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

feat(StealthyFetcher): add the ability to add init_script to context

Browse Source
This commit is contained in:
Karim shoair
2025-09-02 16:19:22 +03:00
parent d49e047736
commit 9543f171aa
3 changed files with 30 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
scrapling/engines/_browsers/_camoufox.py
+14
View File
@@ -60,6 +60,7 @@ class StealthySession:
"timeout",
"page_action",
"wait_selector",
"init_script",
"addons",
"wait_selector_state",
"cookies",
@@ -95,6 +96,7 @@ class StealthySession:
timeout: int | float = 30000,
page_action: Optional[Callable] = None,
wait_selector: Optional[str] = None,
init_script: Optional[str] = None,
addons: Optional[List[str]] = None,
wait_selector_state: SelectorWaitStates = "attached",
cookies: Optional[List[Dict]] = None,
@@ -128,6 +130,7 @@ class StealthySession:
:param timeout: The timeout in milliseconds that is used in all operations and waits through the page. The default is 30,000
: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 init_script: An absolute path to a JavaScript file to be executed on page creation for all pages in this session.
:param geoip: Recommended to use with proxies; Automatically use IP's longitude, latitude, timezone, country, locale, and spoof the WebRTC IP address.
It will also calculate and spoof the browser's language based on the distribution of language speakers in the target region.
:param wait_selector_state: The state to wait for the selector given with `wait_selector`. The default state is `attached`.
@@ -153,6 +156,7 @@ class StealthySession:
"timeout": timeout,
"page_action": page_action,
"wait_selector": wait_selector,
"init_script": init_script,
"addons": addons,
"wait_selector_state": wait_selector_state,
"cookies": cookies,
@@ -180,6 +184,7 @@ class StealthySession:
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
@@ -234,6 +239,9 @@ class StealthySession:
**self.launch_options
)
)
if self.init_script: # pragma: no cover
self.context.add_init_script(path=self.init_script)
if self.cookies: # pragma: no cover
self.context.add_cookies(self.cookies)
@@ -474,6 +482,7 @@ class AsyncStealthySession(StealthySession):
timeout: int | float = 30000,
page_action: Optional[Callable] = None,
wait_selector: Optional[str] = None,
init_script: Optional[str] = None,
addons: Optional[List[str]] = None,
wait_selector_state: SelectorWaitStates = "attached",
cookies: Optional[List[Dict]] = None,
@@ -507,6 +516,7 @@ class AsyncStealthySession(StealthySession):
:param timeout: The timeout in milliseconds that is used in all operations and waits through the page. The default is 30,000
: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 init_script: An absolute path to a JavaScript file to be executed on page creation for all pages in this session.
:param geoip: Recommended to use with proxies; Automatically use IP's longitude, latitude, timezone, country, locale, and spoof the WebRTC IP address.
It will also calculate and spoof the browser's language based on the distribution of language speakers in the target region.
:param wait_selector_state: The state to wait for the selector given with `wait_selector`. The default state is `attached`.
@@ -531,6 +541,7 @@ class AsyncStealthySession(StealthySession):
timeout,
page_action,
wait_selector,
init_script,
addons,
wait_selector_state,
cookies,
@@ -557,6 +568,9 @@ class AsyncStealthySession(StealthySession):
**self.launch_options
)
)
if self.init_script: # pragma: no cover
await self.context.add_init_script(path=self.init_script)
if self.cookies:
await self.context.add_cookies(self.cookies)
scrapling/engines/_browsers/_validators.py
+10
View File
@@ -90,6 +90,7 @@ class CamoufoxConfig(Struct, kw_only=True, frozen=False):
solve_cloudflare: bool = False
wait: int | float = 0
timeout: int | float = 30000
init_script: Optional[str] = None
page_action: Optional[Callable] = None
wait_selector: Optional[str] = None
addons: Optional[List[str]] = None
@@ -131,6 +132,15 @@ class CamoufoxConfig(Struct, kw_only=True, frozen=False):
f"Addon's path is not a folder, you need to pass a folder of the extracted addon: {addon}"
)
if self.init_script is not None:
script_path = Path(self.init_script)
if not script_path.exists():
raise ValueError("Init script path not found")
elif not script_path.is_file():
raise ValueError("Init script is not a file")
elif not script_path.is_absolute():
raise ValueError("Init script is not a absolute path")
if not self.cookies:
self.cookies = []
if self.solve_cloudflare and self.timeout < 60_000:
scrapling/fetchers.py
+6
View File
@@ -62,6 +62,7 @@ class StealthyFetcher(BaseFetcher):
timeout: int | float = 30000,
page_action: Optional[Callable] = None,
wait_selector: Optional[str] = None,
init_script: Optional[str] = None,
addons: Optional[List[str]] = None,
wait_selector_state: SelectorWaitStates = "attached",
cookies: Optional[List[Dict]] = None,
@@ -97,6 +98,7 @@ class StealthyFetcher(BaseFetcher):
:param timeout: The timeout in milliseconds that is used in all operations and waits through the page. The default is 30,000
: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 init_script: An absolute path to a JavaScript file to be executed on page creation with this request.
:param geoip: Recommended to use with proxies; Automatically use IP's longitude, latitude, timezone, country, locale, and spoof the WebRTC IP address.
It will also calculate and spoof the browser's language based on the distribution of language speakers in the target region.
:param wait_selector_state: The state to wait for the selector given with `wait_selector`. The default state is `attached`.
@@ -127,6 +129,7 @@ class StealthyFetcher(BaseFetcher):
disable_ads=disable_ads,
allow_webgl=allow_webgl,
page_action=page_action,
init_script=init_script,
network_idle=network_idle,
block_images=block_images,
block_webrtc=block_webrtc,
@@ -158,6 +161,7 @@ class StealthyFetcher(BaseFetcher):
timeout: int | float = 30000,
page_action: Optional[Callable] = None,
wait_selector: Optional[str] = None,
init_script: Optional[str] = None,
addons: Optional[List[str]] = None,
wait_selector_state: SelectorWaitStates = "attached",
cookies: Optional[List[Dict]] = None,
@@ -193,6 +197,7 @@ class StealthyFetcher(BaseFetcher):
:param timeout: The timeout in milliseconds that is used in all operations and waits through the page. The default is 30,000
: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 init_script: An absolute path to a JavaScript file to be executed on page creation with this request.
:param geoip: Recommended to use with proxies; Automatically use IP's longitude, latitude, timezone, country, locale, and spoof the WebRTC IP address.
It will also calculate and spoof the browser's language based on the distribution of language speakers in the target region.
:param wait_selector_state: The state to wait for the selector given with `wait_selector`. The default state is `attached`.
@@ -223,6 +228,7 @@ class StealthyFetcher(BaseFetcher):
disable_ads=disable_ads,
allow_webgl=allow_webgl,
page_action=page_action,
init_script=init_script,
network_idle=network_idle,
block_images=block_images,
block_webrtc=block_webrtc,