,YACH,aAAa,OAAA,CAAQ,KAAA;AAAA,YACrB,aAAA,EAAe,OAAA,CAAQ,OAAA,IAAW,EAAC;AAAA,YACnC,YAAA,EAAc,MAAA;AAAA,YACd,WAAA,EAAa,aAAa,cAAA;AAAe,WAC3C,CAAE,CAAA;AAEF,UAAA,OAAA,CAAQ,QAAA,CAAS,KAAK,cAAA,EAAgB;AAAA,YACpC,OAAO,OAAA,CAAQ,KAAA;AAAA,YACf,SAAS,MAAA,CAAO,OAAA;AAAA,YAChB,OAAO,MAAA,CAAO;AAAA,WACf,CAAA;AAAA,QACH;AAAA,OACF;AAEA,MAAA,MAAM,eAAA,GAAkB,QAAQ,UAAA,CAAW,eAAA;AAAA,QACzC,iBAAA;AAAA,QACA,MAAM;AACJ,UAAA,MAAM,KAAA,GAAQ,aAAa,SAAA,EAAU;AACrC,UAAA,OAAA,CAAQ,QAAA,CAA0B,MAAA,EAAQ,CAAC,IAAA,MAAU;AAAA,YACnD,GAAG,IAAA;AAAA,YACH,WAAA,EAAa,KAAA;AAAA,YACb,YAAA,EAAc,aAAa,aAAA;AAAc,WAC3C,CAAE,CAAA;AAEF,UAAA,IAAI,KAAA,EAAO;AAET,YAAA,OAAA,CAAQ,GAAA,CAAI,QAAA,CAAS,KAAA,CAAM,SAAS,CAAA;AAAA,UACtC;AAAA,QACF;AAAA,OACF;AAEA,MAAA,MAAM,eAAA,GAAkB,QAAQ,UAAA,CAAW,eAAA;AAAA,QACzC,iBAAA;AAAA,QACA,MAAM;AACJ,UAAA,MAAM,KAAA,GAAQ,aAAa,SAAA,EAAU;AACrC,UAAA,OAAA,CAAQ,QAAA,CAA0B,MAAA,EAAQ,CAAC,IAAA,MAAU;AAAA,YACnD,GAAG,IAAA;AAAA,YACH,WAAA,EAAa,KAAA;AAAA,YACb,YAAA,EAAc,aAAa,aAAA;AAAc,WAC3C,CAAE,CAAA;AAEF,UAAA,IAAI,KAAA,EAAO;AACT,YAAA,OAAA,CAAQ,GAAA,CAAI,QAAA,CAAS,KAAA,CAAM,SAAS,CAAA;AAAA,UACtC;AAAA,QACF;AAAA,OACF;AAIA,MAAA,MAAM,SAAA,GAA2B;AAAA,QAC/B,gBAAgB,SAAA,EAAmB;AACjC,UAAA,OAAA,CAAQ,UAAA,CAAW,QAAA,CAAS,cAAA,EAAgB,EAAE,WAAW,CAAA;AAAA,QAC3D,CAAA;AAAA,QACA,cAAA,GAAiB;AACf,UAAA,MAAM,KAAA,GAAQ,OAAA,CAAQ,KAAA,CAAM,QAAA,EAAS;AACrC,UAAA,KAAA,IAAS,IAAI,CAAA,EAAG,CAAA,GAAI,KAAA,CAAM,KAAA,CAAM,QAAQ,CAAA,EAAA,EAAK;AAC3C,YAAA,OAAA,CAAQ,WAAW,QAAA,CAAS,cAAA,EAAgB,EAAE,SAAA,EAAW,GAAG,CAAA;AAAA,UAC9D;AAAA,QACF,CAAA;AAAA,QACA,MAAA,CAAO,OAAe,OAAA,EAAyB;AAC7C,UAAA,OAAA,CAAQ,WAAW,QAAA,CAAS,aAAA,EAAe,EAAE,KAAA,EAAO,SAAS,CAAA;AAC7D,UAAA,OAAO,aAAa,aAAA,EAAc;AAAA,QACpC,CAAA;AAAA,QACA,UAAA,GAAa;AACX,UAAA,OAAA,CAAQ,UAAA,CAAW,QAAA,CAAS,iBAAA,EAAmB,EAAE,CAAA;AACjD,UAAA,OAAO,aAAa,cAAA,EAAe;AAAA,QACrC,CAAA;AAAA,QACA,UAAA,GAAa;AACX,UAAA,OAAA,CAAQ,UAAA,CAAW,QAAA,CAAS,iBAAA,EAAmB,EAAE,CAAA;AACjD,UAAA,OAAO,aAAa,cAAA,EAAe;AAAA,QACrC,CAAA;AAAA,QACA,WAAA,GAAc;AACZ,UAAA,YAAA,CAAa,QAAA,EAAS;AACtB,UAAA,OAAA,CAAQ,QAAA,CAA0B,QAAQ,OAAO;AAAA,YAC/C,GAAG;AAAA,WACL,CAAE,CAAA;AAAA,QACJ,CAAA;AAAA,QACA,YAAA,GAAe;AACb,UAAA,OAAO,SAAA;AAAA,QACT,CAAA;AAAA,QACA,eAAA,GAAkB;AAChB,UAAA,OAAO,YAAA;AAAA,QACT;AAAA,OACF;AAGA,MAAC,QAAgB,UAAA,GAAa,SAAA;AAE9B,MAAA,OAAO,MAAM;AACX,QAAA,YAAA,EAAa;AACb,QAAA,WAAA,EAAY;AACZ,QAAA,eAAA,EAAgB;AAChB,QAAA,eAAA,EAAgB;AAChB,QAAA,YAAA,CAAa,QAAA,EAAS;AAAA,MACxB,CAAA;AAAA,IACF;AAAA,GACF;AACF","file":"index.js","sourcesContent":["// ─── Text Extractor ───\n//\n// Extracts text content from PDF pages with character, word, and line bounding boxes.\n// Phase 1 implementation uses synthetic extraction; Phase 2 will integrate pdf.js getTextContent().\n\nimport type {\n  PdfTextContent,\n  PdfCharInfo,\n  PdfWordInfo,\n  PdfLineInfo,\n  PdfRect,\n  PdfPageState,\n} from '@gridstorm/pdf-core';\n\n/** Configuration for text extraction. */\nexport interface TextExtractorConfig {\n  /** Threshold for word break detection (in PDF points). */\n  wordSpacingThreshold?: number;\n  /** Threshold for line break detection (in PDF points). */\n  lineSpacingThreshold?: number;\n}\n\nconst DEFAULT_WORD_SPACING = 3;\nconst DEFAULT_LINE_SPACING = 5;\n\n/** Extracts structured text content from PDF text items. */\nexport class TextExtractor {\n  private wordSpacingThreshold: number;\n  private lineSpacingThreshold: number;\n\n  constructor(config: TextExtractorConfig = {}) {\n    this.wordSpacingThreshold = config.wordSpacingThreshold ?? DEFAULT_WORD_SPACING;\n    this.lineSpacingThreshold = config.lineSpacingThreshold ?? DEFAULT_LINE_SPACING;\n  }\n\n  /** Extract text content from raw text items.