Merge branch 'develop' for upgrade to v1.0.17

CHANGELOG.rst

@@ -2,6 +2,11 @@
 CHANGELOG
 *********
 
+`v1.0.17`_ (2019-12-22)
+=======================
+* Enhancement of validation-only speed (~15%)
+* Added *is_valid()* and *iter_errors()* to module API
+
 `v1.0.16`_ (2019-11-18)
 =======================
 * Improved XMLResource class for working with compressed files
@@ -272,3 +277,4 @@ v0.9.6 (2017-05-05)
 .. _v1.0.14: https://github.com/brunato/xmlschema/compare/v1.0.13...v1.0.14
 .. _v1.0.15: https://github.com/brunato/xmlschema/compare/v1.0.14...v1.0.15
 .. _v1.0.16: https://github.com/brunato/xmlschema/compare/v1.0.15...v1.0.16
+.. _v1.0.17: https://github.com/brunato/xmlschema/compare/v1.0.16...v1.0.17

doc/api.rst

@@ -7,6 +7,8 @@ Document level API
 ------------------
 
 .. autofunction:: xmlschema.validate
+.. autofunction:: xmlschema.is_valid
+.. autofunction:: xmlschema.iter_errors
 .. autofunction:: xmlschema.to_dict
 .. autofunction:: xmlschema.to_json
 .. autofunction:: xmlschema.from_json
@@ -47,6 +49,7 @@ Schema level API
     .. automethod:: create_schema
     .. automethod:: create_any_content_group
     .. automethod:: create_any_attribute_group
+    .. automethod:: create_any_type
 
     .. automethod:: get_locations
     .. automethod:: include_schema

doc/conf.py

@@ -62,7 +62,7 @@
 # The short X.Y version.
 version = '1.0'
 # The full version, including alpha/beta/rc tags.
-release = '1.0.16'
+release = '1.0.17'
 
 # The language for content autogenerated by Sphinx. Refer to documentation
 # for a list of supported languages.

publiccode.yml

@@ -6,8 +6,8 @@ publiccodeYmlVersion: '0.2'
 name: xmlschema
 url: 'https://github.com/sissaschool/xmlschema'
 landingURL: 'https://github.com/sissaschool/xmlschema'
-releaseDate: '2019-11-18'
-softwareVersion: v1.0.16
+releaseDate: '2019-12-22'
+softwareVersion: v1.0.17
 developmentStatus: stable
 platforms:
   - linux

setup.py

@@ -38,7 +38,7 @@ def run(self):
 
 setup(
     name='xmlschema',
-    version='1.0.16',
+    version='1.0.17',
     setup_requires=['elementpath~=1.3.0'],
     install_requires=['elementpath~=1.3.0'],
     packages=['xmlschema'],

xmlschema/__init__.py

@@ -21,7 +21,7 @@
     ElementData, XMLSchemaConverter, UnorderedConverter, ParkerConverter,
     BadgerFishConverter, AbderaConverter, JsonMLConverter
 )
-from .documents import validate, to_dict, to_json, from_json
+from .documents import validate, is_valid, iter_errors, to_dict, to_json, from_json
 
 from .validators import (
     XMLSchemaValidatorError, XMLSchemaParseError, XMLSchemaNotBuiltError,
@@ -31,7 +31,7 @@
     XsdGlobals, XMLSchemaBase, XMLSchema, XMLSchema10, XMLSchema11
 )
 
-__version__ = '1.0.16'
+__version__ = '1.0.17'
 __author__ = "Davide Brunato"
 __contact__ = "[email protected]"
 __copyright__ = "Copyright 2016-2019, SISSA"

xmlschema/converters.py

@@ -264,7 +264,8 @@ def element_decode(self, data, xsd_element, level=0):
         if level == 0 and xsd_element.is_global() and not self.strip_namespaces and self:
             schema_namespaces = set(xsd_element.namespaces.values())
             result_dict.update(
-                ('%s:%s' % (self.ns_prefix, k) if k else self.ns_prefix, v) for k, v in self.items()
+                ('%s:%s' % (self.ns_prefix, k) if k else self.ns_prefix, v)
+                for k, v in self._namespaces.items()
                 if v in schema_namespaces or v == XSI_NAMESPACE
             )
 
