xmlschema

• Fix for issues #213, #214, #215 and #218

• Code cleaning and optimizations on schema components

• Reducing and grouping helper functions

Hotfix release for packaging (issue #210)

Improve support for XML resources

• Drop support for Python 3.5
• Add XmlDocument and Wsdl11Document classes
• Refactoring of XMLResource to support ElementTree-like XPath API on both full and lazy modes

• Add schema export API to schema and global maps (issue #187)

• Fix decoding with lax/skip validation modes (issue #204)

• Add keep_unknown optional argument for iter_decode() methods

The internal XML regular expressions parser has been moved to XPath package elementpath (v2.0.2+). From this release the regex patterns are parsed through elementpath's API translate_pattern().

All W3C tests now run successful. Only W3C tests for unavailable features (XML 1.1) or unimplemented features (dynamic schema load) are still failing. This release requires elementpath v2.0.x.

• Fix XPath context for schema nodes

• Fix XPath parser and context for identities

• Fix content type classification (issue #195)

• Make sandbox mode more explicit (PR #191)

• Allow alphanumeric prefixes for the base converter

• Fix XPath issues with default namespace

• Fix W3C tests on XSD identities

• Add ColumnarConverter class

• Add command-line interface utility for document API

• Fix a stable public API for XSD types, elements and attributes

• Add security modes for accessing URLs

• Clean component parsing

• Fix namespace loading for chameleon schemas

• Fix UPA checks with nested choice/all models

• Fixed issues #182 and #183

• Extension of validation tests with XMLSchema11 validator

• Fixed several bugs

• Extended testing with Travis CI

• Change of skip validation mode with errors filtering in decode() or encode()

• Extension of location hints by argument to imported/included schemas

• Fixed lazy validation with identity constraints

• Fixed many W3C instance tests (remain ~100 over 15344 tests)

First Python 3-only release. Unlike the announcements the XMLSchema10 has been kept as default validator for avoiding compatibility problems. It's only a matter of names. For XSD 1.1 still use the class XMLSchema11.

Next release (v1.1.0) will be Python 3.5+ only.

• Enhancement of validation-only speed (~15%)

• Added is_valid() and iter_errors() to module API

• Improved XMLResource class for working with compressed files

• Fix for validation with XSD wildcards and 'lax' process content

• Fix ambiguous items validation for xs:choice and xs:sequence models

• Improved XPath 2.0 bindings

• Added logging for schema initialization and building (handled with argument loglevel)

• Update encoding of collapsed contents with a new model based reordering method

• Removed XLink namespace from meta-schema (loaded from a fallback location like XHTML)

• Fixed half of failed W3C instance tests (remain 255 over 15344 tests)

This is the first release with XSD 1.1 support. XSD 1.0/1.1 are tested with W3C XSD 1.1 suite (almost 100% of tests passed). XML validation is also tested with W3C suite (~97% of tests passed).

• Fix path normalization and tests for Windows platform

• Added XML resource validation in lazy mode (experimental feature)

• Added arguments filler and fill_missing to XSD decode/encode methods

• Added arguments preserve_root, strip_namespaces, force_dict and force_list to XMLSchemaConverter

• Added code coverage and pep8 testing

• Drop support for Python 3.4

• Added a script for running the W3C XSD test suite.

• Check restrictions and model groups UPA violations

• Model groups splitted between two modules for more focusing on models basics

• Added two new exceptions for model group errors

• More control on imported namespaces

• Added use_meta argument to schema classes

• Added includes list and imports dict to schema classes

• Many fixes for passing the W3C's tests for XSD 1.0 schemas

• Added a test for issue #105 and a fix for issue #103

• Fixed Element type mismatch issue when apply SafeXMLParser to schema resources

• More XSD 1.1 features implemented (open content and versioning namespace are missing)

• Programmatic import of ElementTree for avoid module mismatches

• Cleaning and refactoring of test scripts

• Dependency defusedxml package replaced by a custom XMLParser for ElementTree

• Optional decoding of XSD date/time/duration builtin types

• Fixes for issues #93, #96, #97 and #99

• Fixes for issues #87 and #88

• Merged with PR #89 (simpleType restriction annotation parsing)

• XSD 1.1 development: added assertion facet (still to be completed)

• Fixes for issues #85 and #86

• XSD 1.1 development: added explicitTimezone facet and XSD 1.1 builtin types

• Fix for issue #82 and for similar unprotected XSD component lookups

• Added checks for namespace mapping of encoded trees and error messages

• Unification of XSD group decode and encode methods

• Children validation error class improved

• Fixes for issues #77, #79 and #80

• Added test scripts for helpers and ElementTree

• Improved model validation for XSD groups encoding

• Added parent reference to XSD components

• Extended validator errors classes

• Optimized error generation using helper methods

• Improved particle parsing

• Improved ElementTree and XPath API

• Validated data encoding to XML

• Improved converters with decoding/encoding of namespace information

• Added helper functions for encoding and decoding to JSON

• Added XMLResource class for managing access to XML data sources

• Added warnings for failed schema includes and namespace imports

• Schema serialization with pickle for Python 3 (enhancement related to issue #68)

• Data encoding with the default converter

• Improved decoding for xs:union

• First experimental version of data encoding with the default converter

• Fixes for issues #65, #66 and #67

• Extended the tests on lxml XML data

• Fixes for issues #61, #63 and #64

• Encoding of XSD builtin types (strings and numerical)

• Fix for issue #62

• Drop support for Python 3.3

• Add support for preventing XML attacks with the use of the defusedxml package (added defuse argument to schemas)

• Fix for group circularity (issue #58)

• Fix for billion laughs attacks using XSD groups expansion

• Added checks for model restrictions

• Removed XsdAnnotated class

• Added XsdType class as common class for XSD types

• Fixes for issues #55 and #56

• Added XPath 1.0/2.0 full parsing with the derived elementpath package

• Fixes for issues #52 and #54

• Test package improved (tox.ini, other checks with test_package.py)

• Fixes for issues #45, #46, #51

• Added kwargs to iter_decode(), dict_class and list_class arguments have been removed

• Added kwargs to converters initialization in order to push variable keyword arguments from iter_decode()

• Substitution groups support

• Added fetch_schema_locations function to API

• Added locations argument to fetch_schema, validate and to_dict API functions

• A more useful __repr__ for XSD component classes

• Fixes for issues #35, #38, #39

• Fixed issue #34 (min_occurs == 0 check in XsdGroup.is_emptiable)

• Updated copyright information

• Updated schema class creation (now use a metaclass)

• Added index and expected attributes to XMLSchemaChildrenValidationError

• Added locations optional argument to XMLSchema class

• Key/Unique/Keyref constraints partially rewritten

• Fixed ad issue with UCS-2/4 and maxunicode

• UnicodeSubset class rewritten (more speed, less memory)

• Updated unicode_categories.json to Python 3.6 unicodedata

• Added XMLSchemaChildrenValidationError exception

• Some bug fixes

• Code cleaning

• XSD components modules has been merged with schema's modules into 'validators' subpackage

• Improved test scripts with a SchemaObserver class and test line arguments

• Full support for date and time XSD builtin types

• Added identity constraints

• Some bug fix

• Factories code moved to XsdComponent subclasses for simplify parsing and debugging

• All XSD components built from ElementTree elements with a lazy approach

• Implementation of the XSD validation modes ('strict'/'lax'/'skip') both for validating schemas and for validating/decoding XML files

• Defined an XsdBaseComponent class as the common base class for all XSD components, schemas and global maps

• Defined a ValidatorMixin for sharing a common API between validators/decoders classes

• Added built and validity checks for all XSD components

• Added converters for decode/encode data with different conventions

• Modifications on iter_decode() arguments in order to use converters

• Added notations and substitution groups

• Created a subpackage for XSD components

• Documentation extended and tested

• Improved tests for XPath, validation and decoding