added Brave Search document_loader (#6989)

- Added `Brave Search` document loader.
- Refactored BraveSearch wrapper
- Added a Jupyter Notebook example
- Added `Ecosystem/Integrations` BraveSearch page 

Please review:
  - DataLoaders / VectorStores / Retrievers: @rlancemartin, @eyurtsev

docs/extras/ecosystem/integrations/brave_search.mdx

@@ -0,0 +1,36 @@
+# Brave Search
+
+
+>[Brave Search](https://en.wikipedia.org/wiki/Brave_Search) is a search engine developed by Brave Software.
+> - `Brave Search` uses its own web index. As of May 2022, it covered over 10 billion pages and was used to serve 92% 
+> of search results without relying on any third-parties, with the remainder being retrieved 
+> server-side from the Bing API or (on an opt-in basis) client-side from Google. According 
+> to Brave, the index was kept "intentionally smaller than that of Google or Bing" in order to 
+> help avoid spam and other low-quality content, with the disadvantage that "Brave Search is 
+> not yet as good as Google in recovering long-tail queries."
+>- `Brave Search Premium`: As of April 2023 Brave Search is an ad-free website, but it will 
+> eventually switch to a new model that will include ads and premium users will get an ad-free experience.
+> User data including IP addresses won't be collected from its users by default. A premium account 
+> will be required for opt-in data-collection.
+
+
+## Installation and Setup
+
+To get access to the Brave Search API, you need to [create an account and get an API key](https://api.search.brave.com/app/dashboard).
+
+
+## Document Loader
+
+See a [usage example](/docs/modules/data_connection/document_loaders/integrations/brave_search.html).
+
+```python
+from langchain.document_loaders import BraveSearchLoader
+```
+
+## Tool
+
+See a [usage example](/docs/modules/agents/tools/integrations/brave_search.html).
+
+```python
+from langchain.tools import BraveSearch
+```

