Fixed JSON export bugs and added new tests (#987)

Balearica · web-flow · commit 6299aa558cdf · 2024-12-26T01:10:58.000-08:00
diff --git a/package-lock.json b/package-lock.json
diff --git a/package.json b/package.json
@@ -68,7 +68,7 @@
     "node-fetch": "^2.6.9",
     "opencollective-postinstall": "^2.0.3",
     "regenerator-runtime": "^0.13.3",
-    "tesseract.js-core": "^6.0.0-2",
+    "tesseract.js-core": "^6.0.0-3",
     "wasm-feature-detect": "^1.2.11",
     "zlibjs": "^0.3.1"
   },
diff --git a/tests/recognize.test.js b/tests/recognize.test.js
@@ -286,6 +286,24 @@ describe('recognize()', () => {
       expect(blocks[0].paragraphs[0].lines[1].text).to.be('Back \\ Slash\n');
     }).timeout(TIMEOUT);
 
+    it('recongize image with multiple choices', async () => {
+      await workerLegacy.reinitialize('eng');
+      const { data: { blocks } } = await workerLegacy.recognize(`${IMAGE_PATH}/bill.png`, {}, { blocks: true });
+      expect(blocks[0].paragraphs[1].lines[0].words[3].choices.length).to.be(3);
+      expect(blocks[0].paragraphs[1].lines[0].words[3].choices[1].text).to.be('100,000.0ll');
+    }).timeout(TIMEOUT);
+
+    it('recongize image with multiple blocks', async () => {
+      // This also implicitly checks that non-text blocks are ignored,
+      // as otherwise the length would be 5.
+      await worker.reinitialize('eng');
+      await worker.setParameters({
+        tessedit_pageseg_mode: PSM.AUTO,
+      });
+      const { data: { blocks } } = await worker.recognize(`${IMAGE_PATH}/bill.png`, {}, { blocks: true });
+      expect(blocks.length).to.be(4);
+    }).timeout(TIMEOUT);
+
     it('recongize chinese image', async () => {
       await worker.reinitialize('chi_tra');
       const { data: { blocks } } = await worker.recognize(`${IMAGE_PATH}/chinese.png`, {}, { blocks: true });
