def extract(self, source):
"""Extract an image from *source*.
If the image is supported an instance of PIL's Image is returned, otherwise None.
"""
p = Parser()
f = open_pds(source)
if self.log: self.log.debug("Parsing '%s'" % (source))
self.labels = p.parse(f)
if self.log: self.log.debug("Found %d labels" % (len(self.labels)))
if self._check_image_is_supported():
if self.log: self.log.debug("Image in '%s' is supported" % (source))
dim = self._get_image_dimensions()
loc = self._get_image_location()
if self.log: self.log.debug("Image dimensions should be %s" % (str(dim)))
if self.log: self.log.debug("Seeking to image data at %d" % (loc))
f.seek(loc)
if self.log: self.log.debug("Seek successful, reading data")
# rawImageData = f.readline()
# f.seek(-int(self.labels["RECORD_BYTES"]), os.SEEK_CUR)
rawImageData = f.read(dim[0] * dim[1])
if self.log: self.log.debug("Read successful (len: %d), creating Image object" % (len(rawImageData)))
# The frombuffer defaults may change in a future release;
# for portability, change the call to read:
# frombuffer(mode, size, data, 'raw', mode, 0, 1).
img = Image.frombuffer('L', dim, rawImageData, 'raw', 'L', 0, 1)
if self.log: self.log.debug("Image result: %s" % (str(img)))
if self.log: self.log.debug("Image info: %s" % (str(img.info)))
if self.log: self.log.debug("Image size: %s" % (str(img.size)))
else:
if self.log: self.log.error("Image is not supported '%s'" % (source))
img = None
f.close()
return img, self.labels
def extract(self, source):
"""Extract an image from *source*.
If the image is supported an instance of PIL's Image is returned, otherwise None.
"""
p = Parser()
f = open_pds(source)
if self.log: self.log.debug("Parsing '%s'" % (source))
self.labels = p.parse(f)
if self.log: self.log.debug("Found %d labels" % (len(self.labels)))
if self._check_image_is_supported():
if self.log: self.log.debug("Image in '%s' is supported" % (source))
dim = self._get_image_dimensions()
loc = self._get_image_location()
imageSampleBits = int(self.labels['IMAGE']['SAMPLE_BITS'])
imageSampleType = self.labels['IMAGE']['SAMPLE_TYPE']
md5Checksum = self._get_image_checksum()
if self.log: self.log.debug("Image dimensions should be %s" % (str(dim)))
if self.log: self.log.debug("Seeking to image data at %d" % (loc))
f.seek(loc)
if imageSampleBits == 8:
readSize = dim[0] * dim[1]
elif imageSampleBits == 16:
readSize = dim[0] * dim[1] * 2
print readSize
if self.log: self.log.debug("Seek successful, reading data (%s)" % (readSize))
# rawImageData = f.readline()
# f.seek(-int(self.labels["RECORD_BYTES"]), os.SEEK_CUR)
rawImageData = f.read(readSize)
if md5Checksum:
rawImageChecksum = hashlib.md5(rawImageData).hexdigest()
checksumVerificationPassed = rawImageChecksum == md5Checksum and True or False
if not checksumVerificationPassed:
if self.log: self.log.debug("Secure hash verification failed")
if self.raisesChecksumError:
errorMessage = "Verification failed! Expected '%s' but got '%s'." % (md5Checksum, rawImageChecksum)
raise ChecksumError, errorMessage
else:
if self.log: self.log.debug("Secure hash verification passed")
if self.log: self.log.debug("Read successful (len: %d), creating Image object" % (len(rawImageData)))
# The frombuffer defaults may change in a future release;
# for portability, change the call to read:
# frombuffer(mode, size, data, 'raw', mode, 0, 1).
if (imageSampleBits == 16) and imageSampleType == ('MSB_INTEGER'):
#img = Image.frombuffer('I', dim, rawImageData, 'raw', 'I;16BS', 0, 1)
img = Image.frombuffer('F', dim, rawImageData, 'raw', 'F;16B', 0, 1)
img = ImageMath.eval("convert(a/16.0, 'L')", a=img)
else:
img = Image.frombuffer('L', dim, rawImageData, 'raw', 'L', 0, 1)
if self.log:
self.log.debug("Image result: %s" % (str(img)))
self.log.debug("Image info: %s" % (str(img.info)))
self.log.debug("Image mode: %s" % (str(img.mode)))
self.log.debug("Image size: %s" % (str(img.size)))
else:
if self.log: self.log.error("Image is not supported '%s'" % (source))
img = None
f.close()
return img, self.labels
def extract(self, source):
"""Extract an image from *source*.
If the image is supported an instance of PIL's Image is returned, otherwise None.
"""
p = Parser()
f = open_pds(source)
pdsdatadir, pdsfile = os.path.split(source)
if self.log:
self.log.debug("Parsing '%s'" % (source))
self.labels = p.parse(f)
if self.log:
self.log.debug("Found %d labels" % (len(self.labels)))
if self._check_table_is_supported():
if self.log:
self.log.debug("Table in '%s' is supported" % (source))
dim = self._get_table_dimensions()
# Get the location of the table
location = self._get_table_location().strip().replace('"', "")
# location = os.path.join(pdsdatadir,location)
# Get the structure of the table from the pointer
struct_fname = self._get_table_structure().strip().replace('"', "")
structurefile = getPdsFileName(struct_fname, pdsdatadir)
sp = ColumnParser()
s = open_pds(structurefile)
slabels = sp.parse(s)
columns = []
for l in slabels:
columns.append(l["COLUMN"]["NAME"].strip().replace('"', ""))
if self.log:
self.log.debug("Found %d columns" % (len(columns)))
if self.labels["TABLE"]["INTERCHANGE_FORMAT"] == "ASCII":
locationfile = getPdsFileName(location, pdsdatadir)
tbl = csv.DictReader(open(locationfile), fieldnames=columns, delimiter=" ")
else:
if self.log:
self.log.error("Table is not supported '%s'" % (source))
tbl = None
f.close()
return tbl, self.labels
def extract(self, source):
"""Extract an image from *source*.
If the image is supported an instance of PIL's Image is returned, otherwise None.
"""
p = Parser()
f = open_pds(source)
pdsdatadir, pdsfile = os.path.split(source)
if self.log: self.log.debug("Parsing '%s'" % (source))
self.labels = p.parse(f)
if self.log: self.log.debug("Found %d labels" % (len(self.labels)))
if self._check_table_is_supported():
if self.log:
self.log.debug("Table in '%s' is supported" % (source))
dim = self._get_table_dimensions()
# Get the location of the table
location = self._get_table_location().strip().replace("\"", "")
#location = os.path.join(pdsdatadir,location)
# Get the structure of the table from the pointer
struct_fname = self._get_table_structure().strip().replace(
"\"", "")
structurefile = getPdsFileName(struct_fname, pdsdatadir)
sp = ColumnParser()
s = open_pds(structurefile)
slabels = sp.parse(s)
columns = []
for l in slabels:
columns.append(l['COLUMN']['NAME'].strip().replace("\"", ""))
if self.log: self.log.debug("Found %d columns" % (len(columns)))
if self.labels['TABLE']['INTERCHANGE_FORMAT'] == 'ASCII':
locationfile = getPdsFileName(location, pdsdatadir)
tbl = csv.DictReader(open(locationfile),
fieldnames=columns,
delimiter=' ')
else:
if self.log:
self.log.error("Table is not supported '%s'" % (source))
tbl = None
f.close()
return tbl, self.labels
def test_no_exceptions(self): import os from core import open_pds testDataDir = '../../test_data/' outputDir = '../../tmp/' imgExtractor = ImageExtractor(log="ImageExtractor_Unit_Tests") for root, dirs, files in os.walk(testDataDir): for name in files: filename = os.path.join(root, name) print filename img, _ = imgExtractor.extract(open_pds(filename)) try: if img: img.save(outputDir + name + '.jpg') except Exception, e: # Re-raise the exception, causing this test to fail. raise else: # The following is executed if and when control flows off the end of the try clause. assert True
def test_no_exceptions(self): import os from core import open_pds testDataDir = '../../test_data/' outputDir = '../../tmp/' imgExtractor = ImageExtractor(log="ImageExtractor_Unit_Tests") for root, dirs, files in os.walk(testDataDir): for name in files: filename = os.path.join(root, name) print filename img, _ = imgExtractor.extract(open_pds(filename)) try: if img: img.save(outputDir + name + '.jpg') except Exception, e: # Re-raise the exception, causing this test to fail. raise else: # The following is executed if and when control flows off the end of the try clause. assert True
