def test_extract_extracts_right_amount_of_files():
with TemporaryDirectory() as output_dir:
files: List[Path] = get_filenames(CONFIG.test_files_dir / 'test.pdf')
extractor: ITextExtractor = TesseractExtractor(dpi=1, fmt='png')
extractor.batch_extract(files, output_dir)
assert len(sorted(Path(output_dir).glob('*.txt'))) == 8
assert (Path(output_dir) / 'extract.log').exists()
def test_extract_generates_expected_output():
with TemporaryDirectory() as output_dir:
files: List[Path] = get_filenames(CONFIG.test_files_dir / 'pdf')
extractor: ITextExtractor = PDFPlumberExtractor()
extractor.batch_extract(files, output_dir)
assert len(sorted(Path(output_dir).glob('*.txt'))) == 5
assert filecmp.dircmp(output_dir, CONFIG.test_files_dir /
'expected/pdfplumber').diff_files == []
assert (Path(output_dir) / 'extract.log').exists()
def test_extract_returns_correct_number_of_pages(first_page, last_page,
expected):
with TemporaryDirectory() as output_dir:
files: List[Path] = get_filenames(CONFIG.test_files_dir / 'pdf')
extractor: ITextExtractor = PDFPlumberExtractor()
extractor.batch_extract(files,
output_dir,
first_page=first_page,
last_page=last_page)
result = len(sorted(Path(output_dir).glob('*.txt')))
assert result == expected
assert (Path(output_dir) / 'extract.log').exists()
def test_extract_generates_expected_output():
with TemporaryDirectory() as output_dir:
files: List[Path] = get_filenames(CONFIG.test_files_dir / 'test.pdf')
extractor: ITextExtractor = PDFMinerExtractor()
extractor.batch_extract(files, output_dir)
num_extracted = len(sorted(Path(output_dir).glob('*.txt')))
assert num_extracted == 8
# FIXME: Improve
for i in range(0, num_extracted):
text1 = open(sorted(Path(output_dir).glob('*.txt'))[i]).read()
text2 = open(
sorted(
Path(CONFIG.test_files_dir /
'expected/pdfminer').iterdir())[i]).read()
m = SequenceMatcher(None, text1, text2)
assert m.quick_ratio() > 0.99
def test_get_filenames_returns_only_files_with_expected_extension(tmp_path):
pdf_file = tmp_path / 'test.pdf'
pdf_file.touch()
txt_file = tmp_path / 'test.txt'
txt_file.touch()
assert pdf_file in get_filenames(tmp_path)
assert txt_file not in get_filenames(tmp_path)
assert txt_file in get_filenames(tmp_path, 'txt')
assert get_filenames(txt_file) == []
assert get_filenames(pdf_file) == get_filenames(tmp_path) == [pdf_file]
def extract(
input_path: Union[str, os.PathLike],
output_folder: Union[str, os.PathLike],
first_page: int = 1,
last_page: Optional[int] = None,
extractor: str = 'PDFBox',
) -> None:
Path(output_folder).mkdir(exist_ok=True, parents=True)
files: List[Path] = get_filenames(input_path)
if last_page is not None:
last_page = int(last_page)
extractor: ITextExtractor = get_extractor(extractor)
extractor.batch_extract(files,
output_folder,
first_page=first_page,
last_page=last_page)