Merge branch 'master' into edit-node-recipe

src/core/config/Categories.json

@@ -383,6 +383,7 @@
         "ops": [
             "Render Image",
             "Play Media",
+            "Optical Character Recognition",
             "Remove EXIF",
             "Extract EXIF",
             "Split Colour Channels",

src/core/operations/ConvertDataUnits.mjs

@@ -54,7 +54,7 @@ class ConvertDataUnits extends Operation {
 const DATA_UNITS = [
     "Bits (b)", "Nibbles", "Octets", "Bytes (B)",
     "[Binary bits (2^n)]", "Kibibits (Kib)", "Mebibits (Mib)", "Gibibits (Gib)", "Tebibits (Tib)", "Pebibits (Pib)", "Exbibits (Eib)", "Zebibits (Zib)", "Yobibits (Yib)", "[/Binary bits (2^n)]",
-    "[Decimal bits (10^n)]", "Decabits", "Hectobits", "Kilobits (kb)", "Megabits (Mb)", "Gigabits (Gb)", "Terabits (Tb)", "Petabits (Pb)", "Exabits (Eb)", "Zettabits (Zb)", "Yottabits (Yb)", "[/Decimal bits (10^n)]",
+    "[Decimal bits (10^n)]", "Decabits", "Hectobits", "Kilobits (Kb)", "Megabits (Mb)", "Gigabits (Gb)", "Terabits (Tb)", "Petabits (Pb)", "Exabits (Eb)", "Zettabits (Zb)", "Yottabits (Yb)", "[/Decimal bits (10^n)]",
     "[Binary bytes (8 x 2^n)]", "Kibibytes (KiB)", "Mebibytes (MiB)", "Gibibytes (GiB)", "Tebibytes (TiB)", "Pebibytes (PiB)", "Exbibytes (EiB)", "Zebibytes (ZiB)", "Yobibytes (YiB)", "[/Binary bytes (8 x 2^n)]",
     "[Decimal bytes (8 x 10^n)]", "Kilobytes (KB)", "Megabytes (MB)", "Gigabytes (GB)", "Terabytes (TB)", "Petabytes (PB)", "Exabytes (EB)", "Zettabytes (ZB)", "Yottabytes (YB)", "[/Decimal bytes (8 x 10^n)]"
 ];

src/core/operations/OpticalCharacterRecognition.mjs

@@ -0,0 +1,87 @@
+/**
+ * @author n1474335 [n1474335@gmail.com]
+ * @author mshwed [m@ttshwed.com]
+ * @copyright Crown Copyright 2019
+ * @license Apache-2.0
+ */
+
+import Operation from "../Operation.mjs";
+import OperationError from "../errors/OperationError.mjs";
+import { isImage } from "../lib/FileType.mjs";
+import { toBase64 } from "../lib/Base64.mjs";
+import { isWorkerEnvironment } from "../Utils.mjs";
+
+import Tesseract from "tesseract.js";
+const { TesseractWorker } = Tesseract;
+
+import process from "process";
+
+/**
+ * Optical Character Recognition operation
+ */
+class OpticalCharacterRecognition extends Operation {
+
+    /**
+     * OpticalCharacterRecognition constructor
+     */
+    constructor() {
+        super();
+
+        this.name = "Optical Character Recognition";
+        this.module = "OCR";
+        this.description = "Optical character recognition or optical character reader (OCR) is the mechanical or electronic conversion of images of typed, handwritten or printed text into machine-encoded text.<br><br>Supported image formats: png, jpg, bmp, pbm.";
+        this.infoURL = "https://wikipedia.org/wiki/Optical_character_recognition";
+        this.inputType = "ArrayBuffer";
+        this.outputType = "string";
+        this.args = [
+            {
+                name: "Show confidence",
+                type: "boolean",
+                value: true
+            }
+        ];
+    }
+
+    /**
+     * @param {ArrayBuffer} input
+     * @param {Object[]} args
+     * @returns {string}
+     */
+    async run(input, args) {
+        const [showConfidence] = args;
+
+        if (!isWorkerEnvironment()) throw OperationError("This operation only works in a browser");
+
+        const type = isImage(input);
+        if (!type) {
+            throw new OperationError("Invalid File Type");
+        }
+
+        const assetDir = isWorkerEnvironment() ? `${self.docURL}/assets/` : `${process.cwd()}/src/core/vendor/`;
+
+        try {
+            const image = `data:${type};base64,${toBase64(input)}`;
+            const worker = new TesseractWorker({
+                workerPath: `${assetDir}tesseract/worker.min.js`,
+                langPath: `${assetDir}tesseract/lang-data`,
+                corePath: `${assetDir}tesseract/tesseract-core.wasm.js`,
+            });
+            const result = await worker.recognize(image)
+                .progress(progress => {
+                    if (isWorkerEnvironment()) {
+                        self.sendStatusMessage(`Status: ${progress.status} - ${(parseFloat(progress.progress)*100).toFixed(2)}%`);
+                    }
+                });
+
+            if (showConfidence) {
+                return `Confidence: ${result.confidence}%\n\n${result.text}`;
+            } else {
+                return result.text;
+            }
+        } catch (err) {
+            throw new OperationError(`Error performing OCR on image. (${err})`);
+        }
+    }
+}
+
+export default OpticalCharacterRecognition;