sparkocr.utils.svs.tiffparser#

Classes#

Functions#

Attributes#

exception TiffFileError#

Bases: Exception

Exception to indicate invalid TIFF structure.

exception TiffParserError#

Bases: Exception

Exception to indicate parser functionality exceeded

class FileHandle(file, mode='rb', name=None, offset=None, size=None)#

Binary file handle.

A limited, special purpose file handle that can:

• handle embedded files (for CZI within CZI files)

• re-open closed files (for multi-file formats, such as OME-TIFF)

• read and write numpy arrays and records from file like objects

Only ‘rb’ and ‘wb’ modes are supported. Concurrently reading and writing of the same stream is untested.

When initialized from another file handle, do not use it unless this FileHandle is closed.

name#

Name of the file.

Type:

str

path#

Absolute path to file.

Type:

str

size#

Size of file in bytes.

Type:

int

is_file#

If True, file has a filno and can be memory-mapped.

Type:

bool

All attributes are read-only.

is_file = False#

close()#

Close file.

flush()#

Flush write buffers if applicable.

memmap_array(dtype, shape, offset=0, mode='r', order='C')#

open()#

Open or re-open file.

read(size=-1)#

Read ‘size’ bytes from file, or until EOF is reached.

read_array(dtype, count=-1, out=None)#

read_record(dtype, shape=1, byteorder=None)#

Return numpy record from file.

read_segments(offsets, bytecounts, indices=None, sort=True, lock=None, buffersize=None)#

readinto(b)#

Read up to len(b) bytes into b, and return number of bytes read.

seek(offset, whence=0)#

Set file’s current position.

tell()#

Return file’s current position.

write(bytestring)#

Write bytes to file.

write_array(data)#

Write numpy array to binary file.

write_empty(size)#

Append size bytes to file. Position must be at end of file.

class FileSequence(fromfile, files, container=None, sort=None, pattern=None, axesorder=None)#

Series of files containing array data of compatible shape and data type.

files#

List of file names.

Type:

list

shape#

Shape of file series. Excludes shape of individual arrays.

Type:

tuple

axes#

Labels of axes in shape.

Type:

str

axes = ''#

files#

fromfile#

pattern#

shape#

asarray(file=None, ioworkers=1, out=None, **kwargs)#

Read image data from files and return as numpy array.

Raise IndexError or ValueError if array shapes do not match.

Parameters:

• file (int or None) – Index or name of single file to read.

• ioworkers (int or None) – Maximum number of threads to execute the array read function asynchronously. Default: 1. If None, default to the number of processors multiplied by 5. Using threads can significantly improve runtime when reading many small files from a network share.

• out (numpy.ndarray, str, or file-like object) – Buffer where image data will be saved. If None (default), a new array will be created. If numpy.ndarray, a writable array of compatible dtype and shape. If ‘memmap’, create a memory-mapped array in a temporary file. If str or open file, the file name or file object used to create a memory-map to an array stored in a binary file on disk.

• kwargs (dict) – Additional parameters passed to the array read function.

close()#

class LazyConst(cls)#

Class whose attributes are computed on first access from its methods.

class NullContext#

Null context manager.

>>> with NullContext():
...     pass

class OpenFileCache(size, lock=None)#

Keep files open.

files#

lock#

past = []#

size#

clear()#

Close all opened files if not in use.

close(filehandle)#

Close openend file if no longer used.

open(filehandle)#

Re-open file if necessary.

class TIFF#

Namespace for module constants.

