v1.5.0

emunium/emunium.py

@@ -1,29 +1,56 @@
 import asyncio
-import keyboard
-import time
+import math
+import os
 import random
+import tempfile
+import time
+import struct
+
+import keyboard
 import pyautogui
-import math
 import pyclick
 
+def get_image_size(file_path):
+    with open(file_path, 'rb') as file:
+        file.seek(16)
+        width_bytes = file.read(4)
+        height_bytes = file.read(4)
+        width = struct.unpack('>I', width_bytes)[0]
+        height = struct.unpack('>I', height_bytes)[0]
+        return (width, height,)
+
 class EmuniumSelenium:
     def __init__(self, driver):
         self.driver = driver
         self.clicker = pyclick.HumanClicker()
+        self.browser_offsets = ()
+        self.browser_inner_window = ()
 
+    def _get_browser_properties_if_not_found(self):
+        if not self.browser_offsets or not self.browser_inner_window:
+            with tempfile.NamedTemporaryFile(suffix='.png', delete=False) as temp_file:
+                temp_screen_path = temp_file.name
+            self.driver.save_screenshot(temp_screen_path)
+            location = pyautogui.locateOnScreen(temp_screen_path, confidence=0.8)
+            self.browser_offsets = (location.left, location.top,)
+            self.browser_inner_window = get_image_size(temp_screen_path)
+            os.remove(temp_screen_path)
+
     def get_center(self, element):
-        coords = self.driver.execute_script("""
-            var element = arguments[0];
-            var rect = element.getBoundingClientRect();
-            var centerX = rect.left + rect.width / 2 + window.scrollX;
-            var centerY = rect.top + rect.height / 2 + (window.scrollY || window.pageYOffset) + (screen.height - window.innerHeight) / 1.5;
+        self._get_browser_properties_if_not_found()
 
-            return {x: centerX, y: centerY};
-        """, element)
+        element_location = element.location
+        offset_to_screen_x, offset_to_screen_y = self.browser_offsets
+        element_x = element_location['x'] + offset_to_screen_x
+        element_y = element_location['y'] + offset_to_screen_y
 
+        element_size = element.size
+        centered_x = element_x + (element_size['width'] // 2)
+        centered_y = element_y + (element_size['height'] // 2)
+
         return {
-            'x': coords['x'],
-            'y': coords['y']
+            'x': centered_x,
+            'y': centered_y
         }
 
     def find_and_move(self, element, click=False, offset_x=random.uniform(0.0, 1.5), offset_y=random.uniform(0.0, 1.5)):
@@ -52,42 +79,64 @@ def silent_type(self, text, characters_per_minute=280, offset=20):
             time.sleep(delay)
 
     def scroll_smoothly_to_element(self, element):
-        def is_element_in_viewport(element):
-            return self.driver.execute_script("""
-                var rect = arguments[0].getBoundingClientRect();
-                return (
-                    rect.top >= 0 &&
-                    rect.left >= 0 &&
-                    rect.bottom <= (window.innerHeight || document.documentElement.clientHeight) &&
-                    rect.right <= (window.innerWidth || document.documentElement.clientWidth)
-                );
-            """, element)
-
-        def scroll_to_element(element):
-            self.driver.execute_script("arguments[0].scrollIntoView({ behavior: 'smooth', block: 'start' });", element)
-
-        interval = 0.1
-        start_time = time.time()
-
-        while not is_element_in_viewport(element):
-            if time.time() - start_time > 5:
-                break
-
-            scroll_to_element(element)
-            time.sleep(interval)
+        self._get_browser_properties_if_not_found()
+
+        element_rect = element.rect        
+        window_width = self.browser_inner_window[0]
+        window_height = self.browser_inner_window[1]
+
+        scroll_amount = element_rect['y'] - window_height // 2
+        scroll_steps = abs(scroll_amount) // 100
+
+        if scroll_amount > 0:
+            scroll_direction = -1
+        else:
+            scroll_direction = 1
+
+        for _ in range(scroll_steps):
+            pyautogui.scroll(scroll_direction * 100)
+            time.sleep(random.uniform(0.05, 0.1))
+
+        remaining_scroll = scroll_amount % 100
+        if remaining_scroll != 0:
+            pyautogui.scroll(scroll_direction * remaining_scroll)
+            time.sleep(random.uniform(0.05, 0.1))
 
