Skip to content

Latest commit

 

History

History
683 lines (453 loc) · 21.2 KB

CHANGES.rst

File metadata and controls

683 lines (453 loc) · 21.2 KB
Raw

Changes

7.1 (unreleased)

  • Add support for Python 3.12.
  • Drop support for Python 3.7.

7.0.1 (2023-01-02)

  • Fix fallback when zope.i18nmessageid is not installed (regression introduced in 7.0.0).

7.0.0 (2023-01-01)

  • Add support for Python 3.11.
  • Drop support for Python 2.7, 3.5, 3.6.
  • Drop zope.schema._compat module.
  • Fix test deprecation warning on Python 3.11. (#112)

6.2.1 (2022-09-14)

  • Fix outsized integer test values that break tests on newer Python versions. (#115)

6.2.0 (2021-10-18)

  • Add support for Python 3.10.

6.1.1 (2021-10-13)

  • Fix incompatibility introduced in 6.1.0: The Bool field constructor implicitly set required to False if not given. While this is the desired behavior in most common cases, it broke special cases. See issue 104 (scroll down, it is around the reopen).

6.1.0 (2021-02-09)

  • Fix IField.required to not be required by default. See issue 104.

6.0.1 (2021-01-25)

  • Bring branch coverage to 100%.
  • Add support for Python 3.9.
  • Fix FieldUpdateEvent implementation by having an object attribute as the IFieldUpdatedEvent interfaces claims there should be.

6.0.0 (2020-03-21)

  • Require zope.interface 5.0.
  • Ensure the resolution orders of all fields are consistent and make sense. In particular, Bool fields now correctly implement IBool before IFromUnicode. See issue 80.
  • Add support for Python 3.8.
  • Drop support for Python 3.4.

5.0.1 (2020-03-06)

  • Fix: add Text.unicode_normalization = 'NFC' as default, because some are persisting schema fields. Setting that attribute only in __init__ breaks loading old objects.

5.0 (2020-03-06)

  • Set IDecimal attributes min, max and default as Decimal type instead of Number. See issue 88.
  • Enable unicode normalization for Text fields. The default is NFC normalization. Valid forms are 'NFC', 'NFKC', 'NFD', and 'NFKD'. To disable normalization, set unicode_normalization to False or None when calling __init__ of the Text field. See issue 86.

4.9.3 (2018-10-12)

  • Fix a ReST error in getDoc() results when having "subfields" with titles.

4.9.2 (2018-10-11)

  • Make sure that the title for IObject.validate_invariants is a unicode string.

4.9.1 (2018-10-05)

  • Fix SimpleTerm token for non-ASCII bytes values.

4.9.0 (2018-09-24)

  • Make NativeString and NativeStringLine distinct types that implement the newly-distinct interfaces INativeString and INativeStringLine. Previously these were just aliases for either Text (on Python 3) or Bytes (on Python 2).
  • Fix Field.getDoc() when value_type or key_type is present. Previously it could produce ReST that generated Sphinx warnings. See issue 76.
  • Make DottedName accept leading underscores for each segment.
  • Add PythonIdentifier, which accepts one segment of a dotted name, e.g., a python variable or class.

4.8.0 (2018-09-19)

  • Add the interface IFromBytes, which is implemented by the numeric and bytes fields, as well as URI, DottedName, and Id.
  • Fix passing None as the description to a field constructor. See issue 69.

4.7.0 (2018-09-11)

  • Make WrongType have an expected_type field.
  • Add NotAnInterface, an exception derived from WrongType and SchemaNotProvided and raised by the constructor of Object and when validation fails for InterfaceField.
  • Give SchemaNotProvided a schema field.
  • Give WrongContainedType an errors list.
  • Give TooShort, TooLong, TooBig and TooSmall a bound field and the common superclasses LenOutOfBounds, OrderableOutOfBounds, respectively, both of which inherit from OutOfBounds.

4.6.2 (2018-09-10)

  • Fix checking a field's constraint to set the field and value properties if the constraint raises a ValidationError. See issue 66.

4.6.1 (2018-09-10)

  • Fix the Field constructor to again allow MessageID values for the description. This was a regression introduced with the fix for issue 60. See issue 63.

