
    7|h                     N    d dl mZmZmZ d dlZd dlmZ d dlmZ  G d de      Z	y)    )IteratorListUnionN)Document)
BaseLoaderc                   J    e Zd ZdZ	 d	dedeeee   f   defdZde	e
   fdZy)
BrowserlessLoaderz3Load webpages with `Browserless` /content endpoint.	api_tokenurlstext_contentc                 2    || _         	 || _        	 || _        y)z0Initialize with API token and the URLs to scrapeN)r
   r   r   )selfr
   r   r   s       o/var/www/html/test/engine/venv/lib/python3.12/site-packages/langchain_community/document_loaders/browserless.py__init__zBrowserlessLoader.__init__   s      #$	%(    returnc              #   z  K   | j                   D ]  }| j                  rYt        j                  dd| j                  i|ddigd      }t        |j                         d   d   d	   d   d
   d|i       ht        j                  dd| j                  id|i      }t        |j                  d|i        yw)zLazy load Documents from URLs.z$https://chrome.browserless.io/scrapetokenselectorbody)urlelements)paramsjsondatar   resultstextsource)page_contentmetadataz%https://chrome.browserless.io/contentr   N)r   r   requestspostr
   r   r   r   )r   r   responses      r   	lazy_loadzBrowserlessLoader.lazy_load   s      99 &	C  #==:  # !+F% !)!8!;I!Fq!I&!Q #  $==; s !) # C&	s   B9B;N)T)__name__
__module____qualname____doc__strr   r   boolr   r   r   r$    r   r   r	   r	   	   sG    = QU))$)#tCy.$9)IM))8H- )r   r	   )
typingr   r   r   r!   langchain_core.documentsr   )langchain_community.document_loaders.baser   r	   r+   r   r   <module>r/      s     ( (  - @6
 6r   