Create tool.py

tool.py

@@ -0,0 +1,96 @@
+import requests
+from bs4 import BeautifulSoup
+
+class Tool:
+    """Base class for tools."""
+    def __init__(self):
+        self.name = None
+        self.description = None
+        self.inputs = {}
+        self.output_type = None
+
+    def forward(self, *args, **kwargs):
+        raise NotImplementedError("Subclasses must implement this method.")
+
+class SearchInformationTool(Tool):
+    name = "web_search"
+    description = "Perform a web search query and return the search results."
+    inputs = {"query": {"type": "string", "description": "The web search query to perform."}}
+    inputs["filter_year"] = {
+        "type": "string",
+        "description": "[Optional parameter]: filter the search results to only include pages from a specific year.",
+        "nullable": True,
+    }
+    output_type = "string"
+
+    def __init__(self, browser):
+        super().__init__()
+        self.browser = browser
+
+    def forward(self, query: str, filter_year: int | None = None) -> str:
+        self.browser.visit_page(f"https://www.google.com/search?q={query}", filter_year=filter_year)
+        header, content = self.browser._state()
+        return header.strip() + "\n=======================\n" + content
+
+class VisitTool(Tool):
+    name = "visit_page"
+    description = "Visit a webpage at a given URL and return its text."
+    inputs = {"url": {"type": "string", "description": "The relative or absolute URL of the webpage to visit."}}
+    output_type = "string"
+
+    def __init__(self, browser=None):
+        super().__init__()
+        self.browser = browser
+
+    def forward(self, url: str) -> str:
+        self.browser.visit_page(url)
+        header, content = self.browser._state()
+        return header.strip() + "\n=======================\n" + content
+
+class Browser:
+    def __init__(self):
+        self.current_page = None
+
+    def visit_page(self, url: str, filter_year: int = None):
+        try:
+            response = requests.get(url)
+            response.raise_for_status()
+
+            soup = BeautifulSoup(response.text, 'html.parser')
+
+            for script in soup(["script", "style"]):
+                script.decompose()
+
+            text = soup.get_text(separator='\n', strip=True)
+
+            self.current_page = {
+                "url": url,
+                "filter_year": filter_year,
+                "header": f"Header for {url}",
+                "content": text
+            }
+        except requests.RequestException as e:
+            print(f"An error occurred: {e}")
+            self.current_page = {
+                "url": url,
+                "filter_year": filter_year,
+                "header": "Error",
+                "content": f"Failed to retrieve the page: {e}"
+            }
+
+    def _state(self):
+        if self.current_page:
+            return self.current_page["header"], self.current_page["content"]
+        return "", ""
+
+# Example usage:
+if __name__ == "__main__":
+    browser = Browser()
+    search_tool = SearchInformationTool(browser)
+    visit_tool = VisitTool(browser)
+
+    search_result = search_tool.forward("example query", 2023)
+    print(search_result)
+
+    visit_result = visit_tool.forward("https://example.com")
+    print(visit_result)