Add more config properties to xml input filter

ogc/na/input_filters/xml.py

@@ -1,12 +1,18 @@
 """
 XML Input filter for ingest_json.
 
-Processes XML files with [xmltodict](https://pypi.org/project/xmltodict/).
+Processes XML files with [xmltodict](https://pypi.org/project/xmltodict/). Attributes are prefixed with `_` instead
+of `@` by default.
 
 Configuration values:
 
 * `process-namespaces` (default: `False`): Whether to process and expand namespaces (see xmltodict documentation)
 * `namespaces` (default: `None`): Namespace to prefix mappings dict in `url: prefix` format.
+* `attr-prefix` (default: `_`): Prefix that will be used for attributes (in order to avoid potential clashes with
+  element names).
+* `namespace-separator` (default `:`): String that will be used to separate the namespace prefix and the local name
+  when processing namespaces.
+* `text-property` (default: `_`): property name that will be used to put the element's text content into
 """
 from __future__ import annotations
 
@@ -19,6 +25,9 @@
 DEFAULT_CONF = {
     'process-namespaces': False,
     'namespaces': None,
+    'attr-prefix': '_',
+    'namespace-separator': ':',
+    'text-property': '_',
 }
 
 
@@ -34,6 +43,10 @@ def apply_filter(content: bytes, conf: dict[str, Any] | None) -> tuple[dict[str,
     textio = StringIO(content.decode('utf-8'))
     result = xmltodict.parse(textio.read(),
                              process_namespaces=conf['process-namespaces'],
-                             namespaces=conf['namespaces'])
+                             namespaces=conf['namespaces'],
+                             attr_prefix=conf['attr-prefix'],
+                             namespace_separator=conf['namespace-separator'],
+                             cdata_key=conf['text-property'],
+                             )
 
     return result, metadata