Release notes

Version 2.1.2

Changes

  • Changed minimum supported Python version to 3.6.1 (#1270)

Version 2.1.1

Fixes

  • Remove ``py.typed``(#1251)

  • Fix ImportError with Python 3.6.0 (#1253)

  • Fix converting Sequences with Bulk Data when loading from JSON (#1254)

Version 2.1.0

Changelog

  • Dropped support for Python 3.5 (only Python 3.6+ supported)

Enhancements

  • Large testing data is no longer distributed within the pydicom package with the aim to reduce the package download size. These test files will download on-the-fly whenever either the tests are run, or should the file(s) be requested via the data manager functions. For example:

  • Added a new pixel data handler based on pylibjpeg which supports all (non-retired) JPEG transfer syntaxes (#1127)

  • Added apply_rescale() alias

  • Added apply_voi() and apply_windowing()

  • Added prefer_lut keyword parameter to apply_voi_lut() and handle empty VOI LUT module elements (#1234, #1237)

  • Added ability to register external data sources for use with the functions in pydicom.data (#1156)

  • __contains__, __next__ and __iter__ implementations added to PersonName (#1103)

  • Added convenience constants for the MPEG transfer syntaxes to pydicom.uid (#1155)

  • Added support for decoding Waveform Data:

  • When JPEG 2000 image data is unsigned and the Pixel Representation is 1 the image data is converted to signed (#1149)

  • Added keyword property for the new UID keywords in version 2020d of the DICOM Standard

  • Added testing of the variable names used when setting Dataset attributes and INVALID_KEYWORD_BEHAVIOR config option to allow customizing the behavior when a camel case variable name is used that isn’t a known element keyword (#1014)

  • Added INVALID_KEY_BEHAVIOR config option to allow customizing the behavior when an invalid key is used with the Dataset in operator (#1200)

  • Implemented full support (loading, accessing, modifying, writing) of DICOM File-sets and their DICOMDIR files via the FileSet class (#9, #243, #1093)

  • Added AllTransferSyntaxes

  • Added option to turn on pydicom future breaking behavior to allow user code to check itself against the next major version release. Set environment variable “PYDICOM_FUTURE” to “True” or call future_behavior()

  • Added another signature to the bulk_data_uri_handler in from_json to allow for the communication of not just the URI but also the tag and VR to the handler. Previous handlers will work as expected, new signature handlers will get the additional information.

  • pack_bits() can now be used with 2D or 3D input arrays and will pad the packed data to even length by default.

  • Elements with the IS VR accept float strings that are convertible to integers without loss, e.g. “1.0” (#1240)

  • Added encapsulate_extended() function for use when an Extended Offset Table is required (#1178)

Changes

  • Reading and adding unknown non-private tags now does not raise an exception per default, only when enforce_valid_values is set (#1161)

  • Data dictionaries updated to version 2020d of the DICOM Standard

  • Updated a handful of the SOP Class variable names in _storage_sopclass_uids to use the new UID keywords. Variables with Multiframe in them become MultiFrame, those with and in them become And, and DICOSQuadrupoleResonanceQRStorage becomes DICOSQuadrupoleResonanceStorage.

  • The following UID constants are deprecated and will be removed in v2.2:

  • In v3.0 the value for JPEGLossless will change from 1.2.840.10008.1.2.4.70 to 1.2.840.10008.1.2.4.57 to match its UID keyword

  • The following lists of UIDs are deprecated and will be removed in v2.2:

  • DicomDir and the dicomdir module are deprecated and will be removed in v3.0. Use FileSet instead (#1211)

  • pydicom.overlay_data_handlers is deprecated, use pydicom.overlays instead

  • Removed transfer syntax limitations when converting overlays to an ndarray (#1181)

  • The overlay_data_handlers config option is deprecated, the default handler will always be used.

