Merge pull request #51 from JR-1991/include-xmlns

Include XML namespaces

pyproject.toml

@@ -9,8 +9,8 @@ packages = [{include = "sdRDM"}]
 
 [tool.poetry.dependencies]
 python = "^3.9"
-pydantic = "2.5.3"
-pydantic-xml = "^2.7.0"
+pydantic = "^2.6.0"
+pydantic-xml = "^2.9.0"
 numpy = "^1.26.3"
 pandas = "^2.1.4"
 jinja2 = "^3.1.3"
@@ -30,6 +30,7 @@ astropy = "^6.0.0"
 markdown-it-py = "^3.0.0"
 rich = "^13.7.0"
 lxml = "^5.1.0"
+python-frontmatter = "^1.1.0"
 
 [tool.poetry.group.dev.dependencies]
 coverage = "^7.4.0"

sdRDM/generator/classrender.py

@@ -14,6 +14,7 @@ def render_object(
     objects: List[Dict],
     enums: List[Dict],
     inherits: List[Dict],
+    namespaces: Dict,
     repo: Optional[str] = None,
     commit: Optional[str] = None,
     small_types: Dict = {},
@@ -31,6 +32,7 @@ def render_object(
                     objects=all_objects,
                     repo=repo,
                     commit=commit,
+                    namespaces=namespaces,
                 )
                 for subtype in small_types.values()
                 if subtype["origin"] == object["name"]
@@ -46,6 +48,7 @@ def render_object(
         objects=all_objects,
         repo=repo,
         commit=commit,
+        namespaces=namespaces,
     )
 
     methods_part = render_add_methods(
@@ -84,6 +87,7 @@ def render_class(
     object: Dict,
     inherits: List[Dict],
     objects: List[Dict],
+    namespaces: Dict,
     repo: Optional[str] = None,
     commit: Optional[str] = None,
 ) -> str:
@@ -101,6 +105,7 @@ def render_class(
     if filtered and len(filtered) == 1:
         inherit = filtered[0]["parent"]
 
+
     return template.render(
         name=name,
         inherit=inherit,
@@ -115,6 +120,7 @@ def render_class(
         ],
         repo=repo,
         commit=commit,
+        namespaces=namespaces,
     )
 
 

sdRDM/generator/codegen.py

@@ -71,13 +71,14 @@ def generate_api_from_parser(
 
     # Write classes to the directory
     write_classes(
-        libpath,
-        parser.objects,
-        parser.enums,
-        parser.inherits,
-        use_formatter,
-        url,
-        commit,
+        libpath=libpath,
+        objects=parser.objects,
+        enums=parser.enums,
+        inherits=parser.inherits,
+        use_formatter=use_formatter,
+        repo=url,
+        commit=commit,
+        namespaces=parser.namespaces,
     )
 
     # Write init files
@@ -108,6 +109,7 @@ def write_classes(
     enums: List[Dict],
     inherits: List[Dict],
     use_formatter: bool,
+    namespaces: Dict,
     repo: Optional[str] = None,
     commit: Optional[str] = None,
 ) -> None:
@@ -122,13 +124,14 @@ def write_classes(
 
     for object in objects:
         rendered = render_object(
-            object,
-            objects,
-            enums,
-            inherits,
-            repo,
-            commit,
-            small_types,
+            object=object,
+            objects=objects,
+            enums=enums,
+            inherits=inherits,
+            repo=repo,
+            commit=commit,
+            small_types=small_types,
+            namespaces=namespaces,
         )
         path = os.path.join(libpath, "core", f"{object['name'].lower()}.py")
         save_rendered_to_file(rendered, path, use_formatter)

sdRDM/generator/templates/class_template.jinja2

@@ -1,7 +1,15 @@
 @forge_signature
 class {{name}}({% if inherit is not none %}
                 {{inherit}}
-                {% else %}sdRDM.DataModel{% endif %}, {% if repo %}nsmap={"": "{{ repo }}@{{ commit }}#{{ name }}"},{% endif %}):
+                {% else %}sdRDM.DataModel{% endif %},
+                {% if namespaces %}
+                nsmap = {
+                {% for key, value in namespaces.items() %}
+                    "{{key}}": "{{value}}",
+                {% endfor %}
+                },
+                {% endif %}
+                ):
 
