@cj-tech-master/excelts 8.1.2 → 9.0.0

package/dist/cjs/modules/excel/utils/cell-format.js

@@ -522,6 +522,35 @@ function formatNumberPattern(val, fmt) {
     const decimalPlaces = decFmt.replace(/[^0#?]/g, "").length;
     // Round the value
     const roundedVal = roundTo(scaledVal, decimalPlaces);
+    // When value is zero and the format has no required '0' digit placeholders,
+    // '?' placeholders become spaces and '#' placeholders produce nothing.
+    // This handles accounting format zero sections like "-"?? → "- " (dash + spaces).
+    if (roundedVal === 0 && !intFmt.includes("0") && !decFmt.includes("0")) {
+        let result = "";
+        for (const ch of intFmt) {
+            if (ch === "?") {
+                result += " ";
+            }
+            else if (ch !== "#" && ch !== ",") {
+                // Preserve literal characters (already unquoted at this point)
+                result += ch;
+            }
+        }
+        if (decimalPlaces > 0) {
+            // Only emit the decimal point if the decimal format has '?' or '0' placeholders.
+            // Pure '#' decimal digits produce nothing for zero values.
+            const hasDecContent = /[0?]/.test(decFmt);
+            if (hasDecContent) {
+                result += ".";
+                for (const ch of decFmt) {
+                    if (ch === "?") {
+                        result += " ";
+                    }
+                }
+            }
+        }
+        return sign + result;
+    }
     // Split into integer and decimal parts
     const [intPart, decPart = ""] = roundedVal.toString().split(".");
     // Check if format has literal characters mixed with digit placeholders (like "0-0", "000-0000")
@@ -561,16 +590,49 @@ function formatNumberPattern(val, fmt) {
         if (intFmt.includes(",")) {
             formattedInt = commaify(intPart);
         }
-        // Pad integer with leading zeros if needed
+        // Pad integer with leading zeros/spaces if needed
+        // '0' placeholder → pad with "0", '?' placeholder → pad with " "
         const minIntDigits = (intFmt.match(/0/g) ?? []).length;
+        const totalIntSlots = (intFmt.match(/[0?]/g) ?? []).length;
         if (formattedInt.length < minIntDigits) {
             formattedInt = "0".repeat(minIntDigits - formattedInt.length) + formattedInt;
         }
+        if (formattedInt.length < totalIntSlots) {
+            formattedInt = " ".repeat(totalIntSlots - formattedInt.length) + formattedInt;
+        }
+        // '#' integer placeholder: suppress "0" when there are no required '0' or '?' digits
+        // and the integer value is zero (e.g. "#" format with value 0 → empty)
+        if (formattedInt === "0" && minIntDigits === 0 && totalIntSlots === 0) {
+            formattedInt = "";
+        }
     }
     // Format decimal part
     let formattedDec = "";
     if (decimalPlaces > 0) {
-        formattedDec = "." + (decPart + "0".repeat(decimalPlaces)).substring(0, decimalPlaces);
+        const rawDec = (decPart + "0".repeat(decimalPlaces)).substring(0, decimalPlaces);
+        // Process each decimal digit position according to its placeholder:
+        // '0' → always show digit, '?' → show digit or space, '#' → show digit or nothing (trim trailing)
+        const decChars = rawDec.split("");
+        // Walk from the end: '#' trailing zeros are removed, '?' trailing zeros become spaces
+        for (let i = decFmt.length - 1; i >= 0; i--) {
+            if (i >= decChars.length) {
+                continue;
+            }
+            if (decFmt[i] === "#" && decChars[i] === "0") {
+                decChars[i] = "";
+            }
+            else if (decFmt[i] === "?" && decChars[i] === "0") {
+                decChars[i] = " ";
+            }
+            else {
+                break; // stop at first non-zero or '0' placeholder
+            }
+        }
+        const decStr = decChars.join("");
+        // Only emit decimal point if there is content after it
+        if (decStr.length > 0) {
+            formattedDec = "." + decStr;
+        }
     }
     return sign + formattedInt + formattedDec;
 }