@@ -325,6 +326,8 @@ def element_encode(self, obj, xsd_element, level=0):
         if not isinstance(obj, (self.dict, dict)):
             if xsd_element.type.is_simple() or xsd_element.type.has_simple_content():
                 return ElementData(tag, obj, None, {})
+            elif xsd_element.type.mixed and not isinstance(obj, list):
+                return ElementData(tag, obj, None, {})
             else:
                 return ElementData(tag, None, obj, {})
 
@@ -900,7 +903,9 @@ def element_decode(self, data, xsd_element, level=0):
             ])
 
         if level == 0 and xsd_element.is_global() and not self.strip_namespaces and self:
-            attributes.update([('xmlns:%s' % k if k else 'xmlns', v) for k, v in self.items()])
+            attributes.update(
+                [('xmlns:%s' % k if k else 'xmlns', v) for k, v in self._namespaces.items()]
+            )
         if attributes:
             result_list.insert(1, attributes)
         return result_list
@@ -913,7 +918,7 @@ def element_encode(self, obj, xsd_element, level=0):
 
         data_len = len(obj)
         if data_len == 1:
-            if not xsd_element.is_matching(unmap_qname(obj[0]), self.get('')):
+            if not xsd_element.is_matching(unmap_qname(obj[0]), self._namespaces.get('')):
                 raise XMLSchemaValueError("Unmatched tag")
             return ElementData(xsd_element.name, None, None, attributes)
 
@@ -930,7 +935,7 @@ def element_encode(self, obj, xsd_element, level=0):
         else:
             content_index = 2
 
-        if not xsd_element.is_matching(unmap_qname(obj[0]), self.get('')):
+        if not xsd_element.is_matching(unmap_qname(obj[0]), self._namespaces.get('')):
             raise XMLSchemaValueError("Unmatched tag")
 
         if data_len <= content_index:

xmlschema/documents.py

@@ -13,17 +13,20 @@
 
 from .compat import ordered_dict_class
 from .resources import fetch_schema_locations, XMLResource
-from .validators.schema import XMLSchema, XMLSchemaBase
+from .validators import XMLSchema, XMLSchemaBase
 
 
 def get_context(source, schema=None, cls=None, locations=None, base_url=None,
                 defuse='remote', timeout=300, lazy=False):
     """
-    Helper method for obtaining XML document validation/decode context.
-    Return an XMLResource instance and a schema instance.
+    Get the XML document validation/decode context.
+
+    :return: an XMLResource instance and a schema instance.
     """
     if cls is None:
         cls = XMLSchema
+    if not isinstance(source, XMLResource):
+        source = XMLResource(source, base_url, defuse=defuse, timeout=timeout, lazy=lazy)
 
     try:
         schema, locations = fetch_schema_locations(source, locations, base_url=base_url)
@@ -36,9 +39,6 @@ def get_context(source, schema=None, cls=None, locations=None, base_url=None,
     else:
         schema = cls(schema, validation='strict', locations=locations, defuse=defuse, timeout=timeout)
 
-    if not isinstance(source, XMLResource):
-        source = XMLResource(source, defuse=defuse, timeout=timeout, lazy=lazy)
-
     return source, schema
 
 
@@ -75,6 +75,26 @@ def validate(xml_document, schema=None, cls=None, path=None, schema_path=None, u
     schema.validate(source, path, schema_path, use_defaults, namespaces)
 
 
+def is_valid(xml_document, schema=None, cls=None, path=None, schema_path=None, use_defaults=True,
+             namespaces=None, locations=None, base_url=None, defuse='remote', timeout=300, lazy=False):
+    """
+    Like :meth:`validate` except that do not raises an exception but returns ``True`` if
+    the XML document is valid, ``False`` if it's invalid.
+    """
+    source, schema = get_context(xml_document, schema, cls, locations, base_url, defuse, timeout, lazy)
+    return schema.is_valid(source, path, schema_path, use_defaults, namespaces)
+
+
+def iter_errors(xml_document, schema=None, cls=None, path=None, schema_path=None, use_defaults=True,
+                namespaces=None, locations=None, base_url=None, defuse='remote', timeout=300, lazy=False):
+    """
+    Creates an iterator for the errors generated by the validation of an XML document.
+    Takes the same arguments of the function :meth:`validate`.
+    """
+    source, schema = get_context(xml_document, schema, cls, locations, base_url, defuse, timeout, lazy)
+    return schema.iter_errors(source, path, schema_path, use_defaults, namespaces)
+
+
 def to_dict(xml_document, schema=None, cls=None, path=None, process_namespaces=True,
             locations=None, base_url=None, defuse='remote', timeout=300, lazy=False, **kwargs):
     """

xmlschema/etree.py

@@ -25,7 +25,7 @@
 from .compat import PY3
 from .exceptions import XMLSchemaTypeError
 from .namespaces import XSLT_NAMESPACE, HFP_NAMESPACE, VC_NAMESPACE, get_namespace
-from .qnames import get_qname, qname_to_prefixed
+from .qnames import get_qname, qname_to_prefixed, XSI_SCHEMA_LOCATION, XSI_NONS_SCHEMA_LOCATION
 