Fixes

Version 2.0.0

Changelog

  • Dropped support for Python 2 (only Python 3.5+ supported)

  • Changes to Dataset.file_meta

    • file_meta now shown by default in dataset str or repr output; pydicom.config.show_file_meta can be set False to restore previous behavior

    • new FileMetaDataset class that accepts only group 2 data elements

    • Deprecation warning given unless Dataset.file_meta set with a FileMetaDataset object (in pydicom 3, it will be required)

  • Old PersonName class removed; PersonName3 renamed to PersonName. Classes PersonNameUnicode and PersonName3 are aliased to PersonName but are deprecated and will be removed in version 2.1

  • dataelem.isMultiValue (previously deprecated) has been removed. Use dataelem.DataElement.VM instead.

Enhancements

  • Allow PathLike objects for filename argument in dcmread, dcmwrite and Dataset.save_as (#1047)

  • Deflate post-file meta information data when writing a dataset with the Deflated Explicit VR Little Endian transfer syntax UID (#1086)

  • Added config.replace_un_with_known_vr to be able to switch off automatic VR conversion for known tags with VR “UN” (see #1067)

  • Added config.use_DS_numpy and config.use_IS_numpy to have multi-valued data elements with VR of DS or IS return a numpy array (#623) (much faster for bigger arrays). Both default to False to preserve previous behavior

Fixes

  • Fixed reading of datasets with an empty Specific Character Set tag (regression, #1038)

  • Fixed failure to parse dataset with an empty LUT Descriptor or Red/Green/Blue Palette Color LUT Descriptor element. (#1049)

  • Made Dataset.save_as a wrapper for dcmwrite (#1042) rather than having different checks in each

  • Removed 1.2.840.10008.1.2.4.70 - JPEG Lossless (Process 14, SV1) from the Pillow pixel data handler as Pillow doesn’t support JPEG Lossless. (#1053)

  • Fixed error when writing elements with a VR of OF (#1075)

  • Fixed improper conversion when reading elements with a VR of OF (#1075)

  • Fixed apply_voi_lut() and apply_modality_lut() not handling (0028,3006) LUT Data with a VR of OW (#1073)

  • Fixed access to private creator tag in raw datasets (#1078)

  • Fixed description of newly added known private tag (#1082)

  • Fixed update of private blocks after deleting private creator (#1097)

  • Fixed bug in updating pydicom.config.use_DS_Decimal flag in DS_decimal()

Version 1.4.1

Changelog

Fixes

  • Fixed writing of empty sequences (regression, #1030)

Changes

  • In Dataset.to_json and Dataset.to_json_dict, the default of the bulk_data_threshold argument has changed to 1024, and is now ignored if no bulk data handler is set (see #1029)

Version 1.4.0

Changelog

Fixes

  • Fixed handling of VRs AT and PN in json encoding (#915)

  • Fixed handling of binary values in json encoding (#887)

  • Prevent exception if assigning None to UI element (#894)

  • Fixed print output for numeric multi-value elements (#892)

  • Fixed testing PN values for truthiness (#891)

  • Fixed handling of data too large to written in explicit transfer syntax

  • Fixed handling of known tags with VR UN (#899, #968)

  • Fixed assigning of empty values to data elements (#896)

  • Fixed error in unpickling dataset (#947)

  • Fixed error in pickling modified datasets (#951)

  • Fixed improper conversion of the first value of the LUT Descriptor elements (0028,1101-1103) and (0028,3002) (#942)

  • Fixed handling of ISO IR 159 encoding (#917)

  • Fixed propagation of bulk data handler in Dataset.from_json (#971)

  • Correctly handle DICOMDIR files with records in reverse-hierarchical order (#822)

  • Pixel Data encoded using JPEG2000 and decoded using the Pillow handler no longer returns RGB data when the (0028,0004) Photometric Interpretation is YBR_FULL or YBR_FULL_422. (#263, #273, #826)

  • Avoid possible high memory usage while reading sequences (#994)

  • Fixed generate_pixel_data() not returning all available frames when the Basic Offset Table was empty. This may still occur when multiple fragments per frame are present for non-JPEG transfer syntaxes or where no JPEG EOI/EOC marker is present (#685)

  • Fixed possible incorrect switch to explicit VR in sequence items (#999)

  • Fixed JPEG 2000 (UIDs 1.2.840.10008.1.2.4.90 and 1.2.840.10008.1.2.4.91) pixel data with bit depth range 9-16 not producing the correct values with the Pillow handler (#693)

  • Fixed parsing a DICOMDIR file with no records raising an exception (#1004)

Enhancements

  • Added support for converting (60xx,3000) Overlay Data to a numpy ndarray using Dataset.overlay_array() (#912)

  • Added support for deferred reading in file-like objects (#932)

  • Tolerate values with multiple and/or incorrect padding bytes (#940)

  • Added support for uncompressed pixel data with (0028,0004) Photometric Interpretation of YBR_FULL_422 to the numpy pixel data handler.

