freedak/Scrapling
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

refactor/fix(browser fetchers): Solving a logic bug in page rotation

Browse Source 
- Pages that get reused in rotation are possibly contaminated from previous settings used on them, and some are stubborn to remove, so the new approach replaces finished pages with new ones.
- Removed the `max_pages` from sync `StealthySession` to match `DynamicSession` (Doesn't mean anything in sync code)
This commit is contained in:
Karim shoair
2025-09-08 16:08:57 +03:00
parent f298b76b8e
commit ecaec65049
5 changed files with 521 additions and 455 deletions
Download Patch File Download Diff File Expand all files Collapse all files
scrapling/engines/_browsers/_base.py
+403
View File
@@ -0,0 +1,403 @@
from time import time, sleep
from asyncio import sleep as asyncio_sleep, Lock
from camoufox import DefaultAddons
from playwright.sync_api import BrowserContext, Playwright
from playwright.async_api import (
BrowserContext as AsyncBrowserContext,
Playwright as AsyncPlaywright,
)
from camoufox.utils import (
launch_options as generate_launch_options,
installed_verstr as camoufox_version,
)
from scrapling.engines.toolbelt import (
intercept_route,
async_intercept_route,
get_os_name,
)
from ._page import PageInfo, PagePool
from ._config_tools import _compiled_stealth_scripts
from ._validators import validate, PlaywrightConfig, CamoufoxConfig
from ._config_tools import _launch_kwargs, _context_kwargs
from scrapling.core._types import (
Dict,
Optional,
)
__ff_version_str__ = camoufox_version().split(".", 1)[0]
class SyncSession:
def __init__(self, max_pages: int = 1):
self.max_pages = max_pages
self.page_pool = PagePool(max_pages)
self.__max_wait_for_page = 60
self.playwright: Optional[Playwright] = None
self.context: Optional[BrowserContext] = None
self._closed = False
def _get_page(self) -> PageInfo: # pragma: no cover
"""Get a new page to use"""
# Close all finished pages to ensure clean state
self.page_pool.close_all_finished_pages()
# If we're at max capacity after cleanup, wait for busy pages to finish
if self.page_pool.pages_count >= self.max_pages:
start_time = time()
while time() - start_time < self.__max_wait_for_page:
# Wait for any pages to finish, then clean them up
sleep(0.05)
self.page_pool.close_all_finished_pages()
if self.page_pool.pages_count < self.max_pages:
break
else:
raise TimeoutError(
f"No pages finished to clear place in the pool within the {self.__max_wait_for_page}s timeout period"
)
page = self.context.new_page()
timeout = getattr(self, "timeout", 30000)
page.set_default_navigation_timeout(timeout)
page.set_default_timeout(timeout)
if getattr(self, "extra_headers", False):
page.set_extra_http_headers(getattr(self, "extra_headers"))
if getattr(self, "disable_resources", False):
page.route("**/*", intercept_route)
if getattr(self, "stealth", False):
for script in _compiled_stealth_scripts():
page.add_init_script(script=script)
return self.page_pool.add_page(page)
def get_pool_stats(self) -> Dict[str, int]:
"""Get statistics about the current page pool"""
return {
"total_pages": self.page_pool.pages_count,
"busy_pages": self.page_pool.busy_count,
"max_pages": self.max_pages,
}
class AsyncSession(SyncSession):
def __init__(self, max_pages: int = 1):
super().__init__(max_pages)
self.playwright: Optional[AsyncPlaywright] = None
self.context: Optional[AsyncBrowserContext] = None
self._lock = Lock()
async def _get_page(self) -> PageInfo: # pragma: no cover
"""Get a new page to use"""
async with self._lock:
# Close all finished pages to ensure clean state
await self.page_pool.aclose_all_finished_pages()
# If we're at max capacity after cleanup, wait for busy pages to finish
if self.page_pool.pages_count >= self.max_pages:
start_time = time()
while time() - start_time < self.__max_wait_for_page:
# Wait for any pages to finish, then clean them up
await asyncio_sleep(0.05)
await self.page_pool.aclose_all_finished_pages()
if self.page_pool.pages_count < self.max_pages:
break
else:
raise TimeoutError(
f"No pages finished to clear place in the pool within the {self.__max_wait_for_page}s timeout period"
)
page = await self.context.new_page()
timeout = getattr(self, "timeout", 30000)
page.set_default_navigation_timeout(timeout)
page.set_default_timeout(timeout)
if getattr(self, "extra_headers", False):
await page.set_extra_http_headers(getattr(self, "extra_headers"))
if getattr(self, "disable_resources", False):
await page.route("**/*", async_intercept_route)
if getattr(self, "stealth", False):
for script in _compiled_stealth_scripts():
await page.add_init_script(script=script)
return self.page_pool.add_page(page)
class DynamicSessionMixin:
def __validate__(
self,
__max_pages,
headless,
google_search,
hide_canvas,
disable_webgl,
real_chrome,
stealth,
wait,
page_action,
proxy,
locale,
extra_headers,
useragent,
cdp_url,
timeout,
disable_resources,
wait_selector,
init_script,
cookies,
network_idle,
wait_selector_state,
selector_config,
