feat:! Adds pointer tag parsing implementation in parse_xml

xblock/core.py

@@ -24,6 +24,7 @@
 from xblock.internal import class_lazy
 from xblock.plugin import Plugin
 from xblock.validation import Validation
+from xblock.xml import is_pointer_tag
 
 # OrderedDict is used so that namespace attributes are put in predictable order
 # This allows for simple string equality assertions in tests and have no other effects
@@ -32,6 +33,7 @@
     ("block", "http://code.edx.org/xblock/block"),
 ])
 
+
 # __all__ controls what classes end up in the docs.
 __all__ = ['XBlock', 'XBlockAside']
 
@@ -746,6 +748,9 @@
             keys (:class:`.ScopeIds`): The keys identifying where this block
                 will store its data.
         """
+        if is_pointer_tag(node):
+            node, _ = runtime.load_definition_xml(node, keys.def_id)
+
         block = runtime.construct_xblock_from_class(cls, keys)
 
         # The base implementation: child nodes become child blocks.

xblock/runtime.py

@@ -31,6 +31,7 @@
     FieldDataDeprecationWarning,
     UserIdDeprecationWarning,
 )
+from xblock.xml import format_filepath, load_file, name_to_pathname
 
 log = logging.getLogger(__name__)
 
@@ -792,6 +793,24 @@
         child = etree.SubElement(node, "unknown")
         block.add_xml_to_node(child)
 
+    def load_definition_xml(self, node, def_id):
+        """
+        Load an XML definition for a block.
+        This method can be overridden in different runtime environments.
+
+        Args:
+            node: XML element containing attributes for definition loading
+            def_id: Unique identifier for the definition being loaded
+
+        Returns:
+            tuple: A tuple containing the loaded XML definition and the
+            corresponding file path or identifier
+        """
+        url_name = node.get('url_name')
+        filepath = format_filepath(node.tag, name_to_pathname(url_name))
+        definition_xml = load_file(filepath, self.resources_fs, def_id)  # pylint: disable=no-member
+        return definition_xml, filepath
+
     # Rendering
 
     def render(self, block, view_name, context=None):

xblock/test/test_xml.py

@@ -0,0 +1,63 @@
+"""
+Tests for xblock/xml.py
+"""
+
+import unittest
+from io import BytesIO
+from lxml import etree
+
+from xblock.xml import (
+    name_to_pathname,
+    is_pointer_tag,
+    format_filepath,
+    file_to_xml,
+)
+
+
+class TestPointerTagParsing(unittest.TestCase):
+    """
+    Tests for core functions in XBlock.
+    """
+    def test_name_to_pathname(self):
+        self.assertEqual(name_to_pathname("course:subcourse"), "course/subcourse")
+        self.assertEqual(name_to_pathname("module:lesson:part"), "module/lesson/part")
+        self.assertEqual(name_to_pathname("no_colon"), "no_colon")
+
+    def test_is_pointer_tag(self):
+        # Case 1: Valid pointer tag
+        xml_obj = etree.Element("some_tag", url_name="test_url")
+        self.assertTrue(is_pointer_tag(xml_obj))
+
+        # Case 2: Valid course pointer tag
+        xml_obj = etree.Element("course", url_name="test_url", course="test_course", org="test_org")
+        self.assertTrue(is_pointer_tag(xml_obj))
+
+        # Case 3: Invalid case - extra attribute
+        xml_obj = etree.Element("some_tag", url_name="test_url", extra_attr="invalid")
+        self.assertFalse(is_pointer_tag(xml_obj))
+
+        # Case 4: Invalid case - has text
+        xml_obj = etree.Element("some_tag", url_name="test_url")
+        xml_obj.text = "invalid_text"
+        self.assertFalse(is_pointer_tag(xml_obj))
+
+        # Case 5: Invalid case - has children
+        xml_obj = etree.Element("some_tag", url_name="test_url")
+        _ = etree.SubElement(xml_obj, "child")
+        self.assertFalse(is_pointer_tag(xml_obj))
+
+    def test_format_filepath(self):
+        self.assertEqual(format_filepath("course", "test_url"), "course/test_url.xml")
+
+    def test_file_to_xml(self):
+        """Test that `file_to_xml` correctly parses XML from a file object."""
+        # Create a BytesIO object
+        file_obj = BytesIO(b"<root><child>Value</child></root>")
+
+        # Parse the XML
+        result = file_to_xml(file_obj)
+
+        # Verify the result
+        self.assertEqual(result.tag, 'root')
+        self.assertEqual(result[0].tag, 'child')
+        self.assertEqual(result[0].text, 'Value')

xblock/xml.py

@@ -0,0 +1,71 @@
+"""
+XML utilities for XBlock, including parsing and loading XML definitions.
+"""
+from lxml import etree
+
+
+XML_PARSER = etree.XMLParser(dtd_validation=False, load_dtd=False, remove_blank_text=True, encoding='utf-8')
+
+
+def name_to_pathname(name):
+    """
+    Convert a location name for use in a path: replace ':' with '/'.
+    This allows users of the xml format to organize content into directories
+    """
+    return name.replace(':', '/')
+
+
+def is_pointer_tag(xml_obj):
+    """
+    Check if xml_obj is a pointer tag: <blah url_name="something" />.
+    No children, one attribute named url_name, no text.
+
+    Special case for course roots: the pointer is
+      <course url_name="something" org="myorg" course="course">
+
+    xml_obj: an etree Element
+
+    Returns a bool.
+    """
+    if xml_obj.tag != "course":
+        expected_attr = {'url_name'}
+    else:
+        expected_attr = {'url_name', 'course', 'org'}
+
+    actual_attr = set(xml_obj.attrib.keys())
+
+    has_text = xml_obj.text is not None and len(xml_obj.text.strip()) > 0
+
+    return len(xml_obj) == 0 and actual_attr == expected_attr and not has_text
+
+
+def format_filepath(category, name):
+    """
+    Construct a formatted filepath string based on the given category and name.
+    """
+    return f'{category}/{name}.xml'
+
+
+def load_file(filepath, fs, def_id):  # pylint: disable=invalid-name
+    """
+    Open the specified file in fs, and call cls.file_to_xml on it,
+    returning the lxml object.
+
+    Add details and reraise on error.
+    """
+    try:
+        with fs.open(filepath) as xml_file:
+            return file_to_xml(xml_file)
+    except Exception as err:
+        # Add info about where we are, but keep the traceback
+        raise Exception(f'Unable to load file contents at path {filepath} for item {def_id}: {err}') from err
+
+
+def file_to_xml(file_object):
+    """
+    Used when this module wants to parse a file object to xml
+    that will be converted to the definition.
+
+    Returns an lxml Element
+    """
+    return etree.parse(file_object, parser=XML_PARSER).getroot()