  • Added apply_color_lut() function for applying color palette LUTs (#205)

  • Added apply_modality_lut() function for applying modality LUTs.

  • Added get_palette_files() for retrieving well-known palette color datasets.

  • Raise on end of file errors if config.enforce_valid_values is set (#277)

  • Added user warning, or exception in strict mode, if a DICOMDIR has an unexpected transfer syntax (#848)

  • Handle missing offset tags in DICOMDIR (#981)

  • Added optional handler argument to decompress(). This lets you specify a particular handler, rather than following pydicom’s default order (#537)

  • Added apply_voi_lut() function for applying VOI LUTs or windowing operations.

  • Added support for (7fe0,0008) Float Pixel Data and (7fe0,0009) Double Float Pixel Data to pixel_array (#452)

  • JPEG 2000 (1.2.840.10008.1.2.4.91) transfer syntax is supported for data with bit depth > 8 with the Pillow pixel data handler

  • PixelData.is_undefined_length is now set automatically based on whether the Dataset’s Transfer Syntax is compressed (#1006)

  • Updated DICOM dictionary to 2019e edition (#1013)

  • Added support for new VRs OV, SV, UV (#1016)

  • Code dictionaries and Code class for structured reporting added (alpha release only). See the Structured Reporting tutorial for more information

Changes

  • get_frame_offsets() now returns whether the Basic Offset Table is empty and a list of the offsets.

Version 1.3.0

Changelog

Documentation

  • New User Guide page for Python 2 support timeline

  • New User Guide page for working with private data elements

  • example loading set of CT slices and plotting axial, sagittal and coronal (#789)

Changes

  • Removed deprecated uid variables, config.image_handlers and DeferredDataElement (#760)

  • dataelem.isMultiValue is deprecated and will be removed in v1.4. Use dataelem.DataElement.VM instead.

  • dataelem.isStringOrStringList and dataelem.isString functions are removed

  • datadict.add_dict_entry and datadict.add_dict_entries now raise if trying to add a private tag

  • dataset.Dataset.maxBytesToDisplay also limits display length for non-binary VRs (by number of items) (#666)

Enhancements

  • Added datadict.add_private_dict_entry and datadict.add_private_dict_entries to add custom private tags (#799)

  • Added possibility to write into zip file using gzip, by avoiding seek (#753)

  • Added RLE encoding (#730)

  • Added handling of incorrect transfer syntax (explicit vs implicit) (#820)

  • Added creation of Tag instances by DICOM keyword, e.g Tag(“PatientName”)

  • Added possibility to get and add private tags without adding them to the private dictionary

  • Added possibility to use a Dataset in a NumPy array

  • Allow missing padding byte in Pixel Data, issue a warning in this case (#864)

  • Add in-memory image decoding with GDCM

  • Check really used implicit/explicit VR before reading a data set (#819, #820)

  • Added alpha conversion of Datasets to/from JSON format (Python 3 only)

Fixes

  • Correctly handle Dataset.pop and Dataset.setdefault for tuple and keyword arguments (#852)

  • Correctly handle encoding errors when any of the encodings are invalid (not just the first) (#850)

  • Do not raise while resolving an ambiguous VR dependent on PixelRepresentation if both PixelRepresentation and PixelData are not present (#838)

  • Raise exception with specific message if value is too large to be written in explicit transfer syntax (#757)

  • Make hash for PersonName3 behave as expected, make PersonName objects immutable (#785)

  • Fixed generate_uid() returning non-conformant UIDs when prefix=None (#788)

  • Avoid exception if reading from empty file (#810)

  • An invalid encoding is now replaced by the default encoding, if config.enforce_valid_values is not set (#815)

