Fixed bugs with wrong lang data being loaded per #834 and #835

naptha · Oct 3, 2023 · 0cad08a · 0cad08a
1 parent a1031fd
commit 0cad08a
Show file tree

Hide file tree

Showing 7 changed files with 42 additions and 11 deletions.
diff --git a/src/createWorker.js b/src/createWorker.js
@@ -127,7 +127,7 @@ module.exports = async (langs = 'eng', oem = OEM.LSTM_ONLY, _options = {}, confi
         cachePath: options.cachePath,
         cacheMethod: options.cacheMethod,
         gzip: options.gzip,
-        lstmOnly: [OEM.TESSERACT_ONLY, OEM.TESSERACT_LSTM_COMBINED].includes(currentOem)
+        lstmOnly: [OEM.LSTM_ONLY, OEM.TESSERACT_LSTM_COMBINED].includes(currentOem)
           && !options.legacyLang,
       },
     },
@@ -159,13 +159,13 @@ module.exports = async (langs = 'eng', oem = OEM.LSTM_ONLY, _options = {}, confi
     // This logic fails if the user downloaded the LSTM-only English data for a language
     // and then uses `worker.reinitialize` to switch to the Legacy engine.
     // However, the correct data will still be downloaded after initialization fails
-    // and this can be avoided entirely
+    // and this can be avoided entirely if the user loads the correct data ahead of time.
     const langsArr = typeof langs === 'string' ? langs.split('+') : langs;
-    const _langs = langsArr.filter((x) => currentLangs.includes(x));
+    const _langs = langsArr.filter((x) => !currentLangs.includes(x));
     currentLangs.push(_langs);
 
     return loadLanguageInternal(_langs, jobId)
-      .then(() => initializeInternal(_langs, _oem, _config, jobId));
+      .then(() => initializeInternal(langs, _oem, _config, jobId));
   };
 
   const setParameters = (params = {}, jobId) => (

diff --git a/tests/assets/traineddata/chi_tra.traineddata b/tests/assets/traineddata/chi_tra.traineddata
diff --git a/tests/assets/traineddata/chi_tra.traineddata.gz b/tests/assets/traineddata/chi_tra.traineddata.gz
diff --git a/tests/assets/traineddata/eng.traineddata.gz b/tests/assets/traineddata/eng.traineddata.gz
diff --git a/tests/assets/traineddata/osd.traineddata.gz b/tests/assets/traineddata/osd.traineddata.gz
diff --git a/tests/constants.js b/tests/constants.js
diff --git a/tests/recognize.test.js b/tests/recognize.test.js
@@ -2,7 +2,8 @@ const { createWorker, PSM } = Tesseract;
 let worker;
 before(async function cb() {
   this.timeout(0);
-  worker = await createWorker("eng+chi_tra+osd", 1, OPTIONS);
+  worker = await createWorker("eng", 1, OPTIONS);
+  workerLegacy = await createWorker("eng", 0, OPTIONS);
 });
 
 describe('recognize()', () => {
@@ -29,6 +30,19 @@ describe('recognize()', () => {
     ));
   });
 
+  describe('should recognize with Legacy OEM', () => {
+    [
+      { format: 'png', image: SIMPLE_PNG_BASE64, ans: SIMPLE_TEXT_LEGACY },
+      { format: 'jpg', image: SIMPLE_JPG_BASE64, ans: SIMPLE_TEXT_LEGACY },
+    ].forEach(({ format, image, ans }) => (
+      it(`recongize ${format} in base64`, async () => {
+        const { data: { text } } = await workerLegacy.recognize(image);
+        console.log(text);
+        expect(text).to.be(ans);
+      }).timeout(TIMEOUT)
+    ));
+  });
+
   describe('should support orientation metadata', () => {
     [
       { name: 'simple-90.jpg', desc: 'simple', ans: SIMPLE_TEXT },
@@ -125,13 +139,30 @@ describe('recognize()', () => {
     }).timeout(TIMEOUT);
   });
 
-  describe('should support all page seg modes', () => {
+  describe('should support all page seg modes (Legacy)', () => {
+    Object
+      .keys(PSM)
+      .map(name => ({ name, mode: PSM[name] }))
+      .forEach(({ name, mode }) => (
+        it(`support PSM.${name} mode`, async () => {
+          await workerLegacy.reinitialize('eng+osd');
+          await workerLegacy.setParameters({
+            tessedit_pageseg_mode: mode,
+          });
+          const { data } = await workerLegacy.recognize(`${IMAGE_PATH}/simple.png`);
+          expect(Object.keys(data).length).not.to.be(0);
+        }).timeout(TIMEOUT)
+      ));
+  });
+
+  describe('should support all page seg modes except for PSM.OSD_ONLY (LSTM)', () => {
     Object
       .keys(PSM)
+      .filter((x) => x !== 'OSD_ONLY')
       .map(name => ({ name, mode: PSM[name] }))
       .forEach(({ name, mode }) => (
         it(`support PSM.${name} mode`, async () => {
-          await worker.reinitialize('eng');
+          await worker.reinitialize('eng+osd');
           await worker.setParameters({
             tessedit_pageseg_mode: mode,
           });