 
     {% if docstring is not none %}"""{{ docstring }}"""{% endif %}

sdRDM/markdown/markdownparser.py

@@ -1,5 +1,7 @@
+from io import StringIO
 import re
 
+import frontmatter
 from typing import List, Tuple, Dict, IO
 from markdown_it import MarkdownIt
 from markdown_it.token import Token
@@ -17,14 +19,17 @@ class MarkdownParser(BaseModel):
     inherits: List = []
     compositions: List = []
     external_objects: Dict = {}
+    namespaces: Dict = {}
 
     @classmethod
     def parse(cls, handle: IO):
         """Parses a given markdown file to a mermaid schema from which code is generated."""
 
         parser = cls()
 
-        doc = MarkdownIt().parse(cls.clean_html_tags(handle.readlines()))
+        content = parser.clean_html_tags(handle.readlines())
+        doc = MarkdownIt().parse(parser._remove_header(content))
+        parser.namespaces = parser._extract_namespaces(content)
         modules, enumerations = parser.get_objects_and_enumerations(doc)
 
         for module, model in modules.items():
@@ -62,6 +67,36 @@ def add_model(self, parser: "MarkdownParser"):
         self.compositions += parser.compositions
         self.external_objects.update(parser.external_objects)
 
+    @staticmethod
+    def _extract_namespaces(content: str):
+        """Extracts all namespaces from the model.
+
+        Args:
+            content (str): The content of the model.
+
+        Returns:
+            dict: A dictionary containing the extracted namespaces.
+        """
+
+        doc = frontmatter.load(StringIO(content))
+        return doc.get("xmlns", {})
+
+    @staticmethod
+    def _remove_header(text: str):
+        """
+        Removes the header from the given text.
+
+        Args:
+            text (str): The text containing the header.
+
+        Returns:
+            str: The text with the header removed.
+        """
+        header = re.search(r"^---\n.*?\n---\n", text, re.DOTALL)
+        if header:
+            text = text.replace(header.group(), "")
+        return text.strip()
+
     def _has_duplicate_object_names(self, parser):
         """Checks whether there are redundancies within the model"""
 

tests/end2end/test_creation.py

@@ -5,7 +5,7 @@
 
 
 @pytest.mark.e2e
-def test_dataset_creation(model_all, model_all_dataset):
+def test_dataset_creation(model_all):
     """Tests whether the data that is passed to the model is correctly parsed"""
 
     expected = {
@@ -70,3 +70,18 @@ def test_dataset_creation(model_all, model_all_dataset):
     assert (
         dataset.to_dict(mode="python") == expected
     ), "Dataset does not match expected values"
+
+
+@pytest.mark.e2e
+def test_ns_map(model_all):
+    """Tests whether the namespace map is correctly extracted from the model"""
+
+    expected = {
+        "": "http://www.example.com/ns0",
+        "math": "http://www.example.com/math",
+        "chem": "http://www.example.com/chem",
+    }
+
+    assert (
+        model_all.Root.__xml_nsmap__ == expected
+    ), "Namespace map does not match expected values"

tests/end2end/test_serialisation.py

@@ -29,4 +29,4 @@ def test_xml_serialisation(model_all_dataset):
     expected = open("tests/fixtures/static/model_all_expected.xml").read()
     given = model_all_dataset.xml()
 
-    assert given == expected, "YAML serialisation does not match"
+    assert given.strip() == expected.strip(), "XML serialisation does not match"

tests/fixtures/static/model_all.md

@@ -1,3 +1,10 @@
+---
+xmlns:
+    "": "http://www.example.com/ns0"
+    "math": "http://www.example.com/math"
+    "chem": "http://www.example.com/chem"
+---
+
 # Test
 
 This model tests all methods and types that are present within the sdRDM library.

tests/fixtures/static/model_all_expected.xml

@@ -1,5 +1,5 @@
 <?xml version='1.0' encoding='UTF-8'?>
-<Root id="id">
+<Root xmlns="http://www.example.com/ns0" xmlns:math="http://www.example.com/math" xmlns:chem="http://www.example.com/chem" id="id">
   <str_value>string</str_value>
   <float_value>1.5</float_value>
   <int_value>1</int_value>
@@ -27,4 +27,4 @@
       <int_value>1</int_value>
     </nested>
   </nested_multiple_obj>
-</Root>
+</Root>