  • Correctly handle elements with ambiguous VR in sequence items (#804)

  • Fix bug where new DicomDir objects always have is_implicit_VR

  • Fix dataset equality for mixed raw vs converted data elements (#835)

  • Remove excess padding in Pixel Data

  • Fix wrong date format in anonymize example

  • Fix unknown VR exception message when VR isn’t ASCII (#791)

  • Fix jis-x-0201 characters encoding (#856)

Version 1.2.0

Changelog

Changes

  • PIL removed as a fallback if Pillow is not available in the pillow pixel data handler (#722)

  • uid.JPEGBaseLineLossy8bit deprecated and will be removed in v1.3. Use uid.JPEGBaseline instead. (#726)

  • uid.JPEGBaseLineLossy12bit deprecated and will be removed in v1.3. Use uid.JPEGExtended instead. (#726)

  • uid.JPEG2000Lossy deprecated and will be removed in v1.3. Use uid.JPEG2000 instead. (#726)

  • Equality and inequality operator overrides removed from UID.

  • config.image_handlers deprecated and will be removed in v1.3. - use config.pixel_data_handlers instead. There is also a change in behavior in that image_handlers previously used to only contain the pixel data handlers that had their dependencies met. Now pixel_data_handlers contains all handlers no matter whether or not their dependencies are met. To check if a handler is available for use (it has its dependency met) use the handler’s is_available method.

  • DeferredDataElement class deprecated and will be removed in v1.3 (#291)

  • The use of NumPyPy with PyPy is no longer supported, use NumPy instead.

Enhancements

  • Updated DICOM dictionary for 2018c edition (#677)

  • Added possibility to set byte strings as value for VRs that use only the default character set (#624)

  • Functions for encapsulating frames added to encaps module (#696)

  • Added Dataset.fix_meta_info() (#584)

  • Added new function for bit packing pack_bits for use with BitsAllocated = 1 (#715)

  • Added/corrected encoding and decoding of text and person name VRs using character sets with code extensions, added handling of encoding/decoding errors (#716)

  • Handle common spelling errors in Specific Character Set values (#695, #737)

  • Added uid.JPEGLosslessP14 for UID 1.2.840.10008.1.2.4.57

  • Added uid.JPEG2000MultiComponentLossless for UID 1.2.840.10008.1.2.4.92

  • Added uid.JPEG2000MultiComponent for UID 1.2.840.10008.1.2.4.93

  • Added full support for Planar Configuration (#713)

  • Added support for single frame pixel data where BitsAllocated > 8 and SamplesPerPixel > 1 (#713)

  • Small improvement in RLE decoding speed (~10%)

  • Added support for non-conformant RLE segment ordering (#729)

Fixes

  • Removed unused original_string attribute from the DataElement class (#660)

  • Improve performance for Python 3 when dealing with compressed multi-frame Pixel Data with pillow and jpeg-ls (#682)

  • Fixed handling of private tags in repeater range (#689)

  • Fixed Pillow pixel data handler for non-JPEG2k transfer syntax (#663)

  • Fixed handling of elements with ambiguous VR (#700, #728)

  • Adapted pixel handlers where endianess is explicitly adapted (#704)

  • Improve performance of bit unpacking (#715)

  • First character set no longer removed (#707)

  • Fixed RLE decoded data having the wrong byte order (#729)

  • Fixed RLE decoded data having the wrong planar configuration (#729)

  • Fixed numpy arrays returned by the pixel data handlers sometimes being read-only. Read-only arrays are still available for uncompressed transfer syntaxes via a keyword argument for the numpy pixel data handler and should help reduce memory consumption if required. (#717)

  • Fixed deprecation warning in Python 3.7 (#740)

Version 1.1.0

Changelog

Enhancements

  • UID.__str__ no longer returns the UID name (when known). The UID name is still available using the UID.name property.