ALLOCATIONGRANULARITY()#

AXES_LABELS()#

BIG_BE()#

BIG_LE()#

CLASSIC_BE()#

CLASSIC_LE()#

COLORRESPONSEUNIT()#

COMPESSORS()#

COMPRESSION()#

CZ_LSMINFO()#

CZ_LSMINFO_DATATYPES()#

CZ_LSMINFO_DIMENSIONS()#

CZ_LSMINFO_READERS()#

CZ_LSMINFO_SCANINFO_ARRAYS()#

CZ_LSMINFO_SCANINFO_ATTRIBUTES()#

CZ_LSMINFO_SCANINFO_STRUCTS()#

CZ_LSMINFO_SCANTYPE()#

CZ_LSMINFO_TYPEOFDATA()#

DATATYPES()#

DATA_DTYPES()#

DATA_FORMATS()#

DECOMPESSORS()#

EXIF_TAGS()#

EXTRASAMPLE()#

FILEOPEN_FILTER()#

FILETYPE()#

FILE_EXTENSIONS()#

FILE_FLAGS()#

FILLORDER()#

FRAME_ATTRS()#

GEO_CODES()#

GEO_KEYS()#

GPS_TAGS()#

GRAYRESPONSEUNIT()#

GROUP4OPT()#

IOP_TAGS()#

MAXWORKERS()#

MM_DIMENSIONS()#

MM_HEADER()#

NDPI_LE()#

NDPI_TAGS()#

NIH_COLORTABLE_TYPE()#

NIH_CURVEFIT_TYPE()#

NIH_IMAGE_HEADER()#

NIH_LUTMODE_TYPE()#

NIH_UNITS_TYPE()#

OFILETYPE()#

ORIENTATION()#

PHOTOMETRIC()#

PILATUS_HEADER()#

PLANARCONFIG()#

PREDICTOR()#

PREDICTORS()#

RESUNIT()#

SAMPLEFORMAT()#

SAMPLE_DTYPES()#

TAGS()#

TAG_ATTRIBUTES()#

TAG_ENUM()#

TAG_READERS()#

TAG_TUPLE()#

THRESHHOLD()#

TVIPS_HEADER_V1()#

TVIPS_HEADER_V2()#

UIC_TAGS()#

UNPREDICTORS()#

class TiffFile(arg, name=None, offset=None, size=None, multifile=True, _useframes=None, **kwargs)#

Read image and metadata from TIFF file.

TiffFile instances must be closed using the ‘close’ method, which is automatically called when using the ‘with’ context manager.

TiffFile instances are not thread-safe.

pages#

Sequence of TIFF pages in file.

Type:

TiffPages

series#

Sequences of closely related TIFF pages. These are computed from OME, LSM, ImageJ, etc. metadata or based on similarity of page properties such as shape, dtype, and compression.

Type:

list of TiffPageSeries

is_flag#

If True, file is of a certain format. Flags are: bigtiff, uniform, shaped, ome, imagej, stk, lsm, fluoview, nih, vista, micromanager, metaseries, mdgel, mediacy, tvips, fei, sem, scn, svs, scanimage, andor, epics, ndpi, pilatus, qpi.

Type:

bool

All attributes are read-only.

asarray(key=None, series=None, out=None, maxworkers=None)#

close()#

Close open file handle(s).

fei_metadata()#

Return FEI metadata from SFEG or HELIOS tags as dict.

flags()#

Return set of file flags.

fluoview_metadata()#

Return consolidated FluoView metadata as dict.

fstat()#

Return status of file handle as stat_result object.

imagej_metadata()#

Return consolidated ImageJ metadata as dict.

is_mdgel()#

File has MD Gel format.

is_uniform()#

Return if file contains a uniform series of pages.

mdgel_metadata()#

Return consolidated metadata from MD GEL tags as dict.

metaseries_metadata()#

Return MetaSeries metadata from image description as dict.

micromanager_metadata()#

Return consolidated MicroManager metadata as dict.

nih_metadata()#

Return NIH Image metadata from NIHImageHeader tag as dict.

pilatus_metadata()#

Return Pilatus metadata from image description as dict.

scanimage_metadata()#

Return ScanImage non-varying frame and ROI metadata as dict.

series()#

Return related pages as TiffPageSeries.

Side effect: after calling this function, TiffFile.pages might contain TiffPage and TiffFrame instances.

shaped_metadata()#

Return tifffile metadata from JSON descriptions as dicts.

sis_metadata()#

Return Olympus SIS metadata from SIS and INI tags as dict.

stk_metadata()#

Return STK metadata from UIC tags as dict.

class TiffFrame(parent, index, offset=None, keyframe=None, offsets=None, bytecounts=None)#

Lightweight TIFF image file directory (IFD).

Only a limited number of tag values are read from file, e.g. StripOffsets, and StripByteCounts. Other tag values are assumed to be identical with a specified TiffPage instance, the keyframe.

