@xuda.io/drive_module 1.1.952 → 1.1.954

package/index.js

@@ -26,7 +26,7 @@ const crypto = require("node:crypto");
 const unzipper = require("unzipper");
 const fs = require("fs");
 const tesseract = require("node-tesseract-ocr");
-
+const franc = require("franc");
 ////////////////////////////////////////////
 
 [
@@ -1723,15 +1723,18 @@ const ocr_drive_file = async function (app_id, doc) {
     );
 
     var ocr = "";
+    const generalConfig = { lang: "eng", oem: 1, psm: 3 };
 
     switch (doc.mime.split("/")[0]) {
       case "image": {
-        ocr = await tesseract.recognize(target_file, {
-          // lang: "eng",
-          lang: "eng+heb",
-          oem: 1,
-          psm: 3,
-        });
+        const initialText = await tesseract.recognize(
+          target_file,
+          generalConfig
+        );
+
+        // Detect the language from the extracted text
+        const detectedLang = franc(initialText);
+        ocr = await tesseract.recognize(target_file, generalConfig);
 
         break;
       }
@@ -1743,11 +1746,14 @@ const ocr_drive_file = async function (app_id, doc) {
 
             const images = await convertPDF(target_file);
             for await (const imageData of images) {
-              ocr += await tesseract.recognize(imageData.path, {
-                // lang: "eng",
-                oem: 1,
-                psm: 3,
-              });
+              const initialText = await tesseract.recognize(
+                target_file,
+                generalConfig
+              );
+
+              // Detect the language from the extracted text
+              const detectedLang = franc(initialText);
+              ocr += await tesseract.recognize(imageData.path, generalConfig);
             }
             break;
           }

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@xuda.io/drive_module",
-  "version": "1.1.952",
+  "version": "1.1.954",
   "description": "Xuda Drive Server Module",
   "main": "index.js",
   "dependencies": {