  • Dataset equality now only compares the dataset’s DataElements (#464)

  • the codify script now supports VRs OD and OL, and works in Python 3 (#498); documentation has been added for codify

  • the performance for reading and writing datasets has been improved to be better than in pydicom 0.9.9 (#605, #512)

  • added support for bit-packed pixel data (#292)

  • updated DICOM dictionary for 2018b edition

  • added full API documentation to pydicom documentation (#649)

Fixes

  • UID should behave as expected for a python str subclass (#256)

  • group length elements in groups above 0x0006 removed on writing (#32)

  • fixed write_PN raising a TypeError when called with a non-iterable encoding parameter (#489)

  • fixed padding for some odd-sized image data (#599)

  • removed unneeded warning for incorrect date string length (#597)

  • fixed Dataset not slicing correctly when an (0xFFFF,0xFFFF) element is present (#92)

  • use correct VR for unknown private tags and private creators (#620)

  • fixed crash on reading RGB data with implicit VR (#620)

  • parent encoding was not used in sequences without own encoding (#625)

  • fixed error handling for values too large to fit in VR IS (#640)

Other

  • A deprecation warning has been added for UID.__eq__ and UID.__ne__ when comparing UID == [UID name] and UID != [UID name]. Starting in v1.2, UID equality and inequality comparisons will return False when performing the equivalent of UID == [UID name] and UID != [UID name], respectively. E.g. UID(‘1.2.840.10008.1.1’) == ‘Verification SOP Class’ will return False. Use UID.name == [UID name] instead.

Version 1.0.0

This is a major release, with major changes, including backwards-incompatible changes.

Changelog

Major changes

  • full Python 3 compatibility - one code base for both Python 2 and Python 3

  • package name and import name now match – use import pydicom rather than import dicom.

  • added handlers for converting (7fe0,0010) Pixel Data to a numpy.ndarray

    • optional GDCM support for reading files with compressed pixel data (#18)

    • optional Pillow and jpeg_ls support for reading some compressed pixel data files

    • support for decompressing a compressed dataset in-place

  • DICOM dictionary updated to 2017c

  • cleaned up DICOM dictionary code, old non-DICOM-keyword code removed

Other enhancements

  • added util/fixer.py callbacks available to fix non DICOM-compliant values before exceptions thrown

  • added context management methods to Dataset

  • added misc.is_dicom() function to check for DICOM file format

  • added date/time converters (#143)

  • added option to attempt other VRs if translate fails (#197)

  • added heuristics to read files that have no preamble or file meta information

  • support for multi-valued DA, DT, TM data elements (#212)

  • DataElement: added keyword and is_retired properties

  • datadict: added dictionary_is_retired()

  • datadict: added ability to add custom DICOM dictionary items via add_dict_entry() and add_dict_entries()

  • added some support for pickle

  • added support for VRs OD, OL and UC

  • added support for Thai, Japanese and Chinese encodings (#346, #353)

  • added support for slicing to Dataset

  • add/update TransferSyntaxUID when writing standard

  • renamed UID package to uid (conforms to Python standard)

  • added property uid.is_private

  • added definitions for storage SOP Class UIDs (#172)

  • added possibility to read only specific tags (#95)

  • added missing meta elements when writing DICOM file

  • added encaps generator functions to access compressed frames

  • read_file changed to dcmread, write_file to dcmwrite for greater clarity. Previous names still available for backwards compatibility.

Infrastructure

Fixes

  • correctly handle PlanarConfiguration==0 (#151)

  • updated uid generation to ensure uniqueness (#125)

  • handle missing patient data in show_dicomdir

  • assume default transfer syntax if none in file meta (#258)

  • fixed reading/writing of empty tags and tags with bad VR

  • fixed reading AE elements with leading or trailing spaces

  • fixed handling of ambiguous VR elements

  • fixed handling for several error conditions

  • fixed Latin5 (Turkish) character set handling

  • a lot of other small fixes…

Version 0.9.9

Changelog

In addition to bug fixes, pydicom 0.9.9 contains updates for all DICOM dictionaries. New features include DICOMDIR handling, and a utility module which produces python/pydicom source code to recreate a DICOM file.

Enhancements

  • All DICOM dictionaries updated (standard dictionary, UID dictionary, and private dictionaries)

  • Dicom commands also added to dictionary

  • Ability to work with DICOMDIR: read_dicomdir() function and DicomDir class. Example file show_dicomdir.py file added to examples subdirectory.

