Release v1.2.0

Author: jreyesr

README.md

@@ -160,11 +160,15 @@ Initial version, contains the **Extract text** and **Extract boxes** operations.
 * Add a setting to override the detected DPI of the image
 * Add the ability to specify the whitelist and blacklist
 
+### v1.2.0
+
+* Update the Tesseract.JS version to v6.0.0. No user-facing changes. Please [report any issues that you find](https://github.com/jreyesr/n8n-nodes-tesseractjs/issues)!
+
 ## Developer info
 
 ### Releasing changes
 
 1. Bump the version in package.json. We use SemVer.
-2. Add an entry to the top of CHANGELOG.md describing the changes.
+2. Add an entry on the [Version History](#version-history) section above describing the changes.
 3. Push changes, open a PR and merge it to master branch (if developing on another branch)
 4. Create a release. This will kick off the CI which will build and publish the package on NPM

nodes/TesseractNode/operations.ts

@@ -1,5 +1,5 @@
 import {IExecuteFunctions, INodeExecutionData} from "n8n-workflow";
-import type {Worker} from "tesseract.js";
+import type {Worker, Block} from "tesseract.js";
 
 type BoundingBox = {
 	top: number, left: number,
@@ -28,10 +28,22 @@ export async function extractBoxes(this: IExecuteFunctions, worker: Worker, item
 	};
 
 	const data = await this.helpers.getBinaryDataBuffer(itemIndex, imageFieldName)
-	const d = await worker.recognize(data, {rectangle: bbox});
+	const d = await worker.recognize(data, {rectangle: bbox}, {blocks: true});
+
+	// NOTE: since v1.2.0, we start clobbering the nice TS object here, to restore it to the pre-v6 API, which included additional properties
+	// See https://github.com/naptha/tesseract.js/issues/993#issuecomment-2597678687
+	// @ts-ignore
+	d.data.paragraphs = d.data.blocks!.map((block) => block.paragraphs).flat();
+	// @ts-ignore
+	d.data.lines = d.data.blocks!.map((block) => block.paragraphs.map((paragraph) => paragraph.lines)).flat(2);
+	// @ts-ignore
+	d.data.words = d.data.blocks!.map((block) => block.paragraphs.map((paragraph) => paragraph.lines.map((line) => line.words))).flat(3);
+	// @ts-ignore
+	d.data.symbols = d.data.blocks!.map((block) => block.paragraphs.map((paragraph) => paragraph.lines.map((line) => line.words.map((word) => word.symbols)))).flat(4);
 
 	newItem.json = {
-		blocks: d.data[granularity].map(b => ({
+		// @ts-ignore
+		blocks: d.data[granularity].map((b: Block) => ({
 			text: b.text,
 			confidence: b.confidence,
 			bbox: b.bbox,

package.json

@@ -1,6 +1,6 @@
 {
   "name": "n8n-nodes-tesseractjs",
-  "version": "1.1.0",
+  "version": "1.2.0",
   "description": "A n8n module that exposes Tesseract.js, an OCR library that can detect text on images",
   "keywords": [
     "n8n-community-node-package"
@@ -53,6 +53,6 @@
     "n8n-workflow": "*"
   },
   "dependencies": {
-    "tesseract.js": "^5.1.1"
+    "tesseract.js": "^6.0.0"
   }
 }