package/dist/cjs/modules/pdf/excel-bridge.js

@@ -12,7 +12,7 @@
  *
  * const workbook = new Workbook();
  * // ... build workbook ...
- * const pdf = excelToPdf(workbook);
+ * const pdf = await excelToPdf(workbook);
  * ```
  */
 Object.defineProperty(exports, "__esModule", { value: true });
@@ -30,12 +30,13 @@ const types_1 = require("./types");
  *
  * This is a convenience function that converts the Workbook to the PDF module's
  * data model and then generates the PDF.
+ * Yields to the event loop between each output page during layout and rendering.
  *
  * @param workbook - An Excel Workbook instance
  * @param options  - PDF export options
- * @returns PDF file as a Uint8Array
+ * @returns Promise of PDF file as a Uint8Array
  */
-function excelToPdf(workbook, options) {
+async function excelToPdf(workbook, options) {
     const pdfWorkbook = excelWorkbookToPdf(workbook);
     return (0, pdf_exporter_1.exportPdf)(pdfWorkbook, options);
 }
@@ -84,12 +85,24 @@ function convertSheet(ws, workbook) {
                 continue;
             }
             const cells = new Map();
-            row.eachCell({ includeEmpty: false }, cell => {
-                cells.set(cell.col, convertCell(cell));
+            row.eachCell({ includeEmpty: true }, cell => {
+                const hasValue = cell.type !== enums_1.ValueType.Null && cell.type !== enums_1.ValueType.Merge;
+                const hasStyle = cell.style &&
+                    ((cell.style.border &&
+                        (cell.style.border.top ||
+                            cell.style.border.right ||
+                            cell.style.border.bottom ||
+                            cell.style.border.left)) ||
+                        cell.style.fill ||
+                        cell.style.font);
+                if (hasValue || hasStyle) {
+                    cells.set(cell.col, convertCell(cell));
+                }
             });
             rows.set(r, {
                 hidden: row.hidden || undefined,
                 height: row.height ?? undefined,
+                customHeight: row.customHeight || undefined,
                 cells
             });
         }

package/dist/cjs/modules/pdf/index.js

@@ -8,7 +8,7 @@
  * ```typescript
  * import { pdf } from "excelts/pdf";
  *
- * const bytes = pdf([
+ * const bytes = await pdf([
  *   ["Product", "Revenue"],
  *   ["Widget", 1000],
  *   ["Gadget", 2500]
@@ -23,14 +23,14 @@
  * const workbook = new Workbook();
  * const sheet = workbook.addWorksheet("Sales");
  * sheet.addRow(["Product", "Revenue"]);
- * const bytes = excelToPdf(workbook);
+ * const bytes = await excelToPdf(workbook);
  * ```
  *
  * @example Read PDF — extract text, images, and metadata:
  * ```typescript
  * import { readPdf } from "excelts/pdf";
  *
- * const result = readPdf(pdfBytes);
+ * const result = await readPdf(pdfBytes);
  * console.log(result.text);               // All text
  * console.log(result.pages[0].text);      // Page 1 text
  * console.log(result.pages[0].images);    // Page 1 images