  • codify.py: Produce python/pydicom source code from a DICOM file.

  • a number of Python 3 compatibility enhancements

  • setup.py uses ez_setup only if setuptools not already installed

  • exceptions carry tag info with them, to aid in debugging

Contrib file changes

  • pydicom_series: force parameter added (Nil Goyette)

  • dcm_qt_tree: switch to OrderedDict to preserve ordering of tags (Padraig Looney)

Other Contributors

Other than Jonathan and myself, other contributors were: Rickard Holmberg, Julien Lamy, Yaroslav Halchenko, Mark White, Matthew Brett, Dimitri Papadopoulos, videan42 …(sorry if I’ve missed anyone).

Version 0.9.8

Changelog

pydicom 0.9.8 is mainly a consolidation step before moving to official Python 3 compatibility in pydicom 1.0. It also reverts the change to using Decimal for VR of DS (in pydicom 0.9.7), due to performance issues. DS as Decimal is still available, but is off by default.

Major changes

  • Requires Python 2.6 or later, in preparation for Python 3 compatibility

  • experimental Python 3 compatibility (unofficial at this point) – uncomment the two indicated lines in setup.py to use it. Please provide feedback to the issues list.

  • DS values reverted to using float as default (issue 114) due to slow performance using Python Decimal. Speed tests show approx factor of 10 improvement compared with pydicom 0.9.7 (several revisions up to r78ba350a3eb8)

  • streamlined much code internally taking advantage of modern Python constructs: decorators, generators, etc

Bug fixes

  • Fix for duplicate logger from Gunnar Schaefer. Fixes issue 107 (revision 774b7a55db33)

  • Fix rewind behavior in find_bytes (issue 60, revision 6b949a5b925b)

  • Fix error in nested private sequences (issue 113, revision 84af4b240add)

Enhancements

  • UID generator added (Félix C. Morency) (revisions 0197b5846bb5 and 3678b1be6aca, tests in f1ae573d9de5, 0411bab7c985)

  • new PersonName3 class for Python 3: (revision 9b92b336e7d4)

Contrib file changes

  • Fix for pydicom_series for DS decimal (revision e830f30b6781)

  • new dcm_qt_tree.py module - tree display of DICOM files using PySide and Qt. Contributed by Padraig Looney.

Special acknowledgement to Jonathan Suever who contributed most of the Python 3 work and many bug fixes.

Version 0.9.7

Changelog

pydicom 0.9.7 resolves some remaining bugs before moving to Python 3 compatibility. ** It is the last version which will run with Python < 2.6 ** (it will run with Python 2.4 to Python 2.7)

Major changes

  • Added DICOM 2011 keywords. Old “named tags” still work, but will be deprecated in future versions. Most names are identical, but some have changed. For example:

    • SamplesperPixel becomes SamplesPerPixel (capital ‘P’ on ‘Per’)

    • Beams becomes BeamSequence (and similar for all sequences)

    • Decimal and integer strings handled much better (revisions 4ed698a7bfbe and c313d2befb08).

    • New classes for VR of types DS and IS (DS is derived from Python Decimal)

    • New MultiValue class, enforcing all values of same type

    • New config.py file with user-definable parameters:

    • allow_DS_float (default False) for controlling whether float values can be used to construct DS or IS strings.

    • enforce_valid_values (default True) for ensuring IS, DS meet DICOM standard limits To change these, use ‘import dicom.config, then dicom.config.<parameter>={True|False}’ before setting values of data elements

Users are encouraged to switch to the official DICOM keywords, as these are now part of the standard, and promote consistency across programming languages and libraries.

Bug fixes

  • New way to read file meta information, not using the group length, instead reading until end of group 2 data elements. If group length dose not match, log a warning (revision b6b3658f3b14).

