Bring up to date with puppeteer-extra-plugin-stealth

README.md

@@ -50,6 +50,7 @@ stealth(driver,
         webgl_vendor="Intel Inc.",
         renderer="Intel Iris OpenGL Engine",
         fix_hairline=True,
+        hardware_concurrency=4,
         )
 
 url = "https://bot.sannysoft.com/"
@@ -71,6 +72,7 @@ stealth(
     renderer: str = "Intel Iris OpenGL Engine",
     fix_hairline: bool = False,
     run_on_insecure_origins: bool = False,
+    hardware_concurrency: int = 4,
 )
 ```
 

selenium_stealth/__init__.py

@@ -14,6 +14,7 @@
 from .webgl_vendor import webgl_vendor_override
 from .window_outerdimensions import window_outerdimensions
 from .hairline_fix import hairline_fix
+from .navigator_hardware_concurrency import navigator_hardware_concurrency
 
 """
 If user_agent = None then selenium-stealth only remove the 'headless' from userAgent
@@ -32,10 +33,11 @@
 def stealth(driver: Driver, user_agent: str = None,
             languages: [str] = ["en-US", "en"],
             vendor: str = "Google Inc.",
-            platform: str = None,
+            platform: str = "Win32",
             webgl_vendor: str = "Intel Inc.",
             renderer: str = "Intel Iris OpenGL Engine",
             fix_hairline: bool = False,
+            hardware_concurrency: int = 4,
             run_on_insecure_origins: bool = False, **kwargs) -> None:
     if not isinstance(driver, Driver):
         raise ValueError("driver must is selenium.webdriver.Chrome, currently this lib only support Chrome")
@@ -47,6 +49,7 @@ def stealth(driver: Driver, user_agent: str = None,
     chrome_runtime(driver, run_on_insecure_origins, **kwargs)
     iframe_content_window(driver, **kwargs)
     media_codecs(driver, **kwargs)
+    navigator_hardware_concurrency(driver, hardware_concurrency, **kwargs)
     navigator_languages(driver, languages, **kwargs)
     navigator_permissions(driver, **kwargs)
     navigator_plugins(driver, **kwargs)

selenium_stealth/js/iframe.contentWindow.js

@@ -69,7 +69,7 @@
     // Adds a hook to intercept iframe creation events
     const addIframeCreationSniffer = () => {
       /* global document */
-      const createElement = {
+      const createElementHandler = {
         // Make toString() native
         get(target, key) {
           return Reflect.get(target, key)
@@ -86,9 +86,10 @@
         }
       }
       // All this just due to iframes with srcdoc bug
-      document.createElement = new Proxy(
-        document.createElement,
-        createElement
+      utils.replaceWithProxy(
+        document,
+        'createElement',
+        createElementHandler
       )
     }
 

selenium_stealth/js/navigator.hardwareConcurrency.js

@@ -0,0 +1,11 @@
+// https://github.com/berstend/puppeteer-extra/blob/3ea4ebca4237bb45ce402ba6a44d852e3499cb5f/packages/puppeteer-extra-plugin-stealth/evasions/navigator.hardwareConcurrency/index.js
+
+(hardwareConcurrency) => {
+    const patchNavigator = (name, value) =>
+        utils.replaceProperty(Object.getPrototypeOf(navigator), name, {
+            get() {
+                return value
+            }
+        })
+    patchNavigator('hardwareConcurrency', hardwareConcurrency || 4)
+}

selenium_stealth/js/navigator.plugins.js

@@ -66,7 +66,7 @@
           value,
           writable: false,
           enumerable: false, // Important for mimeTypes & plugins: `JSON.stringify(navigator.mimeTypes)`
-          configurable: false
+          configurable: true
         })
 
       // Loop over our fake data and construct items
@@ -78,8 +78,42 @@
           }
           defineProp(item, prop, data[prop])
         }
+        return patchItem(item, data)
+      }
+
+      const patchItem = (item, data) => {
+        let descriptor = Object.getOwnPropertyDescriptors(item)
+
+        // Special case: Plugins have a magic length property which is not enumerable
+        // e.g. `navigator.plugins[i].length` should always be the length of the assigned mimeTypes
+        if (itemProto === Plugin.prototype) {
+          descriptor = {
+            ...descriptor,
+            length: {
+              value: data.__mimeTypes.length,
+              writable: false,
+              enumerable: false,
+              configurable: true // Important to be able to use the ownKeys trap in a Proxy to strip `length`
+            }
+          }
+        }
+
         // We need to spoof a specific `MimeType` or `Plugin` object
-        return Object.create(itemProto, Object.getOwnPropertyDescriptors(item))
+        const obj = Object.create(itemProto, descriptor)
+
+        // Virtually all property keys are not enumerable in vanilla
+        const blacklist = [...Object.keys(data), 'length', 'enabledPlugin']
+        return new Proxy(obj, {
+          ownKeys(target) {
+            return Reflect.ownKeys(target).filter(k => !blacklist.includes(k))
+          },
+          getOwnPropertyDescriptor(target, prop) {
+            if (blacklist.includes(prop)) {
+              return undefined
+            }
+            return Reflect.getOwnPropertyDescriptor(target, prop)
+          }
+        })
       }
 
       const magicArray = []
@@ -144,6 +178,12 @@
           typeProps.forEach((_, i) => keys.push(`${i}`))
           typeProps.forEach(propName => keys.push(propName))
           return keys
+        },
+        getOwnPropertyDescriptor(target, prop) {
+          if (prop === 'length') {
+            return undefined
+          }
+          return Reflect.getOwnPropertyDescriptor(target, prop)
         }
       })
 
@@ -221,12 +261,18 @@
   for (const pluginData of data.plugins) {
     pluginData.__mimeTypes.forEach((type, index) => {
       plugins[pluginData.name][index] = mimeTypes[type]
-      plugins[type] = mimeTypes[type]
-      Object.defineProperty(mimeTypes[type], 'enabledPlugins', {
-        value: JSON.parse(JSON.stringify(plugins[pluginData.name])),
+
+      Object.defineProperty(plugins[pluginData.name], type, {
+        value: mimeTypes[type],
+        writable: false,
+        enumerable: false, // Not enumerable
+        configurable: true
+      })
+      Object.defineProperty(mimeTypes[type], 'enabledPlugin', {
+        value: new Proxy(plugins[pluginData.name], {}), // Prevent circular references
         writable: false,
         enumerable: false, // Important: `JSON.stringify(navigator.plugins)`
-        configurable: false
+        configurable: true
       })
     })
   }

selenium_stealth/navigator_hardware_concurrency.py

@@ -0,0 +1,9 @@
+from pathlib import Path
+from .wrapper import evaluateOnNewDocument
+from selenium.webdriver import Chrome as Driver
+
+
+def navigator_hardware_concurrency(driver: Driver, hardware_concurrency: int, **kwargs) -> None:
+    evaluateOnNewDocument(
+        driver, Path(__file__).parent.joinpath("js/navigator.hardwareConcurrency.js").read_text(), hardware_concurrency
+    )