Removed Default Convert Options

Author: nicolaspearson

coverage/badge-branches.svg

@@ -145,13 +145,6 @@ class Ocr {
 		args.push('-density');
 		args.push(`${options && options.convertDensity ? options.convertDensity : 300}`);
 		args.push(filePath);
-		args.push('-depth');
-		args.push('8');
-		args.push('-strip');
-		args.push('-background');
-		args.push('white');
-		args.push('-alpha');
-		args.push('off');
 		if (options && options.convertArgs) {
 			// Parse all provided options to command line arguments
 			for (const [key, value] of Object.entries(options.convertArgs)) {

coverage/badge-lines.svg

@@ -3,6 +3,18 @@ import * as temp from 'temp';
 
 import { ExtractTextOptions, Ocr } from '../src/index';
 
+const convertArgs = {
+	quality: '100',
+	trim: '',
+	depth: '8',
+	strip: '',
+	background: 'white',
+	alpha: 'off'
+};
+
+// tslint:disable object-literal-key-quotes
+const tesseractArgs = { '-psm': 6, c: 'preserve_interword_spaces=1' };
+
 describe('Extract Text Tests', () => {
 	it('should be able to extract pdf text from single-page.pdf', async (done) => {
 		jest.setTimeout(15 * 1000);
@@ -11,7 +23,10 @@ describe('Extract Text Tests', () => {
 		const pdfPath = path.join(__dirname, relativePath);
 
 		try {
-			const result: string = await Ocr.extractText(pdfPath);
+			const options: ExtractTextOptions = {
+				convertArgs
+			};
+			const result: string = await Ocr.extractText(pdfPath, options);
 			expect(result).toBeDefined();
 			expect(result).toContain('00001-001-0002');
 		} catch (error) {
@@ -29,7 +44,10 @@ describe('Extract Text Tests', () => {
 		const pdfPath = path.join(__dirname, relativePath);
 
 		try {
-			const result: string = await Ocr.extractText(pdfPath);
+			const options: ExtractTextOptions = {
+				convertArgs
+			};
+			const result: string = await Ocr.extractText(pdfPath, options);
 			expect(result).toBeDefined();
 			expect(result).toContain('00001-001-0002');
 		} catch (error) {
@@ -41,13 +59,16 @@ describe('Extract Text Tests', () => {
 	});
 
 	it('should be able to extract pdf text from multi-page.pdf', async (done) => {
-		jest.setTimeout(15 * 1000);
+		jest.setTimeout(25 * 1000);
 		const fileName = 'multi-page.pdf';
 		const relativePath = path.join('sample', fileName);
 		const pdfPath = path.join(__dirname, relativePath);
 
 		try {
-			const options: ExtractTextOptions = { convertDensity: 400, convertArgs: { trim: '' } };
+			const options: ExtractTextOptions = {
+				convertDensity: 600,
+				convertArgs
+			};
 			const result: string = await Ocr.extractText(pdfPath, options);
 			expect(result).toBeDefined();
 			expect(result).toContain('National Airspace System');
@@ -67,7 +88,11 @@ describe('Extract Text Tests', () => {
 		const pdfPath = path.join(__dirname, relativePath);
 
 		try {
-			const options: ExtractTextOptions = { pdfToTextArgs: { f: 1, l: 4 } };
+			const options: ExtractTextOptions = {
+				pdfToTextArgs: { f: 1, l: 4 },
+				convertDensity: 600,
+				convertArgs
+			};
 			const result: string = await Ocr.extractText(pdfPath, options);
 			expect(result).toBeDefined();
 			expect(result).toContain('TraceMonkey');
@@ -81,17 +106,19 @@ describe('Extract Text Tests', () => {
 	});
 
 	it('should be able to extract text from sample.png', async (done) => {
-		jest.setTimeout(15 * 1000);
+		jest.setTimeout(25 * 1000);
 		const fileName = 'sample.png';
 		const relativePath = path.join('sample', fileName);
 		const pngPath = path.join(__dirname, relativePath);
 
 		try {
 			const tmpDir = temp.mkdirSync('tmp');
-			// tslint:disable object-literal-key-quotes
+
 			const options: ExtractTextOptions = {
+				convertDensity: 600,
+				convertArgs,
 				tesseractLang: 'eng',
-				tesseractArgs: { '-psm': 6, c: 'preserve_interword_spaces=1' }
+				tesseractArgs
 			};
 			const result: string = await Ocr.invokeImageOcr(tmpDir, pngPath, options);
 			expect(result).toBeDefined();
@@ -105,17 +132,24 @@ describe('Extract Text Tests', () => {
 	});
 
 	it('should be able to extract text from sample-low.jpg', async (done) => {
-		jest.setTimeout(15 * 1000);
+		jest.setTimeout(25 * 1000);
 		const fileName = 'sample-low.jpg';
 		const relativePath = path.join('sample', fileName);
 		const jpgPath = path.join(__dirname, relativePath);
 
 		try {
 			const options: ExtractTextOptions = {
 				convertDensity: 600,
-				convertArgs: { trim: '' },
+				convertArgs: {
+					...convertArgs,
+					verbose: '',
+					flatten: '',
+					contrast: '',
+					'auto-level': '',
+					sharpen: '0x4.0'
+				},
 				tesseractLang: 'eng',
-				tesseractArgs: { '-psm': 6 }
+				tesseractArgs
 			};
 			const result: string = await Ocr.extractText(jpgPath, options);
 			expect(result).toBeDefined();