@eredzik/calaminejs 0.1.2 → 0.2.0

package/README.md

@@ -0,0 +1,454 @@
+# @eredzik/calaminejs
+
+A high-performance JavaScript/TypeScript library for reading Excel files (XLS/XLSX) and converting them to Parquet format. Built on the Rust [calamine](https://github.com/tafia/calamine) library using WebAssembly for optimal performance.
+
+## Features
+
+- 📊 Read Excel files (XLS and XLSX formats)
+- 🚀 High performance through WebAssembly
+- 🔄 Convert sheets to Parquet format
+- 🎯 Smart header row detection
+- 📈 Progress tracking for large files
+- 🌐 Works in Node.js and browsers
+- 📝 Full TypeScript support
+- 🔢 Preserves data types (strings, numbers, booleans, dates, etc.)
+
+## Installation
+
+```bash
+npm install @eredzik/calaminejs
+```
+
+## Quick Start
+
+### Node.js
+
+```javascript
+import { Workbook } from '@eredzik/calaminejs';
+import { readFileSync } from 'fs';
+
+// Read an Excel file
+const buffer = readFileSync('data.xlsx');
+const workbook = Workbook.from_bytes(new Uint8Array(buffer));
+
+// Get sheet names
+const sheetNames = workbook.sheet_names();
+console.log('Sheets:', sheetNames);
+
+// Access a sheet
+const sheet = workbook.get_sheet(sheetNames[0]);
+console.log(`Rows: ${sheet.row_count()}, Columns: ${sheet.col_count()}`);
+
+// Access cell data
+const rows = sheet.rows;
+console.log('First row:', rows[0]);
+```
+
+### Browser
+
+```javascript
+import { Workbook } from '@eredzik/calaminejs/web';
+
+// From file input
+const file = document.querySelector('input[type="file"]').files[0];
+const arrayBuffer = await file.arrayBuffer();
+const workbook = Workbook.from_bytes(new Uint8Array(arrayBuffer));
+
+// Process the workbook
+const sheet = workbook.get_sheet_by_index(0);
+console.log('Sheet name:', sheet.name);
+```
+
+## API Reference
+
+### Workbook
+
+The main class for working with Excel files.
+
+#### Static Methods
+
+##### `Workbook.from_bytes(data: Uint8Array): Workbook`
+
+Load an Excel file from bytes. Automatically detects XLS or XLSX format.
+
+```javascript
+const workbook = Workbook.from_bytes(new Uint8Array(buffer));
+```
+
+##### `Workbook.from_bytes_with_progress(data: Uint8Array, callback?: Function, interval?: number): Workbook`
+
+Load an Excel file with progress tracking.
+
+```javascript
+const workbook = Workbook.from_bytes_with_progress(
+  new Uint8Array(buffer),
+  (progress) => {
+    console.log(`Processing sheet ${progress.sheetIndex + 1}/${progress.totalSheets}`);
+    console.log(`Sheet: ${progress.sheetName}, Row: ${progress.currentRow}`);
+  },
+  100 // Report progress every 100 rows
+);
+```
+
+**Progress object properties:**
+- `sheetIndex: number` - Current sheet index (0-based)
+- `totalSheets: number` - Total number of sheets
+- `sheetName: string` - Name of current sheet
+- `currentRow: number` - Current row being processed
+- `totalRows: number | null` - Total rows (available when sheet is complete)
+
+#### Instance Methods
+
+##### `sheet_names(): string[]`
+
+Get an array of all sheet names in the workbook.
+
+```javascript
+const names = workbook.sheet_names();
+```
+
+##### `get_sheet(name: string): Sheet | undefined`
+
+Get a sheet by name.
+
+```javascript
+const sheet = workbook.get_sheet('Sheet1');
+```
+
+##### `get_sheet_by_index(index: number): Sheet | undefined`
+
+Get a sheet by index (0-based).
+
+```javascript
+const firstSheet = workbook.get_sheet_by_index(0);
+```
+
+##### `sheet_count(): number`
+
+Get the total number of sheets.
+
+```javascript
+const count = workbook.sheet_count();
+```
+
+### Sheet
+
+Represents a single worksheet in an Excel file.
+
+#### Properties
+
+##### `name: string`
+
+The name of the sheet.
+
+```javascript
+console.log(sheet.name); // "Sheet1"
+```
+
+##### `rows: Array<Array<any>>`
+
+A 2D array of cell values. Values are converted to native JavaScript types:
+- Strings → `string`
+- Numbers → `number`
+- Booleans → `boolean`
+- Empty cells → `null`
+- Dates → `number` (Excel date format)
+- Errors → `string`
+
+```javascript
+const rows = sheet.rows;
+rows.forEach((row, rowIndex) => {
