- Used to import other namespaces explicitly (as disposed by
"xsi:schemaLocation" in an XML instance).
- It's not propagated to included/imported schemas anymore.
- Add loglevel argument to schema init
- Fallback location (for XLink and XHTML) moved at the
end of import tentatives (full fix for issue #137)
- Fix TestGlobalMaps after the remove of XLink from base schemas
- Add custom parser for assertion facet (without position() and last()).
- Move parser initialization to XsdAssert.parse_xpath_test() because
all the components must be defined.
- Add iter_substitutes() to Xsd11Element to fix upa.xsd/upa2.xsd
tests with instances. Now for XsdElement the abstract substitutes
are simply ignored by iter_substitutes().
- ID are not counted for XML document root
- IDREF checked with adding a key with 0 value if missing
(TODO: Add a validation error instance or a reference)
- Put interleave mode after model match for precedence
- Fix regex parsing of character group ending with '-'
- Fix complexType mixed content extension
- Added XMLSchemaNamespaceError for namespace related errors
- version_check moved to schema class
- resolve_qname() now raises KeyError if namespace prefix is not found
- resolve_qname() now raises XMLSchemaNamespaceError if a namespace
is mapped with a prefix but is not loaded by an import
- Checked and improved 'built' properties for XSD components:
* Simple types, attributes and attribute group are now built
withoug cheching on base types.
* Don't check global subcomponents: the scope of built is to
assure that local parts are built.
- Split global maps checks from build phase:
* XsdGlobals._check_schema substituted by check() method with
two opional arguments. The new methos is useful also to check
global maps also after schema building.
- Added validator=None argument to parse_error() method
- In XMLSchema.__init__() the errors generated by check_schema()
are now raised or transformed to parse errors
- Only parse errors are generated during schema building
- More compact errors that omit meta-schema parts
- Keep UnorderedConverter class as the default converter for
encoding unordered data
- Added 'unordered=False' option to iter_encode's arguments
- Added iter_unordered_content() and iter_collapsed_content()
to ModelVisitor class
- Meta schema lazy build: reduce memory usage for meta-schema if
only one schema class is used between XMLSchema10 and XMLSchema11
- Improve import_package() in check_memory.py
- Split decode()/encode() for components and for schemas
- Removed to_dict and to_etree for XSD components
- Updated fetch_schema_locations() to build XMLResource instance
- Added XMLResource.iterfind() for XPath iteration of a resource;
- Validator API refactored: remove path argument from iter_errors()
of components, add validate, is_valid, iter_errors to XMLSchema
class with additional arguments path and schema_path.
- Fix test case patterns.xml (now find also duplicated IDs)
Davide Brunato