phantomwright 0.1.2__tar.gz → 0.1.4__tar.gz

{phantomwright-0.1.2 → phantomwright-0.1.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: phantomwright
-Version: 0.1.2
+Version: 0.1.4
 Summary: Bridging playwright-core patch + extending playwright API for stealth injection & user simulation
 Project-URL: homepage, https://github.com/ai-microsoft/phantom-wright
 Project-URL: changelog, https://github.com/ai-microsoft/phantom-wright/blob/main/CHANGELOG.md
@@ -10,7 +10,7 @@ License-File: LICENSE
 Classifier: Operating System :: OS Independent
 Classifier: Programming Language :: Python :: 3
 Requires-Python: >=3.9
-Requires-Dist: phantomwright-driver==1.57.6
+Requires-Dist: phantomwright-driver==1.57.7
 Provides-Extra: black
 Requires-Dist: black>=25.9.0; extra == 'black'
 Provides-Extra: dev
@@ -183,6 +183,8 @@ async def main():
   }
   ```
 
+> ⚠️ **Note:** To prevent infinite detection loops caused by automatic page refreshes during captcha resolution, manual refreshes on the same URL will **not** trigger the resolve process again.
+
 ## Development
 
 ### Setup & Test
@@ -256,6 +258,10 @@ CDP does not provide endpoints to manipulate WebSocket. Supporting this would re
   await page.route_web_socket("/ws", handler)
   ```
 
+#### `add_init_script` Not Compatible with Edge New Tab Page Prerender
+
+As Edge NTP's prerendered New Tab page won't install Playwright route, `add_init_script` won't work in a prerendered session, plz ensure you disabled Edge's prerender when launch browser.
+
 #### `add_init_script` Timing Issue
 
 `add_init_script` cannot directly call bindings exposed by `expose_function`/`expose_binding`. Init scripts are injected into the HTML document and execute before exposed APIs are available.
@@ -280,7 +286,7 @@ CDP does not provide endpoints to manipulate WebSocket. Supporting this would re
 
 #### `add_init_script` Doesn't Affect Special URLs
 
-Patchright init scripts use routing, which doesn't trigger for `about:blank`, Data-URIs, or `file://` URLs.
+Patchright init scripts use routing, which doesn't trigger for `about:blank`, Data-URIs, `file://` URLs, as well as privilege pages such as `edge://newtab`.
 
 - ❌ **Data-URIs**
   ```python

{phantomwright-0.1.2 → phantomwright-0.1.4}/README.md

@@ -163,6 +163,8 @@ async def main():
   }
   ```
 
+> ⚠️ **Note:** To prevent infinite detection loops caused by automatic page refreshes during captcha resolution, manual refreshes on the same URL will **not** trigger the resolve process again.
+
 ## Development
 
 ### Setup & Test
@@ -236,6 +238,10 @@ CDP does not provide endpoints to manipulate WebSocket. Supporting this would re
   await page.route_web_socket("/ws", handler)
   ```
 
+#### `add_init_script` Not Compatible with Edge New Tab Page Prerender
+
+As Edge NTP's prerendered New Tab page won't install Playwright route, `add_init_script` won't work in a prerendered session, plz ensure you disabled Edge's prerender when launch browser.
+
 #### `add_init_script` Timing Issue
 
 `add_init_script` cannot directly call bindings exposed by `expose_function`/`expose_binding`. Init scripts are injected into the HTML document and execute before exposed APIs are available.
@@ -260,7 +266,7 @@ CDP does not provide endpoints to manipulate WebSocket. Supporting this would re
 
 #### `add_init_script` Doesn't Affect Special URLs
 
-Patchright init scripts use routing, which doesn't trigger for `about:blank`, Data-URIs, or `file://` URLs.
+Patchright init scripts use routing, which doesn't trigger for `about:blank`, Data-URIs, `file://` URLs, as well as privilege pages such as `edge://newtab`.
 
 - ❌ **Data-URIs**
   ```python

phantomwright-0.1.4/phantomwright/_repo_version.py

@@ -0,0 +1 @@
+version = 'v0.1.4'

{phantomwright-0.1.2 → phantomwright-0.1.4}/phantomwright/captcha/cloudfare/solver.py

@@ -1,9 +1,9 @@
 import asyncio
 import json
 import time
-from typing import Callable, Optional, Union
+from typing import Callable, Optional
 
-from phantomwright.async_api import Page, Browser, BrowserContext
+from phantomwright.async_api import Page, BrowserContext
 
 from .utils.consts import ChallengeType
 from .utils.detection import (
@@ -46,7 +46,7 @@ class CloudflareSolverAsync:
 
     def __init__(
         self,
-        context_or_browser: Union[BrowserContext, Browser],
+        context: BrowserContext,
         *,
         max_attempts: int = 3,
         attempt_delay: int = 5,
@@ -56,8 +56,7 @@ class CloudflareSolverAsync:
         Initialize the Cloudflare solver.
 