package/dist/cjs/modules/pdf/pdf.js

@@ -9,7 +9,7 @@
  * ```typescript
  * import { pdf } from "@cj-tech-master/excelts/pdf";
  *
- * const bytes = pdf([
+ * const bytes = await pdf([
  *   ["Product", "Revenue"],
  *   ["Widget", 1000],
  *   ["Gadget", 2500]
@@ -18,7 +18,7 @@
  *
  * @example With options:
  * ```typescript
- * const bytes = pdf([
+ * const bytes = await pdf([
  *   ["Name", "Score"],
  *   ["Alice", 95],
  *   ["Bob", 87]
@@ -27,7 +27,7 @@
  *
  * @example Multiple sheets:
  * ```typescript
- * const bytes = pdf({
+ * const bytes = await pdf({
  *   sheets: [
  *     { name: "Sales", data: [["Product", "Revenue"], ["Widget", 1000]] },
  *     { name: "Costs", data: [["Item", "Amount"], ["Rent", 500]] }
@@ -37,7 +37,7 @@
  *
  * @example With column widths and styles:
  * ```typescript
- * const bytes = pdf({
+ * const bytes = await pdf({
  *   name: "Report",
  *   columns: [{ width: 25 }, { width: 15 }],
  *   data: [
@@ -58,12 +58,13 @@ const pdf_exporter_1 = require("./render/pdf-exporter");
  * Generate a PDF.
  *
  * Accepts anything from a plain 2D array to a multi-sheet workbook.
+ * Yields to the event loop between each output page during layout and rendering.
  *
  * @param input   - 2D array, sheet object, or workbook object
  * @param options - PDF export options (page size, margins, etc.)
- * @returns PDF file as Uint8Array
+ * @returns Promise of PDF file as Uint8Array
  */
-function pdf(input, options) {
+async function pdf(input, options) {
     const workbook = normalizeInput(input);
     return (0, pdf_exporter_1.exportPdf)(workbook, options);
 }