  • Fix bug in copying raw private data elements (issue 98)

  • Force logging level to warning on ‘import dicom’ (issue 102)

  • Deferred read fixed to work with gzipped files (issue 103)

  • Setting individual items in a DS or IS list now saves to file correctly

  • Japanese and Korean encoding fixes (issue 110)

Other Enhancements

  • New Sequence class which verifies items are Datasets (issue 52)

  • Assignment to SQ data element checks value is a Sequence or can be converted to one (issue 111)

  • dir(ds) now includes methods and properties as well as DICOM named tags. Work only on Python >= 2.6 as previous versions do not call __dir__ method (issue 95)

  • Added much more debugging info and simplified reading of data elements (revision b6b3658f3b14)

  • updated example files to DICOM 2011 keywords; fixed bugs

Many of the bug fixes/enhancements were submitted by users. Many thanks to those who contributed.

Version 0.9.6

Changelog

pydicom 0.9.6 updates the dictionary to the DICOM 2011 standard, and has a number of bug fixes

Major changes

  • updated the dictionary to the DICOM 2011 standard’s dictionary.

Bug fixes

  • Fixed bug in Dataset.file_metadata() and deprecated in favor of FileDataset (issue 93)

  • Fixed UID comparisons against non-string values (issue 96)

  • catch exceptions on reading undefined length private data elements (issue 91, issue 97)

  • Fixed bug in raising exception for unknown tag

Other

  • added example file write_new.py to show how to create DICOM files from scratch

  • updated other example files

  • more PEP-8 style changes

Version 0.9.5

Changelog

pydicom 0.9.5 is primarily a bug-fix release but includes some contrib files also.

Major fixes in this release

  • fix for incorrect pixel integer types which could lead to numeric errors (issue 79)

  • By default an InvalidDicomError will be raised when trying to read a non-DICOM file (unless read_file keyword arg {{{force}}} is True) (revision fc790f01f5)

  • fix recursion error on private data elements (issue 81, issue 84)

Other fixes in this release

  • Fix for unicode decode failing with VM > 1 (issue 78)

  • fix for fail of DicomIter on files with Explicit VR Transfer Syntax UID (issue 82)

  • Fix for Python 2.5 and ‘with’ statement (revision 1c32791bf0)

  • Handle ‘OB/OW’ VR as well as ‘OW/OB’ (revision e3ee934bbc)

  • Fix dataset.get(tag) so returns same as dataset[tag] (issue 88)

New ‘Contrib’ files

  • dicom_dao.py by Mike Wallace – CouchDB storage of DICOM info and binary data

  • pydicom_series.py by Almar Klein – Reads files and separates into distinct series.

Other

  • switch to Distribute for packaging

  • preliminary work on Python 3 compatiblity

  • preliminary work on using sphinx for documentation

  • preliminary work on better writing of files from scratch

Version 0.9.4

Changelog

Note

  • there is a backwards incompatible change made to storage of file_meta info. See item below.

  • pydicom 0.9.4 requires Python 2.4 or higher (pydicom 0.9.3 can run under Python 2.3)

Major changes/additions in this version

  • file reading code reorganized substantially

  • significant speed increase for reading DICOM files – approx 3 times faster than 0.9.3

  • partial file reading available – in particular, new optional argument to read_file(), stop_before_pixels, will stop before getting to the pixel data, not reading those into memory. Saves a little time for small images, but could be quite helpful for very large images when the pixel data is not needed.

  • read_file() now returns a !FileDataset object, instead of a plain Dataset. Most user code will not see much difference (except see next bullet on file meta information) but now the information stored in the object has been made explicit – e.g. the endian-ness and whether the file syntax was explicit VR or implicit VR.

  • file meta info has been separated from the main dataset. Logically, this makes more sense, as the file meta is not really part of the dataset, but is specific to the method of storage. This is a backwards-incompatible change, but is easily fixed by changing any references to file-meta data elements from {{{dataset.<name>}}} to {{{dataset.file_meta.<name>}}}. The file_meta is a dataset like any other, all the usual methods for viewing, changing data elements work on it also.

  • private dictionaries file now generated from the GDCM library’s private dictionary – code to convert formats contributed by Daniel Nanz.

  • license has returned to an MIT-based license (with the compatible GDCM also noted for the private dictionary component).

  • contributed files with example code for viewing using wxPython or Tkinter (and PIL) – in dicom.contrib folder. Thanks to Dave Witten, Daniel Nanz and Adit Panchal for these contributions.