4.6.0 (2018-09-07)

  • Add support for Python 3.7.
  • Object instances call their schema's validateInvariants method by default to collect errors from functions decorated with @invariant when validating. This can be disabled by passing validate_invariants=False to the Object constructor. See issue 10.
  • ValidationError can be sorted on Python 3.
  • DottedName and Id consistently handle non-ASCII unicode values on Python 2 and 3 by raising InvalidDottedName and InvalidId in fromUnicode respectively. Previously, a UnicodeEncodeError would be raised on Python 2 while Python 3 would raise the descriptive exception.
  • Field instances are hashable on Python 3, and use a defined hashing algorithm that matches what equality does on all versions of Python. Previously, on Python 2, fields were hashed based on their identity. This violated the rule that equal objects should have equal hashes, and now they do. Since having equal hashes does not imply that the objects are equal, this is not expected to be a compatibility problem. See issue 36.
  • Field instances are only equal when their .interface is equal. In practice, this means that two otherwise identical fields of separate schemas are not equal, do not hash the same, and can both be members of the same dict or set. Prior to this release, when hashing was identity based but only worked on Python 2, that was the typical behaviour. (Field objects that are not members of a schema continue to compare and hash equal if they have the same attributes and interfaces.) See issue 40.
  • Orderable fields, including Int, Float, Decimal, Timedelta, Date and Time, can now have a missing_value without needing to specify concrete min and max values (they must still specify a default value). See issue 9.
  • Choice, SimpleVocabulary and SimpleTerm all gracefully handle using Unicode token values with non-ASCII characters by encoding them with the backslashreplace error handler. See issue 15 and PR 6.
  • All instances of ValidationError have a field and value attribute that is set to the field that raised the exception and the value that failed validation.
  • Float, Int and Decimal fields raise ValidationError subclasses for literals that cannot be parsed. These subclasses also subclass ValueError for backwards compatibility.
  • Add a new exception SchemaNotCorrectlyImplemented, a subclass of WrongContainedType that is raised by the Object field. It has a dictionary (schema_errors) mapping invalid schema attributes to their corresponding exception, and a list (invariant_errors) containing the exceptions raised by validating invariants. See issue 16.
  • Add new fields Mapping and MutableMapping, corresponding to the collections ABCs of the same name; Dict now extends and specializes MutableMapping to only accept instances of dict.
  • Add new fields Sequence and MutableSequence, corresponding to the collections ABCs of the same name; Tuple now extends Sequence and List now extends MutableSequence.
  • Add new field Collection, implementing ICollection. This is the base class of Sequence. Previously this was known as AbstractCollection and was not public. It can be subclassed to add value_type, _type and unique attributes at the class level, enabling a simpler constructor call. See issue 23.
  • Make Object respect a schema attribute defined by a subclass, enabling a simpler constructor call. See issue 23.
  • Add fields and interfaces representing Python's numeric tower. In descending order of generality these are Number, Complex, Real, Rational and Integral. The Int class extends Integral, the Float class extends Real, and the Decimal class extends Number. See issue 49.
  • Make Iterable and Container properly implement IIterable and IContainer, respectively.
  • Make SimpleVocabulary.fromItems accept triples to allow specifying the title of terms. See issue 18.
  • Make TreeVocabulary.fromDict only create ITitledTokenizedTerms when a title is actually provided.
  • Make Choice fields reliably raise a ValidationError when a named vocabulary cannot be found; for backwards compatibility this is also a ValueError. Previously this only worked when the default VocabularyRegistry was in use, not when it was replaced with zope.vocabularyregistry. See issue 55.
  • Make SimpleVocabulary and SimpleTerm have value-based equality and hashing methods.
  • All fields of the schema of an Object field are bound to the top-level value being validated before attempting validation of their particular attribute. Previously only IChoice fields were bound. See issue 17.
  • Share the internal logic of Object field validation and zope.schema.getValidationErrors. See issue 57.
  • Make Field.getDoc() return more information about the properties of the field, such as its required and readonly status. Subclasses can add more information using the new method Field.getExtraDocLines(). This is used to generate Sphinx documentation when using repoze.sphinx.autointerface. See issue 60.

4.5.0 (2017-07-10)

  • Drop support for Python 2.6, 3.2, and 3.3.
  • Add support for Python 3.5 and 3.6.
  • Drop support for 'setup.py test'. Use zope.testrunner instead.

