mirror of
https://github.com/DS4SD/docling.git
synced 2025-08-01 15:02:21 +00:00
0bb1e203b6
- Introduced new explicit exception types instead of `RuntimeError` - Introduced new `ConversionStatus` value for unsupported formats - Tidied up converter member typing & removed asserts Signed-off-by: Panos Vagenas <35837085+vagenas@users.noreply.github.com>
46 lines
1.4 KiB
Python
46 lines
1.4 KiB
Python
from io import BytesIO
|
|
from pathlib import Path
|
|
|
|
import pytest
|
|
|
|
from docling.datamodel.base_models import ConversionStatus, DocumentStream
|
|
from docling.document_converter import ConversionError, DocumentConverter
|
|
|
|
|
|
def get_pdf_path():
|
|
|
|
pdf_path = Path("./tests/data/2305.03393v1-pg9.pdf")
|
|
return pdf_path
|
|
|
|
|
|
@pytest.fixture
|
|
def converter():
|
|
converter = DocumentConverter()
|
|
|
|
return converter
|
|
|
|
|
|
def test_convert_unsupported_doc_format_wout_exception(converter: DocumentConverter):
|
|
result = converter.convert(
|
|
DocumentStream(name="input.xyz", stream=BytesIO(b"xyz")), raises_on_error=False
|
|
)
|
|
assert result.status == ConversionStatus.UNSUPPORTED
|
|
|
|
|
|
def test_convert_unsupported_doc_format_with_exception(converter: DocumentConverter):
|
|
with pytest.raises(ConversionError):
|
|
converter.convert(
|
|
DocumentStream(name="input.xyz", stream=BytesIO(b"xyz")),
|
|
raises_on_error=True,
|
|
)
|
|
|
|
|
|
def test_convert_too_small_filesize_limit_wout_exception(converter: DocumentConverter):
|
|
result = converter.convert(get_pdf_path(), max_file_size=1, raises_on_error=False)
|
|
assert result.status == ConversionStatus.FAILURE
|
|
|
|
|
|
def test_convert_too_small_filesize_limit_with_exception(converter: DocumentConverter):
|
|
with pytest.raises(ConversionError):
|
|
converter.convert(get_pdf_path(), max_file_size=1, raises_on_error=True)
|