mirror/nanobot
1
0
Fork 0
mirror of https://github.com/HKUDS/nanobot.git synced 2026-09-04 02:01:48 +03:00
Code Issues Packages Projects Releases Wiki Activity

feat: add demand-driven document retrieval (#5525)

Browse Source
This commit is contained in:
chengyongru
2026-08-25 15:34:48 +08:00
committed by GitHub
parent 5cf78540a4
commit 4b2965c8f3
9 changed files with 1097 additions and 295 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tests/tools/test_read_enhancements.py
+105 -10
View File
@@ -8,6 +8,19 @@ import pytest
from nanobot.agent.tools import file_state
from nanobot.agent.tools.filesystem import ReadFileTool, WriteFileTool
from nanobot.utils.document import (
DocumentExtractionError,
DocumentLineSource,
LocatedDocumentLine,
)
def _document_source(text: str) -> DocumentLineSource:
lines = (
LocatedDocumentLine(line, line_no, "")
for line_no, line in enumerate(text.splitlines(), 1)
)
return DocumentLineSource(lines)
@pytest.fixture(autouse=True)
@@ -220,6 +233,12 @@ class TestReadPdf:
assert "Invalid page range" in result
out_of_bounds = await tool.execute(path=str(pdf_path), pages="99")
assert out_of_bounds == (
"Error: Invalid page range '99': document has 1 page; "
"use a page number or range within 1-1."
)
@pytest.mark.asyncio
async def test_pdf_file_not_found_error(self, tool, tmp_path):
result = await tool.execute(path=str(tmp_path / "nope.pdf"))
@@ -345,7 +364,10 @@ class TestReadOfficeDocuments:
@pytest.mark.asyncio
async def test_docx_returns_extracted_text(self, tool, tmp_path):
with patch("nanobot.utils.document.extract_text", return_value="Title\n\nParagraph 1"):
with patch(
"nanobot.utils.document.open_document_line_source",
return_value=_document_source("Title\n\nParagraph 1"),
):
f = tmp_path / "test.docx"
f.write_bytes(b"PK")
result = await tool.execute(path=str(f))
@@ -355,16 +377,69 @@ class TestReadOfficeDocuments:
@pytest.mark.asyncio
async def test_xlsx_returns_extracted_text(self, tool, tmp_path):
with patch("nanobot.utils.document.extract_text", return_value="--- Sheet: Sheet1 ---\nName\tAge\nAlice\t30"):
with patch(
"nanobot.utils.document.open_document_line_source",
return_value=_document_source("--- Sheet: Sheet1 ---\nName\tAge\nAlice\t30"),
):
f = tmp_path / "test.xlsx"
f.write_bytes(b"PK")
result = await tool.execute(path=str(f))
assert "Sheet1" in result
assert "Alice" in result
@pytest.mark.asyncio
async def test_office_documents_support_extracted_line_ranges(self, tool, tmp_path):
extracted = "--- Sheet: Sheet1 ---\nName\tAge\nAlice\t30\nBob\t25"
with patch(
"nanobot.utils.document.open_document_line_source",
return_value=_document_source(extracted),
):
f = tmp_path / "test.xlsx"
f.write_bytes(b"PK")
result = await tool.execute(path=str(f), offset=3, limit=1)
assert "3| Alice\t30" in result
assert "Name\tAge" not in result
assert "Use offset=4 to continue" in result
@pytest.mark.asyncio
async def test_office_range_reaches_beyond_attachment_preview_limit(
self,
tool,
tmp_path,
monkeypatch,
):
from openpyxl import Workbook
from nanobot.utils import document as document_utils
workbook_path = tmp_path / "long.xlsx"
workbook = Workbook()
sheet = workbook.active
for row in range(1, 20):
sheet.append([f"ordinary-row-{row}"])
sheet.append(["late-content"])
workbook.save(workbook_path)
workbook.close()
monkeypatch.setattr(document_utils, "_MAX_TEXT_LENGTH", 50)
preview = document_utils.extract_text(workbook_path)
assert preview is not None
assert "late-content" not in preview
result = await tool.execute(path=str(workbook_path), offset=21, limit=1)