docs/extras/modules/agents/tools/integrations/brave_search.ipynb

@@ -12,7 +12,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": null,
    "id": "a4c896e5",
    "metadata": {},
    "outputs": [],
@@ -27,7 +27,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "api_key = \"...\""
+    "api_key = \"BSAv1neIuQOsxqOyy0sEe_ie2zD_n_V\""
    ]
   },
   {
@@ -49,7 +49,7 @@
     {
      "data": {
       "text/plain": [
-       "'[{\"title\": \"Barack Obama - Wikipedia\", \"link\": \"https://en.wikipedia.org/wiki/Barack_Obama\", \"snippet\": \"Outside of politics, <strong>Obama</strong> has published three bestselling books: Dreams from My Father (1995), The Audacity of Hope (2006) and A Promised Land (2020). Rankings by scholars and historians, in which he has been featured since 2010, place him in the <strong>middle</strong> to upper tier of American presidents.\"}, {\"title\": \"Obama\\'s Middle Name -- My Last Name -- is \\'Hussein.\\' So?\", \"link\": \"https://www.cair.com/cair_in_the_news/obamas-middle-name-my-last-name-is-hussein-so/\", \"snippet\": \"Many Americans understand that common names don\\\\u2019t only come in the form of a \\\\u201cSmith\\\\u201d or a \\\\u201cJohnson.\\\\u201d Perhaps, they have a neighbor, mechanic or teacher named Hussein. Or maybe they\\\\u2019ve seen fashion designer Hussein Chalayan in the pages of Vogue or recall <strong>King Hussein</strong>, our ally in the Middle East.\"}, {\"title\": \"What\\'s up with Obama\\'s middle name? - Quora\", \"link\": \"https://www.quora.com/Whats-up-with-Obamas-middle-name\", \"snippet\": \"Answer (1 of 15): A better question would be, \\\\u201cWhat\\\\u2019s up with Obama\\\\u2019s first name?\\\\u201d President <strong>Barack Hussein Obama</strong>\\\\u2019s father\\\\u2019s name was <strong>Barack Hussein Obama</strong>. He was named after his father. Hussein, Obama\\\\u2019s middle name, is a very common Arabic name, meaning &quot;good,&quot; &quot;handsome,&quot; or &quot;beautiful.&quot;\"}]'"
+       "'[{\"title\": \"Obama\\'s Middle Name -- My Last Name -- is \\'Hussein.\\' So?\", \"link\": \"https://www.cair.com/cair_in_the_news/obamas-middle-name-my-last-name-is-hussein-so/\", \"snippet\": \"I wasn\\\\u2019t sure whether to laugh or cry a few days back listening to radio talk show host Bill Cunningham repeatedly scream Barack <strong>Obama</strong>\\\\u2019<strong>s</strong> <strong>middle</strong> <strong>name</strong> \\\\u2014 my last <strong>name</strong> \\\\u2014 as if he had anti-Muslim Tourette\\\\u2019s. \\\\u201cHussein,\\\\u201d Cunningham hissed like he was beckoning Satan when shouting the ...\"}, {\"title\": \"What\\'s up with Obama\\'s middle name? - Quora\", \"link\": \"https://www.quora.com/Whats-up-with-Obamas-middle-name\", \"snippet\": \"Answer (1 of 15): A better question would be, \\\\u201cWhat\\\\u2019s up with <strong>Obama</strong>\\\\u2019s first <strong>name</strong>?\\\\u201d President Barack Hussein <strong>Obama</strong>\\\\u2019s father\\\\u2019s <strong>name</strong> was Barack Hussein <strong>Obama</strong>. He was <strong>named</strong> after his father. Hussein, <strong>Obama</strong>\\\\u2019<strong>s</strong> <strong>middle</strong> <strong>name</strong>, is a very common Arabic <strong>name</strong>, meaning &quot;good,&quot; &quot;handsome,&quot; or ...\"}, {\"title\": \"Barack Obama | Biography, Parents, Education, Presidency, Books, ...\", \"link\": \"https://www.britannica.com/biography/Barack-Obama\", \"snippet\": \"Barack <strong>Obama</strong>, in full Barack Hussein <strong>Obama</strong> II, (born August 4, 1961, Honolulu, Hawaii, U.S.), 44th president of the United States (2009\\\\u201317) and the first African American to hold the office. Before winning the presidency, <strong>Obama</strong> represented Illinois in the U.S.\"}]'"
       ]
      },
      "execution_count": 4,
@@ -86,7 +86,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.9.1"
+   "version": "3.10.6"
   }
  },
  "nbformat": 4,

docs/extras/modules/data_connection/document_loaders/integrations/brave_search.ipynb

@@ -0,0 +1,164 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "3dd292b1-9a73-4ea8-af19-5fa6e3c1a62a",
+   "metadata": {},
+   "source": [
+    "# Brave Search\n",
+    "\n",
+    "\n",
+    ">[Brave Search](https://en.wikipedia.org/wiki/Brave_Search) is a search engine developed by Brave Software.\n",
+    "> - `Brave Search` uses its own web index. As of May 2022, it covered over 10 billion pages and was used to serve 92% \n",
+    "> of search results without relying on any third-parties, with the remainder being retrieved \n",
+    "> server-side from the Bing API or (on an opt-in basis) client-side from Google. According \n",
+    "> to Brave, the index was kept \"intentionally smaller than that of Google or Bing\" in order to \n",
+    "> help avoid spam and other low-quality content, with the disadvantage that \"Brave Search is \n",
+    "> not yet as good as Google in recovering long-tail queries.\"\n",
+    ">- `Brave Search Premium`: As of April 2023 Brave Search is an ad-free website, but it will \n",
+    "> eventually switch to a new model that will include ads and premium users will get an ad-free experience.\n",
+    "> User data including IP addresses won't be collected from its users by default. A premium account \n",
+    "> will be required for opt-in data-collection.\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "26f0888e-3f3e-4b82-ac4a-2df6feeccbe0",
+   "metadata": {},
+   "source": [
+    "## Installation and Setup\n",
+    "\n",
+    "To get access to the Brave Search API, you need to [create an account and get an API key](https://api.search.brave.com/app/dashboard).\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "id": "d7d7be09-58bd-47d7-bf1b-33964564f777",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "api_key = \"...\""
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "b3ac92df-6ff0-4dbb-b32b-a7dc140c48ef",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from langchain.document_loaders import BraveSearchLoader"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "7f483caf-58ef-4138-975a-5b783559dc1b",
+   "metadata": {},
+   "source": [
+    "## Example"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "id": "766634cf-3bc7-4656-939a-cafa218807a6",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "3"
+      ]
+     },
+     "execution_count": 6,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "loader = BraveSearchLoader(query=\"obama middle name\", api_key=api_key, search_kwargs={\"count\": 3})\n",
+    "docs = loader.load()\n",
+    "len(docs)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "id": "f1fcc9f1-cbdc-46b3-89d3-80311d557dc6",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "[{'title': \"Obama's Middle Name -- My Last Name -- is 'Hussein.' So?\",\n",
+       "  'link': 'https://www.cair.com/cair_in_the_news/obamas-middle-name-my-last-name-is-hussein-so/'},\n",
+       " {'title': \"What's up with Obama's middle name? - Quora\",\n",
+       "  'link': 'https://www.quora.com/Whats-up-with-Obamas-middle-name'},\n",
+       " {'title': 'Barack Obama | Biography, Parents, Education, Presidency, Books, ...',\n",
+       "  'link': 'https://www.britannica.com/biography/Barack-Obama'}]"
+      ]
+     },
+     "execution_count": 11,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "[doc.metadata for doc in docs]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "id": "601bfd77-03d3-468e-843f-2523d5e215bd",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "['I wasn’t sure whether to laugh or cry a few days back listening to radio talk show host Bill Cunningham repeatedly scream Barack <strong>Obama</strong>’<strong>s</strong> <strong>middle</strong> <strong>name</strong> — my last <strong>name</strong> — as if he had anti-Muslim Tourette’s. “Hussein,” Cunningham hissed like he was beckoning Satan when shouting the ...',\n",
+       " 'Answer (1 of 15): A better question would be, “What’s up with <strong>Obama</strong>’s first <strong>name</strong>?” President Barack Hussein <strong>Obama</strong>’s father’s <strong>name</strong> was Barack Hussein <strong>Obama</strong>. He was <strong>named</strong> after his father. Hussein, <strong>Obama</strong>’<strong>s</strong> <strong>middle</strong> <strong>name</strong>, is a very common Arabic <strong>name</strong>, meaning &quot;good,&quot; &quot;handsome,&quot; or ...',\n",
+       " 'Barack <strong>Obama</strong>, in full Barack Hussein <strong>Obama</strong> II, (born August 4, 1961, Honolulu, Hawaii, U.S.), 44th president of the United States (2009–17) and the first African American to hold the office. Before winning the presidency, <strong>Obama</strong> represented Illinois in the U.S.']"
+      ]
+     },
+     "execution_count": 10,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "[doc.page_content for doc in docs]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "74a6ba54-9e48-4bac-ab9b-03eabd19eb81",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.10.6"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

langchain/document_loaders/__init__.py

@@ -23,6 +23,7 @@ from langchain.document_loaders.blob_loaders import (
     YoutubeAudioLoader,
 )
 from langchain.document_loaders.blockchain import BlockchainDocumentLoader
+from langchain.document_loaders.brave_search import BraveSearchLoader
 from langchain.document_loaders.chatgpt import ChatGPTLoader
 from langchain.document_loaders.college_confidential import CollegeConfidentialLoader
 from langchain.document_loaders.confluence import ConfluenceLoader
@@ -168,6 +169,7 @@ __all__ = [
     "Blob",
     "BlobLoader",
     "BlockchainDocumentLoader",
+    "BraveSearchLoader",
     "CSVLoader",
     "ChatGPTLoader",
     "CoNLLULoader",