  • updates to pydicom’s DICOM data dictionary contributed by Adit Panchal: CP805/916; Supp 43 and 117 (and UID dict), Supp 119 and 122

Other changes and bug fixes

  • Tag is now a factory function; the class is called !BaseTag. This was part of the file reading speed-up process – a new class !TupleTag was also created, for faster file reading

  • passing a file object to read_file() now works correctly, and also the file closing works as it should (caller needs to close any files passed in) (issue 73)

  • Fix for issue 72 : dataset.get() fails when passed type other than string or Tag. Patch contributed by !NikitaTheSpider

  • Fix for issue 58 : error opening file with unicode. Fix contributed by Pierre Raybaut

  • Fix for issue 42 : catch !AttributeError in property and give proper error message

  • Fix for issue 55 : UI type changed with string operations

  • Tag fixes and enhancements : can create tags with hex string (group, elem). Allow lists as well as tuples (issue 47). Fix arg2=0 bug (issue 64).

Version 0.9.3

Changelog

Major changes

  • changed to MIT-style license

  • option to defer reading of large data element values using read_file()’s new defer_size argument (r102, r103)

  • dictionary of private tags added – descriptive text shown when available (issue36, r97, r110)

  • more conversion to PEP-8 style. Should now use read_file(), save_as(), pixel_array rather than !ReadFile(), !SaveAs(), !PixelArray. Old names kept for now as aliases.

Other Enhancements

  • added DicomFileLike class to simplify and generalize access. Any object that has read, write, seek, tell, and close can now be used. (r105)

  • added dataset.iterall() function to iterate through all items (including inside sequences) (r105)

  • added dataset.formatted_lines() generator to allow custom formatting (r91, r113)

  • made reading tolerant of truncated files – gives a warning, but returns dataset read to that point (r95)

Bug Fixes

  • fixed issue38, name collision for ‘Other Patient Ids’ as both data element and sequence name in DICOM standard (r95, r96)

  • fixed issue40, blank VRs in some DICOM dictionary entries caused NotImplementError on reading (r100)

  • fixed issue41, reading VRs of ‘US or SS’ and similar split on backslash character (r104)

  • fixed bug where TransferSyntaxUID not present when reading file without DICOM header (r109)

  • fixed print recursion bug when printing a UID (r111)

Other

  • many of the example files updated

  • updated anonymize example file to also deal with ‘OtherPatientIDs’ and ‘PatientsBirthDate’ (r98)

Version 0.9.2

Changelog

Major changes

  • Renamed Attribute class and related modules to !DataElement. Old code will continue to work until pydicom 1.0, but with a !DeprecationWarning (issue22, r72, r73)

  • Added support for character sets through Specific Character Set (0008,0005), using Python unicode. Thus foreign languages can display names in Greek, Japanese, Chinese etc characters in environments which support unicode (demonstrated in dicomtree.py example using Tkinter GUI) (r64, r65)

Other Enhancements

  • Added support for auto-completion of dataset elements in ipython; also all environments using Python 2.6 (r69, r70)

  • Added __iter__() to Dataset so returns data elements in DICOM order with “for data_elem in dataset:” (r68)

  • Added dicomtree.py example program showing a DICOM file in a GUI window (Tkinter/Tix).

  • Added !PersonName class to parse components of names more easily (r55)

  • Added UID class to handle UID values. Name rather than UID number shown, UID_dictionary used (r51).

  • Code tested under Python 2.6

  • Added !DataElement.name property; synonym for !DataElement.description() function

Bug Fixes

  • Fixed issue27, sequence with a single empty item read incorrectly

  • Fixed bug that read_OW did not handle !UndefinedLength (r50)

  • Fixed bugs in example files anonymize.py, !DicomInfo.py, and dicomtree.py (r51)

  • Fixed issue33, VR=UN being split on backslash (r70)

  • Fixed issue18, util directory not installed (r45)

Other

  • Added example file myprint.py – shows how to custom format DICOM file information (r67)

  • Reorganized test files and added various new tests

  • added preliminary work on encapsulated data (r50)

  • added some simple files to view or work with pixel data (r46)

  • Dataset.!PixelDataArray() NumPy array changed to property Dataset.!PixelArray

  • changed to setuptools for packaging rather than distutils