TiffFrame is intended to reduce resource usage and speed up reading image data from file, not for introspection of metadata.

index#

is_mdgel = False#

offset = None#

pages = None#

parent#

asarray(*args, **kwargs)#

Read image data from file and return as numpy array.

aspage()#

Return TiffPage from file.

asrgb(*args, **kwargs)#

Read image data from file and return RGB image as numpy array.

class TiffPage(parent, index, keyframe=None)#

TIFF image file directory (IFD).

index#

Index of the page in file.

Type:

int

dtype#

Data type (native byte order) of the image in IFD.

Type:

numpy.dtype or None

shape#

Dimensions of the image in IFD, as returned by asarray.

Type:

tuple of int

axes#

Axes label codes for each dimension in shape: ‘X’ width, ‘Y’ height, ‘S’ sample, ‘I’ image series|page|plane, ‘Z’ depth, ‘C’ color|em-wavelength|channel, ‘E’ ex-wavelength|lambda, ‘T’ time, ‘R’ region|tile, ‘A’ angle, ‘P’ phase, ‘H’ lifetime, ‘L’ exposure, ‘V’ event, ‘Q’ unknown, ‘_’ missing

Type:

str

tags#

Multidict like interface to tags in IFD.

Type:

TiffTags

colormap#

Color look up table, if exists.

Type:

numpy.ndarray

shaped#

Normalized 6 dimensional shape of the image in IFD: 0 : number planes (stk), images (ij), or 1. 1 : separate samplesperpixel or 1. 2 : imagedepth Z (sgi) or 1. 3 : imagelength Y. 4 : imagewidth X. 5 : contig samplesperpixel or 1.

Type:

tuple of int

All attributes are read-only.

axes = ''#

bitspersample = 1#

colormap = None#

compression = 1#

databytecounts = ()#

dataoffsets = ()#

description = ''#

description1 = ''#

dtype = None#

extrasamples = 1#

fillorder = 1#

imagedepth = 1#

imagelength = 0#

imagewidth = 0#

index#

nodata = 0#

offset#

parent#

photometric = 0#

planarconfig = 1#

predictor = 1#

rowsperstrip = 4294967295#

sampleformat = 1#

samplesperpixel = 1#

shape = ()#

shaped = ()#

software = ''#

subfiletype = 0#

tiledepth = 1#

tilelength = 0#

tilewidth = 0#

andor_tags()#

Return consolidated metadata from Andor tags as dict.

asarray(out=None, squeeze=True, lock=None, reopen=True, maxsize=None, maxworkers=None)#

aspage()#

Return self.

asrgb(uint8=False, alpha=None, colormap=None, dmin=None, dmax=None, **kwargs)#

decode()#

epics_tags()#

Return consolidated metadata from EPICS areaDetector tags as dict.

flags()#

Return set of flags.

geotiff_tags()#

Return consolidated metadata from GeoTIFF tags as dict.

hash()#

Return checksum to identify pages in same series.

Pages with the same hash can use the same decode function.

is_contiguous()#

Return offset and size of contiguous data, else None.

Excludes prediction and fill_order.

is_final()#

Return if page’s image data are stored in final form.

Excludes byte-swapping.

is_imagej()#

Return ImageJ description if exists, else None.

is_memmappable()#

Return if page’s image data in file can be memory-mapped.

is_ndpi()#

Page contains NDPI metadata.

is_shaped()#

Return description containing array shape if exists, else None.

maxworkers()#

Return maximum number of threads for decoding strips ot tiles.

ndpi_tags()#

Return consolidated metadata from Hamamatsu NDPI as dict.

pages()#

Return sequence of sub-pages (SubIFDs).

class TiffPageSeries(pages, shape, dtype, axes, parent=None, name=None, transform=None, kind=None, truncated=False)#

Series of TIFF pages with compatible shape and data type.

pages#

Sequence of TiffPages in series.

Type:

list of TiffPage

dtype#

Data type (native byte order) of the image array in series.

Type:

numpy.dtype

shape#

Dimensions of the image array in series.

Type:

tuple

axes#

Labels of axes in shape. See TiffPage.axes.

Type:

str

offset#

Position of image data in file if memory-mappable, else None.

Type:

int or None

axes = ''#

dtype#

index = 0#

kind = ''#

name = ''#

shape#

transform = None#

asarray(**kwargs)#

Return image data from series of TIFF pages as numpy array.

offset()#

Return offset to series data in file, if any.

class TiffPages(parent)#

Sequence of TIFF image file directories (IFD chain).