langchain/document_loaders/brave_search.py

@@ -0,0 +1,32 @@
+from typing import Iterator, List, Optional
+
+from langchain.docstore.document import Document
+from langchain.document_loaders.base import BaseLoader
+from langchain.utilities.brave_search import BraveSearchWrapper
+
+
+class BraveSearchLoader(BaseLoader):
+    """Loads a query result from Brave Search engine into a list of Documents."""
+
+    def __init__(self, query: str, api_key: str, search_kwargs: Optional[dict] = None):
+        """Initializes the BraveLoader.
+
+        Args:
+            query: The query to search for.
+            api_key: The API key to use.
+            search_kwargs: The search kwargs to use.
+        """
+        self.query = query
+        self.api_key = api_key
+        self.search_kwargs = search_kwargs or {}
+
+    def load(self) -> List[Document]:
+        brave_client = BraveSearchWrapper(
+            api_key=self.api_key,
+            search_kwargs=self.search_kwargs,
+        )
+        return brave_client.download_documents(self.query)
+
+    def lazy_load(self) -> Iterator[Document]:
+        for doc in self.load():
+            yield doc

langchain/utilities/brave_search.py

@@ -1,40 +1,68 @@
 import json
+from typing import List
 
 import requests
 from pydantic import BaseModel, Field
 
+from langchain.schema import Document
+
 
 class BraveSearchWrapper(BaseModel):
     api_key: str
     search_kwargs: dict = Field(default_factory=dict)
+    base_url = "https://api.search.brave.com/res/v1/web/search"
 
     def run(self, query: str) -> str:
+        """Query the Brave search engine and return the results as a JSON string.
+
+        Args:
+            query: The query to search for.
+
+        Returns: The results as a JSON string.
+
+        """
+        web_search_results = self._search_request(query=query)
+        final_results = [
+            {
+                "title": item.get("title"),
+                "link": item.get("url"),
+                "snippet": item.get("description"),
+            }
+            for item in web_search_results
+        ]
+        return json.dumps(final_results)
+
+    def download_documents(self, query: str) -> List[Document]:
+        """Query the Brave search engine and return the results as a list of Documents.
+
+        Args:
+            query: The query to search for.
+
+        Returns: The results as a list of Documents.
+
+        """
+        results = self._search_request(query)
+        return [
+            Document(
+                page_content=item.get("description"),
+                metadata={"title": item.get("title"), "link": item.get("url")},
+            )
+            for item in results
+        ]
+
+    def _search_request(self, query: str) -> List[dict]:
         headers = {
             "X-Subscription-Token": self.api_key,
             "Accept": "application/json",
         }
-        base_url = "https://api.search.brave.com/res/v1/web/search"
         req = requests.PreparedRequest()
         params = {**self.search_kwargs, **{"q": query}}
-        req.prepare_url(base_url, params)
+        req.prepare_url(self.base_url, params)
         if req.url is None:
             raise ValueError("prepared url is None, this should not happen")
 
         response = requests.get(req.url, headers=headers)
-
         if not response.ok:
             raise Exception(f"HTTP error {response.status_code}")
 
-        parsed_response = response.json()
+        return response.json().get("web", {}).get("results", [])
-        web_search_results = parsed_response.get("web", {}).get("results", [])
-        final_results = []
-        if isinstance(web_search_results, list):
-            for item in web_search_results:
-                final_results.append(
-                    {
-                        "title": item.get("title"),
-                        "link": item.get("url"),
-                        "snippet": item.get("description"),
-                    }
-                )
-        return json.dumps(final_results)