phantomfetch 0.5.0__tar.gz → 0.5.2__tar.gz

{phantomfetch-0.5.0 → phantomfetch-0.5.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: phantomfetch
-Version: 0.5.0
+Version: 0.5.2
 Summary: High-performance agentic web scraping library combining curl-cffi speed with Playwright browser capabilities
 Keywords: web-scraping,playwright,curl-cffi,async,browser-automation,http-client,agentic,anti-detection
 Author: CosmicBull

{phantomfetch-0.5.0 → phantomfetch-0.5.2}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "phantomfetch"
-version = "0.5.0"
+version = "0.5.2"
 description = "High-performance agentic web scraping library combining curl-cffi speed with Playwright browser capabilities"
 readme = "README.md"
 requires-python = ">=3.13"

{phantomfetch-0.5.0 → phantomfetch-0.5.2}/src/phantomfetch/engines/browser/cdp.py

@@ -1,4 +1,5 @@
 import asyncio
+import random
 import re
 import time
 from typing import TYPE_CHECKING, Any, Literal, Optional
@@ -82,6 +83,8 @@ class CDPEngine:
         cloak_binary_path: str | None = None,
         persistent_context_dir: str | None = None,
         ignore_https_errors: bool = False,
+        max_retries: int = 3,
+        retry_backoff_base: float = 2.0,
     ):
         """
         Args:
@@ -126,7 +129,9 @@ class CDPEngine:
                                    persist across sessions). When set, uses CloakBrowser's
                                    persistent context API. Enables incognito bypass.
             ignore_https_errors: Ignore TLS certificate errors (useful for proxies that
-                                do SSL inspection/MITM). Default False.
+                                 do SSL inspection/MITM). Default False.
+            max_retries: Max retries for CDP endpoint connection attempts. Default 3.
+            retry_backoff_base: Exponential backoff base for retries. Default 2.0.
         """
         self.cdp_endpoint = cdp_endpoint
         self.headless = headless
@@ -148,6 +153,8 @@ class CDPEngine:
         self.cloak_binary_path = cloak_binary_path
         self.persistent_context_dir = persistent_context_dir
         self.ignore_https_errors = ignore_https_errors
+        self.max_retries = max_retries
+        self.retry_backoff_base = retry_backoff_base
 
         self._cloak_browser_available = False
         self._cloak_context: Any = None
@@ -185,11 +192,30 @@ class CDPEngine:
         self._existing_page: Any = None
 
     async def connect(self) -> None:
-        """Initialize Playwright and connect to browser."""
-        if self.cloak_browser:
-            await self._connect_cloakbrowser()
-        else:
-            await self._connect_playwright()
+        """Initialize Playwright and connect to browser with retry."""
+        last_error: Exception | None = None
+        for attempt in range(self.max_retries):
+            try:
+                if self.cloak_browser:
+                    await self._connect_cloakbrowser()
+                else:
+                    await self._connect_playwright()
+                return
+            except Exception as e:
+                last_error = e
+                if attempt < self.max_retries - 1:
+                    wait = self.retry_backoff_base**attempt * (0.5 + random.random())
+                    logger.warning(
+                        f"[cdp] Connect attempt {attempt + 1}/{self.max_retries} failed: {e}. "
+                        f"Retrying in {wait:.2f}s..."
+                    )
+                    await asyncio.sleep(wait)
+                else:
+                    logger.error(
+                        f"[cdp] Connect attempt {attempt + 1}/{self.max_retries} failed: {e}"
+                    )
+        if last_error:
+            raise last_error
 
     async def _connect_cloakbrowser(self) -> None:
         """Connect using CloakBrowser's stealth Chromium binary.

{phantomfetch-0.5.0 → phantomfetch-0.5.2}/src/phantomfetch/fetch.py

@@ -47,7 +47,7 @@ class Fetcher:
         # Advanced CDP
         cdp_use_existing_page: bool = True,
         cdp_connection_type: str = "cdp",
-        backend: Literal["rebrowser", "playwright", "patchright"] = "rebrowser",
+        backend: Literal["rebrowser", "playwright", "patchright"] = "playwright",
         # BrowserForge fingerprinting
         fingerprint: bool = True,
         fingerprint_options: dict[str, Any] | None = None,
@@ -59,6 +59,8 @@ class Fetcher:
         cloak_browser_geoip: bool = False,
         cloak_binary_path: str | None = None,
         persistent_context_dir: str | None = None,
+        browser_max_retries: int = 3,
+        browser_retry_backoff_base: float = 2.0,
     ):
         """
         Initialize the Fetcher.
@@ -143,6 +145,8 @@ class Fetcher:
             cloak_browser_geoip=cloak_browser_geoip,
             cloak_binary_path=cloak_binary_path,
             persistent_context_dir=persistent_context_dir,
+            max_retries=browser_max_retries,
+            retry_backoff_base=browser_retry_backoff_base,
         )
         self._browser = self._cdp_engine
 
@@ -160,11 +164,11 @@ class Fetcher:
         self.max_retries = max_retries
 
     async def __aenter__(self) -> "Fetcher":
-        await self._browser.connect()
         return self
 
     async def __aexit__(self, *args: Any) -> None:
-        await self._browser.disconnect()
+        if self._cdp_engine:
+            await self._cdp_engine.disconnect()
 
     async def start(self) -> None:
         """