4.4.2 (2014-09-04)

  • Fix description of min max field: max value is included, not excluded.

4.4.1 (2014-03-19)

  • Add support for Python 3.4.

4.4.0 (2014-01-22)

  • Add an event on field properties to notify that a field has been updated. This event enables definition of subscribers based on an event, a context and a field. The event contains also the old value and the new value. (also see package zope.schemaevent that define a field event handler)

4.3.3 (2014-01-06)

  • PEP 8 cleanup.
  • Don't raise RequiredMissing if a field's defaultFactory returns the field's missing_value.
  • Update boostrap.py to version 2.2.
  • Add the ability to swallow ValueErrors when rendering a SimpleVocabulary, allowing for cases where vocabulary items may be duplicated (e.g., due to user input).
  • Include the field name in ConstraintNotSatisfied.

4.3.2 (2013-02-24)

  • Fix Python 2.6 support. (Forgot to run tox with all environments before last release.)

4.3.1 (2013-02-24)

  • Make sure that we do not fail during bytes decoding of term token when generated from a bytes value by ignoring all errors. (Another option would have been to hexlify the value, but that would break way too many tests.)

4.3.0 (2013-02-24)

  • Fix a bug where bytes values were turned into tokens inproperly in Python 3.
  • Add zope.schema.fieldproperty.createFieldProperties() function which maps schema fields into FieldProperty instances.

4.2.2 (2012-11-21)

  • Add support for Python 3.3.

4.2.1 (2012-11-09)

  • Fix the default property of fields that have no defaultFactory attribute.

4.2.0 (2012-05-12)

  • Automate build of Sphinx HTML docs and running doctest snippets via tox.
  • Drop explicit support for Python 3.1.
  • Introduce NativeString and NativeStringLine which are equal to Bytes and BytesLine on Python 2 and Text and TextLine on Python 3.
  • Change IURI from a Bytes string to a "native" string. This is a backwards incompatibility which only affects Python 3.
  • Bring unit test coverage to 100%.
  • Move doctests from the package and wired up as normal Sphinx documentation.
  • Add explicit support for PyPy.
  • Add support for continuous integration using tox and jenkins.
  • Drop the external six dependency in favor of a much-trimmed zope.schema._compat module.
  • Ensure tests pass when run under nose.
  • Add setup.py dev alias (runs setup.py develop plus installs nose and coverage).
  • Add setup.py docs alias (installs Sphinx and dependencies).

4.1.1 (2012-03-23)

  • Remove trailing slash in MANIFEST.in, it causes Winbot to crash.

4.1.0 (2012-03-23)

  • Add TreeVocabulary for nested tree-like vocabularies.
  • Fix broken Object field validation where the schema contains a Choice with ICountextSourceBinder source. In this case the vocabulary was not iterable because the field was not bound and the source binder didn't return the real vocabulary. Added simple test for IContextSourceBinder validation. But a test with an Object field with a schema using a Choice with IContextSourceBinder is still missing.

4.0.1 (2011-11-14)

  • Fix bug in fromUnicode method of DottedName which would fail validation on being given unicode. Introduced in 4.0.0.

4.0.0 (2011-11-09)

  • Fix deprecated unittest methods.
  • Port to Python 3. This adds a dependency on six and removes support for Python 2.5.

3.8.1 (2011-09-23)

  • Fix broken Object field validation. Previous version was using a volatile property on object field values which ends in a ForbiddenAttribute error on security proxied objects.

3.8.0 (2011-03-18)

  • Implement a defaultFactory attribute for all fields. It is a callable that can be used to compute default values. The simplest case is:

    Date(defaultFactory=datetime.date.today)

    If the factory needs a context to compute a sensible default value, then it must provide IContextAwareDefaultFactory, which can be used as follows:

    @provider(IContextAwareDefaultFactory)
def today(context):
    return context.today()

Date(defaultFactory=today)

3.7.1 (2010-12-25)

  • Rename the validation token, used in the validation of schema with Object Field to avoid infinite recursion: __schema_being_validated became _v_schema_being_validated, a volatile attribute, to avoid persistency and therefore, read/write conflicts.
  • Don't allow "[]^`" in DottedName. https://bugs.launchpad.net/zope.schema/+bug/191236

3.7.0 (2010-09-12)

  • Improve error messages when term tokens or values are duplicates.
  • Fix the buildout so the tests run.