         Args:
-            context_or_browser: Playwright BrowserContext or Browser to monitor for challenges.
-                If a Browser is provided, all its contexts will be monitored.
+            context: Playwright BrowserContext to monitor for challenges.
             max_attempts: Maximum number of solve attempts per challenge.
             attempt_delay: Delay in seconds between retry attempts.
             log_callback: Optional callback function that receives a JSON string
@@ -71,9 +70,7 @@ class CloudflareSolverAsync:
                 - error: Error message if failed, None otherwise
                 - timestamp: Unix timestamp of the event
         """
-        self._is_browser = hasattr(context_or_browser, 'new_context')
-        self._browser: Optional[Browser] = context_or_browser if self._is_browser else None
-        self._context: Optional[BrowserContext] = None if self._is_browser else context_or_browser
+        self.context = context
         self.max_attempts = max_attempts
         self.attempt_delay = attempt_delay
         self.log = log_callback
@@ -143,7 +140,7 @@ class CloudflareSolverAsync:
                     report["challenge_type"] = challenge_type.name
 
                     if challenge_type is ChallengeType.TURNSTILE:
-                        await page.locator("#cf-turnstile").wait_for()
+                        await page.locator("#cf-turnstile").wait_for(10000)
 
                     cf_iframes = await search_shadow_root_iframes(
                         captcha_container=page,
@@ -203,6 +200,8 @@ class CloudflareSolverAsync:
 
         finally:
             report["duration"] = time.time() - report["start_time"]
+            if report["success"]:
+                report["url"] = None  # Clear URL on success for privacy
             self._log_final_report(report)
 
     # ---------------- callback ----------------
@@ -237,48 +236,20 @@ class CloudflareSolverAsync:
         await page.expose_function("__cf_callback", self._make_on_cf_detected(page))
         page.on("load", lambda: asyncio.create_task(self._rebind(page)))
 
-    def _setup_context(self, context: BrowserContext) -> None:
-        """Set up page monitoring for a single context."""
-        context.on(
-            "page",
-            lambda p: asyncio.create_task(self._setup_page(p)),
-        )
-
     # ---------------- public api ----------------
     def start(self) -> None:
         """
-        Start monitoring the browser or context for Cloudflare challenges.
+        Start monitoring the browser context for Cloudflare challenges.
 
-        This method registers event listeners to automatically detect and solve
-        Cloudflare challenges on any new page. Call this once after creating
-        the solver instance.
-
-        If initialized with a Browser, monitors all existing and new contexts.
-        If initialized with a BrowserContext, monitors only that context.
+        This method registers event listeners on the context to automatically
+        detect and solve Cloudflare challenges on any new page. Call this once
+        after creating the solver instance.
 
         Note:
             This method is synchronous but sets up async handlers internally.
             Challenges will be solved automatically in the background.
         """
-        if self._is_browser:
-            # Track which contexts we've already set up
-            self._monitored_contexts: set = set()
-            
-            def _check_and_setup_contexts():
-                for context in self._browser.contexts:
-                    if context not in self._monitored_contexts:
-                        self._monitored_contexts.add(context)
-                        self._setup_context(context)
-            
-            # Monitor all existing contexts
-            _check_and_setup_contexts()
-            
-            # Start a background task to periodically check for new contexts
-            async def _monitor_contexts():
-                while self._browser.is_connected():
-                    _check_and_setup_contexts()
-                    await asyncio.sleep(0.1)  # Check every 100ms for new contexts
-            
-            self._monitor_task = asyncio.create_task(_monitor_contexts())
-        else:
-            self._setup_context(self._context)
+        self.context.on(
+            "page",
+            lambda p: asyncio.create_task(self._setup_page(p)),
+        )

{phantomwright-0.1.2 → phantomwright-0.1.4}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "phantomwright"
-version = "0.1.2"
+version = "0.1.4"
 description = "Bridging playwright-core patch + extending playwright API for stealth injection & user simulation"
 authors = [
     {name="Hang Yin", email="hangyin@microsoft.com"},
@@ -11,7 +11,7 @@ readme = "README.md"
 requires-python = ">=3.9"
 
 dependencies = [
-    "phantomwright-driver==1.57.6",
+    "phantomwright-driver==1.57.7",
 ]
 
 classifiers = [

phantomwright-0.1.2/phantomwright/_repo_version.py

@@ -1 +0,0 @@
-version = 'v0.1.2'