Merge pull request #4252 from seleniumbase/cdp-and-docker-updates

CDP and Docker updates

Author: mdmintz

Dockerfile

@@ -1,8 +1,9 @@
 # SeleniumBase Docker Image
-FROM ubuntu:22.04
+FROM ubuntu:24.04
 SHELL ["/bin/bash", "-o", "pipefail", "-c"]
 ENV PYTHONUNBUFFERED=1
 ENV PYTHONIOENCODING=UTF-8
+ENV DEBIAN_FRONTEND=noninteractive
 
 #======================
 # Locale Configuration
@@ -23,49 +24,39 @@ RUN locale-gen en_US.UTF-8
 #===========================
 # Fingerprint Configuration
 #===========================
-RUN apt install -y xvfb fonts-liberation fonts-noto-color-emoji libvulkan1 libnss3 libatk-bridge2.0-0 libcups2 libxcomposite1 libxrandr2 libgbm1 libpango-1.0-0 libcairo2 libasound2
+RUN apt-get update
+RUN apt install -y fonts-liberation fonts-noto-color-emoji libvulkan1 libnss3 libatk-bridge2.0-0 libcups2 libxcomposite1 libxrandr2 libgbm1 libpango-1.0-0 libcairo2
 RUN apt install -y fonts-freefont-ttf fonts-dejavu-core fonts-ubuntu fonts-roboto fonts-droid-fallback
 
 #======================
 # Install Common Fonts
 #======================
 RUN apt-get update
 RUN apt-get install -y \
-    fonts-liberation \
     fonts-liberation2 \
     fonts-font-awesome \
-    fonts-ubuntu \
     fonts-terminus \
     fonts-powerline \
     fonts-open-sans \
     fonts-mononoki \
-    fonts-roboto \
     fonts-lato
 
 #============================
 # Install Linux Dependencies
 #============================
 RUN apt-get update
 RUN apt-get install -y \
-    libasound2 \
-    libatk-bridge2.0-0 \
     libatk1.0-0 \
     libatspi2.0-0 \
-    libcups2 \
     libdbus-1-3 \
     libdrm2 \
-    libgbm1 \
     libgtk-3-0 \
     libnspr4 \
-    libnss3 \
     libu2f-udev \
-    libvulkan1 \
     libwayland-client0 \
-    libxcomposite1 \
     libxdamage1 \
     libxfixes3 \
-    libxkbcommon0 \
-    libxrandr2
+    libxkbcommon0
 
 #==========================
 # Install useful utilities
@@ -96,13 +87,20 @@ RUN rm ./google-chrome-stable_current_amd64.deb
 #================
 # Install Python
 #================
+RUN apt-get update && apt-get install -y software-properties-common
+RUN add-apt-repository ppa:deadsnakes/ppa -y
+RUN apt-get update
+RUN apt-get install -y python3.13 python3.13-venv python3.13-dev build-essential
+RUN update-alternatives --install /usr/bin/python3 python3 /usr/bin/python3.13 1
+RUN apt-get clean && rm -rf /var/lib/apt/lists/*
+RUN python3.13 -m ensurepip --upgrade
+RUN python3.13 -m pip install --upgrade pip
 RUN apt-get update
-RUN apt-get install -y python3 python3-pip python3-setuptools python3-dev python3-tk
+RUN apt-get install -y python3.13-tk python3.13-dev
 RUN alias python=python3
 RUN echo "alias python=python3" >> ~/.bashrc
-RUN apt-get -qy --no-install-recommends install python3.10
 RUN rm /usr/bin/python3
-RUN ln -s python3.10 /usr/bin/python3
+RUN ln -s python3.13 /usr/bin/python3
 
 #===============
 # Cleanup Lists
@@ -128,6 +126,8 @@ RUN pip install --upgrade pip setuptools wheel
 RUN cd /SeleniumBase && ls && pip install -r requirements.txt --upgrade
 RUN cd /SeleniumBase && pip install .
 RUN pip install pyautogui
+RUN pip install playwright
+RUN seleniumbase get chromium
 
 #=======================
 # Download chromedriver

README.md

@@ -186,6 +186,16 @@ pytest test_demo_site.py
 
 --------
 
+<p align="left">📓 Here's a high-level stealthy architecture overview of SeleniumBase:</p>
+
+
+
+<img src="https://seleniumbase.github.io/other/sb_stealth.png" width="650" alt="High-Level Stealthy Architecture Overview" title="High-Level Stealthy Architecture Overview" />
+
+(For maximum stealth, use <a translate="no" href="https://github.com/seleniumbase/SeleniumBase/blob/master/examples/cdp_mode/ReadMe.md">CDP Mode</a>, which is used by <a translate="no" href="https://github.com/seleniumbase/SeleniumBase/blob/master/examples/cdp_mode/playwright/ReadMe.md">Stealthy Playwright Mode</a>)
+
+--------
+
 <p align="left"><a href="https://github.com/seleniumbase/SeleniumBase/"><img src="https://seleniumbase.github.io/cdn/img/super_logo_sb3.png" alt="SeleniumBase" title="SeleniumBase" width="232" /></a></p>
 