3.6.4 (2010-06-08)

  • fix validation of schema with Object Field that specify Interface schema.

3.6.3 (2010-04-30)

  • Prefer the standard libraries doctest module to the one from zope.testing.

3.6.2 (2010-04-30)

  • Avoid maximum recursion when validating Object field that points to cycles
  • Make the dependency on zope.i18nmessageid optional.

3.6.1 (2010-01-05)

  • Allow "setup.py test" to run at least a subset of the tests runnable via bin/test (227 for setup.py test vs. 258. for bin/test)
  • Make zope.schema._bootstrapfields.ValidatedProperty descriptor work under Jython.
  • Make "setup.py test" tests pass on Jython.

3.6.0 (2009-12-22)

  • Prefer zope.testing.doctest over doctestunit.
  • Extend validation error to hold the field name.
  • Add FieldProperty class that uses Field.get and Field.set methods instead of storing directly on the instance __dict__.

3.5.4 (2009-03-25)

  • Don't fail trying to validate default value for Choice fields with IContextSourceBinder object given as a source. See https://bugs.launchpad.net/zope3/+bug/340416.
  • Add an interface for DottedName field.
  • Add vocabularyName attribute to the IChoice interface, change "vocabulary" attribute description to be more sensible, making it zope.schema.Field instead of plain zope.interface.Attribute.
  • Make IBool interface of Bool more important than IFromUnicode so adapters registered for IBool take precendence over adapters registered for IFromUnicode.

3.5.3 (2009-03-10)

  • Make Choice and Bool fields implement IFromUnicode interface, because they do provide the fromUnicode method.
  • Change package's mailing list address to zope-dev at zope.org, as zope3-dev at zope.org is now retired.
  • Fix package's documentation formatting. Change package's description.
  • Add buildout part that builds Sphinx-generated documentation.
  • Remove zpkg-related file.

3.5.2 (2009-02-04)

  • Made validation tests compatible with Python 2.5 again (hopefully not breaking Python 2.4)
  • Add an __all__ package attribute to expose documentation.

3.5.1 (2009-01-31)

  • Stop using the old old set type.
  • Make tests compatible and silent with Python 2.4.
  • Fix __cmp__ method in ValidationError. Show some side effects based on the existing __cmp__ implementation. See validation.txt
  • Make 'repr' of the ValidationError and its subclasses more sensible. This may require you to adapt your doctests for the new style, but now it makes much more sense for debugging for developers.

3.5.0a2 (2008-12-11)

  • Move zope.testing to "test" extras_require, as it is not needed for zope.schema itself.
  • Change the order of classes in SET_TYPES tuple, introduced in previous release to one that was in 3.4 (SetType, set), because third-party code could be dependent on that order. The one example is z3c.form's converter.

3.5.0a1 (2008-10-10)

  • Add the doctests to the long description.
  • Remove use of deprecated 'sets' module when running under Python 2.6.
  • Remove spurious doctest failure when running under Python 2.6.
  • Add support to bootstrap on Jython.
  • Add helper methods for schema validation: getValidationErrors and getSchemaValidationErrors.
  • zope.schema now works on Python2.5

3.4.0 (2007-09-28)

Add BeforeObjectAssignedEvent that is triggered before the object field sets a value.

3.3.0 (2007-03-15)

Corresponds to the version of the zope.schema package shipped as part of the Zope 3.3.0 release.

3.2.1 (2006-03-26)

Corresponds to the version of the zope.schema package shipped as part of the Zope 3.2.1 release.

Fix missing import of 'VocabularyRegistryError'. See http://www.zope.org/Collectors/Zope3-dev/544 .

3.2.0 (2006-01-05)

Corresponds to the version of the zope.schema package shipped as part of the Zope 3.2.0 release.

Add "iterable" sources to replace vocabularies, which are now deprecated and scheduled for removal in Zope 3.3.

3.1.0 (2005-10-03)

Corresponds to the version of the zope.schema package shipped as part of the Zope 3.1.0 release.

Allow 'Choice' fields to take either a 'vocabulary' or a 'source' argument (sources are a simpler implementation).

Add 'TimeDelta' and 'ASCIILine' field types.

3.0.0 (2004-11-07)

Corresponds to the version of the zope.schema package shipped as part of the Zope X3.0.0 release.