rusticai-serpapi 0.0.1__tar.gz

rusticai_serpapi-0.0.1/PKG-INFO

@@ -0,0 +1,38 @@
+Metadata-Version: 2.1
+Name: rusticai-serpapi
+Version: 0.0.1
+Summary: Rustic AI module supporting Search through SerpApi
+Home-page: https://www.rustic.ai/
+License: Apache-2.0
+Author: Dragonscale Industries Inc.
+Author-email: dev@dragonscale.ai
+Requires-Python: >=3.12,<4.0
+Classifier: License :: OSI Approved :: Apache Software License
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Provides-Extra: test
+Requires-Dist: rusticai-core (>=0.0.5,<0.1.0)
+Requires-Dist: serpapi (>=0.1.5,<0.2.0)
+Project-URL: Repository, https://github.com/rustic-ai/python-framework
+Project-URL: Rustic AI Core, https://pypi.org/project/rusticai-core/
+Description-Content-Type: text/markdown
+
+# Rustic AI SerpApi
+
+This module provides an agent that can be used to Search the web using [SerpApi](https://serpapi.com/)
+
+
+## Installing
+
+```shell
+pip install rusticai-serpapi
+```
+**Note:** It depends on [rusticai-core](https://pypi.org/project/rusticai-core/)
+
+## Building from Source
+
+```shell
+poetry install --with dev
+poetry build
+```

rusticai_serpapi-0.0.1/README.md

@@ -0,0 +1,18 @@
+# Rustic AI SerpApi
+
+This module provides an agent that can be used to Search the web using [SerpApi](https://serpapi.com/)
+
+
+## Installing
+
+```shell
+pip install rusticai-serpapi
+```
+**Note:** It depends on [rusticai-core](https://pypi.org/project/rusticai-core/)
+
+## Building from Source
+
+```shell
+poetry install --with dev
+poetry build
+```

rusticai_serpapi-0.0.1/pyproject.toml

@@ -0,0 +1,72 @@
+[build-system]
+requires = ["poetry-core"]
+build-backend = "poetry.core.masonry.api"
+
+[tool.poetry]
+name = "rusticai-serpapi"
+version = "0.0.1"
+description = "Rustic AI module supporting Search through SerpApi"
+authors = ["Dragonscale Industries Inc. <dev@dragonscale.ai>"]
+license = "Apache-2.0"
+readme = "README.md"
+homepage = "https://www.rustic.ai/"
+repository = "https://github.com/rustic-ai/python-framework"
+packages = [{ include = "rustic_ai", from = "src" }]
+
+[tool.poetry.urls]
+"Rustic AI Core" = "https://pypi.org/project/rusticai-core/"
+
+[tool.poetry.dependencies]
+python = "^3.12"
+rusticai-core = {   version = "0.0.5"}
+serpapi = "^0.1.5"
+
+[tool.poetry-monorepo.deps]
+
+
+[tool.poetry.group.dev]
+optional = true
+
+[tool.poetry.group.dev.dependencies]
+pytest = "^8.3.4"
+black = { version = "^24.2.0", extras = ["jupyter"] }
+flake8 = "^7.1.2"
+tox = "^4.24.1"
+isort = "^6.0.0"
+mypy = "^1.15.0"
+rusticai-testing = {   version = "0.0.1"}
+
+[tool.poetry.extras]
+test = ["pytest", "rusticai-testing"]
+
+[tool.black]
+line-length = 120
+target-version = ['py312']
+include = '\.pyi?$'
+exclude = '''
+/(
+    \.git
+  | \.mypy_cache
+  | \.tox
+  | \.venv
+  | _build
+  | buck-out
+  | build
+  | dist
+)/
+'''
+
+[tool.mypy]
+python_version = "3.12"
+ignore_missing_imports = true
+check_untyped_defs = true
+plugins = "pydantic.mypy"
+explicit_package_bases = true
+
+
+[tool.isort]
+profile = "black"
+known_third_party = ["pydantic"]
+known_first_party = ["rustic_ai.core", "rustic_ai.serpapi"]
+known_local_folder = ["rustic_ai.testing"]
+

rusticai_serpapi-0.0.1/src/rustic_ai/serpapi/agent.py

@@ -0,0 +1,148 @@
+import os
+from enum import Enum
+from typing import List, Optional
+from urllib.parse import urlparse
+
+import shortuuid
+from pydantic import BaseModel, Field
+
+import serpapi
+from rustic_ai.core.agents.commons.media import MediaLink
+from rustic_ai.core.guild import Agent, AgentMode, AgentSpec, AgentType, agent
+from rustic_ai.core.messaging.core import JsonDict
+
+
+class SearchEngines(Enum):
+    google = {"key": "q"}
+    bing = {"key": "q"}
+    baidu = {"key": "q"}
+    yahoo = {"key": "p"}
+    duckduckgo = {"key": "q"}
+    ebay = {"key": "_nkw"}
+    yandex = {"key": "text"}
+    home_depot = {"key": "keyword"}
+    google_scholar = {"key": "q"}
+    youtube = {"key": "search_query"}
+    walmart = {"key": "query"}
+    google_maps = {"key": "query"}
+    google_patents = {"key": "q"}
+
+    def get_query(self, query):
+        return {self.value["key"]: query}
+
+
+class SERPQuery(BaseModel):
+    """
+    A class representing a search query.
+    """
+
+    engine: str
+    query: str
+    id: str = Field(default="")
+    num: Optional[int] = Field(default=12)
+    start: Optional[int] = Field(default=0)
+
+
+class SERPResults(BaseModel):
+    """
+    A class representing a search result.
+    """
+
+    engine: str
+    query: str
+    count: int
+    id: str = Field(default="")
+    results: List[MediaLink]
+    total_results: Optional[int] = Field(default=None)
+
+
+class SearchError(BaseModel):
+
+    id: str = Field(default="")
+    response: JsonDict
+
+
+class SERPAgent(Agent):
+    def __init__(self, agent_spec: AgentSpec):
+        super().__init__(agent_spec=agent_spec, agent_type=AgentType.BOT, agent_mode=AgentMode.LOCAL)
+
+        self.serp_api_key = os.getenv("SERP_API_KEY", "")
+        assert self.serp_api_key, "SERP_API_KEY environment variable not set"
+        self.client = serpapi.Client(api_key=self.serp_api_key)
+
+    @agent.processor(SERPQuery)
+    def search(self, ctx: agent.ProcessContext[SERPQuery]) -> None:
+        """
+        Handles the received message.
+
+        Args:
+            message (Message): The received message.
+        """
+        search_query = ctx.payload
+
+        # Use the message.payload as search parameters
+        search_params = SearchEngines[search_query.engine].get_query(search_query.query)
+
+        # Execute asynchronous search and add it to the queue
+        results = self.client.search(
+            params=search_params,
+            engine=search_query.engine,
+            num=search_query.num,
+            start=search_query.start,
+        )
+
+        metadata = results["search_metadata"]
+
+        if metadata["status"] == "Success":
+            # Get the search results
+            search_results = results["organic_results"]
+
+            result_links = []
+
+            # Publish the message
+            for result in search_results:
+                result.update({"search_parameters": search_params})
+
+                filepath = urlparse(result["link"]).path
+                filename = os.path.basename(filepath)
+
+                if not filename:
+                    filename = f"{shortuuid.uuid()}.html"
+
+                metadata = {
+                    "title": result["title"],
+                    "favicon": result.get("favicon", ""),
+                    "search_position": result["position"],
+                    "snippet": result["snippet"],
+                    "date": result.get("date", ""),
+                    "query_id": search_query.id,
+                }
+
+                link = MediaLink(
+                    url=result["link"],
+                    name=filename,
+                    metadata=metadata,
+                    mimetype="text/html",
+                    encoding="utf-8",
+                )
+
+                result_links.append(link)
+
+            total_results = None
+            if "search_information" in results and "total_results" in results["search_information"]:
+                total_results = results["search_information"]["total_results"]
+
+            ctx.send(
+                SERPResults(
+                    count=len(result_links),
+                    results=result_links,
+                    total_results=total_results,
+                    id=search_query.id,
+                    query=result["search_parameters"]["q"],
+                    engine=result["search_parameters"]["engine"],
+                ),
+                new_thread=True,
+            )
+        else:
+            # Publish the error message
+            ctx.send(SearchError(response=results["search_metadata"], id=search_query.id))