mirror of
https://github.com/Frooodle/Stirling-PDF.git
synced 2026-05-10 23:10:08 +02:00
3e94157137
# Description of Changes Adds the ability for the Edit agent to request the content of the document before it decides which parameters it needs. This makes it able to process requests like `Split the document after the page containing the "My Section" section`, allowing for document context-based requests for all[^1] tools. I had to make a few changes elsewhere to make this work, including: - Moving the requesting of content out of the Question Agent and into a common location - Added specific API docs for the Split param because the generic ones were not specific enough for the AI to be able to reliably perform the correct operation - Fixed an issue in the tool models generator which caused the Redact params to only be half-generated (causing Pydantic to crash when the AI tried to run Redact) - Added missing logging to a bunch of tools and hooked it up properly so it'll print to stderr - Made the limits for the max pages/chars to extract from PDFs configurable via env var [^1]: Many of the tools can't actually do anything useful with the context at this stage, but will just need the tool API to be extended with new features like page-specific operations to be automatically able to do smart operations without needing to change the Edit agent itself.
87 lines
2.6 KiB
Python
87 lines
2.6 KiB
Python
from __future__ import annotations
|
|
|
|
import pytest
|
|
|
|
from stirling.agents import PdfQuestionAgent
|
|
from stirling.contracts import (
|
|
ExtractedFileText,
|
|
NeedContentResponse,
|
|
PdfQuestionAnswerResponse,
|
|
PdfQuestionNotFoundResponse,
|
|
PdfQuestionRequest,
|
|
PdfTextSelection,
|
|
)
|
|
from stirling.services.runtime import AppRuntime
|
|
|
|
|
|
class StubPdfQuestionAgent(PdfQuestionAgent):
|
|
def __init__(self, runtime: AppRuntime, response: PdfQuestionAnswerResponse | PdfQuestionNotFoundResponse) -> None:
|
|
super().__init__(runtime)
|
|
self.response = response
|
|
|
|
async def _run_answer_agent(
|
|
self,
|
|
request: PdfQuestionRequest,
|
|
) -> PdfQuestionAnswerResponse | PdfQuestionNotFoundResponse:
|
|
return self.response
|
|
|
|
|
|
def invoice_page() -> ExtractedFileText:
|
|
return ExtractedFileText(
|
|
file_name="invoice.pdf",
|
|
pages=[PdfTextSelection(page_number=1, text="Invoice total: 120.00")],
|
|
)
|
|
|
|
|
|
@pytest.mark.anyio
|
|
async def test_pdf_question_agent_requires_extracted_text(runtime: AppRuntime) -> None:
|
|
agent = PdfQuestionAgent(runtime)
|
|
|
|
response = await agent.handle(
|
|
PdfQuestionRequest(question="What is the total?", page_text=[], file_names=["test.pdf"])
|
|
)
|
|
|
|
assert isinstance(response, NeedContentResponse)
|
|
|
|
|
|
@pytest.mark.anyio
|
|
async def test_pdf_question_agent_returns_grounded_answer(runtime: AppRuntime) -> None:
|
|
agent = StubPdfQuestionAgent(
|
|
runtime,
|
|
PdfQuestionAnswerResponse(
|
|
answer="The invoice total is 120.00.",
|
|
evidence=[invoice_page()],
|
|
),
|
|
)
|
|
|
|
response = await agent.handle(
|
|
PdfQuestionRequest(
|
|
question="What is the total?",
|
|
page_text=[invoice_page()],
|
|
file_names=["invoice.pdf"],
|
|
)
|
|
)
|
|
|
|
assert isinstance(response, PdfQuestionAnswerResponse)
|
|
assert response.answer == "The invoice total is 120.00."
|
|
|
|
|
|
@pytest.mark.anyio
|
|
async def test_pdf_question_agent_returns_not_found_when_text_is_insufficient(runtime: AppRuntime) -> None:
|
|
agent = StubPdfQuestionAgent(runtime, PdfQuestionNotFoundResponse(reason="The answer is not present in the text."))
|
|
|
|
response = await agent.handle(
|
|
PdfQuestionRequest(
|
|
question="What is the total?",
|
|
page_text=[
|
|
ExtractedFileText(
|
|
file_name="invoice.pdf",
|
|
pages=[PdfTextSelection(page_number=1, text="This page contains only a shipping address.")],
|
|
)
|
|
],
|
|
file_names=["invoice.pdf"],
|
|
)
|
|
)
|
|
|
|
assert isinstance(response, PdfQuestionNotFoundResponse)
|