ws-bom-robot-app 0.0.58__py3-none-any.whl → 0.0.59__py3-none-any.whl

ws_bom_robot_app/llm/tools/models/main.py

@@ -9,3 +9,5 @@ class ImageGeneratorInput(BaseModel):
   language: str = Field(description="Language of the query. Default is 'it'", default="it")
 class LlmChainInput(BaseModel):
   input: str = Field(description="Input to the LLM chain")
+class SearchOnlineInput(BaseModel):
+  query: str = Field(description="The search query string")

ws_bom_robot_app/llm/tools/tool_manager.py

@@ -4,7 +4,7 @@ from ws_bom_robot_app.llm.models.api import LlmAppTool
 from ws_bom_robot_app.llm.providers.llm_manager import LlmInterface
 from ws_bom_robot_app.llm.vector_store.db.manager import VectorDbManager
 from ws_bom_robot_app.llm.tools.utils import getRandomWaitingMessage, translate_text
-from ws_bom_robot_app.llm.tools.models.main import NoopInput,DocumentRetrieverInput,ImageGeneratorInput,LlmChainInput
+from ws_bom_robot_app.llm.tools.models.main import NoopInput,DocumentRetrieverInput,ImageGeneratorInput,LlmChainInput,SearchOnlineInput
 from pydantic import BaseModel, ConfigDict
 
 class ToolConfig(BaseModel):
@@ -143,6 +143,35 @@ class ToolManager:
           return result
 
 
+    async def search_online(self, query: str):
+        from ws_bom_robot_app.llm.tools.utils import fetch_page, extract_content_with_trafilatura
+        from langchain_community.utilities import DuckDuckGoSearchAPIWrapper
+        import aiohttp, asyncio, ast
+        # Wrapper DuckDuckGo
+        search = DuckDuckGoSearchAPIWrapper(max_results=10)
+        try:
+          raw_results = search.results(query, max_results=10)
+        except Exception as e:
+            print(f"[!] Errore ricerca: {e}")
+        urls = [r["link"] for r in raw_results]
+        async with aiohttp.ClientSession() as session:
+            tasks = [fetch_page(session, url) for url in urls]
+            responses = await asyncio.gather(*tasks)
+        final_results = []
+        for item in responses:
+            url = item["url"]
+            html = item["html"]
+            if html:
+                content = await extract_content_with_trafilatura(html)
+                if content:
+                    final_results.append({"url": url, "content": content})
+                else:
+                    final_results.append({"url": url, "content": "No content found"})
+            else:
+                final_results.append({"url": url, "content": "Page not found"})
+        return final_results
+
+
     #endregion
 
     #class variables (static)
@@ -150,6 +179,7 @@ class ToolManager:
         "document_retriever": ToolConfig(function=document_retriever, model=DocumentRetrieverInput),
         "image_generator": ToolConfig(function=image_generator, model=ImageGeneratorInput),
         "llm_chain": ToolConfig(function=llm_chain, model=LlmChainInput),
+        "search_online": ToolConfig(function=search_online, model=SearchOnlineInput),
     }
 
     #instance methods

ws_bom_robot_app/llm/tools/utils.py

@@ -23,3 +23,19 @@ async def translate_text(llm: LlmInterface, language, text: str, callbacks: list
   prompt = PromptTemplate.from_template(sys_message)
   chain = prompt | llm.get_llm()
   await chain.ainvoke({"language":language, "testo_da_tradurre": text}, {"callbacks": callbacks})
+
+async def fetch_page(session, url):
+    try:
+        async with session.get(url, timeout=10, ssl=False) as response:
+            if response.status == 200:
+                text = await response.text()
+                return {"url": url, "html": text}
+            else:
+                return {"url": url, "html": None}
+    except Exception as e:
+        return {"url": url, "html": None}
+
+async def extract_content_with_trafilatura(html):
+    """Estrae solo il testo principale usando trafilatura"""
+    import trafilatura
+    return trafilatura.extract(html)

{ws_bom_robot_app-0.0.58.dist-info → ws_bom_robot_app-0.0.59.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ws_bom_robot_app
-Version: 0.0.58
+Version: 0.0.59
 Summary: A FastAPI application serving ws bom/robot/llm platform ai.
 Home-page: https://github.com/websolutespa/bom
 Author: Websolute Spa
@@ -17,6 +17,7 @@ Requires-Dist: pydantic==2.10.6
 Requires-Dist: pydantic-settings==2.7.1
 Requires-Dist: fastapi[standard]==0.115.8
 Requires-Dist: chevron==0.14.0
+Requires-Dist: trafilatura==2.0.0
 Requires-Dist: langchain==0.3.25
 Requires-Dist: langchain-community==0.3.24
 Requires-Dist: langchain-core==0.3.59
@@ -48,6 +49,7 @@ Requires-Dist: unstructured-ingest[sharepoint]
 Requires-Dist: unstructured-ingest[slack]
 Requires-Dist: html5lib==1.1
 Requires-Dist: markdownify==0.14.1
+Requires-Dist: duckduckgo-search==8.0.4
 Dynamic: author
 Dynamic: author-email
 Dynamic: classifier

{ws_bom_robot_app-0.0.58.dist-info → ws_bom_robot_app-0.0.59.dist-info}/RECORD

@@ -26,10 +26,10 @@ ws_bom_robot_app/llm/providers/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5
 ws_bom_robot_app/llm/providers/llm_manager.py,sha256=zIkxgTLYQCcup2Ixf4eWap4mNinuJH2YmkjLjZGDyJM,8371
 ws_bom_robot_app/llm/tools/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 ws_bom_robot_app/llm/tools/tool_builder.py,sha256=p_Q32_-OSydcxzj69PgPIuiny816zYv5dVsCHSY0ELc,1188
-ws_bom_robot_app/llm/tools/tool_manager.py,sha256=4Zu4_CjVg-PsrAjb3MHUtwrcGyZw1NZuO02zSH4BSHg,6950
-ws_bom_robot_app/llm/tools/utils.py,sha256=LEId1UolLvJsMYbl-awm7h-gJ9Up9DrrnH6HTglGqVE,1347
+ws_bom_robot_app/llm/tools/tool_manager.py,sha256=adWvaSIur5Ez2gGsuTFqNprZZlQP6ZZj5WknzJjtQ0c,8355
+ws_bom_robot_app/llm/tools/utils.py,sha256=Ba7ScFZPVJ3ke8KLO8ik1wyR2f_zC99Bikqx0OGnKoI,1924
 ws_bom_robot_app/llm/tools/models/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-ws_bom_robot_app/llm/tools/models/main.py,sha256=My-b-2Cr4sRORgHrJVVjUaG59c2y52cvzx3hHUhTPMM,481
+ws_bom_robot_app/llm/tools/models/main.py,sha256=pBQNWPd1OZgZ2xkOnUOawNbujQ5oJXLdyuAex1afLWc,579
 ws_bom_robot_app/llm/utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 ws_bom_robot_app/llm/utils/agent.py,sha256=ISF9faaD5tBi-8sbgQpgfqWT1JIVcgv_lRhyaNAkI2Q,1445
 ws_bom_robot_app/llm/utils/chunker.py,sha256=N7570xBYlObneg-fsvDhPAJ-Pv8C8OaYZOBK6q7LmMI,607
@@ -65,7 +65,7 @@ ws_bom_robot_app/llm/vector_store/loader/__init__.py,sha256=47DEQpj8HBSa-_TImW-5
 ws_bom_robot_app/llm/vector_store/loader/base.py,sha256=L_ugekNuAq0N9O-24wtlHSNHkqSeD-KsJrfGt_FX9Oc,5340
 ws_bom_robot_app/llm/vector_store/loader/docling.py,sha256=yP0zgXLeFAlByaYuj-6cYariuknckrFds0dxdRcnVz8,3456
 ws_bom_robot_app/llm/vector_store/loader/json_loader.py,sha256=LDppW0ZATo4_1hh-KlsAM3TLawBvwBxva_a7k5Oz1sc,858
-ws_bom_robot_app-0.0.58.dist-info/METADATA,sha256=IjxifGGFeeCc_HL_6Z3GHup-x1BNNao9IiTZqjcuU-w,8330
-ws_bom_robot_app-0.0.58.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-ws_bom_robot_app-0.0.58.dist-info/top_level.txt,sha256=Yl0akyHVbynsBX_N7wx3H3ZTkcMLjYyLJs5zBMDAKcM,17
-ws_bom_robot_app-0.0.58.dist-info/RECORD,,
+ws_bom_robot_app-0.0.59.dist-info/METADATA,sha256=hTSpZWSUDrn0IY9jsYvUCGocnzK9kbrEiRINEwiDVe0,8406
+ws_bom_robot_app-0.0.59.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+ws_bom_robot_app-0.0.59.dist-info/top_level.txt,sha256=Yl0akyHVbynsBX_N7wx3H3ZTkcMLjYyLJs5zBMDAKcM,17
+ws_bom_robot_app-0.0.59.dist-info/RECORD,,