def testXMLParsing():
samples = [{
'xml': """<?xml version='1.0' encoding='utf-8'?>
<OME xmlns="http://www.openmicroscopy.org/Schemas/OME/2016-06" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" UUID="urn:uuid:1ae9b229-162c-4431-8be0-1833b52302e5" xsi:schemaLocation="http://www.openmicroscopy.org/Schemas/OME/2016-06 http://www.openmicroscopy.org/Schemas/OME/2016-06/ome.xsd"><Image ID="Image:0"><Pixels BigEndian="false" DimensionOrder="XYZCT" ID="Pixels:0" PhysicalSizeX="0.32499998807907104" PhysicalSizeXUnit="µm" PhysicalSizeY="0.32499998807907104" PhysicalSizeYUnit="µm" SizeC="3" SizeT="1" SizeX="57346" SizeY="54325" SizeZ="1" Type="uint8"><Channel ID="Channel:0:0" Name="Red" SamplesPerPixel="1"><LightPath /></Channel><Channel ID="Channel:0:1" Name="Green" SamplesPerPixel="1"><LightPath /></Channel><Channel ID="Channel:0:2" Name="Blue" SamplesPerPixel="1"><LightPath /></Channel><TiffData IFD="0" PlaneCount="3" /><Plane TheC="0" TheT="0" TheZ="0" /><Plane TheC="1" TheT="0" TheZ="0" /><Plane TheC="2" TheT="0" TheZ="0" /></Pixels></Image></OME>""", # noqa
'checks': {
'frames': 3,
'IndexRange': {
'IndexC': 3
},
'IndexStride': {
'IndexC': 1
},
'channelmap': {
'Blue': 2,
'Green': 1,
'Red': 0
},
'channels': ['Red', 'Green', 'Blue'],
}
}]
# Create a source so we can use internal functions for testing
imagePath = datastore.fetch('sample.ome.tif')
source = large_image_source_ometiff.open(imagePath)
for sample in samples:
xml = ElementTree.fromstring(sample['xml'])
info = etreeToDict(xml)
assert etreeToDict(dictToEtree(info)) == info
source._omeinfo = info['OME']
source._parseOMEInfo()
metadata = source.getMetadata()
for key, value in sample['checks'].items():
if key in {'frames'}:
assert len(metadata[key]) == value
else:
assert metadata[key] == value
def parse_image_description(self, meta=None): # noqa
self._pixelInfo = {}
self._embeddedImages = {}
if not meta:
return
if not isinstance(meta, six.string_types):
meta = meta.decode('utf8', 'ignore')
try:
xml = ElementTree.fromstring(meta)
except Exception:
if 'AppMag = ' in meta:
try:
self._pixelInfo = {
'magnification':
float(
meta.split('AppMag = ')[1].split('|')[0].strip())
}
except Exception:
pass
return
try:
image = xml.find(".//DataObject[@ObjectType='DPScannedImage']")
columns = int(
image.find(".//*[@Name='PIM_DP_IMAGE_COLUMNS']").text)
rows = int(image.find(".//*[@Name='PIM_DP_IMAGE_ROWS']").text)
spacing = [
float(val.strip('"')) for val in image.find(
".//*[@Name='DICOM_PIXEL_SPACING']").text.split()
]
self._pixelInfo = {
'width': columns,
'height': rows,
'mm_x': spacing[0],
'mm_y': spacing[1]
}
except Exception:
pass
# Extract macro and label images
for image in xml.findall(".//*[@ObjectType='DPScannedImage']"):
try:
typestr = image.find(".//*[@Name='PIM_DP_IMAGE_TYPE']").text
datastr = image.find(".//*[@Name='PIM_DP_IMAGE_DATA']").text
except Exception:
continue
if not typestr or not datastr:
continue
typemap = {
'LABELIMAGE': 'label',
'MACROIMAGE': 'macro',
'WSI': 'thumbnail',
}
self._embeddedImages[typemap.get(typestr,
typestr.lower())] = datastr
try:
self._description_record = etreeToDict(xml)
except Exception:
pass
return True
def _parseMetadataXml(self, meta):
if not isinstance(meta, six.string_types):
meta = meta.decode('utf8', 'ignore')
try:
xml = ElementTree.fromstring(meta)
except Exception:
return
self._description_record = etreeToDict(xml)
xml = self._description_record
try:
# Optrascan metadata
scanDetails = xml.get('ScanInfo', xml.get('EncodeInfo'))['ScanDetails']
mag = float(scanDetails['Magnification'])
# In microns; convert to mm
scale = float(scanDetails['PixelResolution']) * 1e-3
self._pixelInfo = {
'magnification': mag,
'mm_x': scale,
'mm_y': scale,
}
except Exception:
pass
def __init__(self, path, **kwargs):
"""
Initialize the tile class. See the base class for other available
parameters.
:param path: a filesystem path for the tile source.
"""
super().__init__(path, **kwargs)
self._largeImagePath = self._getLargeImagePath()
# Read the root dzi file and check that the expected image files exist
try:
with builtins.open(self._largeImagePath) as fptr:
if fptr.read(1024).strip()[:5] != '<?xml':
raise TileSourceError(
'File cannot be opened via deepzoom reader.')
fptr.seek(0)
xml = ElementTree.parse(self._largeImagePath).getroot()
self._info = etreeToDict(xml)['Image']
except (ElementTree.ParseError, KeyError, UnicodeDecodeError):
raise TileSourceError('File cannot be opened via Deepzoom reader.')
except FileNotFoundError:
if not os.path.isfile(self._largeImagePath):
raise TileSourceFileNotFoundError(
self._largeImagePath) from None
raise
# We should now have a dictionary like
# {'Format': 'png', # or 'jpeg'
# 'Overlap': '1',
# 'Size': {'Height': '41784', 'Width': '44998'},
# 'TileSize': '254'}
# and a file structure like
# <rootname>_files/<level>/<x>_<y>.<format>
# images will be TileSize+Overlap square; final images will be
# truncated. Base level is either 0 or probably 8 (level 0 is a 1x1
# pixel tile)
self.sizeX = int(self._info['Size']['Width'])
self.sizeY = int(self._info['Size']['Height'])
self.tileWidth = self.tileHeight = int(self._info['TileSize'])
maxXY = max(self.sizeX, self.sizeY)
self.levels = int(
math.ceil(math.log(maxXY / self.tileWidth) / math.log(2))) + 1
tiledirName = os.path.splitext(os.path.basename(
self._largeImagePath))[0] + '_files'
rootdir = os.path.dirname(self._largeImagePath)
self._tiledir = os.path.join(rootdir, tiledirName)
if not os.path.isdir(self._tiledir):
rootdir = os.path.dirname(rootdir)
self._tiledir = os.path.join(rootdir, tiledirName)
zeroname = '0_0.%s' % self._info['Format']
self._nested = os.path.isdir(os.path.join(self._tiledir, '0',
zeroname))
zeroimg = PIL.Image.open(
os.path.join(self._tiledir, '0', zeroname) if not self._nested else
os.path.join(self._tiledir, '0', zeroname, zeroname))
if zeroimg.size == (1, 1):
self._baselevel = int(
math.ceil(math.log(maxXY) / math.log(2)) -
math.ceil(math.log(maxXY / self.tileWidth) / math.log(2)))
else:
self._baselevel = 0