 ###
 # Programmatic import of xml.etree.ElementTree
@@ -263,6 +263,18 @@ def etree_getpath(elem, root, namespaces=None, relative=True, add_position=False
             return path
 
 
+def etree_iter_location_hints(elem):
+    """Yields schema location hints contained in the attributes of an element."""
+    if XSI_SCHEMA_LOCATION in elem.attrib:
+        locations = elem.attrib[XSI_SCHEMA_LOCATION].split()
+        for ns, url in zip(locations[0::2], locations[1::2]):
+            yield ns, url
+
+    if XSI_NONS_SCHEMA_LOCATION in elem.attrib:
+        for url in elem.attrib[XSI_NONS_SCHEMA_LOCATION].split():
+            yield '', url
+
+
 def etree_elements_assert_equal(elem, other, strict=True, skip_comments=True):
     """
     Tests the equality of two XML Element trees.

xmlschema/namespaces.py

@@ -119,42 +119,74 @@ def clear(self):
 
 class NamespaceMapper(MutableMapping):
     """
-    A class to map/unmap namespace prefixes to URIs. The
+    A class to map/unmap namespace prefixes to URIs. The mapped namespaces are
+    automatically registered when set. Namespaces can be updated overwriting
+    the existing registration or inserted using an alternative prefix.
 
-    :param namespaces: Initial data with namespace prefixes and URIs.
+    :param namespaces: initial data with namespace prefixes and URIs.
+    :param register_namespace: a two-arguments function for registering namespaces \
+    on ElementTree module.
     """
     def __init__(self, namespaces=None, register_namespace=None):
         self._namespaces = {}
         self.register_namespace = register_namespace
         if namespaces is not None:
-            self.update(namespaces)
+            self._namespaces.update(namespaces)
 
-    def __getitem__(self, key):
-        return self._namespaces[key]
+    def __getitem__(self, prefix):
+        return self._namespaces[prefix]
 
-    def __setitem__(self, key, value):
-        self._namespaces[key] = value
+    def __setitem__(self, prefix, uri):
+        self._namespaces[prefix] = uri
         try:
-            self.register_namespace(key, value)
+            self.register_namespace(prefix, uri)
         except (TypeError, ValueError):
             pass
 
-    def __delitem__(self, key):
-        del self._namespaces[key]
+    def __delitem__(self, prefix):
+        del self._namespaces[prefix]
 
     def __iter__(self):
         return iter(self._namespaces)
 
     def __len__(self):
         return len(self._namespaces)
 
+    @property
+    def namespaces(self):
+        return self._namespaces
+
     @property
     def default_namespace(self):
         return self._namespaces.get('')
 
     def clear(self):
         self._namespaces.clear()
 
+    def insert_item(self, prefix, uri):
+        """
+        A method for setting an item that checks the prefix before inserting.
+        In case of collision the prefix is changed adding a numerical suffix.
+        """
+        if not prefix:
+            if '' not in self._namespaces:
+                self._namespaces[prefix] = uri
+                return
+            elif self._namespaces[''] == uri:
+                return
+            prefix = 'default'
+
+        while prefix in self._namespaces:
+            if self._namespaces[prefix] == uri:
+                return
+            match = re.search(r'(\d+)$', prefix)
+            if match:
+                index = int(match.group()) + 1
+                prefix = prefix[:match.span()[0]] + str(index)
+            else:
+                prefix += '0'
+        self._namespaces[prefix] = uri
+
     def map_qname(self, qname):
         """
         Converts an extended QName to the prefixed format. Only registered
@@ -170,7 +202,7 @@ def map_qname(self, qname):
             return qname
 
         qname_uri = get_namespace(qname)
-        for prefix, uri in self.items():
+        for prefix, uri in self._namespaces.items():
             if uri != qname_uri:
                 continue
             if prefix: