Pillow/Tests/test_file_tiff_metadata.py

263 lines
9.8 KiB
Python
Raw Normal View History

import io
import struct
from PIL import Image, TiffImagePlugin, TiffTags
from PIL.TiffImagePlugin import IFDRational, _limit_rational
from .helper import PillowTestCase, hopper
tag_ids = {info.name: info.value for info in TiffTags.TAGS_V2.values()}
2014-06-10 13:10:47 +04:00
class TestFileTiffMetadata(PillowTestCase):
def test_rt_metadata(self):
""" Test writing arbitrary metadata into the tiff image directory
2014-06-10 13:10:47 +04:00
Use case is ImageJ private tags, one numeric, one arbitrary
data. https://github.com/python-pillow/Pillow/issues/291
"""
img = hopper()
2014-06-10 13:10:47 +04:00
# Behaviour change: re #1416
2015-09-12 12:11:10 +03:00
# Pre ifd rewrite, ImageJMetaData was being written as a string(2),
# Post ifd rewrite, it's defined as arbitrary bytes(7). It should
# roundtrip with the actual bytes, rather than stripped text
# of the premerge tests.
#
# For text items, we still have to decode('ascii','replace') because
# the tiff file format can't take 8 bit bytes in that field.
2015-12-10 01:35:35 +03:00
basetextdata = "This is some arbitrary metadata for a text field"
2019-06-13 18:54:11 +03:00
bindata = basetextdata.encode("ascii") + b" \xff"
2015-09-12 12:11:10 +03:00
textdata = basetextdata + " " + chr(255)
reloaded_textdata = basetextdata + " ?"
floatdata = 12.345
doubledata = 67.89
2014-06-10 13:10:47 +04:00
info = TiffImagePlugin.ImageFileDirectory()
2019-06-13 18:54:11 +03:00
ImageJMetaData = tag_ids["ImageJMetaData"]
ImageJMetaDataByteCounts = tag_ids["ImageJMetaDataByteCounts"]
ImageDescription = tag_ids["ImageDescription"]
2015-12-10 01:35:35 +03:00
2015-09-12 12:11:10 +03:00
info[ImageJMetaDataByteCounts] = len(bindata)
info[ImageJMetaData] = bindata
2019-06-13 18:54:11 +03:00
info[tag_ids["RollAngle"]] = floatdata
info.tagtype[tag_ids["RollAngle"]] = 11
info[tag_ids["YawAngle"]] = doubledata
info.tagtype[tag_ids["YawAngle"]] = 12
2015-09-12 12:11:10 +03:00
info[ImageDescription] = textdata
2015-12-10 01:35:35 +03:00
2014-06-10 13:10:47 +04:00
f = self.tempfile("temp.tif")
2014-06-10 13:10:47 +04:00
img.save(f, tiffinfo=info)
Improve handling of file resources Follow Python's file object semantics. User code is responsible for closing resources (usually through a context manager) in a deterministic way. To achieve this, remove __del__ functions. These functions used to closed open file handlers in an attempt to silence Python ResourceWarnings. However, using __del__ has the following drawbacks: - __del__ isn't called until the object's reference count reaches 0. Therefore, resource handlers remain open or in use longer than necessary. - The __del__ method isn't guaranteed to execute on system exit. See the Python documentation: https://docs.python.org/3/reference/datamodel.html#object.__del__ > It is not guaranteed that __del__() methods are called for objects > that still exist when the interpreter exits. - Exceptions that occur inside __del__ are ignored instead of raised. This has the potential of hiding bugs. This is also in the Python documentation: > Warning: Due to the precarious circumstances under which __del__() > methods are invoked, exceptions that occur during their execution > are ignored, and a warning is printed to sys.stderr instead. Instead, always close resource handlers when they are no longer in use. This will close the file handler at a specified point in the user's code and not wait until the interpreter chooses to. It is always guaranteed to run. And, if an exception occurs while closing the file handler, the bug will not be ignored. Now, when code receives a ResourceWarning, it will highlight an area that is mishandling resources. It should not simply be silenced, but fixed by closing resources with a context manager. All warnings that were emitted during tests have been cleaned up. To enable warnings, I passed the `-Wa` CLI option to Python. This exposed some mishandling of resources in ImageFile.__init__() and SpiderImagePlugin.loadImageSeries(), they too were fixed.
2019-05-25 19:30:58 +03:00
with Image.open(f) as loaded:
2015-09-12 00:24:35 +03:00
Improve handling of file resources Follow Python's file object semantics. User code is responsible for closing resources (usually through a context manager) in a deterministic way. To achieve this, remove __del__ functions. These functions used to closed open file handlers in an attempt to silence Python ResourceWarnings. However, using __del__ has the following drawbacks: - __del__ isn't called until the object's reference count reaches 0. Therefore, resource handlers remain open or in use longer than necessary. - The __del__ method isn't guaranteed to execute on system exit. See the Python documentation: https://docs.python.org/3/reference/datamodel.html#object.__del__ > It is not guaranteed that __del__() methods are called for objects > that still exist when the interpreter exits. - Exceptions that occur inside __del__ are ignored instead of raised. This has the potential of hiding bugs. This is also in the Python documentation: > Warning: Due to the precarious circumstances under which __del__() > methods are invoked, exceptions that occur during their execution > are ignored, and a warning is printed to sys.stderr instead. Instead, always close resource handlers when they are no longer in use. This will close the file handler at a specified point in the user's code and not wait until the interpreter chooses to. It is always guaranteed to run. And, if an exception occurs while closing the file handler, the bug will not be ignored. Now, when code receives a ResourceWarning, it will highlight an area that is mishandling resources. It should not simply be silenced, but fixed by closing resources with a context manager. All warnings that were emitted during tests have been cleaned up. To enable warnings, I passed the `-Wa` CLI option to Python. This exposed some mishandling of resources in ImageFile.__init__() and SpiderImagePlugin.loadImageSeries(), they too were fixed.
2019-05-25 19:30:58 +03:00
self.assertEqual(loaded.tag[ImageJMetaDataByteCounts], (len(bindata),))
self.assertEqual(loaded.tag_v2[ImageJMetaDataByteCounts], (len(bindata),))
2015-09-12 00:24:35 +03:00
Improve handling of file resources Follow Python's file object semantics. User code is responsible for closing resources (usually through a context manager) in a deterministic way. To achieve this, remove __del__ functions. These functions used to closed open file handlers in an attempt to silence Python ResourceWarnings. However, using __del__ has the following drawbacks: - __del__ isn't called until the object's reference count reaches 0. Therefore, resource handlers remain open or in use longer than necessary. - The __del__ method isn't guaranteed to execute on system exit. See the Python documentation: https://docs.python.org/3/reference/datamodel.html#object.__del__ > It is not guaranteed that __del__() methods are called for objects > that still exist when the interpreter exits. - Exceptions that occur inside __del__ are ignored instead of raised. This has the potential of hiding bugs. This is also in the Python documentation: > Warning: Due to the precarious circumstances under which __del__() > methods are invoked, exceptions that occur during their execution > are ignored, and a warning is printed to sys.stderr instead. Instead, always close resource handlers when they are no longer in use. This will close the file handler at a specified point in the user's code and not wait until the interpreter chooses to. It is always guaranteed to run. And, if an exception occurs while closing the file handler, the bug will not be ignored. Now, when code receives a ResourceWarning, it will highlight an area that is mishandling resources. It should not simply be silenced, but fixed by closing resources with a context manager. All warnings that were emitted during tests have been cleaned up. To enable warnings, I passed the `-Wa` CLI option to Python. This exposed some mishandling of resources in ImageFile.__init__() and SpiderImagePlugin.loadImageSeries(), they too were fixed.
2019-05-25 19:30:58 +03:00
self.assertEqual(loaded.tag[ImageJMetaData], bindata)
self.assertEqual(loaded.tag_v2[ImageJMetaData], bindata)
2015-09-12 00:24:35 +03:00
Improve handling of file resources Follow Python's file object semantics. User code is responsible for closing resources (usually through a context manager) in a deterministic way. To achieve this, remove __del__ functions. These functions used to closed open file handlers in an attempt to silence Python ResourceWarnings. However, using __del__ has the following drawbacks: - __del__ isn't called until the object's reference count reaches 0. Therefore, resource handlers remain open or in use longer than necessary. - The __del__ method isn't guaranteed to execute on system exit. See the Python documentation: https://docs.python.org/3/reference/datamodel.html#object.__del__ > It is not guaranteed that __del__() methods are called for objects > that still exist when the interpreter exits. - Exceptions that occur inside __del__ are ignored instead of raised. This has the potential of hiding bugs. This is also in the Python documentation: > Warning: Due to the precarious circumstances under which __del__() > methods are invoked, exceptions that occur during their execution > are ignored, and a warning is printed to sys.stderr instead. Instead, always close resource handlers when they are no longer in use. This will close the file handler at a specified point in the user's code and not wait until the interpreter chooses to. It is always guaranteed to run. And, if an exception occurs while closing the file handler, the bug will not be ignored. Now, when code receives a ResourceWarning, it will highlight an area that is mishandling resources. It should not simply be silenced, but fixed by closing resources with a context manager. All warnings that were emitted during tests have been cleaned up. To enable warnings, I passed the `-Wa` CLI option to Python. This exposed some mishandling of resources in ImageFile.__init__() and SpiderImagePlugin.loadImageSeries(), they too were fixed.
2019-05-25 19:30:58 +03:00
self.assertEqual(loaded.tag[ImageDescription], (reloaded_textdata,))
self.assertEqual(loaded.tag_v2[ImageDescription], reloaded_textdata)
2015-09-12 00:24:35 +03:00
Improve handling of file resources Follow Python's file object semantics. User code is responsible for closing resources (usually through a context manager) in a deterministic way. To achieve this, remove __del__ functions. These functions used to closed open file handlers in an attempt to silence Python ResourceWarnings. However, using __del__ has the following drawbacks: - __del__ isn't called until the object's reference count reaches 0. Therefore, resource handlers remain open or in use longer than necessary. - The __del__ method isn't guaranteed to execute on system exit. See the Python documentation: https://docs.python.org/3/reference/datamodel.html#object.__del__ > It is not guaranteed that __del__() methods are called for objects > that still exist when the interpreter exits. - Exceptions that occur inside __del__ are ignored instead of raised. This has the potential of hiding bugs. This is also in the Python documentation: > Warning: Due to the precarious circumstances under which __del__() > methods are invoked, exceptions that occur during their execution > are ignored, and a warning is printed to sys.stderr instead. Instead, always close resource handlers when they are no longer in use. This will close the file handler at a specified point in the user's code and not wait until the interpreter chooses to. It is always guaranteed to run. And, if an exception occurs while closing the file handler, the bug will not be ignored. Now, when code receives a ResourceWarning, it will highlight an area that is mishandling resources. It should not simply be silenced, but fixed by closing resources with a context manager. All warnings that were emitted during tests have been cleaned up. To enable warnings, I passed the `-Wa` CLI option to Python. This exposed some mishandling of resources in ImageFile.__init__() and SpiderImagePlugin.loadImageSeries(), they too were fixed.
2019-05-25 19:30:58 +03:00
loaded_float = loaded.tag[tag_ids["RollAngle"]][0]
self.assertAlmostEqual(loaded_float, floatdata, places=5)
loaded_double = loaded.tag[tag_ids["YawAngle"]][0]
self.assertAlmostEqual(loaded_double, doubledata)
# check with 2 element ImageJMetaDataByteCounts, issue #2006
2018-01-27 09:07:24 +03:00
info[ImageJMetaDataByteCounts] = (8, len(bindata) - 8)
img.save(f, tiffinfo=info)
Improve handling of file resources Follow Python's file object semantics. User code is responsible for closing resources (usually through a context manager) in a deterministic way. To achieve this, remove __del__ functions. These functions used to closed open file handlers in an attempt to silence Python ResourceWarnings. However, using __del__ has the following drawbacks: - __del__ isn't called until the object's reference count reaches 0. Therefore, resource handlers remain open or in use longer than necessary. - The __del__ method isn't guaranteed to execute on system exit. See the Python documentation: https://docs.python.org/3/reference/datamodel.html#object.__del__ > It is not guaranteed that __del__() methods are called for objects > that still exist when the interpreter exits. - Exceptions that occur inside __del__ are ignored instead of raised. This has the potential of hiding bugs. This is also in the Python documentation: > Warning: Due to the precarious circumstances under which __del__() > methods are invoked, exceptions that occur during their execution > are ignored, and a warning is printed to sys.stderr instead. Instead, always close resource handlers when they are no longer in use. This will close the file handler at a specified point in the user's code and not wait until the interpreter chooses to. It is always guaranteed to run. And, if an exception occurs while closing the file handler, the bug will not be ignored. Now, when code receives a ResourceWarning, it will highlight an area that is mishandling resources. It should not simply be silenced, but fixed by closing resources with a context manager. All warnings that were emitted during tests have been cleaned up. To enable warnings, I passed the `-Wa` CLI option to Python. This exposed some mishandling of resources in ImageFile.__init__() and SpiderImagePlugin.loadImageSeries(), they too were fixed.
2019-05-25 19:30:58 +03:00
with Image.open(f) as loaded:
Improve handling of file resources Follow Python's file object semantics. User code is responsible for closing resources (usually through a context manager) in a deterministic way. To achieve this, remove __del__ functions. These functions used to closed open file handlers in an attempt to silence Python ResourceWarnings. However, using __del__ has the following drawbacks: - __del__ isn't called until the object's reference count reaches 0. Therefore, resource handlers remain open or in use longer than necessary. - The __del__ method isn't guaranteed to execute on system exit. See the Python documentation: https://docs.python.org/3/reference/datamodel.html#object.__del__ > It is not guaranteed that __del__() methods are called for objects > that still exist when the interpreter exits. - Exceptions that occur inside __del__ are ignored instead of raised. This has the potential of hiding bugs. This is also in the Python documentation: > Warning: Due to the precarious circumstances under which __del__() > methods are invoked, exceptions that occur during their execution > are ignored, and a warning is printed to sys.stderr instead. Instead, always close resource handlers when they are no longer in use. This will close the file handler at a specified point in the user's code and not wait until the interpreter chooses to. It is always guaranteed to run. And, if an exception occurs while closing the file handler, the bug will not be ignored. Now, when code receives a ResourceWarning, it will highlight an area that is mishandling resources. It should not simply be silenced, but fixed by closing resources with a context manager. All warnings that were emitted during tests have been cleaned up. To enable warnings, I passed the `-Wa` CLI option to Python. This exposed some mishandling of resources in ImageFile.__init__() and SpiderImagePlugin.loadImageSeries(), they too were fixed.
2019-05-25 19:30:58 +03:00
self.assertEqual(
loaded.tag[ImageJMetaDataByteCounts], (8, len(bindata) - 8)
)
self.assertEqual(
loaded.tag_v2[ImageJMetaDataByteCounts], (8, len(bindata) - 8)
)
2014-06-10 13:10:47 +04:00
def test_read_metadata(self):
Improve handling of file resources Follow Python's file object semantics. User code is responsible for closing resources (usually through a context manager) in a deterministic way. To achieve this, remove __del__ functions. These functions used to closed open file handlers in an attempt to silence Python ResourceWarnings. However, using __del__ has the following drawbacks: - __del__ isn't called until the object's reference count reaches 0. Therefore, resource handlers remain open or in use longer than necessary. - The __del__ method isn't guaranteed to execute on system exit. See the Python documentation: https://docs.python.org/3/reference/datamodel.html#object.__del__ > It is not guaranteed that __del__() methods are called for objects > that still exist when the interpreter exits. - Exceptions that occur inside __del__ are ignored instead of raised. This has the potential of hiding bugs. This is also in the Python documentation: > Warning: Due to the precarious circumstances under which __del__() > methods are invoked, exceptions that occur during their execution > are ignored, and a warning is printed to sys.stderr instead. Instead, always close resource handlers when they are no longer in use. This will close the file handler at a specified point in the user's code and not wait until the interpreter chooses to. It is always guaranteed to run. And, if an exception occurs while closing the file handler, the bug will not be ignored. Now, when code receives a ResourceWarning, it will highlight an area that is mishandling resources. It should not simply be silenced, but fixed by closing resources with a context manager. All warnings that were emitted during tests have been cleaned up. To enable warnings, I passed the `-Wa` CLI option to Python. This exposed some mishandling of resources in ImageFile.__init__() and SpiderImagePlugin.loadImageSeries(), they too were fixed.
2019-05-25 19:30:58 +03:00
with Image.open("Tests/images/hopper_g4.tif") as img:
self.assertEqual(
{
"YResolution": IFDRational(4294967295, 113653537),
"PlanarConfiguration": 1,
"BitsPerSample": (1,),
"ImageLength": 128,
"Compression": 4,
"FillOrder": 1,
"RowsPerStrip": 128,
"ResolutionUnit": 3,
"PhotometricInterpretation": 0,
"PageNumber": (0, 1),
"XResolution": IFDRational(4294967295, 113653537),
"ImageWidth": 128,
"Orientation": 1,
"StripByteCounts": (1968,),
"SamplesPerPixel": 1,
"StripOffsets": (8,),
},
img.tag_v2.named(),
)
self.assertEqual(
{
"YResolution": ((4294967295, 113653537),),
"PlanarConfiguration": (1,),
"BitsPerSample": (1,),
"ImageLength": (128,),
"Compression": (4,),
"FillOrder": (1,),
"RowsPerStrip": (128,),
"ResolutionUnit": (3,),
"PhotometricInterpretation": (0,),
"PageNumber": (0, 1),
"XResolution": ((4294967295, 113653537),),
"ImageWidth": (128,),
"Orientation": (1,),
"StripByteCounts": (1968,),
"SamplesPerPixel": (1,),
"StripOffsets": (8,),
},
img.tag.named(),
)
2014-06-10 13:10:47 +04:00
def test_write_metadata(self):
""" Test metadata writing through the python code """
2019-06-13 18:54:11 +03:00
img = Image.open("Tests/images/hopper.tif")
2019-06-13 18:54:11 +03:00
f = self.tempfile("temp.tiff")
2014-06-10 13:10:47 +04:00
img.save(f, tiffinfo=img.tag)
Improve handling of file resources Follow Python's file object semantics. User code is responsible for closing resources (usually through a context manager) in a deterministic way. To achieve this, remove __del__ functions. These functions used to closed open file handlers in an attempt to silence Python ResourceWarnings. However, using __del__ has the following drawbacks: - __del__ isn't called until the object's reference count reaches 0. Therefore, resource handlers remain open or in use longer than necessary. - The __del__ method isn't guaranteed to execute on system exit. See the Python documentation: https://docs.python.org/3/reference/datamodel.html#object.__del__ > It is not guaranteed that __del__() methods are called for objects > that still exist when the interpreter exits. - Exceptions that occur inside __del__ are ignored instead of raised. This has the potential of hiding bugs. This is also in the Python documentation: > Warning: Due to the precarious circumstances under which __del__() > methods are invoked, exceptions that occur during their execution > are ignored, and a warning is printed to sys.stderr instead. Instead, always close resource handlers when they are no longer in use. This will close the file handler at a specified point in the user's code and not wait until the interpreter chooses to. It is always guaranteed to run. And, if an exception occurs while closing the file handler, the bug will not be ignored. Now, when code receives a ResourceWarning, it will highlight an area that is mishandling resources. It should not simply be silenced, but fixed by closing resources with a context manager. All warnings that were emitted during tests have been cleaned up. To enable warnings, I passed the `-Wa` CLI option to Python. This exposed some mishandling of resources in ImageFile.__init__() and SpiderImagePlugin.loadImageSeries(), they too were fixed.
2019-05-25 19:30:58 +03:00
with Image.open(f) as loaded:
Improve handling of file resources Follow Python's file object semantics. User code is responsible for closing resources (usually through a context manager) in a deterministic way. To achieve this, remove __del__ functions. These functions used to closed open file handlers in an attempt to silence Python ResourceWarnings. However, using __del__ has the following drawbacks: - __del__ isn't called until the object's reference count reaches 0. Therefore, resource handlers remain open or in use longer than necessary. - The __del__ method isn't guaranteed to execute on system exit. See the Python documentation: https://docs.python.org/3/reference/datamodel.html#object.__del__ > It is not guaranteed that __del__() methods are called for objects > that still exist when the interpreter exits. - Exceptions that occur inside __del__ are ignored instead of raised. This has the potential of hiding bugs. This is also in the Python documentation: > Warning: Due to the precarious circumstances under which __del__() > methods are invoked, exceptions that occur during their execution > are ignored, and a warning is printed to sys.stderr instead. Instead, always close resource handlers when they are no longer in use. This will close the file handler at a specified point in the user's code and not wait until the interpreter chooses to. It is always guaranteed to run. And, if an exception occurs while closing the file handler, the bug will not be ignored. Now, when code receives a ResourceWarning, it will highlight an area that is mishandling resources. It should not simply be silenced, but fixed by closing resources with a context manager. All warnings that were emitted during tests have been cleaned up. To enable warnings, I passed the `-Wa` CLI option to Python. This exposed some mishandling of resources in ImageFile.__init__() and SpiderImagePlugin.loadImageSeries(), they too were fixed.
