mirror of
https://github.com/crewAIInc/crewAI.git
synced 2026-01-25 08:08:14 +00:00
Cutting new version with improved scrapping
This commit is contained in:
@@ -19,7 +19,11 @@ class ScrapeElementFromWebsiteTool(BaseTool):
|
|||||||
args_schema: Type[BaseModel] = ScrapeElementFromWebsiteToolSchema
|
args_schema: Type[BaseModel] = ScrapeElementFromWebsiteToolSchema
|
||||||
website_url: Optional[str] = None
|
website_url: Optional[str] = None
|
||||||
css_element: Optional[str] = None
|
css_element: Optional[str] = None
|
||||||
headers: Optional[dict] = {'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/58.0.3029.110 Safari/537.3'}
|
headers: Optional[dict] = {
|
||||||
|
'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/58.0.3029.110 Safari/537.3',
|
||||||
|
'Accept-Language': 'en-US,en;q=0.5',
|
||||||
|
'Referer': 'https://www.google.com/'
|
||||||
|
}
|
||||||
|
|
||||||
def __init__(self, website_url: Optional[str] = None, css_element: Optional[str] = None, **kwargs):
|
def __init__(self, website_url: Optional[str] = None, css_element: Optional[str] = None, **kwargs):
|
||||||
super().__init__(**kwargs)
|
super().__init__(**kwargs)
|
||||||
|
|||||||
@@ -17,7 +17,11 @@ class ScrapeWebsiteTool(BaseTool):
|
|||||||
description: str = "A tool that can be used to read a website content."
|
description: str = "A tool that can be used to read a website content."
|
||||||
args_schema: Type[BaseModel] = ScrapeWebsiteToolSchema
|
args_schema: Type[BaseModel] = ScrapeWebsiteToolSchema
|
||||||
website_url: Optional[str] = None
|
website_url: Optional[str] = None
|
||||||
headers: Optional[dict] = {'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/58.0.3029.110 Safari/537.3'}
|
headers: Optional[dict] = {
|
||||||
|
'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/58.0.3029.110 Safari/537.3',
|
||||||
|
'Accept-Language': 'en-US,en;q=0.5',
|
||||||
|
'Referer': 'https://www.google.com/'
|
||||||
|
}
|
||||||
|
|
||||||
def __init__(self, website_url: Optional[str] = None, **kwargs):
|
def __init__(self, website_url: Optional[str] = None, **kwargs):
|
||||||
super().__init__(**kwargs)
|
super().__init__(**kwargs)
|
||||||
|
|||||||
Reference in New Issue
Block a user