package/dist/cjs/modules/pdf/reader/pdf-reader.js

@@ -17,18 +17,18 @@
  * - Cross-reference tables and streams (PDF 1.5+)
  * - Incremental updates and xref recovery
  *
- * @example Basic text extraction:
+ * @example Text extraction:
  * ```typescript
  * import { readPdf } from "excelts/pdf";
  *
- * const pdf = readPdf(pdfBytes);
+ * const pdf = await readPdf(pdfBytes);
  * console.log(pdf.text);           // All text from all pages
  * console.log(pdf.pages[0].text);  // Text from page 1
  * ```
  *
  * @example Image extraction:
  * ```typescript
- * const pdf = readPdf(pdfBytes);
+ * const pdf = await readPdf(pdfBytes);
  * for (const image of pdf.pages[0].images) {
  *   console.log(image.format, image.width, image.height);
  *   fs.writeFileSync(`image.${image.format}`, image.data);
@@ -37,7 +37,7 @@
  *
  * @example Metadata:
  * ```typescript
- * const pdf = readPdf(pdfBytes);
+ * const pdf = await readPdf(pdfBytes);
  * console.log(pdf.metadata.title);
  * console.log(pdf.metadata.author);
  * console.log(pdf.metadata.pageCount);
@@ -45,7 +45,7 @@
  *
  * @example Encrypted PDF:
  * ```typescript
- * const pdf = readPdf(pdfBytes, { password: "secret" });
+ * const pdf = await readPdf(pdfBytes, { password: "secret" });
  * ```
  */
 Object.defineProperty(exports, "__esModule", { value: true });
@@ -59,19 +59,36 @@ const annotation_extractor_1 = require("./annotation-extractor");
 const form_extractor_1 = require("./form-extractor");
 const metadata_reader_1 = require("./metadata-reader");
 const errors_1 = require("../errors");
+const utils_base_1 = require("../../../utils/utils.base.js");
 // =============================================================================
 // Public API
 // =============================================================================
 /**
  * Read a PDF file and extract text, images, and metadata.
+ * Yields to the event loop between pages to avoid blocking.
  *
  * @param data - Raw PDF file bytes
  * @param options - Extraction options
- * @returns Extracted content
+ * @returns Promise of extracted content
  * @throws {PdfStructureError} If the PDF structure is invalid
  * @throws {PdfError} If decryption fails (wrong password)
  */
-function readPdf(data, options) {
+async function readPdf(data, options) {
+    const { doc, opts, metadata, pagesInfo, pageIndicesToProcess } = prepareRead(data, options);
+    const pages = [];
+    for (let i = 0; i < pageIndicesToProcess.length; i++) {
+        const pageIdx = pageIndicesToProcess[i];
+        pages.push(processPage(pagesInfo[pageIdx].dict, pageIdx, doc, opts));
+        if (i < pageIndicesToProcess.length - 1) {
+            await (0, utils_base_1.yieldToEventLoop)();
+        }
+    }
+    return finalizeRead(pages, pagesInfo.length, metadata, opts, doc);
+}
+/**
+ * Shared setup: parse document, handle encryption, extract metadata, resolve pages.
+ */
+function prepareRead(data, options) {
     const opts = {
         password: options?.password ?? "",
         pages: options?.pages,
@@ -81,86 +98,81 @@ function readPdf(data, options) {
         extractAnnotations: options?.extractAnnotations ?? true,
         extractFormFields: options?.extractFormFields ?? true
     };
-    // Parse document structure
     const doc = new pdf_document_1.PdfDocument(data);
-    // Handle encryption
     if ((0, pdf_decrypt_1.isEncrypted)(doc)) {
         const success = (0, pdf_decrypt_1.initDecryption)(doc, opts.password);
         if (!success) {
             throw new errors_1.PdfStructureError("Failed to decrypt PDF: incorrect password");
         }
     }
-    // Extract metadata
     const metadata = opts.extractMetadata ? (0, metadata_reader_1.extractMetadata)(doc) : createEmptyMetadata();
-    // Get pages (with object identity for correct decryption)
     const pagesInfo = doc.getPagesWithObjInfo();
     const pageIndicesToProcess = opts.pages
         ? opts.pages.map(p => p - 1).filter(p => p >= 0 && p < pagesInfo.length)
         : Array.from({ length: pagesInfo.length }, (_, i) => i);
-    // Process each page
-    const pages = [];
-    for (const pageIdx of pageIndicesToProcess) {
-        const { dict: pageDict } = pagesInfo[pageIdx];
-        const pageNumber = pageIdx + 1;
-        const warnings = [];
-        // Extract text
-        let text = "";
-        let textLines = [];
-        let textFragments = [];
-        if (opts.extractText) {
-            try {
-                textFragments = (0, content_interpreter_1.extractTextFromPage)(pageDict, doc);
-                text = (0, text_reconstruction_1.reconstructText)(textFragments);
-                textLines = (0, text_reconstruction_1.reconstructTextLines)(textFragments);
-            }
-            catch (err) {
-                const msg = err instanceof Error ? err.message : String(err);
-                warnings.push(`Text extraction failed on page ${pageNumber}: ${msg}`);
-            }
+    return { doc, opts, metadata, pagesInfo, pageIndicesToProcess };
+}
+/**
+ * Process a single page: extract text, images, annotations, and dimensions.
+ */
+function processPage(pageDict, pageIdx, doc, opts) {
+    const pageNumber = pageIdx + 1;
+    const warnings = [];
+    let text = "";
+    let textLines = [];
+    let textFragments = [];
+    if (opts.extractText) {
+        try {
+            textFragments = (0, content_interpreter_1.extractTextFromPage)(pageDict, doc);
+            text = (0, text_reconstruction_1.reconstructText)(textFragments);
+            textLines = (0, text_reconstruction_1.reconstructTextLines)(textFragments);
         }
-        // Extract images
-        let images = [];
-        if (opts.extractImages) {
-            try {
-                images = (0, image_extractor_1.extractImagesFromPage)(pageDict, doc);
-            }
-            catch (err) {
-                const msg = err instanceof Error ? err.message : String(err);
-                warnings.push(`Image extraction failed on page ${pageNumber}: ${msg}`);
-            }
+        catch (err) {
+            const msg = err instanceof Error ? err.message : String(err);
+            warnings.push(`Text extraction failed on page ${pageNumber}: ${msg}`);
         }
-        // Extract annotations
-        let annotations = [];
-        if (opts.extractAnnotations) {
-            try {
-                annotations = (0, annotation_extractor_1.extractAnnotationsFromPage)(pageDict, doc);
-            }
-            catch (err) {
-                const msg = err instanceof Error ? err.message : String(err);
-                warnings.push(`Annotation extraction failed on page ${pageNumber}: ${msg}`);
-            }
+    }
+    let images = [];
+    if (opts.extractImages) {
+        try {
+            images = (0, image_extractor_1.extractImagesFromPage)(pageDict, doc);
+        }
+        catch (err) {
+            const msg = err instanceof Error ? err.message : String(err);
+            warnings.push(`Image extraction failed on page ${pageNumber}: ${msg}`);
+        }
+    }
+    let annotations = [];
+    if (opts.extractAnnotations) {
+        try {
+            annotations = (0, annotation_extractor_1.extractAnnotationsFromPage)(pageDict, doc);
+        }
+        catch (err) {
+            const msg = err instanceof Error ? err.message : String(err);
+            warnings.push(`Annotation extraction failed on page ${pageNumber}: ${msg}`);
         }
-        // Get page dimensions
-        const { width, height } = getPageDimensions(pageDict, doc);
-        pages.push({
-            pageNumber,
-            text,
-            textLines,
-            textFragments,
-            images,
-            annotations,
-            width,
-            height,
-            warnings
-        });
     }
-    // Concatenate all page text
+    const { width, height } = getPageDimensions(pageDict, doc);
+    return {
+        pageNumber,
+        text,
+        textLines,
+        textFragments,
+        images,
+        annotations,
+        width,
+        height,
+        warnings
+    };
+}
+/**
+ * Finalize: concatenate text, update metadata page count, extract form fields.
+ */
+function finalizeRead(pages, totalPageCount, metadata, opts, doc) {
     const allText = pages.map(p => p.text).join("\n\n");
-    // Update page count in metadata
     if (opts.extractMetadata) {
-        metadata.pageCount = pagesInfo.length;
+        metadata.pageCount = totalPageCount;
     }
-    // Extract form fields (document-level, not per-page)
     let formFields = [];
     if (opts.extractFormFields) {
         try {
@@ -170,12 +182,7 @@ function readPdf(data, options) {
             // Non-fatal — just return empty
         }
     }
-    return {
-        text: allText,
-        pages,
-        metadata,
-        formFields
-    };
+    return { text: allText, pages, metadata, formFields };
 }
 // =============================================================================
 // Helpers

