@xuda.io/drive_module 1.1.975 → 1.1.977

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 countryLanguage = require("country-language");
 ////////////////////////////////////////////
 
 [
@@ -1723,21 +1723,31 @@ const ocr_drive_file = async function (app_id, doc) {
     );
 
     var ocr = "";
-    const generalConfig = { oem: 1, psm: 3 };
-    const { franc } = await import("franc");
 
+    function detectLanguageByCountry(cfCountry) {
+      const countryData = countryLanguage.getCountry(cfCountry);
+
+      // Check if the country data is available
+      if (
+        countryData &&
+        countryData.languages &&
+        countryData.languages.length > 0
+      ) {
+        // Return the first official language of the country
+        return countryData.languages[0].iso639_1;
+      }
+
+      // Default to English if no language found
+      return "en";
+    }
+    let lang = detectLanguageByCountry(doc.country);
+    if (lang !== "en") {
+      lang += "+en";
+    }
+    const generalConfig = { oem: 1, psm: 3, lang };
     switch (doc.mime.split("/")[0]) {
       case "image": {
-        const initialText = await tesseract.recognize(
-          target_file,
-          generalConfig
-        );
-
-        // Detect the language from the extracted text
-        const detectedLang = franc(initialText);
-        debugger;
-        const tesseractLang = detectedLang || "eng";
-        ocr = await tesseract.recognize(target_file, tesseractLang);
+        ocr = await tesseract.recognize(target_file, generalConfig);
 
         break;
       }
@@ -1749,15 +1759,7 @@ const ocr_drive_file = async function (app_id, doc) {
 
             const images = await convertPDF(target_file);
             for await (const imageData of images) {
-              const initialText = await tesseract.recognize(
-                imageData.path,
-                generalConfig
-              );
-
-              // Detect the language from the extracted text
-              const detectedLang = franc(initialText);
-              const tesseractLang = detectedLang || "eng";
-              ocr += await tesseract.recognize(imageData.path, tesseractLang);
+              ocr += await tesseract.recognize(imageData.path, generalConfig);
             }
             break;
           }

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@xuda.io/drive_module",
-  "version": "1.1.975",
+  "version": "1.1.977",
   "description": "Xuda Drive Server Module",
   "main": "index.js",
   "dependencies": {