 <blockquote>

examples/cdp_mode/playwright/raw_idealista_nested.py

@@ -0,0 +1,25 @@
+"""(Bypasses the DataDome slider CAPTCHA)"""
+from playwright.sync_api import sync_playwright
+from seleniumbase import SB
+
+with SB(uc=True, locale="es") as sb:
+    url = "https://www.idealista.com/venta-viviendas/barcelona-provincia/"
+    sb.activate_cdp_mode(url)
+    sb.sleep(1)
+    sb.solve_captcha()
+    sb.sleep(2)
+    endpoint_url = sb.cdp.get_endpoint_url()
+
+    with sync_playwright() as p:
+        browser = p.chromium.connect_over_cdp(endpoint_url)
+        context = browser.contexts[0]
+        page = context.pages[0]
+        page.click("button#didomi-notice-agree-button")
+        page.wait_for_timeout(1000)
+        print("*** " + page.locator("h1").inner_text())
+        items = page.locator("div.item-info-container")
+        for i in range(items.count()):
+            item = items.nth(i)
+            print(item.locator("a.item-link").text_content().strip())
+            print(item.locator("span.item-price").text_content().strip())
+            item_text = items.nth(i)

examples/cdp_mode/raw_idealista.py

@@ -0,0 +1,15 @@
+"""(Bypasses the DataDome slider CAPTCHA)"""
+from seleniumbase import SB
+
+with SB(uc=True, test=True, locale="es") as sb:
+    url = "https://www.idealista.com/venta-viviendas/barcelona-provincia/"
+    sb.activate_cdp_mode(url)
+    sb.sleep(1)
+    sb.solve_captcha()
+    sb.sleep(2)
+    sb.click("button#didomi-notice-agree-button")
+    print("*** " + sb.get_text("h1"))
+    items = sb.find_elements("div.item-info-container")
+    for item in items:
+        print(item.querySelector("a.item-link").text)
+        print(item.querySelector("span.item-price").text)

mkdocs_build/requirements.txt

@@ -1,9 +1,9 @@
 # mkdocs dependencies for generating the seleniumbase.io website
 # Minimum Python version: 3.10 (for generating docs only)
 
-regex>=2026.1.15
-pymdown-extensions>=10.20.1
-pipdeptree>=2.30.0
+regex>=2026.2.19
+pymdown-extensions>=10.21
+pipdeptree>=2.31.0
 python-dateutil>=2.8.2
 Markdown==3.10.2
 click==8.3.1

requirements.txt

@@ -9,18 +9,18 @@ exceptiongroup>=1.3.1
 websockets~=15.0.1;python_version<"3.10"
 websockets>=16.0;python_version>="3.10"
 filelock~=3.19.1;python_version<"3.10"
-filelock>=3.20.3;python_version>="3.10"
+filelock>=3.24.3;python_version>="3.10"
 fasteners>=0.20
 mycdp>=1.3.2
 pynose>=1.5.5
 platformdirs~=4.4.0;python_version<"3.10"
-platformdirs>=4.5.1;python_version>="3.10"
+platformdirs>=4.9.2;python_version>="3.10"
 typing-extensions>=4.15.0
 sbvirtualdisplay>=1.4.0
 MarkupSafe>=3.0.3
 Jinja2>=3.1.6
 six>=1.17.0
-parse>=1.21.0
+parse>=1.21.1
 parse-type>=0.6.6
 colorama>=0.4.6
 pyyaml>=6.0.3
@@ -38,13 +38,13 @@ sniffio==1.3.1
 h11==0.16.0
 outcome==1.3.0.post0
 trio>=0.31.0,<1;python_version<"3.10"
-trio>=0.32.0,<1;python_version>="3.10"
+trio>=0.33.0,<1;python_version>="3.10"
 trio-websocket~=0.12.2
 wsproto==1.2.0;python_version<"3.10"
 wsproto~=1.3.2;python_version>="3.10"
 websocket-client~=1.9.0
 selenium==4.32.0;python_version<"3.10"
-selenium==4.40.0;python_version>="3.10"
+selenium==4.41.0;python_version>="3.10"
 cssselect==1.3.0;python_version<"3.10"
 cssselect>=1.4.0,<2;python_version>="3.10"
 nest-asyncio==1.6.0
@@ -72,7 +72,7 @@ PyAutoGUI>=0.9.54;platform_system=="Linux"
 markdown-it-py==3.0.0;python_version<"3.10"
 markdown-it-py==4.0.0;python_version>="3.10"
 mdurl==0.1.2
-rich>=14.3.2,<15
+rich>=14.3.3,<15
 
