De-duplicate element names in ILIAS crawler

This prevents any conflicts caused by multiple files with the same name.
Conflicts may still arise due to transforms, but that is out of our
control and a user error.

PFERD/crawl/ilias/kit_ilias_html.py

@@ -461,3 +461,55 @@ def _tomorrow() -> date:
 
 def _sanitize_path_name(name: str) -> str:
     return name.replace("/", "-").replace("\\", "-").strip()
+
+
+def deduplicate_element_names(elements: List[IliasPageElement]) -> List[IliasPageElement]:
+    """
+    De-duplicates element names by appending an incrementing number to later elements:
+      test.pdf
+      test.pdf
+    would result in
+      test.pdf
+      test_1.pdf
+
+    It is also space-aware:
+      "te st.pdf"
+      "te st.pdf"
+    would result in
+      "tes st.pdf"
+      "tes st 1.pdf"
+    """
+    known_names = dict()
+    result_elements = []
+
+    for element in elements:
+        # This file is new - add it and mark its name as used
+        if element.name not in known_names:
+            known_names[element.name] = 1
+            result_elements.append(element)
+            continue
+
+        # This file is a duplicate. Find a suitable suffix
+        current_counter = known_names[element.name]
+        adjusted_element = _append_number(element, current_counter)
+        # increment the counter so the next duplicate does not conflict
+        known_names[element.name] += 1
+        # also block the new name, so another file with the *renamed* name gets renamed as well
+        known_names[adjusted_element.name] = 1
+
+        result_elements.append(adjusted_element)
+
+    return result_elements
+
+
+def _append_number(element: IliasPageElement, number: int) -> IliasPageElement:
+    extension_index = element.name.rfind(".")
+    suffix = f" {number}" if " " in element.name else f"_{number}"
+    if extension_index < 0:
+        new_name = element.name + suffix
+    else:
+        new_name = element.name[:extension_index] + suffix + element.name[extension_index:]
+
+    return IliasPageElement(
+        element.type, element.url, new_name, element.mtime, element.description
+    )