\n   *  In Phase 2 this will accept pdf.js TextItem[]; for now it works with PdfCharInfo[]. */\n  extract(chars: PdfCharInfo[], _page: PdfPageState): PdfTextContent {\n    if (chars.length === 0) {\n      return { chars, words: [], lines: [] };\n    }\n\n    const words = this.segmentWords(chars);\n    const lines = this.segmentLines(words, chars);\n\n    return { chars, words, lines };\n  }\n\n  /** Build PdfCharInfo from a simple text string (for testing/placeholder). */\n  buildCharsFromString(\n    text: string,\n    startX: number,\n    startY: number,\n    fontSize: number,\n    fontName = 'Helvetica',\n  ): PdfCharInfo[] {\n    const chars: PdfCharInfo[] = [];\n    let x = startX;\n    const charWidth = fontSize * 0.6; // Approximate monospace width\n\n    for (let i = 0; i < text.length; i++) {\n      const char = text[i]!;\n      const rect: PdfRect = [x, startY, x + charWidth, startY + fontSize];\n\n      chars.push({\n        char,\n        rect,\n        fontName,\n        fontSize,\n        transform: [fontSize, 0, 0, fontSize, x, startY],\n      });\n\n      x += charWidth;\n    }\n\n    return chars;\n  }\n\n  /** Segment characters into words based on spacing. */\n  private segmentWords(chars: PdfCharInfo[]): PdfWordInfo[] {\n    if (chars.length === 0) return [];\n\n    const words: PdfWordInfo[] = [];\n    let wordStart = 0;\n    let wordChars: PdfCharInfo[] = [chars[0]!];\n\n    for (let i = 1; i < chars.length; i++) {\n      const prev = chars[i - 1]!;\n      const curr = chars[i]!;\n\n      const gap = curr.rect[0] - prev.rect[2]; // x1 of current - x2 of previous\n      const isSpace = curr.char === ' ' || prev.char === ' ';\n      const isNewLine = Math.abs(curr.rect[1] - prev.rect[1]) > this.lineSpacingThreshold;\n      const isWordBreak = gap > this.wordSpacingThreshold || isSpace || isNewLine;\n\n      if (isWordBreak) {\n        // Complete current word (skip if only spaces)\n        const text = wordChars.map((c) => c.char).join('').trim();\n        if (text.length > 0) {\n          words.push({\n            text,\n            rect: this.boundingRect(wordChars),\n            charIndices: [wordStart, wordStart + wordChars.length - 1],\n          });\n        }\n\n        // Skip space characters\n        if (curr.char !== ' ') {\n          wordStart = i;\n          wordChars = [curr];\n        } else {\n          wordStart = i + 1;\n          wordChars = [];\n        }\n      } else {\n        wordChars.push(curr);\n      }\n    }\n\n    // Final word\n    if (wordChars.length > 0) {\n      const text = wordChars.map((c) => c.char).join('').trim();\n      if (text.length > 0) {\n        words.push({\n          text,\n          rect: this.boundingRect(wordChars),\n          charIndices: [wordStart, wordStart + wordChars.length - 1],\n        });\n      }\n    }\n\n    return words;\n  }\n\n  /** Segment words into lines based on vertical position. */\n  private segmentLines(words: PdfWordInfo[], _chars: PdfCharInfo[]): PdfLineInfo[] {\n    if (words.length === 0) return [];\n\n    const lines: PdfLineInfo[] = [];\n    let lineStart = 0;\n    let lineWords: PdfWordInfo[] = [words[0]!];\n    let lineY = words[0]!.rect[1]; // y1 of first word\n\n    for (let i = 1; i < words.length; i++) {\n      const word = words[i]!;\n      const yDiff = Math.abs(word.rect[1] - lineY);\n\n      if (yDiff > this.lineSpacingThreshold) {\n        // New line\n        lines.push(this.createLine(lineWords, lineStart));\n        lineStart = i;\n        lineWords = [word];\n        lineY = word.rect[1];\n      } else {\n        lineWords.push(word);\n      }\n    }\n\n    // Final line\n    if (lineWords.length > 0) {\n      lines.push(this.createLine(lineWords, lineStart));\n    }\n\n    return lines;\n  }\n\n  private createLine(words: PdfWordInfo[], startIndex: number): PdfLineInfo {\n    const text = words.map((w) => w.text).join(' ');\n    const rects = words.map((w) => w.rect);\n    const rect: PdfRect = [\n      Math.min(...rects.map((r) => r[0])),\n      Math.min(...rects.map((r) => r[1])),\n      Math.max(...rects.map((r) => r[2])),\n      Math.max(...rects.map((r) => r[3])),\n    ];\n    return {\n      text,\n      rect,\n      wordIndices: [startIndex, startIndex + words.length - 1],\n    };\n  }\n\n  /** Compute bounding rect for a set of characters. */\n  private boundingRect(chars: PdfCharInfo[]): PdfRect {\n    let minX = Infinity,\n      minY = Infinity,\n      maxX = -Infinity,\n      maxY = -Infinity;\n\n    for (const c of chars) {\n      if (c.rect[0] < minX) minX = c.rect[0];\n      if (c.rect[1] < minY) minY = c.rect[1];\n      if (c.rect[2] > maxX) maxX = c.rect[2];\n      if (c.rect[3] > maxY) maxY = c.rect[3];\n    }\n\n    return [minX, minY, maxX, maxY];\n  }\n}\n","// ─── Search Engine ───\n//\n// Full-text search across PDF pages with match navigation.\n\nimport type { PdfTextContent, PdfRect } from '@gridstorm/pdf-core';\n\n/** Search options. */\nexport interface SearchOptions {\n  /** Case-sensitive search. */\n  caseSensitive?: boolean;\n  /** Match whole words only. */\n  wholeWord?: boolean;\n  /** Use regex pattern. */\n  regex?: boolean;\n}\n\n/** A single search match. */\nexport interface SearchMatch {\n  /** Page index where the match was found. */\n  pageIndex: number;\n  /** Start character index within the page text. */\n  charStart: number;\n  /** End character index (exclusive). */\n  charEnd: number;\n  /** Bounding rectangles covering the match (may span multiple lines). */\n  rects: PdfRect[];\n  /** The matched text. */\n  text: string;\n}\n\n/** Search result set. */\nexport interface SearchResult {\n  /** All matches across all pages. */\n  matches: SearchMatch[];\n  /** Total match count. */\n  totalCount: number;\n  /** Currently active match index (-1 if none). */\n  activeIndex: number;\n}\n\n/** Search engine for finding text in PDF pages. */\nexport class SearchEngine {\n  private textContents = new Map<number, PdfTextContent>();\n  private lastResult: SearchResult = {\n    matches: [],\n    totalCount: 0,\n    activeIndex: -1,\n  };\n\n  /** Set the text content for a page. */\n  setPageTextContent(pageIndex: number, textContent: PdfTextContent): void {\n    this.textContents.set(pageIndex, textContent);\n  }\n\n  /** Clear text content for a page. */\n  clearPageTextContent(pageIndex: number): void {\n    this.textContents.delete(pageIndex);\n  }\n\n  /** Clear all cached text content. */\n  clearAll(): void {\n    this.textContents.clear();\n    this.lastResult = { matches: [], totalCount: 0, activeIndex: -1 };\n  }\n\n  /** Search for a query across all loaded pages. */\n  search(query: string, options: SearchOptions = {}): SearchResult {\n    if (!query) {\n      this.lastResult = { matches: [], totalCount: 0, activeIndex: -1 };\n      return this.lastResult;\n    }\n\n    const matches: SearchMatch[] = [];\n\n    // Sort pages by index for consistent ordering\n    const sortedPages = [...this.textContents.entries()].sort(\n      ([a], [b]) => a - b,\n    );\n\n    for (const [pageIndex, textContent] of sortedPages) {\n      const pageMatches = this.searchPage(\n        pageIndex,\n        textContent,\n        query,\n        options,\n      );\n      matches.push(...pageMatches);\n    }\n\n    this.lastResult = {\n      matches,\n      totalCount: matches.length,\n      activeIndex: matches.length > 0 ? 0 : -1,\n    };\n\n    return this.lastResult;\n  }\n\n  /** Navigate to the next search match. */\n  nextMatch(): SearchMatch | null {\n    if (this.lastResult.totalCount === 0) return null;\n\n    this.lastResult.activeIndex =\n      (this.lastResult.activeIndex + 1) % this.lastResult.totalCount;\n    return this.lastResult.matches[this.lastResult.activeIndex] ?? null;\n  }\n\n  /** Navigate to the previous search match. */\n  prevMatch(): SearchMatch | null {\n    if (this.lastResult.totalCount === 0) return null;\n\n    this.lastResult.activeIndex =\n      (this.lastResult.activeIndex - 1 + this.lastResult.totalCount) %\n      this.lastResult.totalCount;\n    return this.lastResult.matches[this.lastResult.activeIndex] ?? null;\n  }\n\n  /** Get the current active match. */\n  getActiveMatch(): SearchMatch | null {\n    if (\n      this.lastResult.activeIndex < 0 ||\n      this.lastResult.activeIndex >= this.lastResult.totalCount\n    ) {\n      return null;\n    }\n    return this.lastResult.matches[this.lastResult.activeIndex] ?? null;\n  }\n\n  /** Get the last search result. */\n  getLastResult(): SearchResult {\n    return this.lastResult;\n  }\n\n  private searchPage(\n    pageIndex: number,\n    textContent: PdfTextContent,\n    query: string,\n    options: SearchOptions,\n  ): SearchMatch[] {\n    const matches: SearchMatch[] = [];\n\n    // Build full page text from lines\n    const fullText = textContent.lines.map((l) => l.text).join('\\n');\n\n    let pattern: RegExp;\n    try {\n      if (options.regex) {\n        const flags = options.caseSensitive ? 'g' : 'gi';\n        pattern = new RegExp(query, flags);\n      } else {\n        const escaped = query.replace(/[.*+?^${}()|[\\]\\\\]/g, '\\\\$&');\n        const flags = options.caseSensitive ? 