def test_pdf_open(tmp_path):
# fail on a buffer full of null bytes
with pytest.raises(PdfParser.PdfFormatError):
PdfParser.PdfParser(buf=bytearray(65536))
# make an empty PDF object
with PdfParser.PdfParser() as empty_pdf:
assert len(empty_pdf.pages) == 0
assert len(empty_pdf.info) == 0
assert not empty_pdf.should_close_buf
assert not empty_pdf.should_close_file
# make a PDF file
pdf_filename = helper_save_as_pdf(tmp_path, "RGB")
# open the PDF file
with PdfParser.PdfParser(filename=pdf_filename) as hopper_pdf:
assert len(hopper_pdf.pages) == 1
assert hopper_pdf.should_close_buf
assert hopper_pdf.should_close_file
# read a PDF file from a buffer with a non-zero offset
with open(pdf_filename, "rb") as f:
content = b"xyzzy" + f.read()
with PdfParser.PdfParser(buf=content, start_offset=5) as hopper_pdf:
assert len(hopper_pdf.pages) == 1
assert not hopper_pdf.should_close_buf
assert not hopper_pdf.should_close_file
# read a PDF file from an already open file
with open(pdf_filename, "rb") as f:
with PdfParser.PdfParser(f=f) as hopper_pdf:
assert len(hopper_pdf.pages) == 1
assert hopper_pdf.should_close_buf
assert not hopper_pdf.should_close_file
def test_pdf_append(self):
# make a PDF file
pdf_filename = self.helper_save_as_pdf("RGB", producer="PdfParser")
# open it, check pages and info
with PdfParser.PdfParser(pdf_filename, mode="r+b") as pdf:
self.assertEqual(len(pdf.pages), 1)
self.assertEqual(len(pdf.info), 4)
self.assertEqual(
pdf.info.Title,
os.path.splitext(os.path.basename(pdf_filename))[0])
self.assertEqual(pdf.info.Producer, "PdfParser")
self.assertIn(b"CreationDate", pdf.info)
self.assertIn(b"ModDate", pdf.info)
self.check_pdf_pages_consistency(pdf)
# append some info
pdf.info.Title = "abc"
pdf.info.Author = "def"
pdf.info.Subject = u"ghi\uABCD"
pdf.info.Keywords = "qw)e\\r(ty"
pdf.info.Creator = "hopper()"
pdf.start_writing()
pdf.write_xref_and_trailer()
# open it again, check pages and info again
with PdfParser.PdfParser(pdf_filename) as pdf:
self.assertEqual(len(pdf.pages), 1)
self.assertEqual(len(pdf.info), 8)
self.assertEqual(pdf.info.Title, "abc")
self.assertIn(b"CreationDate", pdf.info)
self.assertIn(b"ModDate", pdf.info)
self.check_pdf_pages_consistency(pdf)
# append two images
mode_CMYK = hopper("CMYK")
mode_P = hopper("P")
mode_CMYK.save(pdf_filename,
append=True,
save_all=True,
append_images=[mode_P])
# open the PDF again, check pages and info again
with PdfParser.PdfParser(pdf_filename) as pdf:
self.assertEqual(len(pdf.pages), 3)
self.assertEqual(len(pdf.info), 8)
self.assertEqual(PdfParser.decode_text(pdf.info[b"Title"]), "abc")
self.assertEqual(pdf.info.Title, "abc")
self.assertEqual(pdf.info.Producer, "PdfParser")
self.assertEqual(pdf.info.Keywords, "qw)e\\r(ty")
self.assertEqual(pdf.info.Subject, u"ghi\uABCD")
self.assertIn(b"CreationDate", pdf.info)
self.assertIn(b"ModDate", pdf.info)
self.check_pdf_pages_consistency(pdf)
def test_pdf_append(tmp_path):
# make a PDF file
pdf_filename = helper_save_as_pdf(tmp_path, "RGB", producer="PdfParser")
# open it, check pages and info
with PdfParser.PdfParser(pdf_filename, mode="r+b") as pdf:
assert len(pdf.pages) == 1
assert len(pdf.info) == 4
assert pdf.info.Title == os.path.splitext(
os.path.basename(pdf_filename))[0]
assert pdf.info.Producer == "PdfParser"
assert b"CreationDate" in pdf.info
assert b"ModDate" in pdf.info
check_pdf_pages_consistency(pdf)
# append some info
pdf.info.Title = "abc"
pdf.info.Author = "def"
pdf.info.Subject = "ghi\uABCD"
pdf.info.Keywords = "qw)e\\r(ty"
pdf.info.Creator = "hopper()"
pdf.start_writing()
pdf.write_xref_and_trailer()
# open it again, check pages and info again
with PdfParser.PdfParser(pdf_filename) as pdf:
assert len(pdf.pages) == 1
assert len(pdf.info) == 8
assert pdf.info.Title == "abc"
assert b"CreationDate" in pdf.info
assert b"ModDate" in pdf.info
check_pdf_pages_consistency(pdf)
# append two images
mode_cmyk = hopper("CMYK")
mode_p = hopper("P")
mode_cmyk.save(pdf_filename,
append=True,
save_all=True,
append_images=[mode_p])
# open the PDF again, check pages and info again
with PdfParser.PdfParser(pdf_filename) as pdf:
assert len(pdf.pages) == 3
assert len(pdf.info) == 8
assert PdfParser.decode_text(pdf.info[b"Title"]) == "abc"
assert pdf.info.Title == "abc"
assert pdf.info.Producer == "PdfParser"
assert pdf.info.Keywords == "qw)e\\r(ty"
assert pdf.info.Subject == "ghi\uABCD"
assert b"CreationDate" in pdf.info
assert b"ModDate" in pdf.info
check_pdf_pages_consistency(pdf)
def test_pdf_info(self):
# make a PDF file
pdf_filename = self.helper_save_as_pdf(
"RGB",
title="title",
author="author",
subject="subject",
keywords="keywords",
creator="creator",
producer="producer",
creationDate=time.strptime("2000", "%Y"),
modDate=time.strptime("2001", "%Y"))
# open it, check pages and info
with PdfParser.PdfParser(pdf_filename) as pdf:
self.assertEqual(len(pdf.info), 8)
self.assertEqual(pdf.info.Title, "title")
self.assertEqual(pdf.info.Author, "author")
self.assertEqual(pdf.info.Subject, "subject")
self.assertEqual(pdf.info.Keywords, "keywords")
self.assertEqual(pdf.info.Creator, "creator")
self.assertEqual(pdf.info.Producer, "producer")
self.assertEqual(pdf.info.CreationDate,
time.strptime("2000", "%Y"))
self.assertEqual(pdf.info.ModDate, time.strptime("2001", "%Y"))
self.check_pdf_pages_consistency(pdf)
def helper_save_as_pdf(self, mode, **kwargs):
# Arrange
im = hopper(mode)
outfile = self.tempfile("temp_" + mode + ".pdf")
# Act
im.save(outfile, **kwargs)
# Assert
self.assertTrue(os.path.isfile(outfile))
self.assertGreater(os.path.getsize(outfile), 0)
with PdfParser.PdfParser(outfile) as pdf:
if kwargs.get("append_images", False) or \
kwargs.get("append", False):
self.assertGreater(len(pdf.pages), 1)
else:
self.assertGreater(len(pdf.pages), 0)
with open(outfile, 'rb') as fp:
contents = fp.read()
size = tuple(
int(d) for d in contents.split(b'/MediaBox [ 0 0 ')[1].split(b']')
[0].split())
self.assertEqual(im.size, size)
return outfile
def test_redos(newline):
malicious = b" trailer<<>>" + newline * 3456
# This particular exception isn't relevant here.
# The important thing is it doesn't timeout, cause a ReDoS (CVE-2021-25292).
with pytest.raises(PdfParser.PdfFormatError):
PdfParser.PdfParser(buf=malicious)
def test_pdf_info(tmp_path):
# make a PDF file
pdf_filename = helper_save_as_pdf(
tmp_path,
"RGB",
title="title",
author="author",
subject="subject",
keywords="keywords",
creator="creator",
producer="producer",
creationDate=time.strptime("2000", "%Y"),
modDate=time.strptime("2001", "%Y"),
)
# open it, check pages and info
with PdfParser.PdfParser(pdf_filename) as pdf:
assert len(pdf.info) == 8
assert pdf.info.Title == "title"
assert pdf.info.Author == "author"
assert pdf.info.Subject == "subject"
assert pdf.info.Keywords == "keywords"
assert pdf.info.Creator == "creator"
assert pdf.info.Producer == "producer"
assert pdf.info.CreationDate == time.strptime("2000", "%Y")
assert pdf.info.ModDate == time.strptime("2001", "%Y")
check_pdf_pages_consistency(pdf)
def helper_save_as_pdf(self, mode, **kwargs):
# Arrange
im = hopper(mode)
outfile = self.tempfile("temp_" + mode + ".pdf")
# Act
im.save(outfile, **kwargs)
# Assert
assert os.path.isfile(outfile)
assert os.path.getsize(outfile) > 0
with PdfParser.PdfParser(outfile) as pdf:
if kwargs.get("append_images", False) or kwargs.get(
"append", False):
assert len(pdf.pages) > 1
else:
assert len(pdf.pages) > 0
with open(outfile, "rb") as fp:
contents = fp.read()
size = tuple(
int(d) for d in contents.split(b"/MediaBox [ 0 0 ")[1].split(b"]")
[0].split())
assert im.size == size
return outfile
def test_pdf_append(self):
# make a PDF file
pdf_filename = self.helper_save_as_pdf("RGB", producer="PdfParser")
# open it, check pages and info
with PdfParser.PdfParser(pdf_filename, mode="r+b") as pdf:
self.assertEqual(len(pdf.pages), 1)
self.assertEqual(len(pdf.info), 4)
self.assertEqual(pdf.info.Title, os.path.splitext(
os.path.basename(pdf_filename)
)[0])
self.assertEqual(pdf.info.Producer, "PdfParser")
self.assertIn(b"CreationDate", pdf.info)
self.assertIn(b"ModDate", pdf.info)
self.check_pdf_pages_consistency(pdf)
# append some info
pdf.info.Title = "abc"
pdf.info.Author = "def"
pdf.info.Subject = u"ghi\uABCD"
pdf.info.Keywords = "qw)e\\r(ty"
pdf.info.Creator = "hopper()"
pdf.start_writing()
pdf.write_xref_and_trailer()
# open it again, check pages and info again
with PdfParser.PdfParser(pdf_filename) as pdf:
self.assertEqual(len(pdf.pages), 1)
self.assertEqual(len(pdf.info), 8)
self.assertEqual(pdf.info.Title, "abc")
self.assertIn(b"CreationDate", pdf.info)
self.assertIn(b"ModDate", pdf.info)
self.check_pdf_pages_consistency(pdf)
# append two images
mode_CMYK = hopper("CMYK")
mode_P = hopper("P")
mode_CMYK.save(pdf_filename,
append=True, save_all=True, append_images=[mode_P])
# open the PDF again, check pages and info again
with PdfParser.PdfParser(pdf_filename) as pdf:
self.assertEqual(len(pdf.pages), 3)
self.assertEqual(len(pdf.info), 8)
self.assertEqual(PdfParser.decode_text(pdf.info[b"Title"]), "abc")
self.assertEqual(pdf.info.Title, "abc")
self.assertEqual(pdf.info.Producer, "PdfParser")
self.assertEqual(pdf.info.Keywords, "qw)e\\r(ty")
self.assertEqual(pdf.info.Subject, u"ghi\uABCD")
self.assertIn(b"CreationDate", pdf.info)
self.assertIn(b"ModDate", pdf.info)
self.check_pdf_pages_consistency(pdf)
def test_pdf_open(self):
# fail on a buffer full of null bytes
self.assertRaises(PdfParser.PdfFormatError,
PdfParser.PdfParser,
buf=bytearray(65536))
# make an empty PDF object
with PdfParser.PdfParser() as empty_pdf:
self.assertEqual(len(empty_pdf.pages), 0)
self.assertEqual(len(empty_pdf.info), 0)
self.assertFalse(empty_pdf.should_close_buf)
self.assertFalse(empty_pdf.should_close_file)
# make a PDF file
pdf_filename = self.helper_save_as_pdf("RGB")
# open the PDF file
with PdfParser.PdfParser(filename=pdf_filename) as hopper_pdf:
self.assertEqual(len(hopper_pdf.pages), 1)
self.assertTrue(hopper_pdf.should_close_buf)
self.assertTrue(hopper_pdf.should_close_file)
# read a PDF file from a buffer with a non-zero offset
with open(pdf_filename, "rb") as f:
content = b"xyzzy" + f.read()
with PdfParser.PdfParser(buf=content, start_offset=5) as hopper_pdf:
self.assertEqual(len(hopper_pdf.pages), 1)
self.assertFalse(hopper_pdf.should_close_buf)
self.assertFalse(hopper_pdf.should_close_file)
# read a PDF file from an already open file
with open(pdf_filename, "rb") as f:
with PdfParser.PdfParser(f=f) as hopper_pdf:
self.assertEqual(len(hopper_pdf.pages), 1)
self.assertTrue(hopper_pdf.should_close_buf)
self.assertFalse(hopper_pdf.should_close_file)
def helper_save_as_pdf(self, mode, **kwargs):
# Arrange
im = hopper(mode)
outfile = self.tempfile("temp_" + mode + ".pdf")
# Act
im.save(outfile, **kwargs)
# Assert
self.assertTrue(os.path.isfile(outfile))
self.assertGreater(os.path.getsize(outfile), 0)
with PdfParser.PdfParser(outfile) as pdf:
if kwargs.get("append_images", False) or \
kwargs.get("append", False):
self.assertGreater(len(pdf.pages), 1)
else:
self.assertGreater(len(pdf.pages), 0)
return outfile