):
params = {
"max_pages": __max_pages,
"headless": headless,
"google_search": google_search,
"hide_canvas": hide_canvas,
"disable_webgl": disable_webgl,
"real_chrome": real_chrome,
"stealth": stealth,
"wait": wait,
"page_action": page_action,
"proxy": proxy,
"locale": locale,
"extra_headers": extra_headers,
"useragent": useragent,
"timeout": timeout,
"selector_config": selector_config,
"disable_resources": disable_resources,
"wait_selector": wait_selector,
"init_script": init_script,
"cookies": cookies,
"network_idle": network_idle,
"wait_selector_state": wait_selector_state,
"cdp_url": cdp_url,
}
config = validate(params, PlaywrightConfig)
self.max_pages = config.max_pages
self.headless = config.headless
self.hide_canvas = config.hide_canvas
self.disable_webgl = config.disable_webgl
self.real_chrome = config.real_chrome
self.stealth = config.stealth
self.google_search = config.google_search
self.wait = config.wait
self.proxy = config.proxy
self.locale = config.locale
self.extra_headers = config.extra_headers
self.useragent = config.useragent
self.timeout = config.timeout
self.cookies = config.cookies
self.disable_resources = config.disable_resources
self.cdp_url = config.cdp_url
self.network_idle = config.network_idle
self.wait_selector = config.wait_selector
self.init_script = config.init_script
self.wait_selector_state = config.wait_selector_state
self.selector_config = config.selector_config
self.page_action = config.page_action
self._headers_keys = (
set(map(str.lower, self.extra_headers.keys()))
if self.extra_headers
else set()
)
self.__initiate_browser_options__()
def __initiate_browser_options__(self):
if not self.cdp_url:
# `launch_options` is used with persistent context
self.launch_options = dict(
_launch_kwargs(
self.headless,
self.proxy,
self.locale,
tuple(self.extra_headers.items())
if self.extra_headers
else tuple(),
self.useragent,
self.real_chrome,
self.stealth,
self.hide_canvas,
self.disable_webgl,
)
)
self.launch_options["extra_http_headers"] = dict(
self.launch_options["extra_http_headers"]
)
self.launch_options["proxy"] = dict(self.launch_options["proxy"]) or None
self.context_options = dict()
else:
# while `context_options` is left to be used when cdp mode is enabled
self.launch_options = dict()
self.context_options = dict(
_context_kwargs(
self.proxy,
self.locale,
tuple(self.extra_headers.items())
if self.extra_headers
else tuple(),
self.useragent,
self.stealth,
)
)
self.context_options["extra_http_headers"] = dict(
self.context_options["extra_http_headers"]
)
self.context_options["proxy"] = dict(self.context_options["proxy"]) or None
class StealthySessionMixin:
def __validate__(
self,
max_pages,
headless,
block_images,
disable_resources,
block_webrtc,
allow_webgl,
network_idle,
humanize,
solve_cloudflare,
wait,
timeout,
page_action,
wait_selector,
init_script,
addons,
wait_selector_state,
cookies,
google_search,
extra_headers,
proxy,
os_randomize,
disable_ads,
geoip,
selector_config,
additional_args,
):
params = {
"max_pages": max_pages,
"headless": headless,
"block_images": block_images,
"disable_resources": disable_resources,
"block_webrtc": block_webrtc,
"allow_webgl": allow_webgl,
"network_idle": network_idle,
"humanize": humanize,
"solve_cloudflare": solve_cloudflare,
"wait": wait,
"timeout": timeout,
"page_action": page_action,
"wait_selector": wait_selector,
"init_script": init_script,
"addons": addons,
"wait_selector_state": wait_selector_state,
"cookies": cookies,
"google_search": google_search,
"extra_headers": extra_headers,
"proxy": proxy,
"os_randomize": os_randomize,
"disable_ads": disable_ads,
"geoip": geoip,
"selector_config": selector_config,
"additional_args": additional_args,
}
config = validate(params, CamoufoxConfig)
self.max_pages = config.max_pages
self.headless = config.headless
self.block_images = config.block_images
self.disable_resources = config.disable_resources
self.block_webrtc = config.block_webrtc
self.allow_webgl = config.allow_webgl
self.network_idle = config.network_idle
self.humanize = config.humanize
self.solve_cloudflare = config.solve_cloudflare
self.wait = config.wait
self.timeout = config.timeout
self.page_action = config.page_action
self.wait_selector = config.wait_selector
self.init_script = config.init_script
self.addons = config.addons
self.wait_selector_state = config.wait_selector_state
self.cookies = config.cookies
self.google_search = config.google_search
self.extra_headers = config.extra_headers
self.proxy = config.proxy
self.os_randomize = config.os_randomize
self.disable_ads = config.disable_ads
self.geoip = config.geoip
self.selector_config = config.selector_config
self.additional_args = config.additional_args
self.selector_config = config.selector_config
self.page_action = config.page_action
self._headers_keys = (
set(map(str.lower, self.extra_headers.keys()))
if self.extra_headers
else set()
)
self.__initiate_browser_options__()
def __initiate_browser_options__(self):
"""Initiate browser options."""
self.launch_options = generate_launch_options(
**{
"geoip": self.geoip,
"proxy": dict(self.proxy) if self.proxy else self.proxy,
"addons": self.addons,
"exclude_addons": [] if self.disable_ads else [DefaultAddons.UBO],
"headless": self.headless,
"humanize": True if self.solve_cloudflare else self.humanize,
"i_know_what_im_doing": True, # To turn warnings off with the user configurations
"allow_webgl": self.allow_webgl,
"block_webrtc": self.block_webrtc,
"block_images": self.block_images, # Careful! it makes some websites don't finish loading at all like stackoverflow even in headful mode.
"os": None if self.os_randomize else get_os_name(),
"user_data_dir": "",
"ff_version": __ff_version_str__,
"firefox_user_prefs": {
# This is what enabling `enable_cache` does internally, so we do it from here instead
"browser.sessionhistory.max_entries": 10,
"browser.sessionhistory.max_total_viewers": -1,
"browser.cache.memory.enable": True,
"browser.cache.disk_cache_ssl": True,
"browser.cache.disk.smart_size.enabled": True,
},
**self.additional_args,
}
)
@staticmethod
def _detect_cloudflare(page_content: str) -> str | None:
"""
Detect the type of Cloudflare challenge present in the provided page content.
This function analyzes the given page content to identify whether a specific
type of Cloudflare challenge is present. It checks for three predefined
challenge types: non-interactive, managed, and interactive. If a challenge
type is detected, it returns the corresponding type as a string. If no
challenge type is detected, it returns None.
Args:
page_content (str): The content of the page to analyze for Cloudflare
challenge types.
Returns:
str: A string representing the detected Cloudflare challenge type, if
found. Returns None if no challenge matches.
"""
challenge_types = (
"non-interactive",
"managed",
"interactive",
)
for ctype in challenge_types:
if f"cType: '{ctype}'" in page_content:
return ctype
return None
scrapling/engines/_browsers/_camoufox.py
+39 -230
View File
@@ -1,17 +1,8 @@
from time import time, sleep
from re import compile as re_compile
from asyncio import sleep as asyncio_sleep, Lock
from camoufox import DefaultAddons
from camoufox.utils import (
launch_options as generate_launch_options,
installed_verstr as camoufox_version,
)
from playwright.sync_api import (
Response as SyncPlaywrightResponse,
sync_playwright,
BrowserContext,
Playwright,
Locator,
Page,
)
@@ -24,9 +15,8 @@ from playwright.async_api import (
Page as async_Page,
)
from ._base import SyncSession, AsyncSession, StealthySessionMixin
from scrapling.core.utils import log
from ._page import PageInfo, PagePool
from ._validators import validate, CamoufoxConfig
from scrapling.core._types import (
Dict,
List,
@@ -37,17 +27,13 @@ from scrapling.core._types import (
from scrapling.engines.toolbelt import (
Response,
ResponseFactory,
async_intercept_route,
generate_convincing_referer,
get_os_name,
intercept_route,
)
__CF_PATTERN__ = re_compile("challenges.cloudflare.com/cdn-cgi/challenge-platform/.*")
__ff_version_str__ = camoufox_version().split(".", 1)[0]
class StealthySession:
class StealthySession(StealthySessionMixin, SyncSession):
"""A Stealthy session manager with page pooling."""
__slots__ = (
@@ -87,7 +73,7 @@ class StealthySession:
def __init__(
self,
max_pages: int = 1,
__max_pages: int = 1,
headless: bool = True, # noqa: F821
block_images: bool = False,
disable_resources: bool = False,
@@ -141,107 +127,38 @@ class StealthySession:
:param google_search: Enabled by default, Scrapling will set the referer header to be as if this request came from a Google search of this website's domain name.
:param extra_headers: A dictionary of extra headers to add to the request. _The referer set by the `google_search` argument takes priority over the referer set here if used together._
:param proxy: The proxy to be used with requests, it can be a string or a dictionary with the keys 'server', 'username', and 'password' only.
:param max_pages: The maximum number of tabs to be opened at the same time. It will be used in rotation through a PagePool.
:param selector_config: The arguments that will be passed in the end while creating the final Selector's class.
:param additional_args: Additional arguments to be passed to Camoufox as additional settings, and it takes higher priority than Scrapling's settings.
"""
params = {
"max_pages": max_pages,
"headless": headless,
"block_images": block_images,
"disable_resources": disable_resources,
"block_webrtc": block_webrtc,
"allow_webgl": allow_webgl,
"network_idle": network_idle,
"humanize": humanize,
"solve_cloudflare": solve_cloudflare,
"wait": wait,
"timeout": timeout,
"page_action": page_action,
"wait_selector": wait_selector,
"init_script": init_script,
"addons": addons,
"wait_selector_state": wait_selector_state,
"cookies": cookies,
"google_search": google_search,
"extra_headers": extra_headers,
"proxy": proxy,
"os_randomize": os_randomize,
"disable_ads": disable_ads,
"geoip": geoip,
"selector_config": selector_config,
"additional_args": additional_args,
}
config = validate(params, CamoufoxConfig)
self.max_pages = config.max_pages
self.headless = config.headless
self.block_images = config.block_images
self.disable_resources = config.disable_resources
self.block_webrtc = config.block_webrtc
self.allow_webgl = config.allow_webgl
self.network_idle = config.network_idle
self.humanize = config.humanize
self.solve_cloudflare = config.solve_cloudflare
self.wait = config.wait
self.timeout = config.timeout
self.page_action = config.page_action
self.wait_selector = config.wait_selector
self.init_script = config.init_script
self.addons = config.addons
self.wait_selector_state = config.wait_selector_state
self.cookies = config.cookies
self.google_search = config.google_search
self.extra_headers = config.extra_headers
self.proxy = config.proxy
self.os_randomize = config.os_randomize
self.disable_ads = config.disable_ads
self.geoip = config.geoip
self.selector_config = config.selector_config
self.additional_args = config.additional_args
self.playwright: Optional[Playwright] = None
self.context: Optional[BrowserContext] = None
self.page_pool = PagePool(self.max_pages)
self._closed = False
self.selector_config = config.selector_config
self.page_action = config.page_action
self._headers_keys = (
set(map(str.lower, self.extra_headers.keys()))
if self.extra_headers
else set()
)
self.__initiate_browser_options__()
def __initiate_browser_options__(self):
"""Initiate browser options."""
self.launch_options = generate_launch_options(
**{
"geoip": self.geoip,
"proxy": dict(self.proxy) if self.proxy else self.proxy,
"addons": self.addons,
"exclude_addons": [] if self.disable_ads else [DefaultAddons.UBO],
"headless": self.headless,
"humanize": True if self.solve_cloudflare else self.humanize,
"i_know_what_im_doing": True, # To turn warnings off with the user configurations
"allow_webgl": self.allow_webgl,
"block_webrtc": self.block_webrtc,
"block_images": self.block_images, # Careful! it makes some websites don't finish loading at all like stackoverflow even in headful mode.
"os": None if self.os_randomize else get_os_name(),
"user_data_dir": "",
"ff_version": __ff_version_str__,
"firefox_user_prefs": {
# This is what enabling `enable_cache` does internally, so we do it from here instead
"browser.sessionhistory.max_entries": 10,
"browser.sessionhistory.max_total_viewers": -1,
"browser.cache.memory.enable": True,
"browser.cache.disk_cache_ssl": True,
"browser.cache.disk.smart_size.enabled": True,
},
**self.additional_args,
}
self.__validate__(
__max_pages,
headless,
block_images,
disable_resources,
block_webrtc,
allow_webgl,
network_idle,
humanize,
solve_cloudflare,
wait,
timeout,
page_action,
wait_selector,
init_script,
addons,
wait_selector_state,
cookies,
google_search,
extra_headers,
proxy,
os_randomize,
disable_ads,
geoip,
selector_config,
additional_args,
)
super().__init__(max_pages=self.max_pages)
def __create__(self):
"""Create a browser for this instance and context."""
@@ -284,68 +201,6 @@ class StealthySession:
self._closed = True
def _get_or_create_page(self) -> PageInfo: # pragma: no cover
"""Get an available page or create a new one"""
# Try to get a ready page first
page_info = self.page_pool.get_ready_page()
if page_info:
return page_info
# Create a new page if under limit
if self.page_pool.pages_count < self.max_pages:
page = self.context.new_page()
page.set_default_navigation_timeout(self.timeout)
page.set_default_timeout(self.timeout)
if self.extra_headers:
page.set_extra_http_headers(self.extra_headers)
if self.disable_resources:
page.route("**/*", intercept_route)
return self.page_pool.add_page(page)
# Wait for a page to become available
max_wait = 30
start_time = time()
while time() - start_time < max_wait:
page_info = self.page_pool.get_ready_page()
if page_info:
return page_info
sleep(0.05)
raise TimeoutError("No pages available within timeout period")
@staticmethod
def _detect_cloudflare(page_content):
"""
Detect the type of Cloudflare challenge present in the provided page content.
This function analyzes the given page content to identify whether a specific
type of Cloudflare challenge is present. It checks for three predefined
challenge types: non-interactive, managed, and interactive. If a challenge
type is detected, it returns the corresponding type as a string. If no
challenge type is detected, it returns None.
Args:
page_content (str): The content of the page to analyze for Cloudflare
challenge types.
Returns:
str: A string representing the detected Cloudflare challenge type, if
found. Returns None if no challenge matches.
"""
challenge_types = (
"non-interactive",
"managed",
"interactive",
)
for ctype in challenge_types:
if f"cType: '{ctype}'" in page_content:
return ctype
return None
def _solve_cloudflare(self, page: Page) -> None: # pragma: no cover
"""Solve the cloudflare challenge displayed on the playwright page passed
@@ -416,7 +271,7 @@ class StealthySession:
):
final_response = finished_response
page_info = self._get_or_create_page()
page_info = self._get_page()
page_info.mark_busy(url=url)
try: # pragma: no cover
@@ -462,8 +317,8 @@ class StealthySession:
page_info.page, first_response, final_response, self.selector_config
)
# Mark the page as ready for next use
page_info.mark_ready()
# Mark the page as finished for next use
page_info.mark_finished()
return response
@@ -471,17 +326,8 @@ class StealthySession:
page_info.mark_error()
raise e
def get_pool_stats(self) -> Dict[str, int]:
"""Get statistics about the current page pool"""
return {
"total_pages": self.page_pool.pages_count,
"ready_pages": self.page_pool.ready_count,
"busy_pages": self.page_pool.busy_count,
"max_pages": self.max_pages,
}
class AsyncStealthySession(StealthySession):
class AsyncStealthySession(StealthySessionMixin, AsyncSession):
"""A Stealthy session manager with page pooling."""
def __init__(
@@ -544,7 +390,7 @@ class AsyncStealthySession(StealthySession):
:param selector_config: The arguments that will be passed in the end while creating the final Selector's class.
:param additional_args: Additional arguments to be passed to Camoufox as additional settings, and it takes higher priority than Scrapling's settings.
"""
super().__init__(
self.__validate__(
max_pages,
headless,
block_images,
@@ -571,11 +417,7 @@ class AsyncStealthySession(StealthySession):
selector_config,
additional_args,
)
self.playwright: Optional[AsyncPlaywright] = None
self.context: Optional[AsyncBrowserContext] = None
self._lock = Lock()
self.__enter__ = None
self.__exit__ = None
super().__init__(max_pages=self.max_pages)
async def __create__(self):
"""Create a browser for this instance and context."""
@@ -618,39 +460,6 @@ class AsyncStealthySession(StealthySession):
self._closed = True
async def _get_or_create_page(self) -> PageInfo:
"""Get an available page or create a new one"""
async with self._lock:
# Try to get a ready page first
page_info = self.page_pool.get_ready_page()
if page_info:
return page_info
# Create a new page if under limit
if self.page_pool.pages_count < self.max_pages:
page = await self.context.new_page()
page.set_default_navigation_timeout(self.timeout)
page.set_default_timeout(self.timeout)
if self.extra_headers:
await page.set_extra_http_headers(self.extra_headers)
if self.disable_resources:
await page.route("**/*", async_intercept_route)
return self.page_pool.add_page(page)
# Wait for a page to become available
max_wait = 30
start_time = time()
while time() - start_time < max_wait: # pragma: no cover
page_info = self.page_pool.get_ready_page()
if page_info:
return page_info
await asyncio_sleep(0.05)
raise TimeoutError("No pages available within timeout period")
async def _solve_cloudflare(self, page: async_Page):
"""Solve the cloudflare challenge displayed on the playwright page passed. The async version
@@ -723,7 +532,7 @@ class AsyncStealthySession(StealthySession):
):
final_response = finished_response
page_info = await self._get_or_create_page()
page_info = await self._get_page()
page_info.mark_busy(url=url)
try:
@@ -771,8 +580,8 @@ class AsyncStealthySession(StealthySession):
page_info.page, first_response, final_response, self.selector_config
)
# Mark the page as ready for next use
page_info.mark_ready()
# Mark the page as finished for next use
page_info.mark_finished()
return response
scrapling/engines/_browsers/_controllers.py
+40 -209
View File
@@ -1,10 +1,6 @@
from time import time, sleep
from asyncio import sleep as asyncio_sleep, Lock
from playwright.sync_api import (
Response as SyncPlaywrightResponse,
sync_playwright,
BrowserContext,
Playwright,
Locator,
)
@@ -21,9 +17,7 @@ from rebrowser_playwright.async_api import (
)
from scrapling.core.utils import log
from ._page import PageInfo, PagePool
from ._validators import validate, PlaywrightConfig
from ._config_tools import _compiled_stealth_scripts, _launch_kwargs, _context_kwargs
from ._base import SyncSession, AsyncSession, DynamicSessionMixin
from scrapling.core._types import (
Dict,
List,
@@ -35,12 +29,10 @@ from scrapling.engines.toolbelt import (
Response,
ResponseFactory,
generate_convincing_referer,
intercept_route,
async_intercept_route,
)
class DynamicSession:
class DynamicSession(DynamicSessionMixin, SyncSession):
"""A Browser session manager with page pooling."""
__slots__ = (
@@ -127,108 +119,31 @@ class DynamicSession:
:param proxy: The proxy to be used with requests, it can be a string or a dictionary with the keys 'server', 'username', and 'password' only.
:param selector_config: The arguments that will be passed in the end while creating the final Selector's class.
"""
params = {
"max_pages": __max_pages,
"headless": headless,
"google_search": google_search,
"hide_canvas": hide_canvas,
"disable_webgl": disable_webgl,
"real_chrome": real_chrome,
"stealth": stealth,
"wait": wait,
"page_action": page_action,
"proxy": proxy,
"locale": locale,
"extra_headers": extra_headers,
"useragent": useragent,
"timeout": timeout,
"selector_config": selector_config,
"disable_resources": disable_resources,
"wait_selector": wait_selector,
"init_script": init_script,
"cookies": cookies,
"network_idle": network_idle,
"wait_selector_state": wait_selector_state,
"cdp_url": cdp_url,
}
config = validate(params, PlaywrightConfig)
self.max_pages = config.max_pages
self.headless = config.headless
self.hide_canvas = config.hide_canvas
self.disable_webgl = config.disable_webgl
self.real_chrome = config.real_chrome
self.stealth = config.stealth
self.google_search = config.google_search
self.wait = config.wait
self.proxy = config.proxy
self.locale = config.locale
self.extra_headers = config.extra_headers
self.useragent = config.useragent
self.timeout = config.timeout
self.cookies = config.cookies
self.disable_resources = config.disable_resources
self.cdp_url = config.cdp_url
self.network_idle = config.network_idle
self.wait_selector = config.wait_selector
self.init_script = config.init_script
self.wait_selector_state = config.wait_selector_state
self.playwright: Optional[Playwright] = None
self.context: Optional[BrowserContext] = None
self.page_pool = PagePool(self.max_pages)
self._closed = False
self.selector_config = config.selector_config
self.page_action = config.page_action
self._headers_keys = (
set(map(str.lower, self.extra_headers.keys()))
if self.extra_headers
else set()
self.__validate__(
__max_pages,
headless,
google_search,
hide_canvas,
disable_webgl,
real_chrome,
stealth,
wait,
page_action,
proxy,
locale,
extra_headers,
useragent,
cdp_url,
timeout,
disable_resources,
wait_selector,
init_script,
cookies,
network_idle,
wait_selector_state,
selector_config,
)
self.__initiate_browser_options__()
def __initiate_browser_options__(self):
if not self.cdp_url:
# `launch_options` is used with persistent context
self.launch_options = dict(
_launch_kwargs(
self.headless,
self.proxy,
self.locale,
tuple(self.extra_headers.items())
if self.extra_headers
else tuple(),
self.useragent,
self.real_chrome,
self.stealth,
self.hide_canvas,
self.disable_webgl,
)
)
self.launch_options["extra_http_headers"] = dict(
self.launch_options["extra_http_headers"]
)
self.launch_options["proxy"] = dict(self.launch_options["proxy"]) or None
self.context_options = dict()
else:
# while `context_options` is left to be used when cdp mode is enabled
self.launch_options = dict()
self.context_options = dict(
_context_kwargs(
self.proxy,
self.locale,
tuple(self.extra_headers.items())
if self.extra_headers
else tuple(),
self.useragent,
self.stealth,
)
)
self.context_options["extra_http_headers"] = dict(
self.context_options["extra_http_headers"]
)
self.context_options["proxy"] = dict(self.context_options["proxy"]) or None
super().__init__(max_pages=self.max_pages)
def __create__(self):
"""Create a browser for this instance and context."""
@@ -237,7 +152,7 @@ class DynamicSession:
# Because rebrowser_playwright doesn't play well with real browsers
sync_context = sync_playwright
self.playwright = sync_context().start()
self.playwright: Playwright = sync_context().start()
if self.cdp_url: # pragma: no cover
self.context = self.playwright.chromium.connect_over_cdp(
@@ -280,43 +195,10 @@ class DynamicSession:
self._closed = True
def _get_or_create_page(self) -> PageInfo: # pragma: no cover
"""Get an available page or create a new one"""
# Try to get a ready page first
page_info = self.page_pool.get_ready_page()
if page_info:
return page_info
# Create a new page if under limit
if self.page_pool.pages_count < self.max_pages:
page = self.context.new_page()
page.set_default_navigation_timeout(self.timeout)
page.set_default_timeout(self.timeout)
if self.extra_headers:
page.set_extra_http_headers(self.extra_headers)
if self.disable_resources:
page.route("**/*", intercept_route)
if self.stealth:
for script in _compiled_stealth_scripts():
page.add_init_script(script=script)
return self.page_pool.add_page(page)
# Wait for a page to become available
max_wait = 30
start_time = time()
while time() - start_time < max_wait:
page_info = self.page_pool.get_ready_page()
if page_info:
return page_info
sleep(0.05)
raise TimeoutError("No pages available within timeout period")
def fetch(self, url: str) -> Response:
def fetch(
self,
url: str,
) -> Response:
"""Opens up the browser and do your request based on your chosen options.
:param url: The Target url.
@@ -340,7 +222,7 @@ class DynamicSession:
):
final_response = finished_response
page_info = self._get_or_create_page()
page_info = self._get_page()
page_info.mark_busy(url=url)
try: # pragma: no cover
@@ -380,8 +262,8 @@ class DynamicSession:
page_info.page, first_response, final_response, self.selector_config
)
# Mark the page as ready for next use
page_info.mark_ready()
# Mark the page as finished for next use
page_info.mark_finished()
return response
@@ -389,17 +271,8 @@ class DynamicSession:
page_info.mark_error()
raise e
def get_pool_stats(self) -> Dict[str, int]:
"""Get statistics about the current page pool"""
return {
"total_pages": self.page_pool.pages_count,
"ready_pages": self.page_pool.ready_count,
"busy_pages": self.page_pool.busy_count,
"max_pages": self.max_pages,
}
class AsyncDynamicSession(DynamicSession):
class AsyncDynamicSession(DynamicSessionMixin, AsyncSession):
"""An async Browser session manager with page pooling, it's using a persistent browser Context by default with a temporary user profile directory."""
def __init__(
@@ -455,7 +328,7 @@ class AsyncDynamicSession(DynamicSession):
:param selector_config: The arguments that will be passed in the end while creating the final Selector's class.
"""
super().__init__(
self.__validate__(
max_pages,
headless,
google_search,
@@ -479,12 +352,7 @@ class AsyncDynamicSession(DynamicSession):
wait_selector_state,
selector_config,
)
self.playwright: Optional[AsyncPlaywright] = None
self.context: Optional[AsyncBrowserContext] = None
self._lock = Lock()
self.__enter__ = None
self.__exit__ = None
super().__init__(max_pages=self.max_pages)
async def __create__(self):
"""Create a browser for this instance and context."""
@@ -541,43 +409,6 @@ class AsyncDynamicSession(DynamicSession):
self._closed = True
async def _get_or_create_page(self) -> PageInfo:
"""Get an available page or create a new one"""
async with self._lock:
# Try to get a ready page first
page_info = self.page_pool.get_ready_page()
if page_info:
return page_info
# Create a new page if under limit
if self.page_pool.pages_count < self.max_pages:
page = await self.context.new_page()
page.set_default_navigation_timeout(self.timeout)
page.set_default_timeout(self.timeout)
if self.extra_headers:
await page.set_extra_http_headers(self.extra_headers)
if self.disable_resources:
await page.route("**/*", async_intercept_route)
if self.stealth:
for script in _compiled_stealth_scripts():
await page.add_init_script(script=script)
return self.page_pool.add_page(page)
# Wait for a page to become available
max_wait = 30 # seconds
start_time = time()
while time() - start_time < max_wait: # pragma: no cover
page_info = self.page_pool.get_ready_page()
if page_info:
return page_info
await asyncio_sleep(0.05)
raise TimeoutError("No pages available within timeout period")
async def fetch(self, url: str) -> Response:
"""Opens up the browser and do your request based on your chosen options.
@@ -602,7 +433,7 @@ class AsyncDynamicSession(DynamicSession):
):
final_response = finished_response
page_info = await self._get_or_create_page()
page_info = await self._get_page()
page_info.mark_busy(url=url)
try:
@@ -642,8 +473,8 @@ class AsyncDynamicSession(DynamicSession):
page_info.page, first_response, final_response, self.selector_config
)
# Mark the page as ready for next use
page_info.mark_ready()
# Mark the page as finished for next use
page_info.mark_finished()
return response
scrapling/engines/_browsers/_page.py
+39 -15
View File
@@ -6,7 +6,9 @@ from playwright.async_api import Page as AsyncPage
from scrapling.core._types import Optional, List, Literal
PageState = Literal["ready", "busy", "error"] # States that a page can be in
PageState = Literal[
"finished", "ready", "busy", "error"
] # States that a page can be in
@dataclass
@@ -23,9 +25,9 @@ class PageInfo:
self.state = "busy"
self.url = url
def mark_ready(self):
"""Mark the page as ready for new requests"""
self.state = "ready"
def mark_finished(self):
"""Mark the page as finished for new requests"""
self.state = "finished"
self.url = ""
def mark_error(self):
@@ -62,24 +64,16 @@ class PagePool:
self.pages.append(page_info)
return page_info
def get_ready_page(self) -> Optional[PageInfo]:
"""Get a page that's ready for use"""
with self._lock:
for page_info in self.pages:
if page_info.state == "ready":
return page_info
return None
@property
def pages_count(self) -> int:
"""Get the total number of pages"""
return len(self.pages)
@property
def ready_count(self) -> int:
"""Get the number of ready pages"""
def finished_count(self) -> int:
"""Get the number of finished pages"""
with self._lock:
return sum(1 for p in self.pages if p.state == "ready")
return sum(1 for p in self.pages if p.state == "finished")
@property
def busy_count(self) -> int:
@@ -91,3 +85,33 @@ class PagePool:
"""Remove pages in error state"""
with self._lock:
self.pages = [p for p in self.pages if p.state != "error"]
def close_all_finished_pages(self):
"""Close all pages in finished state and remove them from the pool"""
with self._lock:
pages_to_remove = []
for page_info in self.pages:
if page_info.state == "finished":
try:
page_info.page.close()
except Exception:
pass
pages_to_remove.append(page_info)
for page_info in pages_to_remove:
self.pages.remove(page_info)
async def aclose_all_finished_pages(self):
"""Async version: Close all pages in finished state and remove them from the pool"""
with self._lock:
pages_to_remove = []
for page_info in self.pages:
if page_info.state == "finished":
try:
await page_info.page.close()
except Exception:
pass
pages_to_remove.append(page_info)
for page_info in pages_to_remove:
self.pages.remove(page_info)
scrapling/fetchers.py
-1
View File
@@ -118,7 +118,6 @@ class StealthyFetcher(BaseFetcher):
with StealthySession(
wait=wait,
max_pages=1,
proxy=proxy,
geoip=geoip,
addons=addons,