'g' : 'gi';\n        const word = options.wholeWord ? `\\\\b${escaped}\\\\b` : escaped;\n        pattern = new RegExp(word, flags);\n      }\n    } catch {\n      return matches;\n    }\n\n    let match: RegExpExecArray | null;\n    while ((match = pattern.exec(fullText)) !== null) {\n      const matchText = match[0];\n      const matchStart = match.index;\n      const matchEnd = matchStart + matchText.length;\n\n      // Find covering word rects\n      const rects = this.findMatchRects(textContent, matchStart, matchEnd);\n\n      matches.push({\n        pageIndex,\n        charStart: matchStart,\n        charEnd: matchEnd,\n        rects,\n        text: matchText,\n      });\n\n      // Avoid infinite loop for zero-length matches\n      if (matchText.length === 0) {\n        pattern.lastIndex++;\n      }\n    }\n\n    return matches;\n  }\n\n  /** Find bounding rects for a text range. */\n  private findMatchRects(\n    textContent: PdfTextContent,\n    _start: number,\n    _end: number,\n  ): PdfRect[] {\n    // In Phase 1 we use word-level rects; Phase 2 will use char-level precision.\n    // For now, return rects for all words that overlap the match range.\n    const rects: PdfRect[] = [];\n\n    // Simple approach: iterate lines and match by character offset\n    let charOffset = 0;\n    for (const line of textContent.lines) {\n      const lineEnd = charOffset + line.text.length;\n\n      if (lineEnd > _start && charOffset < _end) {\n        // This line overlaps the match — use word-level rects within the line\n        for (let wi = line.wordIndices[0]; wi <= line.wordIndices[1]; wi++) {\n          const word = textContent.words[wi];\n          if (word) {\n            rects.push(word.rect);\n          }\n        }\n      }\n\n      charOffset = lineEnd + 1; // +1 for \\n\n    }\n\n    return rects.length > 0 ? rects : [[0, 0, 0, 0]];\n  }\n}\n","// ─── Text Plugin ───\n//\n// GridStorm PDF plugin for text extraction and search.\n\nimport type {\n  PdfPlugin,\n  PdfPluginContext,\n  PdfPluginDisposer,\n} from '@gridstorm/pdf-core';\nimport { TextExtractor } from './text-extractor';\nimport { SearchEngine } from './search-engine';\nimport type { SearchOptions, SearchResult, SearchMatch } from './search-engine';\n\n/** Plugin state for text operations. */\nexport interface TextPluginState {\n  /** Current search query. */\n  searchQuery: string;\n  /** Search options. */\n  searchOptions: SearchOptions;\n  /** Last search result. */\n  searchResult: SearchResult | null;\n  /** Active search match. */\n  activeMatch: SearchMatch | null;\n}\n\n/** Public API exposed by the text plugin. */\nexport interface TextPluginApi {\n  /** Extract text from a specific page. */\n  extractPageText(pageIndex: number): void;\n  /** Extract text from all pages. */\n  extractAllText(): void;\n  /** Search for text across all pages. */\n  search(query: string, options?: SearchOptions): SearchResult;\n  /** Navigate to next search match. */\n  searchNext(): SearchMatch | null;\n  /** Navigate to previous search match. */\n  searchPrev(): SearchMatch | null;\n  /** Clear search results. */\n  clearSearch(): void;\n  /** Get the text extractor instance. */\n  getExtractor(): TextExtractor;\n  /** Get the search