 class EmuniumPpeteer:
     def __init__(self, page):
         self.page = page        
         self.clicker = pyclick.HumanClicker()
+        self.browser_offsets = ()
+        self.browser_inner_window = ()
+
+    async def _get_browser_properties_if_not_found(self):
+        if not self.browser_offsets or not self.browser_inner_window:
+            with tempfile.NamedTemporaryFile(suffix='.png', delete=False) as temp_file:
+                temp_screen_path = temp_file.name
+            await self.page.screenshot(path=temp_screen_path)
+            location = pyautogui.locateOnScreen(temp_screen_path, confidence=0.8)
+            self.browser_offsets = (location.left, location.top,)
+            self.browser_inner_window = get_image_size(temp_screen_path)
+            os.remove(temp_screen_path)
 
     async def get_center(self, element):
-        return await self.page.evaluate('''(element) => {
-            const rect = element.getBoundingClientRect();
-            const centerX = rect.left + rect.width / 2 + window.scrollX;
-            const centerY = rect.top + rect.height / 2 + (window.scrollY || window.pageYOffset) + (screen.height - window.innerHeight) / 1.4;
-            return {x: centerX, y: centerY};
-        }''', element)
+        await self._get_browser_properties_if_not_found()
+
+        rect = await element.boundingBox()
+        if rect is None:
+            return None
+
+        offset_to_screen_x, offset_to_screen_y = self.browser_offsets
+        element_x = rect['x'] + offset_to_screen_x
+        element_y = rect['y'] + offset_to_screen_y
+        element_width = rect['width']
+        element_height = rect['height']
+        centered_x = element_x + (element_width // 2)
+        centered_y = element_y + (element_height // 2)
+        return {
+            'x': centered_x,
+            'y': centered_y
+        }
 
     async def find_and_move(self, element, click=False, offset_x=random.uniform(0.0, 1.5), offset_y=random.uniform(0.0, 1.5)):
         center = await self.get_center(element)        
@@ -115,26 +164,28 @@ async def silent_type(self, text, characters_per_minute=280, offset=20):
             await asyncio.sleep(delay)
 
     async def scroll_smoothly_to_element(self, element):
-        is_element_in_viewport = await self.page.evaluate('''(element) => {
-            const rect = element.getBoundingClientRect();
-            return (
-                rect.top >= 0 &&
-                rect.left >= 0 &&
-                rect.bottom <= (window.innerHeight || document.documentElement.clientHeight) &&
-                rect.right <= (window.innerWidth || document.documentElement.clientWidth)
-            );
-        }''', element)
-
-        scroll_to_element = await self.page.evaluate('''(element) => {
-            element.scrollIntoView({ behavior: "smooth", block: "start" });
-        }''', element)
-
-        interval = 0.1
-        start_time = time.time()
-
-        while not is_element_in_viewport:
-            if time.time() - start_time > 5:
-                break
-
-            await scroll_to_element
-            await asyncio.sleep(interval)
+        await self._get_browser_properties_if_not_found()
+
+        element_rect = await element.boundingBox()
+        if element_rect is None:
+            return None
+
+        window_width = self.browser_inner_window[0]
+        window_height = self.browser_inner_window[1]
+
+        scroll_amount = element_rect['y'] - window_height // 2
+        scroll_steps = abs(scroll_amount) // 100
+
+        if scroll_amount > 0:
+            scroll_direction = -1
+        else:
+            scroll_direction = 1
+
+        for _ in range(scroll_steps):
+            pyautogui.scroll(scroll_direction * 100)
+            await asyncio.sleep(random.uniform(0.05, 0.1))
+
+        remaining_scroll = scroll_amount % 100
+        if remaining_scroll != 0:
+            pyautogui.scroll(scroll_direction * remaining_scroll)
+            await asyncio.sleep(random.uniform(0.05, 0.1))

setup.py

@@ -5,7 +5,7 @@
 
 setup(
     name='emunium',
-    version='1.0.2',
+    version='1.5.0',
     author='Maehdakvan',
     author_email='visitanimation@google.com',
     description='A Python module for automating interactions to mimic human behavior in browsers when using Selenium or Pyppeteer. Provides utilities to programmatically move the mouse cursor, click on page elements, type text, and scroll as if performed by a human user.',