+  row.forEach((cell, colIndex) => {
+    console.log(`Cell [${rowIndex}, ${colIndex}]:`, cell);
+  });
+});
+```
+
+#### Methods
+
+##### `get_cell(row: number, col: number): CellValue | undefined`
+
+Get a specific cell with type information.
+
+```javascript
+const cell = sheet.get_cell(0, 0);
+if (cell.is_string) {
+  console.log('String value:', cell.to_string_value());
+}
+```
+
+##### `row_count(): number`
+
+Get the number of rows in the sheet.
+
+```javascript
+const rowCount = sheet.row_count();
+```
+
+##### `col_count(): number`
+
+Get the maximum number of columns in the sheet.
+
+```javascript
+const colCount = sheet.col_count();
+```
+
+##### `infer_header_row(): HeaderInfo | undefined`
+
+Automatically detect which row contains the table header using heuristics:
+- Headers typically contain string values in most columns
+- Headers are followed by rows with data
+- Headers have multiple non-empty cells
+- Prioritizes rows in the first 20 rows
+
+```javascript
+const headerInfo = sheet.infer_header_row();
+if (headerInfo) {
+  console.log('Header found at row:', headerInfo.row_index);
+  console.log('Column names:', headerInfo.column_names);
+}
+```
+
+##### `to_parquet(): Uint8Array`
+
+Convert the sheet to Parquet format. Column types are automatically inferred:
+- All booleans → Boolean column
+- All integers → Int64 column
+- All floats/numbers → Float64 column
+- All dates → Datetime column (millisecond precision)
+- Mixed or strings → String column
+
+```javascript
+const parquetBytes = sheet.to_parquet();
+// Save to file or process further
+```
+
+##### `to_parquet_with_names(columnNames: string[]): Uint8Array`
+
+Convert the sheet to Parquet format with custom column names.
+
+```javascript
+const columnNames = ['ID', 'Name', 'Age', 'Email'];
+const parquetBytes = sheet.to_parquet_with_names(columnNames);
+```
+
+### CellValue
+
+Detailed cell information with type checking and conversion methods.
+
+#### Type Checking Properties
+
+- `is_empty: boolean` - Check if cell is empty
+- `is_string: boolean` - Check if cell contains a string
+- `is_float: boolean` - Check if cell contains a float
+- `is_int: boolean` - Check if cell contains an integer
+- `is_bool: boolean` - Check if cell contains a boolean
+- `is_error: boolean` - Check if cell contains an error
+- `is_datetime: boolean` - Check if cell contains a date/time
+- `is_duration: boolean` - Check if cell contains a duration
+
+#### Conversion Methods
+
+##### `to_string_value(): string | undefined`
+
+Convert cell to string representation.
+
+```javascript
+const cell = sheet.get_cell(0, 0);
+const str = cell.to_string_value();
+```
+
+##### `to_float_value(): number | undefined`
+
+Convert cell to float (works for numbers, booleans, dates).
+
+```javascript
+const num = cell.to_float_value();
+```
+
+##### `to_int_value(): number | undefined`
+
+Convert cell to integer (works for integers, floats, booleans).
+
+```javascript
+const int = cell.to_int_value();
+```
+
+##### `to_bool_value(): boolean | undefined`
+
+Get boolean value (only works for boolean cells).
+
+```javascript
+const bool = cell.to_bool_value();
+```
+
+### HeaderInfo
+
+Information about detected header row.
+
+#### Properties
+
+- `row_index: number` - The index of the header row (0-based)
+- `column_names: string[]` - Array of column names extracted from the header
+
+## Examples
+
+### Reading and Processing Data
+
+```javascript
+import { Workbook } from '@eredzik/calaminejs';
+import { readFileSync } from 'fs';
+
+const buffer = readFileSync('sales.xlsx');
+const workbook = Workbook.from_bytes(new Uint8Array(buffer));
+const sheet = workbook.get_sheet('Sales Data');
+
+// Detect header
+const headerInfo = sheet.infer_header_row();
+if (headerInfo) {
+  console.log('Columns:', headerInfo.column_names);
+  
+  // Process data rows (skip header)
+  const dataRows = sheet.rows.slice(headerInfo.row_index + 1);
+  dataRows.forEach(row => {
+    console.log('Row data:', row);
+  });
+}
+```
+
+### Converting to Parquet