2019-05-25 19:30:58 +03:00
original = img.tag_v2.named()
reloaded = loaded.tag_v2.named()
2016-02-05 01:57:13 +03:00
for k, v in original.items():
if isinstance(v, IFDRational):
2019-06-13 18:54:11 +03:00
original[k] = IFDRational(*_limit_rational(v, 2 ** 31))
2018-08-26 06:52:02 +03:00
elif isinstance(v, tuple) and isinstance(v[0], IFDRational):
2019-06-13 18:54:11 +03:00
original[k] = tuple(
IFDRational(*_limit_rational(elt, 2 ** 31)) for elt in v
)
2019-06-13 18:54:11 +03:00
ignored = ["StripByteCounts", "RowsPerStrip", "PageNumber", "StripOffsets"]
2014-06-10 13:10:47 +04:00
for tag, value in reloaded.items():
2016-02-05 01:57:13 +03:00
if tag in ignored:
continue
2019-06-13 18:54:11 +03:00
if isinstance(original[tag], tuple) and isinstance(
original[tag][0], IFDRational
):
2015-12-27 14:27:18 +03:00
# Need to compare element by element in the tuple,
# not comparing tuples of object references
2019-06-13 18:54:11 +03:00
self.assert_deep_equal(
original[tag],
value,
"{} didn't roundtrip, {}, {}".format(tag, original[tag], value),
2019-06-13 18:54:11 +03:00
)
2015-12-27 14:27:18 +03:00
else:
2019-06-13 18:54:11 +03:00
self.assertEqual(
original[tag],
value,
"{} didn't roundtrip, {}, {}".format(tag, original[tag], value),
2019-06-13 18:54:11 +03:00
)
2014-06-10 13:10:47 +04:00
for tag, value in original.items():
if tag not in ignored:
2019-06-13 18:54:11 +03:00
self.assertEqual(value, reloaded[tag], "%s didn't roundtrip" % tag)
def test_no_duplicate_50741_tag(self):
2019-06-13 18:54:11 +03:00
self.assertEqual(tag_ids["MakerNoteSafety"], 50741)
self.assertEqual(tag_ids["BestQualityScale"], 50780)
def test_empty_metadata(self):
2019-06-13 18:54:11 +03:00
f = io.BytesIO(b"II*\x00\x08\x00\x00\x00")
head = f.read(8)
info = TiffImagePlugin.ImageFileDirectory(head)
# Should not raise struct.error.
self.assert_warning(UserWarning, info.load, f)
def test_iccprofile(self):
# https://github.com/python-pillow/Pillow/issues/1462
2019-06-13 18:54:11 +03:00
im = Image.open("Tests/images/hopper.iccprofile.tif")
out = self.tempfile("temp.tiff")
im.save(out)
Improve handling of file resources Follow Python's file object semantics. User code is responsible for closing resources (usually through a context manager) in a deterministic way. To achieve this, remove __del__ functions. These functions used to closed open file handlers in an attempt to silence Python ResourceWarnings. However, using __del__ has the following drawbacks: - __del__ isn't called until the object's reference count reaches 0. Therefore, resource handlers remain open or in use longer than necessary. - The __del__ method isn't guaranteed to execute on system exit. See the Python documentation: https://docs.python.org/3/reference/datamodel.html#object.__del__ > It is not guaranteed that __del__() methods are called for objects > that still exist when the interpreter exits. - Exceptions that occur inside __del__ are ignored instead of raised. This has the potential of hiding bugs. This is also in the Python documentation: > Warning: Due to the precarious circumstances under which __del__() > methods are invoked, exceptions that occur during their execution > are ignored, and a warning is printed to sys.stderr instead. Instead, always close resource handlers when they are no longer in use. This will close the file handler at a specified point in the user's code and not wait until the interpreter chooses to. It is always guaranteed to run. And, if an exception occurs while closing the file handler, the bug will not be ignored. Now, when code receives a ResourceWarning, it will highlight an area that is mishandling resources. It should not simply be silenced, but fixed by closing resources with a context manager. All warnings that were emitted during tests have been cleaned up. To enable warnings, I passed the `-Wa` CLI option to Python. This exposed some mishandling of resources in ImageFile.__init__() and SpiderImagePlugin.loadImageSeries(), they too were fixed.
2019-05-25 19:30:58 +03:00
with Image.open(out) as reloaded:
self.assertNotIsInstance(im.info["icc_profile"], tuple)
self.assertEqual(im.info["icc_profile"], reloaded.info["icc_profile"])
def test_iccprofile_binary(self):
# https://github.com/python-pillow/Pillow/issues/1526
2018-09-27 13:35:00 +03:00
# We should be able to load this,
# but probably won't be able to save it.
Improve handling of file resources Follow Python's file object semantics. User code is responsible for closing resources (usually through a context manager) in a deterministic way. To achieve this, remove __del__ functions. These functions used to closed open file handlers in an attempt to silence Python ResourceWarnings. However, using __del__ has the following drawbacks: - __del__ isn't called until the object's reference count reaches 0. Therefore, resource handlers remain open or in use longer than necessary. - The __del__ method isn't guaranteed to execute on system exit. See the Python documentation: https://docs.python.org/3/reference/datamodel.html#object.__del__ > It is not guaranteed that __del__() methods are called for objects > that still exist when the interpreter exits. - Exceptions that occur inside __del__ are ignored instead of raised. This has the potential of hiding bugs. This is also in the Python documentation: > Warning: Due to the precarious circumstances under which __del__() > methods are invoked, exceptions that occur during their execution > are ignored, and a warning is printed to sys.stderr instead. Instead, always close resource handlers when they are no longer in use. This will close the file handler at a specified point in the user's code and not wait until the interpreter chooses to. It is always guaranteed to run. And, if an exception occurs while closing the file handler, the bug will not be ignored. Now, when code receives a ResourceWarning, it will highlight an area that is mishandling resources. It should not simply be silenced, but fixed by closing resources with a context manager. All warnings that were emitted during tests have been cleaned up. To enable warnings, I passed the `-Wa` CLI option to Python. This exposed some mishandling of resources in ImageFile.__init__() and SpiderImagePlugin.loadImageSeries(), they too were fixed.
2019-05-25 19:30:58 +03:00
with Image.open("Tests/images/hopper.iccprofile_binary.tif") as im:
self.assertEqual(im.tag_v2.tagtype[34675], 1)
self.assertTrue(im.info["icc_profile"])
def test_iccprofile_save_png(self):
2019-06-13 18:54:11 +03:00
im = Image.open("Tests/images/hopper.iccprofile.tif")
outfile = self.tempfile("temp.png")
im.save(outfile)
def test_iccprofile_binary_save_png(self):
2019-06-13 18:54:11 +03:00
im = Image.open("Tests/images/hopper.iccprofile_binary.tif")
outfile = self.tempfile("temp.png")
im.save(outfile)
def test_exif_div_zero(self):
im = hopper()
info = TiffImagePlugin.ImageFileDirectory_v2()
2016-02-05 01:57:13 +03:00
info[41988] = TiffImagePlugin.IFDRational(0, 0)
2019-06-13 18:54:11 +03:00
out = self.tempfile("temp.tiff")
im.save(out, tiffinfo=info, compression="raw")
Improve handling of file resources Follow Python's file object semantics. User code is responsible for closing resources (usually through a context manager) in a deterministic way. To achieve this, remove __del__ functions. These functions used to closed open file handlers in an attempt to silence Python ResourceWarnings. However, using __del__ has the following drawbacks: - __del__ isn't called until the object's reference count reaches 0. Therefore, resource handlers remain open or in use longer than necessary. - The __del__ method isn't guaranteed to execute on system exit. See the Python documentation: https://docs.python.org/3/reference/datamodel.html#object.__del__ > It is not guaranteed that __del__() methods are called for objects > that still exist when the interpreter exits. - Exceptions that occur inside __del__ are ignored instead of raised. This has the potential of hiding bugs. This is also in the Python documentation: > Warning: Due to the precarious circumstances under which __del__() > methods are invoked, exceptions that occur during their execution > are ignored, and a warning is printed to sys.stderr instead. Instead, always close resource handlers when they are no longer in use. This will close the file handler at a specified point in the user's code and not wait until the interpreter chooses to. It is always guaranteed to run. And, if an exception occurs while closing the file handler, the bug will not be ignored. Now, when code receives a ResourceWarning, it will highlight an area that is mishandling resources. It should not simply be silenced, but fixed by closing resources with a context manager. All warnings that were emitted during tests have been cleaned up. To enable warnings, I passed the `-Wa` CLI option to Python. This exposed some mishandling of resources in ImageFile.__init__() and SpiderImagePlugin.loadImageSeries(), they too were fixed.
2019-05-25 19:30:58 +03:00
with Image.open(out) as reloaded:
self.assertEqual(0, reloaded.tag_v2[41988].numerator)
self.assertEqual(0, reloaded.tag_v2[41988].denominator)
2019-08-29 12:02:19 +03:00
def test_empty_values(self):
data = io.BytesIO(
2019-06-13 18:54:11 +03:00
b"II*\x00\x08\x00\x00\x00\x03\x00\x1a\x01\x05\x00\x00\x00\x00\x00"
b"\x00\x00\x00\x00\x1b\x01\x05\x00\x00\x00\x00\x00\x00\x00\x00\x00"
b"\x98\x82\x02\x00\x07\x00\x00\x002\x00\x00\x00\x00\x00\x00\x00a "
b"text\x00\x00"
)
head = data.read(8)
info = TiffImagePlugin.ImageFileDirectory_v2(head)
info.load(data)
# Should not raise ValueError.
info = dict(info)
self.assertIn(33432, info)
2017-09-14 19:25:11 +03:00
def test_PhotoshopInfo(self):
2019-06-13 18:54:11 +03:00
im = Image.open("Tests/images/issue_2278.tif")
2017-09-14 19:25:11 +03:00
2019-08-19 14:12:16 +03:00
self.assertEqual(len(im.tag_v2[34377]), 1)
self.assertIsInstance(im.tag_v2[34377][0], bytes)
2019-06-13 18:54:11 +03:00
out = self.tempfile("temp.tiff")
2017-09-14 19:25:11 +03:00
im.save(out)
Improve handling of file resources Follow Python's file object semantics. User code is responsible for closing resources (usually through a context manager) in a deterministic way. To achieve this, remove __del__ functions. These functions used to closed open file handlers in an attempt to silence Python ResourceWarnings. However, using __del__ has the following drawbacks: - __del__ isn't called until the object's reference count reaches 0. Therefore, resource handlers remain open or in use longer than necessary. - The __del__ method isn't guaranteed to execute on system exit. See the Python documentation: https://docs.python.org/3/reference/datamodel.html#object.__del__ > It is not guaranteed that __del__() methods are called for objects > that still exist when the interpreter exits. - Exceptions that occur inside __del__ are ignored instead of raised. This has the potential of hiding bugs. This is also in the Python documentation: > Warning: Due to the precarious circumstances under which __del__() > methods are invoked, exceptions that occur during their execution > are ignored, and a warning is printed to sys.stderr instead. Instead, always close resource handlers when they are no longer in use. This will close the file handler at a specified point in the user's code and not wait until the interpreter chooses to. It is always guaranteed to run. And, if an exception occurs while closing the file handler, the bug will not be ignored. Now, when code receives a ResourceWarning, it will highlight an area that is mishandling resources. It should not simply be silenced, but fixed by closing resources with a context manager. All warnings that were emitted during tests have been cleaned up. To enable warnings, I passed the `-Wa` CLI option to Python. This exposed some mishandling of resources in ImageFile.__init__() and SpiderImagePlugin.loadImageSeries(), they too were fixed.
2019-05-25 19:30:58 +03:00
with Image.open(out) as reloaded:
self.assertEqual(len(reloaded.tag_v2[34377]), 1)
self.assertIsInstance(reloaded.tag_v2[34377][0], bytes)
2017-09-19 13:35:14 +03:00
def test_too_many_entries(self):
ifd = TiffImagePlugin.ImageFileDirectory_v2()
# 277: ("SamplesPerPixel", SHORT, 1),
2019-06-13 18:54:11 +03:00
ifd._tagdata[277] = struct.pack("hh", 4, 4)
2017-09-19 13:35:14 +03:00
ifd.tagtype[277] = TiffTags.SHORT
# Should not raise ValueError.
self.assert_warning(UserWarning, lambda: ifd[277])