 # --- Testing Requirements --- #
 # ("pip install -r requirements.txt" also installs this, but "pip install -e ." won't.)

seleniumbase/__version__.py

@@ -1,2 +1,2 @@
 # seleniumbase package
-__version__ = "4.46.5"
+__version__ = "4.47.0"

seleniumbase/core/browser_launcher.py

@@ -1501,9 +1501,12 @@ def _uc_gui_click_captcha(
                 ):
                     frame = '[data-callback="onCaptchaSuccess"]'
                 elif driver.is_element_present(
-                    "div:not([class]) > div:not([class])"
+                    "div:not([class]):not([id]) > div:not([class]):not([id])"
                 ):
-                    frame = "div:not([class]) > div:not([class])"
+                    frame = (
+                        "div:not([class]):not([id]) > "
+                        "div:not([class]):not([id])"
+                    )
                 else:
                     return False
             if (

seleniumbase/core/sb_cdp.py

@@ -1981,6 +1981,16 @@ def _on_an_incapsula_hcaptcha_page(self, *args, **kwargs):
             return True
         return False
 
+    def _on_a_datadome_slider_page(self, *args, **kwargs):
+        self.loop.run_until_complete(self.page.wait())
+        if (
+            self.is_element_visible(
+                'body > iframe[src*="/geo.captcha-delivery.com/captcha/"]'
+            )
+        ):
+            return True
+        return False
+
     def _on_a_g_recaptcha_page(self, *args, **kwargs):
         time.sleep(0.4)  # reCAPTCHA may need a moment to appear
         self.loop.run_until_complete(self.page.wait())
@@ -2054,6 +2064,26 @@ def __gui_click_recaptcha(self, use_cdp=False):
             return True
         return False
 
+    def __gui_slide_datadome_captcha(self):
+        iframe = 'body > iframe[src*="/geo.captcha-delivery.com/captcha/"]'
+        if not self.is_element_visible(iframe):
+            return False
+        src = self.get_attribute(iframe, "src")
+        tab = self.get_active_tab()
+        self.open_new_tab(url=src)
+        time.sleep(0.41)
+        self.loop.run_until_complete(self.page.wait())
+        time.sleep(0.25)
+        x1, y1 = self.get_gui_element_center("div.slider")
+        x2, y2 = self.get_gui_element_center("div.sliderTarget")
+        self.close_active_tab()
+        self.switch_to_tab(tab)
+        self.gui_drag_drop_points(x1, y1, x2, y2, timeframe=0.55)
+        time.sleep(0.25)
+        self.loop.run_until_complete(self.page.wait())
+        time.sleep(0.15)
+        return True
+
     def __cdp_click_incapsula_hcaptcha(self):
         selector = "iframe[data-hcaptcha-widget-id]"
         if self.is_element_visible('iframe[src*="_Incapsula_Resource?"]'):
@@ -2128,6 +2158,9 @@ def __click_captcha(self, use_cdp=False):
         elif self._on_an_incapsula_hcaptcha_page():
             result = self.__cdp_click_incapsula_hcaptcha()
             return result
+        elif self._on_a_datadome_slider_page():
+            result = self.__gui_slide_datadome_captcha()
+            return result
         else:
             return False
         selector = None
@@ -2174,9 +2207,11 @@ def __click_captcha(self, use_cdp=False):
         ):
             selector = '[data-callback="onCaptchaSuccess"]'
         elif self.is_element_present(
-            "div:not([class]) > div:not([class])"
+            "div:not([class]):not([id]) > div:not([class]):not([id])"
         ):
-            selector = "div:not([class]) > div:not([class])"
+            selector = (
+                "div:not([class]):not([id]) > div:not([class]):not([id])"
+            )
         else:
             return False
         if not selector:

seleniumbase/undetected/cdp_driver/tab.py

@@ -1669,9 +1669,11 @@ async def solve_captcha(self):
         ):
             selector = '[data-callback="onCaptchaSuccess"]'
         elif await self.is_element_present(
-            "div:not([class]) > div:not([class])"
+            "div:not([class]):not([id]) > div:not([class]):not([id])"
         ):
-            selector = "div:not([class]) > div:not([class])"
+            selector = (
+                "div:not([class]):not([id]) > div:not([class]):not([id])"
+            )
         else:
             return False
         if not selector: