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

feat(PlaywrightFetcher): Add async support for PlaywrightFetcher

Browse Source
This commit is contained in:
Karim shoair
2024-12-16 00:15:47 +02:00
parent 889c111855
commit af4f2c0f74
7 changed files with 170 additions and 35 deletions
Download Patch File Download Diff File Expand all files Collapse all files
pytest.ini
+2
View File
@@ -1,2 +1,4 @@
[pytest]
asyncio_mode = auto
asyncio_default_fixture_loop_scope = function
addopts = -p no:warnings --doctest-modules --ignore=setup.py --verbose
scrapling/engines/camo.py
+5 -4
View File
@@ -6,7 +6,7 @@ from scrapling.core._types import (Callable, Dict, List, Literal, Optional,
from scrapling.core.utils import log
from scrapling.engines.toolbelt import (Response, StatusText,
check_type_validity,
construct_proxy_dict, do_nothing,
construct_proxy_dict,
generate_convincing_referer,
get_os_name, intercept_route)
@@ -15,7 +15,7 @@ class CamoufoxEngine:
def __init__(
self, headless: Optional[Union[bool, Literal['virtual']]] = True, block_images: Optional[bool] = False, disable_resources: Optional[bool] = False,
block_webrtc: Optional[bool] = False, allow_webgl: Optional[bool] = True, network_idle: Optional[bool] = False, humanize: Optional[Union[bool, float]] = True,
timeout: Optional[float] = 30000, page_action: Callable = do_nothing, wait_selector: Optional[str] = None, addons: Optional[List[str]] = None,
timeout: Optional[float] = 30000, page_action: Callable = None, wait_selector: Optional[str] = None, addons: Optional[List[str]] = None,
wait_selector_state: str = 'attached', google_search: Optional[bool] = True, extra_headers: Optional[Dict[str, str]] = None,
proxy: Optional[Union[str, Dict[str, str]]] = None, os_randomize: Optional[bool] = None, disable_ads: Optional[bool] = True,
geoip: Optional[bool] = False,
@@ -65,7 +65,7 @@ class CamoufoxEngine:
if callable(page_action):
self.page_action = page_action
else:
self.page_action = do_nothing
self.page_action = None
log.error('[Ignored] Argument "page_action" must be callable')
self.wait_selector = wait_selector
@@ -106,7 +106,8 @@ class CamoufoxEngine:
if self.network_idle:
page.wait_for_load_state('networkidle')
page = self.page_action(page)
if self.page_action is not None:
page = self.page_action(page)
if self.wait_selector and type(self.wait_selector) is str:
waiter = page.locator(self.wait_selector)
scrapling/engines/pw.py
+81 -10
View File
@@ -5,9 +5,9 @@ from scrapling.core.utils import log, lru_cache
from scrapling.engines.constants import (DEFAULT_STEALTH_FLAGS,
NSTBROWSER_DEFAULT_QUERY)
from scrapling.engines.toolbelt import (Response, StatusText,
async_intercept_route,
check_type_validity, construct_cdp_url,
construct_proxy_dict, do_nothing,
do_nothing_async,
construct_proxy_dict,
generate_convincing_referer,
generate_headers, intercept_route,
js_bypass_path)
@@ -20,7 +20,7 @@ class PlaywrightEngine:
useragent: Optional[str] = None,
network_idle: Optional[bool] = False,
timeout: Optional[float] = 30000,
page_action: Callable = do_nothing,
page_action: Callable = None,
wait_selector: Optional[str] = None,
locale: Optional[str] = 'en-US',
wait_selector_state: Optional[str] = 'attached',
@@ -75,10 +75,10 @@ class PlaywrightEngine:
self.cdp_url = cdp_url
self.useragent = useragent
self.timeout = check_type_validity(timeout, [int, float], 30000)
if callable(page_action):
if page_action is not None and callable(page_action):
self.page_action = page_action
else:
self.page_action = do_nothing
self.page_action = None
log.error('[Ignored] Argument "page_action" must be callable')
self.wait_selector = wait_selector
@@ -225,7 +225,8 @@ class PlaywrightEngine:
if self.network_idle:
page.wait_for_load_state('networkidle')
page = self.page_action(page)
if self.page_action is not None:
page = self.page_action(page)
if self.wait_selector and type(self.wait_selector) is str:
waiter = page.locator(self.wait_selector)
@@ -238,11 +239,8 @@ class PlaywrightEngine:
# This will be parsed inside `Response`
encoding = res.headers.get('content-type', '') or 'utf-8' # default encoding
status_text = res.status_text
# PlayWright API sometimes give empty status text for some reason!
if not status_text:
status_text = StatusText.get(res.status)
status_text = res.status_text or StatusText.get(res.status)
response = Response(
url=res.url,
@@ -258,3 +256,76 @@ class PlaywrightEngine:
)
page.close()
return response
async def async_fetch(self, url: str) -> Response:
"""Async version of `fetch`
:param url: Target url.
:return: A `Response` object that is the same as `Adaptor` object except it has these added attributes: `status`, `reason`, `cookies`, `headers`, and `request_headers`
"""
if not self.stealth or self.real_chrome:
# Because rebrowser_playwright doesn't play well with real browsers
from playwright.async_api import async_playwright
else:
from rebrowser_playwright.async_api import async_playwright
async with async_playwright() as p:
# Creating the browser
if self.cdp_url:
cdp_url = self._cdp_url_logic()
browser = await p.chromium.connect_over_cdp(endpoint_url=cdp_url)
else:
browser = await p.chromium.launch(**self.__launch_kwargs())
context = await browser.new_context(**self.__context_kwargs())
# Finally we are in business
page = await 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 self.__stealth_scripts():
await page.add_init_script(path=script)
res = await page.goto(url, referer=generate_convincing_referer(url) if self.google_search else None)
await page.wait_for_load_state(state="domcontentloaded")
if self.network_idle:
await page.wait_for_load_state('networkidle')
if self.page_action is not None:
page = await self.page_action(page)
if self.wait_selector and type(self.wait_selector) is str:
waiter = page.locator(self.wait_selector)
await waiter.first.wait_for(state=self.wait_selector_state)
# Wait again after waiting for the selector, helpful with protections like Cloudflare
await page.wait_for_load_state(state="load")
await page.wait_for_load_state(state="domcontentloaded")
if self.network_idle:
await page.wait_for_load_state('networkidle')
# This will be parsed inside `Response`
encoding = res.headers.get('content-type', '') or 'utf-8' # default encoding
# PlayWright API sometimes give empty status text for some reason!
status_text = res.status_text or StatusText.get(res.status)
response = Response(
url=res.url,
text=await page.content(),
body=(await page.content()).encode('utf-8'),
status=res.status,
reason=status_text,
encoding=encoding,
cookies={cookie['name']: cookie['value'] for cookie in await page.context.cookies()},
headers=await res.all_headers(),
request_headers=await res.request.all_headers(),
**self.adaptor_arguments
)
await page.close()
return response
scrapling/engines/toolbelt/__init__.py
+3 -4
View File
@@ -1,7 +1,6 @@
from .custom import (BaseFetcher, Response, StatusText, check_if_engine_usable,
check_type_validity, do_nothing, do_nothing_async,
get_variable_name)
check_type_validity, get_variable_name)
from .fingerprints import (generate_convincing_referer, generate_headers,
get_os_name)
from .navigation import (construct_cdp_url, construct_proxy_dict,
intercept_route, js_bypass_path)
from .navigation import (async_intercept_route, construct_cdp_url,
construct_proxy_dict, intercept_route, js_bypass_path)
scrapling/engines/toolbelt/custom.py
-11
View File
@@ -296,14 +296,3 @@ def check_type_validity(variable: Any, valid_types: Union[List[Type], None], def
return default_value
return variable
# Pew Pew
def do_nothing(page):
# Just works as a filler for `page_action` argument in browser engines
return page
async def do_nothing_async(page):
# Just works as a filler for `page_action` argument in browser engines
return page
scrapling/engines/toolbelt/navigation.py
+16 -3
View File
@@ -4,6 +4,7 @@ Functions related to files and URLs
import os
from urllib.parse import urlencode, urlparse
from playwright.async_api import Route as async_Route
from playwright.sync_api import Route
from scrapling.core._types import Dict, Optional, Union
@@ -11,7 +12,7 @@ from scrapling.core.utils import log, lru_cache
from scrapling.engines.constants import DEFAULT_DISABLED_RESOURCES
def intercept_route(route: Route) -> Union[Route, None]:
def intercept_route(route: Route):
"""This is just a route handler but it drops requests that its type falls in `DEFAULT_DISABLED_RESOURCES`
:param route: PlayWright `Route` object of the current page
@@ -19,8 +20,20 @@ def intercept_route(route: Route) -> Union[Route, None]:
"""
if route.request.resource_type in DEFAULT_DISABLED_RESOURCES:
log.debug(f'Blocking background resource "{route.request.url}" of type "{route.request.resource_type}"')
return route.abort()
return route.continue_()
route.abort()
route.continue_()
async def async_intercept_route(route: async_Route):
"""This is just a route handler but it drops requests that its type falls in `DEFAULT_DISABLED_RESOURCES`
:param route: PlayWright `Route` object of the current page
:return: PlayWright `Route` object
"""
if route.request.resource_type in DEFAULT_DISABLED_RESOURCES:
log.debug(f'Blocking background resource "{route.request.url}" of type "{route.request.resource_type}"')
await route.abort()
await route.continue_()
def construct_proxy_dict(proxy_string: Union[str, Dict[str, str]]) -> Union[Dict, None]:
scrapling/fetchers.py
+63 -3
View File
@@ -2,7 +2,7 @@ from scrapling.core._types import (Callable, Dict, List, Literal, Optional,
Union)
from scrapling.engines import (CamoufoxEngine, PlaywrightEngine, StaticEngine,
check_if_engine_usable)
from scrapling.engines.toolbelt import BaseFetcher, Response, do_nothing
from scrapling.engines.toolbelt import BaseFetcher, Response
class Fetcher(BaseFetcher):
@@ -175,7 +175,7 @@ class StealthyFetcher(BaseFetcher):
def fetch(
self, url: str, headless: Optional[Union[bool, Literal['virtual']]] = True, block_images: Optional[bool] = False, disable_resources: Optional[bool] = False,
block_webrtc: Optional[bool] = False, allow_webgl: Optional[bool] = True, network_idle: Optional[bool] = False, addons: Optional[List[str]] = None,
timeout: Optional[float] = 30000, page_action: Callable = do_nothing, wait_selector: Optional[str] = None, humanize: Optional[Union[bool, float]] = True,
timeout: Optional[float] = 30000, page_action: Callable = None, wait_selector: Optional[str] = None, humanize: Optional[Union[bool, float]] = True,
wait_selector_state: str = 'attached', google_search: Optional[bool] = True, extra_headers: Optional[Dict[str, str]] = None, proxy: Optional[Union[str, Dict[str, str]]] = None,
os_randomize: Optional[bool] = None, disable_ads: Optional[bool] = True, geoip: Optional[bool] = False,
) -> Response:
@@ -250,7 +250,7 @@ class PlayWrightFetcher(BaseFetcher):
def fetch(
self, url: str, headless: Union[bool, str] = True, disable_resources: bool = None,
useragent: Optional[str] = None, network_idle: Optional[bool] = False, timeout: Optional[float] = 30000,
page_action: Optional[Callable] = do_nothing, wait_selector: Optional[str] = None, wait_selector_state: Optional[str] = 'attached',
page_action: Optional[Callable] = None, wait_selector: Optional[str] = None, wait_selector_state: Optional[str] = 'attached',
hide_canvas: Optional[bool] = False, disable_webgl: Optional[bool] = False, extra_headers: Optional[Dict[str, str]] = None, google_search: Optional[bool] = True,
proxy: Optional[Union[str, Dict[str, str]]] = None, locale: Optional[str] = 'en-US',
stealth: Optional[bool] = False, real_chrome: Optional[bool] = False,
@@ -307,6 +307,66 @@ class PlayWrightFetcher(BaseFetcher):
)
return engine.fetch(url)
async def async_fetch(
self, url: str, headless: Union[bool, str] = True, disable_resources: bool = None,
useragent: Optional[str] = None, network_idle: Optional[bool] = False, timeout: Optional[float] = 30000,
page_action: Optional[Callable] = None, wait_selector: Optional[str] = None, wait_selector_state: Optional[str] = 'attached',
hide_canvas: Optional[bool] = False, disable_webgl: Optional[bool] = False, extra_headers: Optional[Dict[str, str]] = None, google_search: Optional[bool] = True,
proxy: Optional[Union[str, Dict[str, str]]] = None, locale: Optional[str] = 'en-US',
stealth: Optional[bool] = False, real_chrome: Optional[bool] = False,
cdp_url: Optional[str] = None,
nstbrowser_mode: Optional[bool] = False, nstbrowser_config: Optional[Dict] = None,
) -> Response:
"""Opens up a browser and do your request based on your chosen options below.