assert "21| late-content" in result
assert "beyond end" not in result
@pytest.mark.asyncio
async def test_pptx_returns_extracted_text(self, tool, tmp_path):
with patch("nanobot.utils.document.extract_text", return_value="--- Slide 1 ---\nWelcome\n--- Slide 2 ---\nContent"):
with patch(
"nanobot.utils.document.open_document_line_source",
return_value=_document_source(
"--- Slide 1 ---\nWelcome\n--- Slide 2 ---\nContent"
),
):
f = tmp_path / "test.pptx"
f.write_bytes(b"PK")
result = await tool.execute(path=str(f))
@@ -373,7 +448,10 @@ class TestReadOfficeDocuments:
@pytest.mark.asyncio
async def test_docx_missing_library(self, tool, tmp_path):
with patch("nanobot.utils.document.extract_text", return_value="[error: python-docx not installed]"):
with patch(
"nanobot.utils.document.open_document_line_source",
side_effect=DocumentExtractionError("python-docx not installed"),
):
f = tmp_path / "test.docx"
f.write_bytes(b"PK")
result = await tool.execute(path=str(f))
@@ -382,7 +460,10 @@ class TestReadOfficeDocuments:
@pytest.mark.asyncio
async def test_docx_corrupt_file(self, tool, tmp_path):
with patch("nanobot.utils.document.extract_text", return_value="[error: failed to extract DOCX: bad zip]"):
with patch(
"nanobot.utils.document.open_document_line_source",
side_effect=DocumentExtractionError("failed to extract DOCX: bad zip"),
):
f = tmp_path / "test.docx"
f.write_bytes(b"not-a-zip")
result = await tool.execute(path=str(f))
@@ -391,7 +472,7 @@ class TestReadOfficeDocuments:
@pytest.mark.asyncio
async def test_unsupported_extension(self, tool, tmp_path):
with patch("nanobot.utils.document.extract_text", return_value=None):
with patch("nanobot.utils.document.open_document_line_source", return_value=None):
f = tmp_path / "test.docx"
f.write_bytes(b"PK")
result = await tool.execute(path=str(f))
@@ -400,7 +481,10 @@ class TestReadOfficeDocuments:
@pytest.mark.asyncio
async def test_empty_document_returns_descriptive_message(self, tool, tmp_path):
with patch("nanobot.utils.document.extract_text", return_value=""):
with patch(
"nanobot.utils.document.open_document_line_source",
return_value=_document_source(""),
):
f = tmp_path / "empty.docx"
f.write_bytes(b"PK")
result = await tool.execute(path=str(f))
@@ -415,7 +499,10 @@ class TestOfficeDocTruncation:
@pytest.mark.asyncio
async def test_large_document_truncated(self, tool, tmp_path):
with patch("nanobot.utils.document.extract_text", return_value="x" * 200_000):
with patch(
"nanobot.utils.document.open_document_line_source",
return_value=_document_source("x" * 200_000),
):
f = tmp_path / "large.docx"
f.write_bytes(b"PK")
result = await tool.execute(path=str(f))
@@ -424,7 +511,10 @@ class TestOfficeDocTruncation:
@pytest.mark.asyncio
async def test_small_document_not_truncated(self, tool, tmp_path):
with patch("nanobot.utils.document.extract_text", return_value="Hello world"):
with patch(
"nanobot.utils.document.open_document_line_source",
return_value=_document_source("Hello world"),
):
f = tmp_path / "small.docx"
f.write_bytes(b"PK")
result = await tool.execute(path=str(f))
@@ -433,7 +523,12 @@ class TestOfficeDocTruncation:
@pytest.mark.asyncio
async def test_error_response_not_truncated(self, tool, tmp_path):
with patch("nanobot.utils.document.extract_text", return_value="[error: failed to extract DOCX: something went wrong]"):
with patch(
"nanobot.utils.document.open_document_line_source",
side_effect=DocumentExtractionError(
"failed to extract DOCX: something went wrong"
),
):
f = tmp_path / "bad.docx"
f.write_bytes(b"PK")
result = await tool.execute(path=str(f))