Add NX docstring as attribute

.gitmodules

@@ -1,3 +1,3 @@
 [submodule "src/pynxtools/definitions"]
 	path = src/pynxtools/definitions
-	url = https://github.com/FAIRmat-NFDI/nexus_definitions.git
+	url = https://github.com/FAIRmat-NFDI/nexus_definitions.git

src/pynxtools/data/NXtest.nxdl.xml

@@ -13,7 +13,9 @@
         <field name="program_name"/>
         <field name="definition">
             <doc>This is a dummy NXDL to test out the dataconverter.</doc>
-            <attribute name="version"/>
+            <attribute name="version">
+                <doc>This is the version of the definition.</doc>
+            </attribute>
             <enumeration>
                 <item value="NXTEST"/>
                 <item value="NXtest"/>

src/pynxtools/dataconverter/convert.py

@@ -237,7 +237,13 @@ def convert(
     )
 
     helpers.add_default_root_attributes(data=data, filename=os.path.basename(output))
-    Writer(data=data, nxdl_f_path=nxdl_f_path, output_path=output).write()
+
+    write_docs = kwargs.pop("write_docs", False)
+    docs_format = kwargs.pop("docs_format", None)
+    Writer(data=data, nxdl_f_path=nxdl_f_path, output_path=output).write(
+        write_docs=write_docs,
+        docs_format=docs_format,
+    )
 
     logger.info(f"The output file generated: {output}.")
 
@@ -350,7 +356,21 @@ def main_cli():
     default=None,
     help="A json config file for the reader",
 )
-# pylint: disable=too-many-arguments
+@click.option(
+    "--write-docs",
+    is_flag=True,
+    default=False,
+    help="Write docs for the individual NeXus concepts as HDF5 attributes.",
+)
+@click.option(
+    "--docs-format",
+    type=click.Choice(["default", "html", "html5", "xml", "pseudoxml"]),
+    default=None,
+    help=(
+        "Optionally specify the format in which the docs for the individual NeXus concepts is generated. "
+        "By default, the docs are formatted as in the NXDL file."
+    ),
+)
 def convert_cli(
     files: Tuple[str, ...],
     input_file: Tuple[str, ...],
@@ -363,6 +383,8 @@ def convert_cli(
     mapping: str,
     config_file: str,
     fail: bool,
+    write_docs: bool,
+    docs_format: str,
     **kwargs,
 ):
     """This command allows you to use the converter functionality of the dataconverter."""
@@ -390,6 +412,18 @@ def convert_cli(
     if config_file:
         kwargs["config_file"] = config_file
 
+    if write_docs:
+        kwargs["write_docs"] = write_docs
+        if not docs_format:
+            kwargs["docs_format"] = "default"
+        else:
+            kwargs["docs_format"] = docs_format
+
+    elif docs_format is not None:
+        raise click.UsageError(
+            "Error: --docs-format can only be used with --write-docs."
+        )
+
     file_list = []
     for file in files:
         if os.path.isdir(file):

src/pynxtools/dataconverter/helpers.py

@@ -790,6 +790,64 @@ def get_concept_basepath(path: str) -> str:
     return "/" + "/".join(concept_path)
 
 
+def get_concept_path_from_elem(elem: ET.Element) -> str:
+    """
+    Process individual XML element to generate the NeXus concept path.
+
+    Output is e.g. "NXexperiment:/NXentry/NXinstrument/NXdetector".
+    """
+
+    name = elem.attrib.get("name", "")
+    elem_type = elem.attrib.get("type", "")
+    nxdlbase = elem.attrib.get("nxdlbase", "")  # .split("/")[-1]
+    nxdlbase_class = elem.attrib.get("nxdlbase_class", "")
+    nxdlpath = elem.attrib.get("nxdlpath", "")
+    category = elem.attrib.get("category", "")
+    # optional = elem.attrib.get("optional", "")
+    # extends = elem.attrib.get("extends", "")
+
+    # print(f"tag: {tag}")
+    # print(f"name: {name}")
+    # print(f"elem_type: {elem_type}")
+    # print(f"nxdlbase: {nxdlbase}")
+    # print(f"nxdlbase_class: {nxdlbase_class}")
+    # print(f"nxdlpath: {nxdlpath}")
+    # # print(f"optional: {optional}")
+    # # print(f"extends: {extends}")
+    # print("\n")
+
+    concept_path = ""
+
+    if elem.tag.endswith("group"):
+        if nxdlbase_class and nxdlbase_class == "application":
+            concept_path += "NXmpes:"
+            concept_path += nxdlpath  # + = f"(elem_type)"
+
+        else:
+            if nxdlbase:
+                concept_path += nxdlbase.replace(".nxdl.xml", "").split(os.path.sep)[-1]
+            concept_path += nxdlpath  # + = f"(elem_type)"
+
+    elif elem.tag.endswith("field"):
+        pass
+
+    elif elem.tag.endswith("attribute"):
+        pass
+    elif elem.tag.endswith("definition"):
+        concept_path += name
+
+    return concept_path
+
+    # if nxdlpath:
+    #     # Split the nxdlpath and construct the string
+    #     path_parts = nxdlpath.strip("/").split("/")
+    #     formatted_path = "/".join(path_parts)
+    #     return f"{formatted_path}({elem_type})"
+    # else:
+    #     # For elements with no path, return the name and type
+    #     return f"{name}({elem_type})"
+
+
 def remove_namespace_from_tag(tag):
     """Helper function to remove the namespace from an XML tag."""