:param url: Target url.
:param headless: Run the browser in headless/hidden (default), or headful/visible mode.
:param disable_resources: Drop requests of unnecessary resources for speed boost. It depends but it made requests ~25% faster in my tests for some websites.
Requests dropped are of type `font`, `image`, `media`, `beacon`, `object`, `imageset`, `texttrack`, `websocket`, `csp_report`, and `stylesheet`.
This can help save your proxy usage but be careful with this option as it makes some websites never finish loading.
:param useragent: Pass a useragent string to be used. Otherwise the fetcher will generate a real Useragent of the same browser and use it.
:param network_idle: Wait for the page until there are no network connections for at least 500 ms.
:param timeout: The timeout in milliseconds that is used in all operations and waits through the page. The default is 30000
:param locale: Set the locale for the browser if wanted. The default value is `en-US`.
:param page_action: Added for automation. A function that takes the `page` object, does the automation you need, then returns `page` again.
:param wait_selector: Wait for a specific css selector to be in a specific state.
:param wait_selector_state: The state to wait for the selector given with `wait_selector`. Default state is `attached`.
:param stealth: Enables stealth mode, check the documentation to see what stealth mode does currently.
:param real_chrome: If you have chrome browser installed on your device, enable this and the Fetcher will launch an instance of your browser and use it.
:param hide_canvas: Add random noise to canvas operations to prevent fingerprinting.
:param disable_webgl: Disables WebGL and WebGL 2.0 support entirely.
:param google_search: Enabled by default, Scrapling will set the referer header to be as if this request came from a Google search for 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 cdp_url: Instead of launching a new browser instance, connect to this CDP URL to control real browsers/NSTBrowser through CDP.
:param nstbrowser_mode: Enables NSTBrowser mode, it have to be used with `cdp_url` argument or it will get completely ignored.
:param nstbrowser_config: The config you want to send with requests to the NSTBrowser. If left empty, Scrapling defaults to an optimized NSTBrowser's docker browserless config.
:return: A `Response` object that is the same as `Adaptor` object except it has these added attributes: `status`, `reason`, `cookies`, `headers`, and `request_headers`
"""
engine = PlaywrightEngine(
proxy=proxy,
locale=locale,
timeout=timeout,
stealth=stealth,
cdp_url=cdp_url,
headless=headless,
useragent=useragent,
real_chrome=real_chrome,
page_action=page_action,
hide_canvas=hide_canvas,
network_idle=network_idle,
google_search=google_search,
extra_headers=extra_headers,
wait_selector=wait_selector,
disable_webgl=disable_webgl,
nstbrowser_mode=nstbrowser_mode,
nstbrowser_config=nstbrowser_config,
disable_resources=disable_resources,
wait_selector_state=wait_selector_state,
adaptor_arguments=self.adaptor_arguments,
)
return await engine.async_fetch(url)
class CustomFetcher(BaseFetcher):
def fetch(self, url: str, browser_engine, **kwargs) -> Response: