aiagents4pharma 1.27.1__py3-none-any.whl → 1.28.0__py3-none-any.whl

aiagents4pharma/talk2scholars/agents/__init__.py

@@ -5,3 +5,4 @@ This file is used to import all the modules in the package.
 from . import main_agent
 from . import s2_agent
 from . import zotero_agent
+from . import pdf_agent

aiagents4pharma/talk2scholars/agents/pdf_agent.py

@@ -0,0 +1,106 @@
+#!/usr/bin/env python3
+"""
+Agent for interacting with PDF documents via question and answer.
+
+This module initializes and compiles a LangGraph application that enables users to query PDF
+documents using a question_and_answer tool. It integrates a language model and follows
+the ReAct pattern to process and answer queries related to PDF content.
+
+Usage:
+    >>> app = get_app("unique_thread_id")
+    >>> response = app.invoke(initial_state)
+"""
+
+import logging
+import hydra
+from langchain_core.language_models.chat_models import BaseChatModel
+from langgraph.graph import START, StateGraph
+from langgraph.prebuilt import create_react_agent, ToolNode
+from langgraph.checkpoint.memory import MemorySaver
+from ..state.state_talk2scholars import Talk2Scholars
+from ..tools.pdf.question_and_answer import question_and_answer_tool
+from ..tools.s2.query_results import query_results
+
+# Initialize logger
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
+
+def get_app(
+    uniq_id,
+    llm_model: BaseChatModel
+):
+    """
+    Initializes and returns the LangGraph application for the PDF agent.
+
+    This function sets up the PDF agent by loading configuration settings via Hydra,
+    initializing a model, and creating a workflow graph that incorporates
+    PDF-specific tools. The agent is built using the ReAct pattern to facilitate interactive
+    querying and processing of PDF documents.
+
+    Args:
+        uniq_id (str): A unique identifier for the current conversation session or thread.
+        llm_model (BaseChatModel, optional): The language model instance to be used. 
+            Defaults to ChatOpenAI(model="gpt-4o-mini", temperature=0).
+
+    Returns:
+        StateGraph: A compiled LangGraph application capable of handling PDF interactions.
+
+    Example:
+        >>> app = get_app("thread_123")
+        >>> result = app.invoke(initial_state)
+    """
+    # Load configuration using Hydra.
+    with hydra.initialize(version_base=None, config_path="../configs"):
+        cfg = hydra.compose(
+            config_name="config",
+            overrides=["agents/talk2scholars/pdf_agent=default"],
+        )
+        cfg = cfg.agents.talk2scholars.pdf_agent
+        logger.info("Loaded pdf_agent configuration.")
+
+    def agent_pdf_node(state: Talk2Scholars):
+        """
+        Processes the current state by invoking the language model for PDF question and answer.
+
+        Args:
+            state (Talk2Scholars): The current conversation state containing query details
+            and context.
+
+        Returns:
+            Any: The response generated by the language model after processing the state.
+        """
+        logger.info("Creating Agent_PDF node with thread_id %s", uniq_id)
+        response = model.invoke(
+            state,
+            {"configurable": {"thread_id": uniq_id}}
+        )
+        return response
+
+    # Define the tool node that includes the PDF QnA tool.
+    tools = ToolNode([question_and_answer_tool, query_results])
+
+    logger.info("Using OpenAI model %s", llm_model)
+
+    # Create the agent using the provided BaseChatModel instance.
+    model = create_react_agent(
+        llm_model,
+        tools=tools,
+        state_schema=Talk2Scholars,
+        prompt=cfg.pdf_agent,
+        checkpointer=MemorySaver(),
+    )
+
+    # Define a new workflow graph with the state schema.
+    workflow = StateGraph(Talk2Scholars)
+    workflow.add_node("agent_pdf", agent_pdf_node)
+    workflow.add_edge(START, "agent_pdf")
+
+    # Initialize memory to persist state between runs.
+    checkpointer = MemorySaver()
+
+    # Compile the graph into a runnable app.
+    app = workflow.compile(checkpointer=checkpointer)
+    logger.info("Compiled the PDF agent graph.")
+
+    return app

aiagents4pharma/talk2scholars/configs/agents/talk2scholars/__init__.py

@@ -4,4 +4,5 @@ Import all the modules in the package
 
 from . import s2_agent
 from . import main_agent
+from . import pdf_agent
 from . import zotero_agent

aiagents4pharma/talk2scholars/configs/agents/talk2scholars/pdf_agent/__init__.py

@@ -0,0 +1,3 @@
+"""
+Import all the modules in the package
+"""

aiagents4pharma/talk2scholars/configs/config.yaml

@@ -4,8 +4,10 @@ defaults:
   - agents/talk2scholars/s2_agent: default
   - agents/talk2scholars/zotero_agent: default
   - app/frontend: default
+  - agents/talk2scholars/pdf_agent: default
   - tools/search: default
   - tools/single_paper_recommendation: default
   - tools/multi_paper_recommendation: default
   - tools/retrieve_semantic_scholar_paper_id: default
+  - tools/question_and_answer: default
   - tools/zotero_read: default

aiagents4pharma/talk2scholars/configs/tools/__init__.py

@@ -5,4 +5,5 @@ Import all the modules in the package
 from . import search
 from . import single_paper_recommendation
 from . import multi_paper_recommendation
+from . import question_and_answer
 from . import zotero_read

aiagents4pharma/talk2scholars/configs/tools/question_and_answer/__init__.py

@@ -0,0 +1,3 @@
+"""
+Import all the modules in the package
+"""

aiagents4pharma/talk2scholars/state/state_talk2scholars.py

@@ -60,5 +60,6 @@ class Talk2Scholars(AgentState):
     last_displayed_papers: Annotated[Dict[str, Any], replace_dict]
     papers: Annotated[Dict[str, Any], replace_dict]
     multi_papers: Annotated[Dict[str, Any], replace_dict]
+    pdf_data: Annotated[Dict[str, Any], replace_dict]
     zotero_read: Annotated[Dict[str, Any], replace_dict]
     llm_model: BaseChatModel

aiagents4pharma/talk2scholars/tests/test_pdf_agent.py

@@ -0,0 +1,126 @@
+"""
+Unit Tests for the PDF agent.
+"""
+
+# pylint: disable=redefined-outer-name
+from unittest import mock
+import pytest
+from langchain_core.messages import HumanMessage, AIMessage
+from ..agents.pdf_agent import get_app
+from ..state.state_talk2scholars import Talk2Scholars
+
+
+@pytest.fixture(autouse=True)
+def mock_hydra_fixture():
+    """Mock Hydra configuration to prevent external dependencies."""
+    with mock.patch("hydra.initialize"), mock.patch("hydra.compose") as mock_compose:
+        # Create a mock configuration with a pdf_agent section.
+        cfg_mock = mock.MagicMock()
+        # The pdf_agent config will be accessed as cfg.agents.talk2scholars.pdf_agent in get_app.
+        cfg_mock.agents.talk2scholars.pdf_agent.some_property = "Test prompt"
+        mock_compose.return_value = cfg_mock
+        yield mock_compose
+
+
+@pytest.fixture
+def mock_tools_fixture():
+    """Mock PDF agent tools to prevent execution of real API calls."""
+    with (
+        mock.patch(
+            "aiagents4pharma.talk2scholars.agents.pdf_agent.question_and_answer_tool"
+        ) as mock_question_and_answer_tool,
+        mock.patch(
+            "aiagents4pharma.talk2scholars.agents.pdf_agent.query_results"
+        ) as mock_query_results,
+    ):
+        mock_question_and_answer_tool.return_value = {
+            "result": "Mock Question and Answer Result"
+        }
+        mock_query_results.return_value = {"result": "Mock Query Result"}
+        yield [mock_question_and_answer_tool, mock_query_results]
+
+
+@pytest.fixture
+def mock_llm():
+    """Provide a dummy language model to pass into get_app."""
+    return mock.Mock()
+
+
+@pytest.mark.usefixtures("mock_hydra_fixture")
+def test_pdf_agent_initialization(mock_llm):
+    """Test that PDF agent initializes correctly with mock configuration."""
+    thread_id = "test_thread"
+    with mock.patch(
+        "aiagents4pharma.talk2scholars.agents.pdf_agent.create_react_agent"
+    ) as mock_create:
+        mock_create.return_value = mock.Mock()
+        app = get_app(thread_id, mock_llm)
+        assert app is not None
+        assert mock_create.called
+
+
+def test_pdf_agent_invocation(mock_llm):
+    """Test that the PDF agent processes user input and returns a valid response."""
+    thread_id = "test_thread"
+    # Create a sample state with a human message.
+    mock_state = Talk2Scholars(
+        messages=[HumanMessage(content="Extract key data from PDF")]
+    )
+    with mock.patch(
+        "aiagents4pharma.talk2scholars.agents.pdf_agent.create_react_agent"
+    ) as mock_create:
+        mock_agent = mock.Mock()
+        mock_create.return_value = mock_agent
+        # Simulate a response from the PDF agent.
+        mock_agent.invoke.return_value = {
+            "messages": [
+                AIMessage(content="PDF content extracted successfully")
+            ],
+            "pdf_data": {"page": 1, "text": "Sample PDF text"},
+        }
+        app = get_app(thread_id, mock_llm)
+        result = app.invoke(
+            mock_state,
+            config={
+                "configurable": {
+                    "thread_id": thread_id,
+                    "checkpoint_ns": "test_ns",
+                    "checkpoint_id": "test_checkpoint",
+                }
+            },
+        )
+        assert "messages" in result
+        assert "pdf_data" in result
+        assert result["pdf_data"]["page"] == 1
+
+
+def test_pdf_agent_tools_assignment(request, mock_llm):
+    """Ensure that the correct tools are assigned to the PDF agent."""
+    thread_id = "test_thread"
+    mock_tools = request.getfixturevalue("mock_tools_fixture")
+    with (
+        mock.patch(
+            "aiagents4pharma.talk2scholars.agents.pdf_agent.create_react_agent"
+        ) as mock_create,
+        mock.patch(
+            "aiagents4pharma.talk2scholars.agents.pdf_agent.ToolNode"
+        ) as mock_toolnode,
+    ):
+        mock_agent = mock.Mock()
+        mock_create.return_value = mock_agent
+        mock_tool_instance = mock.Mock()
+        # For the PDF agent, we expect two tools: question_and_answer_tool and query_results.
+        mock_tool_instance.tools = mock_tools
+        mock_toolnode.return_value = mock_tool_instance
+        get_app(thread_id, mock_llm)
+        assert mock_toolnode.called
+        assert len(mock_tool_instance.tools) == 2
+
+
+def test_pdf_agent_hydra_failure(mock_llm):
+    """Test exception handling when Hydra fails to load config for PDF agent."""
+    thread_id = "test_thread"
+    with mock.patch("hydra.initialize", side_effect=Exception("Hydra error")):
+        with pytest.raises(Exception) as exc_info:
+            get_app(thread_id, mock_llm)
+        assert "Hydra error" in str(exc_info.value)

aiagents4pharma/talk2scholars/tests/test_question_and_answer_tool.py

@@ -0,0 +1,186 @@
+"""
+Unit tests for question_and_answer tool functionality.
+"""
+
+from langchain.docstore.document import Document
+
+from ..tools.pdf import question_and_answer
+from ..tools.pdf.question_and_answer import (
+    extract_text_from_pdf_data,
+    question_and_answer_tool,
+    generate_answer,
+)
+
+
+def test_extract_text_from_pdf_data():
+    """
+    Test that extract_text_from_pdf_data returns text containing 'Hello World'.
+    """
+    extracted_text = extract_text_from_pdf_data(DUMMY_PDF_BYTES)
+    assert "Hello World" in extracted_text
+
+
+DUMMY_PDF_BYTES = (
+    b"%PDF-1.4\n"
+    b"%\xe2\xe3\xcf\xd3\n"
+    b"1 0 obj\n"
+    b"<< /Type /Catalog /Pages 2 0 R >>\n"
+    b"endobj\n"
+    b"2 0 obj\n"
+    b"<< /Type /Pages /Count 1 /Kids [3 0 R] >>\n"
+    b"endobj\n"
+    b"3 0 obj\n"
+    b"<< /Type /Page /Parent 2 0 R /MediaBox [0 0 612 792] /Contents 4 0 R "
+    b"/Resources << /Font << /F1 5 0 R >> >> >>\n"
+    b"endobj\n"
+    b"4 0 obj\n"
+    b"<< /Length 44 >>\n"
+    b"stream\nBT\n/F1 24 Tf\n72 712 Td\n(Hello World) Tj\nET\nendstream\n"
+    b"endobj\n"
+    b"5 0 obj\n"
+    b"<< /Type /Font /Subtype /Type1 /BaseFont /Helvetica >>\n"
+    b"endobj\n"
+    b"xref\n0 6\n0000000000 65535 f \n0000000010 00000 n \n0000000053 00000 n \n"
+    b"0000000100 00000 n \n0000000150 00000 n \n0000000200 00000 n \n"
+    b"trailer\n<< /Size 6 /Root 1 0 R >>\nstartxref\n250\n%%EOF\n"
+)
+
+
+def fake_generate_answer(question, pdf_bytes, _llm_model):
+    """
+    Fake generate_answer function to bypass external dependencies.
+    """
+    return {
+        "answer": "Mock answer",
+        "question": question,
+        "pdf_bytes_length": len(pdf_bytes),
+    }
+
+
+def test_question_and_answer_tool_success(monkeypatch):
+    """
+    Test that question_and_answer_tool returns the expected result on success.
+    """
+    monkeypatch.setattr(
+        question_and_answer, "generate_answer", fake_generate_answer
+    )
+    # Create a valid state with pdf_data containing both pdf_object and pdf_url,
+    # and include a dummy llm_model.
+    state = {
+        "pdf_data": {"pdf_object": DUMMY_PDF_BYTES, "pdf_url": "http://dummy.url"},
+        "llm_model": object(),  # Provide a dummy LLM model instance.
+    }
+    question = "What is in the PDF?"
+    # Call the underlying function directly via .func to bypass the StructuredTool wrapper.
+    result = question_and_answer_tool.func(
+        question=question, tool_call_id="test_call_id", state=state
+    )
+    assert result["answer"] == "Mock answer"
+    assert result["question"] == question
+    assert result["pdf_bytes_length"] == len(DUMMY_PDF_BYTES)
+
+
+def test_question_and_answer_tool_no_pdf_data():
+    """
+    Test that an error is returned if the state lacks the 'pdf_data' key.
+    """
+    state = {}  # pdf_data key is missing.
+    question = "Any question?"
+    result = question_and_answer_tool.func(
+        question=question, tool_call_id="test_call_id", state=state
+    )
+    messages = result.update["messages"]
+    assert any("No pdf_data found in state." in msg.content for msg in messages)
+
+
+def test_question_and_answer_tool_no_pdf_object():
+    """
+    Test that an error is returned if the pdf_object is missing within pdf_data.
+    """
+    state = {"pdf_data": {"pdf_object": None}}
+    question = "Any question?"
+    result = question_and_answer_tool.func(
+        question=question, tool_call_id="test_call_id", state=state
+    )
+    messages = result.update["messages"]
+    assert any(
+        "PDF binary data is missing in the pdf_data from state." in msg.content
+        for msg in messages
+    )
+
+
+def test_question_and_answer_tool_no_llm_model():
+    """
+    Test that an error is returned if the LLM model is missing in the state.
+    """
+    state = {
+        "pdf_data": {"pdf_object": DUMMY_PDF_BYTES, "pdf_url": "http://dummy.url"}
+        # Note: llm_model is intentionally omitted.
+    }
+    question = "What is in the PDF?"
+    result = question_and_answer_tool.func(
+        question=question, tool_call_id="test_call_id", state=state
+    )
+    assert result == {"error": "No LLM model found in state."}
+
+
+def test_generate_answer(monkeypatch):
+    """
+    Test generate_answer function with controlled monkeypatched dependencies.
+    """
+
+    def fake_split_text(_self, _text):
+        """Fake split_text method that returns controlled chunks."""
+        return ["chunk1", "chunk2"]
+
+    monkeypatch.setattr(
+        question_and_answer.CharacterTextSplitter, "split_text", fake_split_text
+    )
+
+    def fake_annoy_from_documents(_documents, _embeddings):
+        """
+        Fake Annoy.from_documents function that returns a fake vector store.
+        """
+        # pylint: disable=too-few-public-methods, unused-argument
+        class FakeVectorStore:
+            """Fake vector store for similarity search."""
+            def similarity_search(self, _question, k):
+                """Return a list with a single dummy Document."""
+                return [Document(page_content="dummy content")]
+        return FakeVectorStore()
+
+    monkeypatch.setattr(
+        question_and_answer.Annoy, "from_documents", fake_annoy_from_documents
+    )
+
+    def fake_load_qa_chain(_llm, chain_type):  # chain_type matches the keyword argument
+        """
+        Fake load_qa_chain function that returns a fake QA chain.
+        """
+        # pylint: disable=too-few-public-methods, unused-argument
+        class FakeChain:
+            """Fake QA chain for testing generate_answer."""
+            def invoke(self, **kwargs):
+                """
+                Fake invoke method that returns a mock answer.
+                """
+                input_data = kwargs.get("input")
+                return {
+                    "answer": "real mock answer",
+                    "question": input_data.get("question"),
+                }
+        return FakeChain()
+
+    monkeypatch.setattr(question_and_answer, "load_qa_chain", fake_load_qa_chain)
+    # Set dummy configuration values so that generate_answer can run.
+    question_and_answer.cfg.chunk_size = 1000
+    question_and_answer.cfg.chunk_overlap = 0
+    question_and_answer.cfg.openai_api_key = "dummy_key"
+    question_and_answer.cfg.num_retrievals = 1
+    question_and_answer.cfg.qa_chain_type = "dummy-chain"
+
+    question = "What is in the PDF?"
+    dummy_llm_model = object()  # A dummy model placeholder.
+    answer = generate_answer(question, DUMMY_PDF_BYTES, dummy_llm_model)
+    assert answer["answer"] == "real mock answer"
+    assert answer["question"] == question

aiagents4pharma/talk2scholars/tools/__init__.py

@@ -5,4 +5,5 @@ Import statements
 """
 
 from . import s2
+from . import pdf
 from . import zotero

aiagents4pharma/talk2scholars/tools/pdf/__init__.py

@@ -0,0 +1,5 @@
+'''
+This file is used to import all the modules in the package.
+'''
+
+from . import question_and_answer

aiagents4pharma/talk2scholars/tools/pdf/question_and_answer.py

@@ -0,0 +1,170 @@
+#!/usr/bin/env python3
+"""
+question_and_answer: Tool for performing Q&A on PDF documents using retrieval augmented generation.
+
+This module provides functionality to extract text from PDF binary data, split it into 
+chunks, retrieve relevant segments via a vector store, and generate an answer to a 
+user-provided question using a language model chain.
+"""
+
+import io
+import logging
+from typing import Annotated, Dict, Any, List
+
+from PyPDF2 import PdfReader
+from pydantic import BaseModel, Field
+import hydra
+
+from langchain.chains.question_answering import load_qa_chain
+from langchain.docstore.document import Document
+from langchain.text_splitter import CharacterTextSplitter
+from langchain_community.vectorstores import Annoy
+from langchain_openai import OpenAIEmbeddings
+from langchain_core.language_models.chat_models import BaseChatModel
+
+from langchain_core.messages import ToolMessage
+from langchain_core.tools import tool
+from langchain_core.tools.base import InjectedToolCallId
+from langgraph.types import Command
+from langgraph.prebuilt import InjectedState
+
+# Set up logging.
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+logger.setLevel(logging.INFO)
+
+# Load configuration using Hydra.
+with hydra.initialize(version_base=None, config_path="../../configs"):
+    cfg = hydra.compose(config_name="config", overrides=["tools/question_and_answer=default"])
+    cfg = cfg.tools.question_and_answer
+    logger.info("Loaded Question and Answer tool configuration.")
+
+class QuestionAndAnswerInput(BaseModel):
+    """
+    Input schema for the PDF Question and Answer tool.
+
+    Attributes:
+        question (str): The question to ask regarding the PDF content.
+        tool_call_id (str): Unique identifier for the tool call, injected automatically.
+    """
+    question: str = Field(
+        description="The question to ask regarding the PDF content."
+    )
+    tool_call_id: Annotated[str, InjectedToolCallId]
+    state: Annotated[dict, InjectedState]
+
+def extract_text_from_pdf_data(pdf_bytes: bytes) -> str:
+    """
+    Extract text content from PDF binary data.
+
+    This function uses PyPDF2 to read the provided PDF bytes and concatenates the text
+    extracted from each page.
+
+    Args:
+        pdf_bytes (bytes): The binary data of the PDF document.
+
+    Returns:
+        str: The complete text extracted from the PDF.
+    """
+    reader = PdfReader(io.BytesIO(pdf_bytes))
+    text = ""
+    for page in reader.pages:
+        page_text = page.extract_text() or ""
+        text += page_text
+    return text
+
+def generate_answer(question: str, pdf_bytes: bytes, llm_model: BaseChatModel) -> Dict[str, Any]:
+    """
+    Generate an answer for a question using retrieval augmented generation on PDF content.
+
+    This function extracts text from the PDF data, splits the text into manageable chunks,
+    performs a similarity search to retrieve the most relevant segments, and then uses a
+    question-answering chain (built using the provided llm_model) to generate an answer.
+
+    Args:
+        question (str): The question to be answered.
+        pdf_bytes (bytes): The binary content of the PDF document.
+        llm_model (BaseChatModel): The language model instance to use for answering.
+
+    Returns:
+        Dict[str, Any]: A dictionary containing the answer generated by the language model.
+    """
+    text = extract_text_from_pdf_data(pdf_bytes)
+    logger.info("Extracted text from PDF.")
+    text_splitter = CharacterTextSplitter(
+        separator="\n",
+        chunk_size=cfg.chunk_size,
+        chunk_overlap=cfg.chunk_overlap
+    )
+    chunks = text_splitter.split_text(text)
+    documents: List[Document] = [Document(page_content=chunk) for chunk in chunks]
+    logger.info("Split PDF text into %d chunks.", len(documents))
+
+    embeddings = OpenAIEmbeddings(openai_api_key=cfg.openai_api_key)
+    vector_store = Annoy.from_documents(documents, embeddings)
+    search_results = vector_store.similarity_search(
+        question,
+        k=cfg.num_retrievals
+    )
+    logger.info("Retrieved %d relevant document chunks.", len(search_results))
+    # Use the provided llm_model to build the QA chain.
+    qa_chain = load_qa_chain(llm_model, chain_type=cfg.qa_chain_type)
+    answer = qa_chain.invoke(
+        input={"input_documents": search_results, "question": question}
+    )
+    return answer
+
+@tool(args_schema=QuestionAndAnswerInput)
+def question_and_answer_tool(
+    question: str,
+    tool_call_id: Annotated[str, InjectedToolCallId],
+    state: Annotated[dict, InjectedState],
+) -> Dict[str, Any]:
+    """
+    Answer a question using PDF content stored in the state via retrieval augmented generation.
+
+    This tool retrieves the PDF binary data from the state (under the key "pdf_data"), extracts its
+    textual content, and generates an answer to the specified question. It also extracts the 
+    llm_model (of type BaseChatModel) from the state to use for answering.
+
+    Args:
+        question (str): The question regarding the PDF content.
+        tool_call_id (str): Unique identifier for the current tool call.
+        state (dict): A dictionary representing the current state, expected to contain PDF data
+                      under the key "pdf_data" with a sub-key "pdf_object" for the binary content,
+                      and a key "llm_model" holding the language model instance.
+
+    Returns:
+        Dict[str, Any]: A dictionary containing the generated answer or an error message.
+    """
+    logger.info("Starting PDF Question and Answer tool using PDF data from state.")
+    pdf_state = state.get("pdf_data")
+    if not pdf_state:
+        error_msg = "No pdf_data found in state."
+        logger.error(error_msg)
+        return Command(
+            update={
+                "messages": [
+                    ToolMessage(content=error_msg, tool_call_id=tool_call_id)
+                ]
+            }
+        )
+    pdf_bytes = pdf_state.get("pdf_object")
+    if not pdf_bytes:
+        error_msg = "PDF binary data is missing in the pdf_data from state."
+        logger.error(error_msg)
+        return Command(
+            update={
+                "messages": [
+                    ToolMessage(content=error_msg, tool_call_id=tool_call_id)
+                ]
+            }
+        )
+    # Retrieve llm_model from state; use a default if not provided.
+    llm_model = state.get("llm_model")
+    if not llm_model:
+        logger.error("Missing LLM model instance in state.")
+        return {"error": "No LLM model found in state."}
+    answer = generate_answer(question, pdf_bytes, llm_model)
+    logger.info("Generated answer: %s", answer)
+    return answer

aiagents4pharma/talk2scholars/tools/s2/query_results.py

@@ -44,7 +44,7 @@ def query_results(question: str, state: Annotated[dict, InjectedState]) -> str:
         raise NoPapersFoundError(
             "No papers found. A search needs to be performed first."
         )
-    context_key = state.get("last_displayed_papers")
+    context_key = state.get("last_displayed_papers","pdf_data")
     dic_papers = state.get(context_key)
     df_papers = pd.DataFrame.from_dict(dic_papers, orient="index")
     df_agent = create_pandas_dataframe_agent(

{aiagents4pharma-1.27.1.dist-info → aiagents4pharma-1.28.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: aiagents4pharma
-Version: 1.27.1
+Version: 1.28.0
 Summary: AI Agents for drug discovery, drug development, and other pharmaceutical R&D.
 Classifier: Programming Language :: Python :: 3
 Classifier: License :: OSI Approved :: MIT License

{aiagents4pharma-1.27.1.dist-info → aiagents4pharma-1.28.0.dist-info}/RECORD

@@ -135,16 +135,18 @@ aiagents4pharma/talk2knowledgegraphs/utils/enrichments/pubchem_strings.py,sha256
 aiagents4pharma/talk2knowledgegraphs/utils/extractions/__init__.py,sha256=7gwwtfzKhB8GuOBD47XRi0NprwEXkOzwNl5eeu-hDTI,86
 aiagents4pharma/talk2knowledgegraphs/utils/extractions/pcst.py,sha256=m5p0yoJb7I19ua5yeQfXPf7c4r6S1XPwttsrM7Qoy94,9336
 aiagents4pharma/talk2scholars/__init__.py,sha256=gphERyVKZHvOnMQsml7TIHlaIshHJ75R1J3FKExkfuY,120
-aiagents4pharma/talk2scholars/agents/__init__.py,sha256=WxEauzCzLEGyhdIRkxSBpNW5c_Uzf7iJUdM57IQkXH8,144
+aiagents4pharma/talk2scholars/agents/__init__.py,sha256=ZwFiHOlDGJk1601J5xEZDy0btPzqiOk2UCocKxohde8,168
 aiagents4pharma/talk2scholars/agents/main_agent.py,sha256=nZIhOyEUSHECM4-wEHbDrfHRLkqoxW0H4fy6-MpA6N8,9397
+aiagents4pharma/talk2scholars/agents/pdf_agent.py,sha256=c9-_z5qp5Zkgh6piEIlgI4uo4OMXD3janZNmfYwnFCg,3729
 aiagents4pharma/talk2scholars/agents/s2_agent.py,sha256=ZiXtQVX2UbIyMOSXajuloWepEm7DKs6ZpPS0HgHzw0g,4492
 aiagents4pharma/talk2scholars/agents/zotero_agent.py,sha256=flIvg1ORaMiQpGEbsRM4zJHRNXi6UUv7emHDjH5HVY4,3961
 aiagents4pharma/talk2scholars/configs/__init__.py,sha256=tf2gz8n7M4ko6xLdX_C925ELVIxoP6SgkPcbeh59ad4,151
-aiagents4pharma/talk2scholars/configs/config.yaml,sha256=IBrHX_mACNb7R4rrI_zbWgscAMUdIAkOg9LDgLN1o28,386
+aiagents4pharma/talk2scholars/configs/config.yaml,sha256=Lk5kZSDENqCMhushMxDIyLCzLtH7IpvVP_9f5BaUAMQ,469
 aiagents4pharma/talk2scholars/configs/agents/__init__.py,sha256=yyh7PB2oY_JulnpSQCWS4wwCH_uzIdt47O2Ay48x_oU,75
-aiagents4pharma/talk2scholars/configs/agents/talk2scholars/__init__.py,sha256=MI4RmoQQ2P-JQgCJ8XEucqF6g2VlNNi37yxcsXFj9Oo,122
+aiagents4pharma/talk2scholars/configs/agents/talk2scholars/__init__.py,sha256=64GEWAoKOd_YHLi27eSOcOC5eSLK0IG_FNra3ZBt02Y,146
 aiagents4pharma/talk2scholars/configs/agents/talk2scholars/main_agent/__init__.py,sha256=fqQQ-GlRcbzru2KmEk3oMma0R6_SzGM8dOXzYeU4oVA,46
 aiagents4pharma/talk2scholars/configs/agents/talk2scholars/main_agent/default.yaml,sha256=wHbFTQSfdRc0JO2lbzXXHliFNz40Oza-mGmVphNOoPw,2615
+aiagents4pharma/talk2scholars/configs/agents/talk2scholars/pdf_agent/__init__.py,sha256=fqQQ-GlRcbzru2KmEk3oMma0R6_SzGM8dOXzYeU4oVA,46
 aiagents4pharma/talk2scholars/configs/agents/talk2scholars/s2_agent/__init__.py,sha256=fqQQ-GlRcbzru2KmEk3oMma0R6_SzGM8dOXzYeU4oVA,46
 aiagents4pharma/talk2scholars/configs/agents/talk2scholars/s2_agent/default.yaml,sha256=WQOHG1WwnoQSUyIRfEEK6LLGwmWy2gaZNXpb12WsgNk,1975
 aiagents4pharma/talk2scholars/configs/agents/talk2scholars/zotero_agent/__init__.py,sha256=fqQQ-GlRcbzru2KmEk3oMma0R6_SzGM8dOXzYeU4oVA,46
@@ -152,9 +154,10 @@ aiagents4pharma/talk2scholars/configs/agents/talk2scholars/zotero_agent/default.
 aiagents4pharma/talk2scholars/configs/app/__init__.py,sha256=JoSZV6N669kGMv5zLDszwf0ZjcRHx9TJfIqGhIIdPXE,70
 aiagents4pharma/talk2scholars/configs/app/frontend/__init__.py,sha256=fqQQ-GlRcbzru2KmEk3oMma0R6_SzGM8dOXzYeU4oVA,46
 aiagents4pharma/talk2scholars/configs/app/frontend/default.yaml,sha256=wsELBdRLv6UqZ9QZfwpS7K4xfMj5s-a99-aXqIs6WEI,868
-aiagents4pharma/talk2scholars/configs/tools/__init__.py,sha256=z5PzYS1WyoC4PiWVIIOGICldps2yMh2E_2HFZA9ltpQ,177
+aiagents4pharma/talk2scholars/configs/tools/__init__.py,sha256=NDXBZVtEYpOVL0EMm69ffoFAZw9G4tQiwsSFdxRrxLQ,211
 aiagents4pharma/talk2scholars/configs/tools/multi_paper_recommendation/__init__.py,sha256=fqQQ-GlRcbzru2KmEk3oMma0R6_SzGM8dOXzYeU4oVA,46
 aiagents4pharma/talk2scholars/configs/tools/multi_paper_recommendation/default.yaml,sha256=iEsEW89MlQwKsAW4ZAxLt4pDBwA1qxImYQ2dfONIf6c,442
+aiagents4pharma/talk2scholars/configs/tools/question_and_answer/__init__.py,sha256=fqQQ-GlRcbzru2KmEk3oMma0R6_SzGM8dOXzYeU4oVA,46
 aiagents4pharma/talk2scholars/configs/tools/retrieve_semantic_scholar_paper_id/__init__.py,sha256=fqQQ-GlRcbzru2KmEk3oMma0R6_SzGM8dOXzYeU4oVA,46
 aiagents4pharma/talk2scholars/configs/tools/retrieve_semantic_scholar_paper_id/default.yaml,sha256=HG-N8yRjlX9zFwbIBvaDI9ndKjfL-gqPTCCPMLgdUpw,271
 aiagents4pharma/talk2scholars/configs/tools/search/__init__.py,sha256=fqQQ-GlRcbzru2KmEk3oMma0R6_SzGM8dOXzYeU4oVA,46
@@ -164,30 +167,34 @@ aiagents4pharma/talk2scholars/configs/tools/single_paper_recommendation/default.
 aiagents4pharma/talk2scholars/configs/tools/zotero_read/__init__.py,sha256=fqQQ-GlRcbzru2KmEk3oMma0R6_SzGM8dOXzYeU4oVA,46
 aiagents4pharma/talk2scholars/configs/tools/zotero_read/default.yaml,sha256=iILspz9EvN8jpVHzMsW3L9BDEST5eqOUO7TnhxwXBrI,468
 aiagents4pharma/talk2scholars/state/__init__.py,sha256=S6SxlszIMZSIMJehjevPF9sKyR-PAwWb5TEdo6xWXE8,103
-aiagents4pharma/talk2scholars/state/state_talk2scholars.py,sha256=DoCtKP2qd69mXPwfOb-aYw9Hq2fYmx6b76S-HlsVSNo,2382
+aiagents4pharma/talk2scholars/state/state_talk2scholars.py,sha256=E0epqlBurzNcMzN4WV6nh--YkVAesbxQTuBBlJhESVA,2436
 aiagents4pharma/talk2scholars/tests/__init__.py,sha256=U3PsTiUZaUBD1IZanFGkDIOdFieDVJtGKQ5-woYUo8c,45
 aiagents4pharma/talk2scholars/tests/test_call_s2.py,sha256=ZL5HmnYNVyaBJgPGQi9JnbD1d1rtWnWusVxVRVW3aHc,3375
 aiagents4pharma/talk2scholars/tests/test_call_zotero.py,sha256=N4g6Pt2vuaxIhHQbIqlMaDUF4O7vIvRqa7pPIkpL8FI,3314
 aiagents4pharma/talk2scholars/tests/test_llm_main_integration.py,sha256=SAMG-Kb2S9sei8Us5vUWCUJikTKXPZVKQ6aJJPEhJsc,1880
 aiagents4pharma/talk2scholars/tests/test_main_agent.py,sha256=8FKujCVhkurCe5IE6OGPTmz1p4eH1CDi467vM6VtM5A,4318
+aiagents4pharma/talk2scholars/tests/test_pdf_agent.py,sha256=TN4Sq5-SCxv-9VfFyq7sOlBlxbekmnWuB7-qh4MrhkA,4656
+aiagents4pharma/talk2scholars/tests/test_question_and_answer_tool.py,sha256=TpCDiGfsC2y6bOkm0ZTXjT1Vp8D-Po25wiEH5aDT_DA,6491
 aiagents4pharma/talk2scholars/tests/test_routing_logic.py,sha256=AZrvaEBDk51KL6edrZY3GpQ_N6VbrlADqXFeg_jxDoQ,2284
 aiagents4pharma/talk2scholars/tests/test_s2_agent.py,sha256=BhW1wGc-wUPS4fwNBQRtBXJaJ_i7L6t_G9Bq57fK7rI,7784
 aiagents4pharma/talk2scholars/tests/test_s2_tools.py,sha256=QEwraJk9_Kp6ZSGYyYDXWH62wIjSwi1Pptwwbx1fuG0,13176
 aiagents4pharma/talk2scholars/tests/test_state.py,sha256=_iHXvoZnU_eruf8l1sQKBSCIVnxNkH_9VzkVtZZA6bY,384
 aiagents4pharma/talk2scholars/tests/test_zotero_agent.py,sha256=3TKz6yjNfYulaQv-MBv1zXCmR9xh9g3ju4Ge5HDdt1o,6136
 aiagents4pharma/talk2scholars/tests/test_zotero_tool.py,sha256=LI7KBTxPga7E-841pugjpNqtWgoIz0mDIJEZzdIL9eI,5759
-aiagents4pharma/talk2scholars/tools/__init__.py,sha256=-9iXVIGzFLak6a14Ib8yDg1bfiHgJz2nAhwWEk1jhOk,89
+aiagents4pharma/talk2scholars/tools/__init__.py,sha256=UtGutYNNaRcr2nOmT_XqbTiaJpgVYKo3KVGVPFVrX2Y,107
+aiagents4pharma/talk2scholars/tools/pdf/__init__.py,sha256=WOm-o-fFzyjFZBaHg658Gjzdiu1Kt-h9xvzvw0hR7aE,103
+aiagents4pharma/talk2scholars/tools/pdf/question_and_answer.py,sha256=22JvT7F0rY11TF40pBfe9Cn2Y-6Tx73NfWDt4NJv700,6639
 aiagents4pharma/talk2scholars/tools/s2/__init__.py,sha256=wytqCmGm8Fbl8y5qLdIkxhhG8VHLYMifCGjbH_LK2Fc,258
 aiagents4pharma/talk2scholars/tools/s2/display_results.py,sha256=UR0PtEHGDpOhPH0Di5HT8-Fip2RkEMTJgzROsChb1gc,2959
 aiagents4pharma/talk2scholars/tools/s2/multi_paper_rec.py,sha256=QM30Oq3518cuEWwpfA5R7NzNmNklYUkt9Y1D5jdjmG4,5430
-aiagents4pharma/talk2scholars/tools/s2/query_results.py,sha256=EUfzRh5Qc_tMl5fDIFb9PIsQkkrU4Xb5MR0sud_X5-c,2017
+aiagents4pharma/talk2scholars/tools/s2/query_results.py,sha256=S4yBNtg1loDu4ckLPrW4H8GAswriPaRU4U08cOuw2HE,2028
 aiagents4pharma/talk2scholars/tools/s2/retrieve_semantic_scholar_paper_id.py,sha256=Lg1L4HQCN2LaQEyWtLD73O67PMoXkPHi-Y8rCzHS0A4,2499
 aiagents4pharma/talk2scholars/tools/s2/search.py,sha256=i5KMFJWK31CjYtVT1McJpLzgcwvyTHZe2aHZlscfK3Q,4667
 aiagents4pharma/talk2scholars/tools/s2/single_paper_rec.py,sha256=7PoZfcstxDThWX6NYOgxN_9M_nwgMPAALch8OmjraVY,5568
 aiagents4pharma/talk2scholars/tools/zotero/__init__.py,sha256=1UW4r5ECvAwYpo1Fjf7lQPO--M8I85baYCHocFOAq4M,53
 aiagents4pharma/talk2scholars/tools/zotero/zotero_read.py,sha256=NJ65fAJ4u2Zq15uvEajVOhI4QnNvyqA6FHPaEDqvMw0,4321
-aiagents4pharma-1.27.1.dist-info/LICENSE,sha256=IcIbyB1Hyk5ZDah03VNQvJkbNk2hkBCDqQ8qtnCvB4Q,1077
-aiagents4pharma-1.27.1.dist-info/METADATA,sha256=kTCOwxjPuhVeXqavYrNYu08GDiJsykcb_8Vz7_xK_l0,13252
-aiagents4pharma-1.27.1.dist-info/WHEEL,sha256=jB7zZ3N9hIM9adW7qlTAyycLYW9npaWKLRzaoVcLKcM,91
-aiagents4pharma-1.27.1.dist-info/top_level.txt,sha256=-AH8rMmrSnJtq7HaAObS78UU-cTCwvX660dSxeM7a0A,16
-aiagents4pharma-1.27.1.dist-info/RECORD,,
+aiagents4pharma-1.28.0.dist-info/LICENSE,sha256=IcIbyB1Hyk5ZDah03VNQvJkbNk2hkBCDqQ8qtnCvB4Q,1077
+aiagents4pharma-1.28.0.dist-info/METADATA,sha256=jG17DZJJ8a8hrmwOECb_eDodny-pGm93rTuu9nYnacc,13252
+aiagents4pharma-1.28.0.dist-info/WHEEL,sha256=jB7zZ3N9hIM9adW7qlTAyycLYW9npaWKLRzaoVcLKcM,91
+aiagents4pharma-1.28.0.dist-info/top_level.txt,sha256=-AH8rMmrSnJtq7HaAObS78UU-cTCwvX660dSxeM7a0A,16
+aiagents4pharma-1.28.0.dist-info/RECORD,,