engine instance. */\n  getSearchEngine(): SearchEngine;\n}\n\nconst INITIAL_STATE: TextPluginState = {\n  searchQuery: '',\n  searchOptions: {},\n  searchResult: null,\n  activeMatch: null,\n};\n\n/** Create the text extraction and search plugin. */\nexport function createTextPlugin(): PdfPlugin {\n  return {\n    id: 'text',\n    name: 'Text Extraction & Search',\n    version: '0.1.0',\n\n    install(context: PdfPluginContext): PdfPluginDisposer {\n      const extractor = new TextExtractor();\n      const searchEngine = new SearchEngine();\n\n      // Register plugin state\n      context.registerState<TextPluginState>('text', { ...INITIAL_STATE });\n\n      // ─── Command Handlers ───\n\n      const unsubExtract = context.commandBus.registerHandler(\n        'text:extract',\n        (payload: { pageIndex: number }) => {\n          const state = context.store.getState();\n          const page = state.pages[payload.pageIndex];\n          if (!page) return;\n\n          // In Phase 1, use placeholder chars; Phase 2 will use pdf.js\n          const chars = extractor.buildCharsFromString(\n            `Sample text for page ${payload.pageIndex + 1}`,\n            72, // 1 inch margin\n            72,\n            12,\n          );\n\n          const textContent = extractor.extract(chars, page);\n\n          // Update page state with extracted text\n          context.store.setState((prev) => {\n            const pages = [...prev.pages];\n            const existing = pages[payload.pageIndex];\n            if (existing) {\n              pages[payload.pageIndex] = { ...existing, textContent };\n            }\n            return { ...prev, pages };\n          });\n\n          // Feed text to search engine\n          searchEngine.setPageTextContent(payload.pageIndex, textContent);\n\n          context.eventBus.emit('text:extracted', {\n            pageIndex: payload.pageIndex,\n            textContent,\n          });\n        },\n      );\n\n      const unsubSearch = context.commandBus.registerHandler(\n        'text:search',\n        (payload: { query: string; options?: SearchOptions }) => {\n          const result = searchEngine.search(payload.query, payload.options);\n\n          context.setState<TextPluginState>('text', (prev) => ({\n            ...prev,\n            searchQuery: payload.query,\n            searchOptions: payload.options ?? {},\n            searchResult: result,\n            activeMatch: searchEngine.getActiveMatch(),\n          }));\n\n          context.eventBus.emit('search:found', {\n            query: payload.query,\n            matches: result.matches,\n            total: result.totalCount,\n          });\n        },\n      );\n\n      const unsubSearchNext = context.commandBus.registerHandler(\n        'text:searchNext',\n        () => {\n          const match = searchEngine.nextMatch();\n          context.setState<TextPluginState>('text', (prev) => ({\n            ...prev,\n            activeMatch: match,\n            searchResult: searchEngine.getLastResult(),\n          }));\n\n          if (match) {\n            // Navigate to the match page\n            context.api.goToPage(match.pageIndex);\n          }\n        },\n      );\n\n      const unsubSearchPrev = context.commandBus.registerHandler(\n        'text:searchPrev',\n        () => {\n          const match = searchEngine.prevMatch();\n          context.setState<TextPluginState>('text', (prev) => ({\n            ...prev,\n            activeMatch: match,\n            searchResult: searchEngine.getLastResult(),\n          }));\n\n          if (match) {\n            context.api.goToPage(match.pageIndex);\n          }\n        },\n      );\n\n      // ─── Plugin API ───\n\n      const pluginApi: TextPluginApi = {\n        extractPageText(pageIndex: number) {\n          context.commandBus.dispatch('text:extract', { pageIndex });\n        },\n        extractAllText() {\n          const state = context.store.getState();\n          for (let i = 0; i < state.pages.length; i++) {\n            context.commandBus.dispatch('text:extract', { pageIndex: i });\n          }\n        },\n        search(query: string, options?: SearchOptions) {\n          context.commandBus.dispatch('text:search', { query, options });\n          return searchEngine.getLastResult();\n        },\n        searchNext() {\n          context.commandBus.dispatch('text:searchNext', {});\n          return searchEngine.getActiveMatch();\n        },\n        searchPrev() {\n          context.commandBus.dispatch('text:searchPrev', {});\n          return searchEngine.getActiveMatch();\n        },\n        clearSearch() {\n          searchEngine.clearAll();\n          context.setState<TextPluginState>('text', () => ({\n            ...INITIAL_STATE,\n          }));\n        },\n        getExtractor() {\n          return extractor;\n        },\n        getSearchEngine() {\n          return searchEngine;\n        },\n      };\n\n      // Expose API via plugin context (can be retrieved via api.getPluginApi('text'))\n      (context as any)._pluginApi = pluginApi;\n\n      return () => {\n        unsubExtract();\n        unsubSearch();\n        unsubSearchNext();\n        unsubSearchPrev();\n        searchEngine.clearAll();\n      };\n    },\n  };\n}\n"]}

package/package.json

@@ -0,0 +1,47 @@
+{
+  "name": "@gridstorm/pdf-plugin-text",
+  "version": "0.1.2",
+  "description": "Text extraction and search plugin for GridStorm PDF",
+  "license": "MIT",
+  "type": "module",
+  "main": "./dist/index.cjs",
+  "module": "./dist/index.js",
+  "types": "./dist/index.d.ts",
+  "exports": {
+    ".": {
+      "types": "./dist/index.d.ts",
+      "import": "./dist/index.js",
+      "require": "./dist/index.cjs"
+    }
+  },
+  "files": [
+    "dist"
+  ],
+  "scripts": {
+    "build": "tsup",
+    "dev": "tsup --watch"
+  },
+  "dependencies": {
+    "@gridstorm/pdf-core": "workspace:*"
+  },
+  "devDependencies": {
+    "tsup": "^8.0.0",
+    "typescript": "^5.5.0"
+  },
+  "publishConfig": {
+    "access": "public"
+  },
+  "repository": {
+    "type": "git",
+    "url": "https://github.com/007krcs/grid-data.git",
+    "directory": "packages/pdf-plugin-text"
+  },
+  "homepage": "https://grid-data-analytics-explorer.vercel.app/",
+  "bugs": {
+    "url": "https://github.com/007krcs/grid-data/issues"
+  },
+  "engines": {
+    "node": ">=18.0.0"
+  },
+  "sideEffects": false
+}