def test_text_filter(resources, outdir):
input_pdf = resources / 'veraPDF test suite 6-2-10-t02-pass-a.pdf'
# Ensure the test PDF has detect we can find
proc = run(['pdftotext', str(input_pdf), '-'],
check=True,
stdout=PIPE,
encoding='utf-8')
assert proc.stdout.strip() != '', "Need input test file that contains text"
pdf = Pdf.open(input_pdf)
page = pdf.pages[0]
keep = []
for operands, command in parse_content_stream(page):
if command == Operator('Tj'):
print("skipping Tj")
continue
keep.append((operands, command))
new_stream = Stream(pdf, keep)
print(new_stream.read_bytes()) # pylint: disable=no-member
page['/Contents'] = new_stream
page['/Rotate'] = 90
pdf.save(outdir / 'notext.pdf', True)
proc = run(
['pdftotext', str(outdir / 'notext.pdf'), '-'],
check=True,
stdout=PIPE,
encoding='utf-8',
)
assert proc.stdout.strip() == '', "Expected text to be removed"
def test_stream_dict_oneshot():
pdf = pikepdf.new()
stream1 = Stream(pdf, b'12345', One=1, Two=2)
stream2 = Stream(pdf, b'67890', {'/Three': 3, '/Four': 4})
stream3 = pdf.make_stream(b'abcdef', One=1, Two=2)
assert stream1.One == 1
assert stream1.read_bytes() == b'12345'
assert stream2.Three == 3
assert stream3.One == 1
def test_data_decoding_errors(filter_: str, data: bytes, msg: str):
p = Pdf.new()
st = Stream(p, data, Filter=Name(filter_))
with pytest.raises(DataDecodingError, match=msg):
st.read_bytes()