Instances of TiffPages have a state (cache, keyframe, etc.) and are not thread-safe.

pages = []#

parent = None#

class TiffTag(parent, tagheader, tagoffset)#

TIFF tag structure.

name#

Name of tag, TIFF.TAGS[code].

Type:

string

code#

Decimal code of tag.

Type:

int

dtype#

Datatype of tag data. One of TIFF DATA_FORMATS.

Type:

str

count#

Number of values.

Type:

int

value#

Tag data as Python object.

Type:

various types

offset#

Location of tag structure in file.

Type:

int

valueoffset#

Location of value in file.

Type:

int

All attributes are read-only.

code#

count#

dtype#

offset#

value#

valueoffset#

class TiffTags#

Multidict like interface to TiffTag instances in TiffPage.

Differences to a regular dict:

• values are instances of TiffTag.

• keys are TiffTag.code (int).

• multiple values can be stored per key.

• can be indexed with TiffTag.name (str), although slower than by key.

• iter() returns values instead of keys.

• values() and items() contain all values sorted by offset stored in file.

• len() returns the number of all values.

• get() takes an optional index argument.

• some functions are not implemented, e.g. update, setdefault, pop.

add(tag)#

Add a tag.

get(key, default=None, index=None)#

Return tag of code or name if exists, else default.

getall(key, default=None)#

Return list of all tags of code or name if exists, else default.

items()#

Return all (code, tag) pairs in order tags are stored in file.

keys()#

Return new view of all codes.

values()#

Return all tags in order they are stored in file.

class dtype(s)#

Bases: object

itemsize#

class lazyattr(func)#

Attribute whose value is computed on first access.

func#

apply_colormap(image, colormap, contig=True)#

asbool(value, true=(b'true', 'true'), false=(b'false', 'false'))#

Return string as bool if possible, else raise TypeError.

>>> asbool(b' False ')
False

askopenfilename(**kwargs)#

astype(value, types=None)#

Return argument as one of types if possible.

>>> astype('42')
42
>>> astype('3.14')
3.14
>>> astype('True')
True
>>> astype(b'Neee-Wom')
'Neee-Wom'

byteorder_isnative(byteorder)#

Return if byteorder matches the system’s byteorder.

>>> byteorder_isnative('=')
True

bytes2str(b, encoding=None, errors='strict')#

Return unicode string from encoded bytes.

bytestr(s, encoding='cp1252')#

Return bytes from unicode string, else pass through.

clean_whitespace(string, compact=False)#

Return string with compressed whitespace.

create_output(out, shape, dtype, mode='w+', suffix=None)#

enumarg(enum, arg)#

Return enum member from its name or value.

>>> enumarg(TIFF.PHOTOMETRIC, 2)
<PHOTOMETRIC.RGB: 2>
>>> enumarg(TIFF.PHOTOMETRIC, 'RGB')
<PHOTOMETRIC.RGB: 2>

enumstr(enum)#

Return short string representation of Enum instance.

excel_datetime(timestamp, epoch=None)#

Return datetime object from timestamp in Excel serial format.

Convert LSM time stamps.

>>> excel_datetime(40237.029999999795)
datetime.datetime(2010, 2, 28, 0, 43, 11, 999982)

fluoview_description_metadata(description, ignoresections=None)#

Return metatata from FluoView image description as dict.

The FluoView image description format is unspecified. Expect failures.

>>> descr = ('[Intensity Mapping]\nMap Ch0: Range=00000 to 02047\n'
...          '[Intensity Mapping End]')
>>> fluoview_description_metadata(descr)
{'Intensity Mapping': {'Map Ch0: Range': '00000 to 02047'}}

format_size(size, threshold=1536)#

Return file size as string from byte size.

>>> format_size(1234)
'1234 B'
>>> format_size(12345678901)
'11.50 GiB'

getdtype(d)#

hexdump(bytestr, width=75, height=24, snipat=-2, modulo=2, ellipsis=None)#

Return hexdump representation of bytes.

>>> hexdump(binascii.unhexlify('49492a00080000000e00fe0004000100'))
'49 49 2a 00 08 00 00 00 0e 00 fe 00 04 00 01 00 II*.............'

identityfunc(arg, *args, **kwargs)#

Single argument identity function.

>>> identityfunc('arg')
'arg'

imagej_description(shape, rgb=None, colormaped=False, version=None, hyperstack=None, mode=None, loop=None, **kwargs)#

Return ImageJ image description from data shape.

ImageJ can handle up to 6 dimensions in order TZCYXS.

>>> imagej_description((51, 5, 2, 196, 171))  
ImageJ=1.11a
images=510
channels=2
slices=5
frames=51
hyperstack=true
mode=grayscale
loop=false

imagej_description_metadata(description)#

Return metatata from ImageJ image description as dict.

Raise ValueError if not a valid ImageJ description.

>>> description = 'ImageJ=1.11a\nimages=510\nhyperstack=true\n'
>>> imagej_description_metadata(description)  
{'ImageJ': '1.11a', 'images': 510, 'hyperstack': True}

imagej_metadata(data, bytecounts, byteorder)#

Return IJMetadata tag value as dict.

The ‘Info’ string can have multiple formats, e.g. OIF or ScanImage, that might be parsed into dicts using the matlabstr2py or oiffile.SettingsFile functions. ‘ROI’ and ‘Overlays’ are returned as bytes, which can be parsed with the ImagejRoi.frombytes() function of the roifile package.

imagej_metadata_tag(metadata, byteorder)#

Return IJMetadata and IJMetadataByteCounts tags from metadata dict.

The tags can be passed to TiffWriter.save() as extratags.

The metadata dict may contain the following keys and values:

Infostr

Human-readable information as string.

Labelssequence of str

Human-readable labels for each channel.

Rangessequence of doubles

Lower and upper values for each channel.

LUTssequence of (3, 256) uint8 ndarrays

Color palettes for each channel.

Plotbytes

Undocumented ImageJ internal format.

ROI: bytes

Undocumented ImageJ internal region of interest format.

Overlaysbytes

Undocumented ImageJ internal format.

imagej_shape(shape, rgb=None)#

Return shape normalized to 6D ImageJ hyperstack TZCYXS.

Raise ValueError if not a valid ImageJ hyperstack shape.

>>> imagej_shape((2, 3, 4, 5, 3), False)
(2, 3, 4, 5, 3, 1)

imshow(data, photometric=None, planarconfig=None, bitspersample=None, nodata=0, interpolation=None, cmap=None, vmin=None, vmax=None, figure=None, title=None, dpi=96, subplot=None, maxdim=None, **kwargs)#

isprintable(string)#

Return if all characters in string are printable.

>>> isprintable('abc')
True
>>> isprintable(b'
')
False

json_description(shape, **metadata)#

Return JSON image description from data shape and other metadata.

Return UTF-8 encoded JSON.

>>> json_description((256, 256, 3), axes='YXS')  
b'{"shape": [256, 256, 3], "axes": "YXS"}'

json_description_metadata(description)#

Return metatata from JSON formated image description as dict.

Raise ValuError if description is of unknown format.

>>> description = '{"shape": [256, 256, 3], "axes": "YXS"}'
>>> json_description_metadata(description)  
{'shape': [256, 256, 3], 'axes': 'YXS'}
>>> json_description_metadata('shape=(256, 256, 3)')
{'shape': (256, 256, 3)}

julian_datetime(julianday, milisecond=0)#

Return datetime from days since 1/1/4713 BC and ms since midnight.

Convert Julian dates according to MetaMorph.

>>> julian_datetime(2451576, 54362783)
datetime.datetime(2000, 2, 2, 15, 6, 2, 783)

log_warning(msg, *args, **kwargs)#

Log message with level WARNING.

lsm2bin(lsmfile, binfile=None, tile=None, verbose=True)#

matlabstr2py(string)#

metaseries_description_metadata(description)#

Return metatata from MetaSeries image description as dict.

natural_sorted(iterable)#

Return human sorted list of strings.

E.g. for sorting file names.

>>> natural_sorted(['f1', 'f2', 'f10'])
['f1', 'f2', 'f10']

nullfunc(*args, **kwargs)#

Null function.

>>> nullfunc('arg', kwarg='kwarg')

olympusini_metadata(inistr)#

Return OlympusSIS metadata from INI string.

No documentation is available.

parse_filenames(files, pattern, axesorder=None)#

parse_kwargs(kwargs, *keys, **keyvalues)#

Return dict with keys from keys|keyvals and values from kwargs|keyvals.

Existing keys are deleted from kwargs.

>>> kwargs = {'one': 1, 'two': 2, 'four': 4}
>>> kwargs2 = parse_kwargs(kwargs, 'two', 'three', four=None, five=5)
>>> kwargs == {'one': 1}
True
>>> kwargs2 == {'two': 2, 'four': 4, 'five': 5}
True

pformat(arg, width=79, height=24, compact=True)#

pformat_xml(xml)#

Return pretty formatted XML.

pilatus_description_metadata(description)#

Return metatata from Pilatus image description as dict.

Return metadata from Pilatus pixel array detectors by Dectris, created by camserver or TVX software.

>>> pilatus_description_metadata('# Pixel_size 172e-6 m x 172e-6 m')
{'Pixel_size': (0.000172, 0.000172)}

product(iterable)#

Return product of sequence of numbers.

Equivalent of functools.reduce(operator.mul, iterable, 1). Multiplying numpy integers might overflow.

>>> product([2**8, 2**30])
274877906944
>>> product([])
1

read_bytes(fh, byteorder, dtype, count, offsetsize)#

Read tag data from file and return as bytes.

read_colormap(fh, byteorder, dtype, count, offsetsize)#

Read ColorMap data from file and return as numpy array.

read_cz_lsminfo(fh, byteorder, dtype, count, offsetsize)#

Read CZ_LSMINFO tag from file and return as dict.

read_cz_sem(fh, byteorder, dtype, count, offsetsize)#

Read Zeiss SEM tag and return as dict.

See https://sourceforge.net/p/gwyddion/mailman/message/29275000/ for unnamed values.

read_exif_ifd(fh, byteorder, dtype, count, offsetsize)#

Read EXIF tags from file and return as dict.

read_fei_metadata(fh, byteorder, dtype, count, offsetsize)#

Read FEI SFEG/HELIOS headers and return as dict.

read_gps_ifd(fh, byteorder, dtype, count, offsetsize)#

Read GPS tags from file and return as dict.

read_interoperability_ifd(fh, byteorder, dtype, count, offsetsize)#

Read Interoperability tags from file and return as dict.

read_json(fh, byteorder, dtype, count, offsetsize)#

Read JSON tag data from file and return as object.

read_lsm_channelcolors(fh)#

Read LSM ChannelColors structure from file and return as dict.

read_lsm_eventlist(fh)#

Read LSM events from file and return as list of (time, type, text).

read_lsm_floatpairs(fh)#

Read LSM sequence of float pairs from file and return as list.

read_lsm_positions(fh)#

Read LSM positions from file and return as list.

read_lsm_scaninfo(fh)#

Read LSM ScanInfo structure from file and return as dict.

read_lsm_timestamps(fh)#

Read LSM time stamps from file and return as list.

read_metaseries_catalog(fh)#

Read MetaSeries non-TIFF hint catalog from file.

Raise ValueError if the file does not contain a valid hint catalog.

read_micromanager_metadata(fh)#

Read MicroManager non-TIFF settings from open file and return as dict.

The settings can be used to read image data without parsing the TIFF file.

Raise ValueError if the file does not contain valid MicroManager metadata.

read_mm_header(fh, byteorder, dtype, count, offsetsize)#

Read FluoView mm_header tag from file and return as dict.

read_mm_stamp(fh, byteorder, dtype, count, offsetsize)#

Read FluoView mm_stamp tag from file and return as numpy.ndarray.

read_nih_image_header(fh, byteorder, dtype, count, offsetsize)#

Read NIH_IMAGE_HEADER tag from file and return as dict.

read_numpy(fh, byteorder, dtype, count, offsetsize)#

Read tag data from file and return as numpy array.

read_scanimage_metadata(fh)#

Read ScanImage BigTIFF v3 static and ROI metadata from open file.

Return non-varying frame data as dict and ROI group data as JSON.

The settings can be used to read image data and metadata without parsing the TIFF file.

Raise ValueError if file does not contain valid ScanImage v3 metadata.

read_sis(fh, byteorder, dtype, count, offsetsize)#

Read OlympusSIS structure and return as dict.

No specification is avaliable. Only few fields are known.

read_sis_ini(fh, byteorder, dtype, count, offsetsize)#

Read OlympusSIS INI string and return as dict.

read_tags(fh, byteorder, offsetsize, tagnames, customtags=None, maxifds=None)#

Read tags from chain of IFDs and return as list of dicts.

The file handle position must be at a valid IFD header.

read_tvips_header(fh, byteorder, dtype, count, offsetsize)#

Read TVIPS EM-MENU headers and return as dict.

read_uic1tag(fh, byteorder, dtype, count, offsetsize, planecount=None)#

Read MetaMorph STK UIC1Tag from file and return as dict.

Return empty dictionary if planecount is unknown.

read_uic2tag(fh, byteorder, dtype, planecount, offsetsize)#

Read MetaMorph STK UIC2Tag from file and return as dict.

read_uic3tag(fh, byteorder, dtype, planecount, offsetsize)#

Read MetaMorph STK UIC3Tag from file and return as dict.

read_uic4tag(fh, byteorder, dtype, planecount, offsetsize)#

Read MetaMorph STK UIC4Tag from file and return as dict.

read_uic_image_property(fh)#

Read UIC ImagePropertyEx tag from file and return as dict.

read_uic_tag(fh, tagid, planecount, offset)#

Read a single UIC tag value from file and return tag name and value.

UIC1Tags use an offset.

read_utf8(fh, byteorder, dtype, count, offsetsize)#

Read tag data from file and return as unicode string.

recarray2dict(recarray)#

Return numpy.recarray as dict.

reorient(image, orientation)#

repeat_nd(a, repeats)#

reshape_axes(axes, shape, newshape, unknown=None)#

reshape_nd(data_or_shape, ndim)#

scanimage_artist_metadata(artist)#

Return metatata from ScanImage artist tag as dict.

scanimage_description_metadata(description)#

Return metatata from ScanImage image description as dict.

sequence(value)#

Return tuple containing value if value is not a tuple or list.

>>> sequence(1)
(1,)
>>> sequence([1])
[1]
>>> sequence('ab')
('ab',)

snipstr(string, width=79, snipat=None, ellipsis=None)#

Return string cut to specified length.

>>> snipstr('abcdefghijklmnop', 8)
'abc...op'

squeeze_axes(shape, axes, skip=None)#

stack_pages(pages, out=None, maxworkers=None, **kwargs)#

stk_description_metadata(description)#

Return metadata from MetaMorph image description as list of dict.

The MetaMorph image description format is unspecified. Expect failures.

stripascii(string)#

Return string truncated at last byte that is 7-bit ASCII.

Clean NULL separated and terminated TIFF strings.

>>> stripascii(b'string\x00string\n\x01\x00')
b'string\x00string\n'
>>> stripascii(b'\x00')
b''

stripnull(string, null=b'\x00')#

Return string truncated at first null character.

Clean NULL terminated C strings. For unicode strings use null=’0’.

>>> stripnull(b'string\x00')
b'string'
>>> stripnull('string\x00', null='\0')
'string'

svs_description_metadata(description)#

Return metatata from Aperio image description as dict.

The Aperio image description format is unspecified. Expect failures.

>>> svs_description_metadata('Aperio Image Library v1.0')
{'Aperio Image Library': 'v1.0'}

transpose_axes(image, axes, asaxes=None)#

unpack_rgb(data, dtype=None, bitspersample=None, rescale=True)#

update_kwargs(kwargs, **keyvalues)#

Update dict with keys and values if keys do not already exist.

>>> kwargs = {'one': 1, }
>>> update_kwargs(kwargs, one=None, two=2)
>>> kwargs == {'one': 1, 'two': 2}
True

validate_jhove(filename, jhove=None, ignore=None)#

Validate TIFF file using jhove -m TIFF-hul.

Raise ValueError if jhove outputs an error message unless the message contains one of the strings in ‘ignore’.

JHOVE does not support bigtiff or more than 50 IFDs.

See JHOVE TIFF-hul Module

xml2dict(xml, sanitize=True, prefix=None)#

Return XML as dict.

>>> xml2dict('<?xml version="1.0" ?><root attr="name"><key>1</key></root>')
{'root': {'key': 1, 'attr': 'name'}}
>>> xml2dict('<level1><level2>3.5322</level2></level1>')
{'level1': {'level2': 3.5322}}

itemsizes#