def test_process_raw(app, mock_grobid_response, pdf_file):
"""Test process with raw content."""
with open(pdf_file, 'rb') as file:
content = file.read()
pdf_extractor = PDFExtractor()
output = pdf_extractor.process_raw(content)
assert 'teiHeader' in output
def test_extract_metadata_api_error_response(app, mock_grobid_error_response,
pdf_file):
"""Test metadata extraction with error on API."""
pdf_extractor = PDFExtractor()
# Test error on api during extraction
with pytest.raises(Exception) as exception:
pdf_extractor.extract_metadata(pdf_file)
assert str(exception.value) == 'Metadata extraction failed'
def test_api_is_alive(app, monkeypatch):
"""Test if api is alive."""
pdf_extractor = PDFExtractor()
# Test API is up
assert pdf_extractor.api_is_alive()
# Test API is down
monkeypatch.setattr(PDFExtractor, 'do_request', lambda *args: ('', 503))
assert not pdf_extractor.api_is_alive()
# Test API raise exception
monkeypatch.setattr(PDFExtractor, 'do_request', lambda *args: Exception)
assert not pdf_extractor.api_is_alive()
def metadata():
"""Extract PDF metadata and return as a json object."""
try:
if 'file' not in request.files:
raise Exception('File not found')
# Get the file posted
pdf_file = request.files['file']
pdf_extractor = PDFExtractor()
# Extract metadata from PDF
return jsonify(pdf_extractor.process_raw(pdf_file.read()))
except Exception as exception:
return jsonify(dict(error=str(exception))), 400
def extract_metadata(pid=None):
"""Publish a deposit or send a message for review."""
deposit = DepositRecord.get_record_by_pid(pid)
if not deposit:
abort(400)
main_file = [
file for file in deposit.files
if file['category'] == 'main' and file.mimetype == 'application/pdf'
]
if not main_file:
abort(500)
# Get file content
with main_file[0].file.storage().open() as pdf_file:
content = pdf_file.read()
# Extract data from pdf
pdf_extractor = PDFExtractor()
pdf_metadata = format_extracted_data(pdf_extractor.process_raw(content))
return make_response(jsonify(pdf_metadata))
def test_extract_metadata(app, mock_grobid_response, pdf_file):
"""Test metadata extraction."""
pdf_extractor = PDFExtractor()
# Test valid extraction
output = pdf_extractor.extract_metadata(pdf_file)
assert output.startswith('<?xml version="1.0" encoding="UTF-8"?>')
# Test non existing file
with pytest.raises(ValueError) as exception:
pdf_extractor.extract_metadata('not_existing_file.pdf')
assert str(exception.value) == 'Input file does not exist'
# Test non valid pdf
input_file = os.path.dirname(os.path.abspath(__file__)) + '/data/test.doc'
with pytest.raises(ValueError) as exception:
pdf_extractor.extract_metadata(input_file)
assert str(exception.value) == 'Input file is not a valid PDF file'
def test_do_request(app, mock_grobid_response, pdf_file):
"""Test request to API."""
pdf_extractor = PDFExtractor()
# Test valid call
assert pdf_extractor.do_request('isalive', 'get') == (b'true', 200)
# Test unexisting endpoint
assert pdf_extractor.do_request('unexisting', 'get')[1] == 404
# Test invalid request type
with pytest.raises(ValueError):
pdf_extractor.do_request('isalive', 'invalid')
# Test post request
assert pdf_extractor.do_request(
'processFulltextDocument',
'post',
files={'input': (pdf_file, open(pdf_file,
'rb'), 'application/pdf')})[1] == 200
def test_process(app, mock_grobid_response, pdf_file):
"""Test process method."""
pdf_extractor = PDFExtractor()
# Test output as XML
output = pdf_extractor.process(pdf_file, dict_output=False)
assert output.startswith('<?xml version="1.0" encoding="UTF-8"?>')
# Test output as dictionary
output = pdf_extractor.process(pdf_file, dict_output=True)
assert 'teiHeader' in output
# Test output as XML in a file
with tempfile.NamedTemporaryFile(mode='w+b', suffix=".pdf") as temp_file:
pdf_extractor.process(pdf_file, output_file=temp_file.name)
with open(temp_file.name, 'r') as output_file:
output = output_file.read()
assert output.startswith('<?xml version="1.0" encoding="UTF-8"?>')
def test_load_config(app, monkeypatch):
"""Test configuration loading."""
monkeypatch.setattr(PDFExtractor, 'api_is_alive', lambda *args: False)
with pytest.raises(ConnectionRefusedError):
PDFExtractor()