From 7edbeba448122c882c8b7e51b698cb6bcdfabfdc Mon Sep 17 00:00:00 2001
From: G <52905881+giga-a@users.noreply.github.com>
Date: Fri, 5 Feb 2021 20:22:29 -0800
Subject: [PATCH] [auto-f] fixed ocr issues - ref 450ab97b

---
 modules/engine.js    | 50 ++++++++++++++++++++++++++------------------
 modules/slow-scan.js |  2 +-
 2 files changed, 31 insertions(+), 21 deletions(-)

diff --git a/modules/engine.js b/modules/engine.js
index 3b2718d..afece0a 100644
--- a/modules/engine.js
+++ b/modules/engine.js
@@ -1,5 +1,6 @@
 var helper = require('./helper.js')
 var tesseract = require("tesseract.js")
+const { createWorker } = require('tesseract.js');
 
 function merge_dicts(temp_dict){
   result = {}
@@ -53,29 +54,38 @@ async function detect_logic(type, uuid, username, options, site, source = "", sc
         detections_count += 1
         temp_detected.count += 1
         var temp_found = "false"
-        if (detection.type == "ocr" && screen_shot != "" && options.includes("FindUserProfilesSlowTempDisable")) {
+        if (detection.type == "ocr" && screen_shot != "" && options.includes("FindUserProfilesSlow")) {
           const temp_buffer_image = Buffer.from(screen_shot, "base64")
-          await tesseract.recognize(temp_buffer_image, "eng").then(result => {
-              text = result.data.text.replace(/[^A-Za-z0-9]/gi, "");
-              detection.string = detection.string.replace(/[^A-Za-z0-9]/gi, "");
-              if (text != "") {
-                if (text.toLowerCase().includes(detection.string.toLowerCase())) {
-                  temp_found = "true";
-                }
-                if (detection.return == temp_found) {
-                  temp_profile.found += 1
-                  temp_detected.ocr += 1
-                  if (detection.return == 'true') {
-                    temp_detected.true += 1
-                  } else {
-                    temp_detected.false += 1
-                  }
+          const ocr_worker = createWorker();
+          try{
+            await ocr_worker.load();
+            await ocr_worker.loadLanguage('eng');
+            await ocr_worker.initialize('eng');
+            const { data: { text } } = await ocr_worker.recognize(temp_buffer_image);
+            await ocr_worker.terminate();
+            if (text != "") {
+              if (text.toLowerCase().includes(detection.string.toLowerCase())) {
+                temp_found = "true";
+              }
+              if (detection.return == temp_found) {
+                temp_profile.found += 1
+                temp_detected.ocr += 1
+                if (detection.return == 'true') {
+                  temp_detected.true += 1
+                } else {
+                  temp_detected.false += 1
                 }
               }
-            })
-            .catch(err => {
-              helper.verbose && console.log(err);
-            })
+            }
+            else{
+              detections_count -= 1
+              temp_detected.count -= 1
+            }
+          }
+          catch(err){
+            detections_count -= 1
+            temp_detected.count -= 1
+          }
         } else if (detection.type == "normal" && source != "") {
           if (source.toLowerCase().includes(detection.string.replace("{username}", username).toLowerCase())) {
             temp_found = "true";
diff --git a/modules/slow-scan.js b/modules/slow-scan.js
index b6719a7..368d2c6 100644
--- a/modules/slow-scan.js
+++ b/modules/slow-scan.js
@@ -84,7 +84,7 @@ async function find_username_site(uuid, username, options, site) {
       var text_only = "unavailable";
       var title = "unavailable";
       var link = site.url.replace("{username}", username);
-      await driver.manage().setTimeouts(timeouts);
+      //await driver.manage().setTimeouts(timeouts);
       await driver.get(link);;
       source = await driver.getPageSource();
       screen_shot = await driver.takeScreenshot();