+
+```javascript
+import { Workbook } from '@eredzik/calaminejs';
+import { writeFileSync, readFileSync } from 'fs';
+
+const buffer = readFileSync('data.xlsx');
+const workbook = Workbook.from_bytes(new Uint8Array(buffer));
+const sheet = workbook.get_sheet_by_index(0);
+
+// Option 1: Auto-generated column names
+const parquet1 = sheet.to_parquet();
+writeFileSync('output1.parquet', parquet1);
+
+// Option 2: Custom column names
+const headerInfo = sheet.infer_header_row();
+if (headerInfo) {
+  // Skip header row and convert data
+  const dataSheet = {
+    ...sheet,
+    rows: sheet.rows.slice(headerInfo.row_index + 1)
+  };
+  const parquet2 = dataSheet.to_parquet_with_names(headerInfo.column_names);
+  writeFileSync('output2.parquet', parquet2);
+}
+```
+
+### Working with Cell Types
+
+```javascript
+const sheet = workbook.get_sheet_by_index(0);
+
+for (let row = 0; row < sheet.row_count(); row++) {
+  for (let col = 0; col < sheet.col_count(); col++) {
+    const cell = sheet.get_cell(row, col);
+    
+    if (cell.is_string) {
+      console.log(`String: ${cell.to_string_value()}`);
+    } else if (cell.is_int) {
+      console.log(`Integer: ${cell.to_int_value()}`);
+    } else if (cell.is_float) {
+      console.log(`Float: ${cell.to_float_value()}`);
+    } else if (cell.is_bool) {
+      console.log(`Boolean: ${cell.to_bool_value()}`);
+    } else if (cell.is_datetime) {
+      const excelDate = cell.to_float_value();
+      // Convert Excel date to JavaScript Date
+      const jsDate = new Date((excelDate - 25569) * 86400 * 1000);
+      console.log(`Date: ${jsDate.toISOString()}`);
+    }
+  }
+}
+```
+
+### Progress Tracking for Large Files
+
+```javascript
+import { Workbook } from '@eredzik/calaminejs';
+import { readFileSync } from 'fs';
+
+const buffer = readFileSync('large-file.xlsx');
+
+console.log('Loading workbook...');
+const workbook = Workbook.from_bytes_with_progress(
+  new Uint8Array(buffer),
+  (progress) => {
+    const percent = ((progress.currentRow / (progress.totalRows || progress.currentRow)) * 100).toFixed(1);
+    console.log(`[${progress.sheetName}] Processing: ${percent}%`);
+  },
+  500 // Report every 500 rows
+);
+
+console.log('Workbook loaded successfully!');
+```
+
+## Browser vs Node.js
+
+The package provides separate builds optimized for each environment:
+
+```javascript
+// Node.js (default)
+import { Workbook } from '@eredzik/calaminejs';
+// or
+import { Workbook } from '@eredzik/calaminejs/node';
+
+// Browser
+import { Workbook } from '@eredzik/calaminejs/web';
+```
+
+## Performance Tips
+
+1. **Use progress callbacks for large files** to provide user feedback and avoid blocking
+2. **Process sheets on-demand** instead of loading all sheets at once
+3. **Use `get_cell()` for sparse data** instead of accessing the full `rows` array
+4. **Infer header once** and reuse the result instead of calling it multiple times
+5. **Convert to Parquet** for efficient storage and further processing with data tools
+
+## Requirements
+
+- Node.js >= 16.0.0 (for Node.js usage)
+- Modern browser with WebAssembly support (for browser usage)
+
+## Supported File Formats
+
+- **XLSX** - Excel 2007+ (.xlsx)
+- **XLS** - Excel 97-2003 (.xls)
+
+## Supported Data Types
+
+The library preserves Excel data types:
+
+- **Empty** - Empty cells
+- **String** - Text values
+- **Float** - Floating-point numbers
+- **Int** - Integer numbers
+- **Bool** - Boolean values (TRUE/FALSE)
+- **Error** - Excel error values (#N/A, #REF!, etc.)
+- **DateTime** - Date and time values (stored as Excel serial numbers)
+- **Duration** - Duration values
+
+## License
+
+MIT
+
+## Repository
+
+GitHub: [https://github.com/eredzik/calaminejs](https://github.com/eredzik/calaminejs)
+
+## Issues
+
+Report issues: [https://github.com/eredzik/calaminejs/issues](https://github.com/eredzik/calaminejs/issues)
+
+## Credits
+
+Built on top of the excellent [calamine](https://github.com/tafia/calamine) Rust library and [Polars](https://www.pola.rs/) for Parquet conversion.
+

package/dist/node/calamine_js.d.ts

@@ -31,8 +31,14 @@ export class Sheet {
   /**
    * Convert sheet to Parquet format
    * Returns the Parquet file as bytes
+   * unmerge_cells: if true, spread merged cell values to fill empty cells horizontally
    */
-  to_parquet(): Uint8Array;
+  to_parquet(unmerge_cells: boolean): Uint8Array;
+  /**
+   * Get the merged regions in this sheet
+   * Returns an array of arrays, where each inner array is [start_row, start_col, end_row, end_col]
+   */
+  merged_regions(): Array<any>;
   /**
    * Infer which row is the table header
    * Returns HeaderInfo with the row index and column names, or None if no header is found
@@ -46,8 +52,9 @@ export class Sheet {
   /**
    * Convert sheet to Parquet format with custom column names
    * column_names: array of column names (must match col_count)
+   * unmerge_cells: if true, spread merged cell values to fill empty cells horizontally
    */
-  to_parquet_with_names(column_names: string[]): Uint8Array;
+  to_parquet_with_names(column_names: string[], unmerge_cells: boolean): Uint8Array;
   get_cell(row: number, col: number): CellValue | undefined;
   col_count(): number;
   row_count(): number;

package/dist/node/calamine_js.js

@@ -338,10 +338,12 @@ class Sheet {
     /**
      * Convert sheet to Parquet format
      * Returns the Parquet file as bytes
+     * unmerge_cells: if true, spread merged cell values to fill empty cells horizontally
+     * @param {boolean} unmerge_cells
      * @returns {Uint8Array}
      */
-    to_parquet() {
-        const ret = wasm.sheet_to_parquet(this.__wbg_ptr);
+    to_parquet(unmerge_cells) {
+        const ret = wasm.sheet_to_parquet(this.__wbg_ptr, unmerge_cells);
         if (ret[3]) {
             throw takeFromExternrefTable0(ret[2]);
         }
@@ -349,6 +351,15 @@ class Sheet {
         wasm.__wbindgen_free(ret[0], ret[1] * 1, 1);
         return v1;
     }
+    /**
+     * Get the merged regions in this sheet
+     * Returns an array of arrays, where each inner array is [start_row, start_col, end_row, end_col]
+     * @returns {Array<any>}
+     */
+    merged_regions() {
+        const ret = wasm.sheet_merged_regions(this.__wbg_ptr);
+        return ret;
+    }
     /**
      * Infer which row is the table header
      * Returns HeaderInfo with the row index and column names, or None if no header is found
@@ -366,13 +377,15 @@ class Sheet {
     /**
      * Convert sheet to Parquet format with custom column names
      * column_names: array of column names (must match col_count)
+     * unmerge_cells: if true, spread merged cell values to fill empty cells horizontally
      * @param {string[]} column_names
+     * @param {boolean} unmerge_cells
      * @returns {Uint8Array}
      */
-    to_parquet_with_names(column_names) {
+    to_parquet_with_names(column_names, unmerge_cells) {
         const ptr0 = passArrayJsValueToWasm0(column_names, wasm.__wbindgen_malloc);
         const len0 = WASM_VECTOR_LEN;
-        const ret = wasm.sheet_to_parquet_with_names(this.__wbg_ptr, ptr0, len0);
+        const ret = wasm.sheet_to_parquet_with_names(this.__wbg_ptr, ptr0, len0, unmerge_cells);
         if (ret[3]) {
             throw takeFromExternrefTable0(ret[2]);
         }

package/dist/node/calamine_js_bg.wasm.d.ts

@@ -22,11 +22,12 @@ export const headerinfo_row_index: (a: number) => number;
 export const sheet_col_count: (a: number) => number;
 export const sheet_get_cell: (a: number, b: number, c: number) => number;
 export const sheet_infer_header_row: (a: number) => number;
+export const sheet_merged_regions: (a: number) => any;
 export const sheet_name: (a: number) => [number, number];
 export const sheet_row_count: (a: number) => number;
 export const sheet_rows: (a: number) => any;
-export const sheet_to_parquet: (a: number) => [number, number, number, number];
-export const sheet_to_parquet_with_names: (a: number, b: number, c: number) => [number, number, number, number];
+export const sheet_to_parquet: (a: number, b: number) => [number, number, number, number];
+export const sheet_to_parquet_with_names: (a: number, b: number, c: number, d: number) => [number, number, number, number];
 export const workbook_from_bytes: (a: number, b: number) => [number, number, number];
 export const workbook_from_bytes_with_progress: (a: number, b: number, c: number, d: number) => [number, number, number];
 export const workbook_get_sheet: (a: number, b: number, c: number) => number;

package/dist/web/calamine_js.d.ts

@@ -31,8 +31,14 @@ export class Sheet {
   /**
    * Convert sheet to Parquet format
    * Returns the Parquet file as bytes
+   * unmerge_cells: if true, spread merged cell values to fill empty cells horizontally
    */
-  to_parquet(): Uint8Array;
+  to_parquet(unmerge_cells: boolean): Uint8Array;
+  /**
+   * Get the merged regions in this sheet
+   * Returns an array of arrays, where each inner array is [start_row, start_col, end_row, end_col]
+   */
+  merged_regions(): Array<any>;
   /**
    * Infer which row is the table header
    * Returns HeaderInfo with the row index and column names, or None if no header is found
@@ -46,8 +52,9 @@ export class Sheet {
   /**
    * Convert sheet to Parquet format with custom column names
    * column_names: array of column names (must match col_count)
+   * unmerge_cells: if true, spread merged cell values to fill empty cells horizontally
    */
-  to_parquet_with_names(column_names: string[]): Uint8Array;
+  to_parquet_with_names(column_names: string[], unmerge_cells: boolean): Uint8Array;
   get_cell(row: number, col: number): CellValue | undefined;
   col_count(): number;
   row_count(): number;
@@ -91,11 +98,12 @@ export interface InitOutput {
   readonly sheet_col_count: (a: number) => number;
   readonly sheet_get_cell: (a: number, b: number, c: number) => number;
   readonly sheet_infer_header_row: (a: number) => number;
+  readonly sheet_merged_regions: (a: number) => any;
   readonly sheet_name: (a: number) => [number, number];
   readonly sheet_row_count: (a: number) => number;
   readonly sheet_rows: (a: number) => any;
-  readonly sheet_to_parquet: (a: number) => [number, number, number, number];
-  readonly sheet_to_parquet_with_names: (a: number, b: number, c: number) => [number, number, number, number];
+  readonly sheet_to_parquet: (a: number, b: number) => [number, number, number, number];
+  readonly sheet_to_parquet_with_names: (a: number, b: number, c: number, d: number) => [number, number, number, number];
   readonly workbook_from_bytes: (a: number, b: number) => [number, number, number];
   readonly workbook_from_bytes_with_progress: (a: number, b: number, c: number, d: number) => [number, number, number];
   readonly workbook_get_sheet: (a: number, b: number, c: number) => number;