package/dist/cjs/modules/pdf/render/constants.js

@@ -0,0 +1,33 @@
+"use strict";
+/**
+ * Shared rendering constants used by both the layout engine and page renderer.
+ *
+ * Keeping these in one place ensures row-height computation and text rendering
+ * use exactly the same values, preventing clipped or overlapping content.
+ */
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.PX_TO_PT = exports.EXCEL_COLUMN_PADDING_PX = exports.MAX_DIGIT_WIDTH_PX = exports.INDENT_WIDTH = exports.LINE_HEIGHT_FACTOR = exports.CELL_PADDING_V = exports.CELL_PADDING_H = void 0;
+/** Horizontal cell padding in points (left + right = 2 × CELL_PADDING_H). */
+exports.CELL_PADDING_H = 3;
+/** Vertical cell padding in points (top + bottom = 2 × CELL_PADDING_V). */
+exports.CELL_PADDING_V = 2;
+/**
+ * Line-height multiplier applied to the font size.
+ *
+ * Excel's default row height for an 11pt font is 15pt, which after removing
+ * vertical padding (2 × 2 = 4pt) leaves 11pt × 1.0 — but Excel also adds
+ * internal leading. A factor of 1.2 matches standard PDF/typographic practice
+ * and keeps text readable without inflating row heights.
+ */
+exports.LINE_HEIGHT_FACTOR = 1.2;
+/** Width of one indent level in points (~3 characters at 11pt). */
+exports.INDENT_WIDTH = 10;
+/**
+ * Excel column widths are measured in characters of the default font's digit width.
+ * For Calibri 11pt (the default), maxDigitWidth ≈ 7 pixels at 96 DPI.
+ * Excel adds 5 pixels of padding per column (4px text margin + 1px gridline).
+ * To convert to PDF points: (charWidth × 7 + 5) × (72/96).
+ */
+exports.MAX_DIGIT_WIDTH_PX = 7;
+exports.EXCEL_COLUMN_PADDING_PX = 5;
+exports.PX_TO_PT = 72 / 96; // 0.75