def testIsObjectFree(self):
"""
Tests the ``PdfFileReader.isObjectFree()` method.
"""
# TO-DO Find PDF files that feature free-entry lists. We are checking
# isObjectFree() only against used items.
inputFiles = (
"jpeg.pdf",
"Seige_of_Vicksburg_Sample_OCR.pdf",
"SF424_page2.pdf",
)
for filename in inputFiles:
filepath = join(self.localDataRoot, filename)
r = PdfFileReader(join(TEST_DATA_ROOT, filename))
expItems = self._parseXRefTable(filepath, (0, 1, 3))
actualItems = list()
for ref in r.objects(PdfFileReader.R_XTABLE, True):
actualItems.append(
# This is where isObjectFree() gets invoked
(ref.idnum, ref.generation, r.isObjectFree(ref)))
r.close()
expItems = sorted(expItems)
actualItems = sorted(actualItems)
self.assertListEqual(expItems, actualItems)
def testXRefTableObjects(self):
"""
Ensures that after ``PdfFileReader._parsePdfFile()`` all the indirect
references from the XRef-Table *only* have been loaded as expected.
Objects from the free entries list are included as well in the test.
This case tests the part of ``PdfFileReader.objects()`` responsible for
generating the Cross-Reference Table entries too.
"""
self.maxDiff = None
inputFiles = ("jpeg.pdf", "Seige_of_Vicksburg_Sample_OCR.pdf",
"SF424_page2.pdf")
for filename in inputFiles:
filepath = join(TEST_DATA_ROOT, filename)
xtablepath = join(self.localDataRoot, filename)
r = PdfFileReader(filepath)
# The two below are (id, gen, byte offset)-valued lists
actualItems = list()
expItems = list()
for ref in r.objects(PdfFileReader.R_XTABLE, True):
actualItems.append(
(ref.idnum, ref.generation,
r._xrefTable[ref.generation][ref.idnum][0]))
r.close()
# We artificially read the XRef Table entries that we know belong
# to filepath, and store them into expItems.
expItems = sorted(self._parseXRefTable(xtablepath, (0, 1, 2)))
actualItems = sorted(actualItems)
expItems = sorted(expItems)
self.assertListEqual(expItems, actualItems,
"Differences found in " + filename)
def testXTableAgainstXStream(self):
"""
In section 7.5.8.4 of ISO 32000, "Compatibility with Applications That
Do Not Support Compressed Reference Streams", the standard describes a
means of crafting PDF files designed for versions 1.5+ that can be
opened nevertheless by readers that support older versions.
This test case verifies that all the items hidden by the XRef Table in
non-conforming readers are *all and exactly* loaded into the XRef
Stream by readers that support PDF 1.5+.
"""
self.maxDiff = None
# TO-DO Possibly add a few other files to this test case
inputFiles = ("GeoBase_NHNC1_Data_Model_UML_EN.pdf", )
for filename in inputFiles:
filepath = join(self.localDataRoot, filename)
expItems = {
e[0]: e[1:]
for e in self._parseXRefTable(filepath, (0, 2, 3))
}
actualItems = list()
r = PdfFileReader(join(TEST_DATA_ROOT, filename))
for ref in r.objects(PdfFileReader.R_XSTREAM, True):
actualItems.append(ref)
r.close()
actualItems = sorted(actualItems, key=lambda e: e.idnum)
expKeys = sorted(expItems.keys())
actualKeys = list(map(lambda e: e.idnum, actualItems))
self.assertListEqual(expKeys, actualKeys,
"Lists of item IDs are not identical")
for e, a in zip(expKeys, actualItems):
self.assertEqual(e, a.idnum, "Items ID does not correspond")
# If an item is in use in the XRef Stream, ensure then that it
# is marked free in the XRef Table.
if r._xrefStm[a.idnum][0] in (2, ):
self.assertTrue(
expItems[e][-1],
"Item %d should be hid by the XRef Table, but it was "
"not." % e,
)
def testXRefStreamObjects(self):
"""
Like ``PdfReaderTestCases.testXRefTableObjects()``, except that it
tests objects referenced by the Cross-Reference Stream.
``PdfFileReader.objects()`` second part (dealing with XStream objects)
is invoked and implicitly tested.
"""
inputFiles = ("crazyones.pdf", )
for filename in inputFiles:
filepath = join(self.localDataRoot, filename)
r = PdfFileReader(join(TEST_DATA_ROOT, filename))
# Two lists of tuples as explained by Table 18
actualItems = list()
expItems = list()
with open(filepath, "r") as instream:
for line in instream:
if not line or line.isspace() or line.startswith("%"):
continue
this_type, field2, field3 = (int(f) for f in line.split())
expItems.append((this_type, field2, field3))
for item in r.objects(PdfFileReader.R_XSTREAM, True):
priv8Item = r._xrefStm[item.idnum]
if priv8Item[0] in {0, 1}:
self.assertEqual(priv8Item[2], item.generation)
elif priv8Item[0] == 2:
self.assertEqual(item.generation, 0)
actualItems.append(priv8Item)
r.close()
actualItems = sorted(actualItems)
expItems = sorted(expItems)
self.assertListEqual(
expItems,
actualItems,
"Didn't correctly read the Cross-Reference Stream",
)