package/dist/web/calamine_js.js

@@ -340,10 +340,12 @@ export class Sheet {
     /**
      * Convert sheet to Parquet format
      * Returns the Parquet file as bytes
+     * unmerge_cells: if true, spread merged cell values to fill empty cells horizontally
+     * @param {boolean} unmerge_cells
      * @returns {Uint8Array}
      */
-    to_parquet() {
-        const ret = wasm.sheet_to_parquet(this.__wbg_ptr);
+    to_parquet(unmerge_cells) {
+        const ret = wasm.sheet_to_parquet(this.__wbg_ptr, unmerge_cells);
         if (ret[3]) {
             throw takeFromExternrefTable0(ret[2]);
         }
@@ -351,6 +353,15 @@ export class Sheet {
         wasm.__wbindgen_free(ret[0], ret[1] * 1, 1);
         return v1;
     }
+    /**
+     * Get the merged regions in this sheet
+     * Returns an array of arrays, where each inner array is [start_row, start_col, end_row, end_col]
+     * @returns {Array<any>}
+     */
+    merged_regions() {
+        const ret = wasm.sheet_merged_regions(this.__wbg_ptr);
+        return ret;
+    }
     /**
      * Infer which row is the table header
      * Returns HeaderInfo with the row index and column names, or None if no header is found
@@ -368,13 +379,15 @@ export class Sheet {
     /**
      * Convert sheet to Parquet format with custom column names
      * column_names: array of column names (must match col_count)
+     * unmerge_cells: if true, spread merged cell values to fill empty cells horizontally
      * @param {string[]} column_names
+     * @param {boolean} unmerge_cells
      * @returns {Uint8Array}
      */
-    to_parquet_with_names(column_names) {
+    to_parquet_with_names(column_names, unmerge_cells) {
         const ptr0 = passArrayJsValueToWasm0(column_names, wasm.__wbindgen_malloc);
         const len0 = WASM_VECTOR_LEN;
-        const ret = wasm.sheet_to_parquet_with_names(this.__wbg_ptr, ptr0, len0);
+        const ret = wasm.sheet_to_parquet_with_names(this.__wbg_ptr, ptr0, len0, unmerge_cells);
         if (ret[3]) {
             throw takeFromExternrefTable0(ret[2]);
         }

package/dist/web/calamine_js_bg.wasm.d.ts

@@ -22,11 +22,12 @@ export const headerinfo_row_index: (a: number) => number;
 export const sheet_col_count: (a: number) => number;
 export const sheet_get_cell: (a: number, b: number, c: number) => number;
 export const sheet_infer_header_row: (a: number) => number;
+export const sheet_merged_regions: (a: number) => any;
 export const sheet_name: (a: number) => [number, number];
 export const sheet_row_count: (a: number) => number;
 export const sheet_rows: (a: number) => any;
-export const sheet_to_parquet: (a: number) => [number, number, number, number];
-export const sheet_to_parquet_with_names: (a: number, b: number, c: number) => [number, number, number, number];
+export const sheet_to_parquet: (a: number, b: number) => [number, number, number, number];
+export const sheet_to_parquet_with_names: (a: number, b: number, c: number, d: number) => [number, number, number, number];
 export const workbook_from_bytes: (a: number, b: number) => [number, number, number];
 export const workbook_from_bytes_with_progress: (a: number, b: number, c: number, d: number) => [number, number, number];
 export const workbook_get_sheet: (a: number, b: number, c: number) => number;

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@eredzik/calaminejs",
-  "version": "0.1.2",
+  "version": "0.2.0",
   "description": "Rust calamine library bindings for JavaScript/TypeScript - Excel file reading and Parquet conversion",
   "main": "./dist/node/calamine_js.js",
   "module": "./dist/web/calamine_js.js",
@@ -45,10 +45,14 @@
     "build:node:size": "wasm-pack build --target nodejs --out-dir dist/node --profile release-size",
     "build:web:size": "wasm-pack build --target web --out-dir dist/web --profile release-size",
     "postbuild": "rm -f dist/node/.gitignore dist/web/.gitignore dist/node/package.json dist/web/package.json dist/node/README.md dist/web/README.md dist/node/LICENSE dist/web/LICENSE",
-    "clean": "rm -rf dist target"
+    "clean": "rm -rf dist target",
+    "test": "vitest run",
+    "test:watch": "vitest"
   },
   "devDependencies": {
-    "npm-run-all": "^4.1.5"
+    "npm-run-all": "^4.1.5",
+    "vitest": "^2.1.8",
+    "@types/node": "^22.10.5"
   },
   "keywords": [
     "excel",