feat: normalize/sanitize Anwendungsfall.beschreibung

src/fundamend/reader/ahbreader.py

@@ -31,7 +31,7 @@
     _is_segment_group,
     _is_uebertragungsdatei,
 )
-from fundamend.utils import lstrip, strip
+from fundamend.utils import lstrip, remove_linebreaks_and_hyphens, strip
 
 # pylint:disable=duplicate-code
 # yes, it's very similar to the MigReader
@@ -252,7 +252,7 @@ def _read_anwendungsfall(self, original_element: ET.Element) -> Anwendungsfall:
             format_element = next((child for child in original_element[0] if child.tag.startswith("M_")))
         return Anwendungsfall(
             pruefidentifikator=original_element.attrib["Pruefidentifikator"],
-            beschreibung=original_element.attrib["Beschreibung"].strip(),
+            beschreibung=remove_linebreaks_and_hyphens(original_element.attrib["Beschreibung"]),
             kommunikation_von=original_element.attrib["Kommunikation_von"].strip(),
             format=EdifactFormat(lstrip("M_", format_element.tag)),
             elements=tuple(segments_and_groups),

src/fundamend/utils.py

@@ -46,3 +46,31 @@ def strip(prefix: str, text: str, suffix: str) -> str:
         The text with the prefix and suffix stripped.
     """
     return lstrip(prefix, rstrip(text, suffix))
+
+
+_replacements: dict[str, str] = {
+    "-\r\n": "",
+    "\r\n": " ",
+    "\r": "",
+    "\n": "",
+}
+
+
+def remove_linebreaks_and_hyphens(original: str) -> str:
+    """
+    Normalize a multi line string by stripping leading and trailing whitespace and removing line breaks.
+
+    Args:
+        original: The string to normalize.
+
+    Returns:
+        The normalized string.
+    """
+    result = original
+    for old, new in _replacements.items():
+        result = result.replace(old, new)
+    # if you add more replacement rules, please also add a unit test in bltest_utils.py
+    return " ".join(result.strip().split())
+
+
+__all__ = ["lstrip", "rstrip", "strip", "remove_linebreaks_and_hyphens"]

unittests/test_utils.py

@@ -0,0 +1,22 @@
+import pytest
+
+from fundamend.utils import remove_linebreaks_and_hyphens
+
+
+@pytest.mark.parametrize(
+    "original, expected",
+    [
+        pytest.param("foo", "foo", id="no change"),
+        pytest.param("foo ", "foo", id="trailing whitespace"),
+        pytest.param(" foo", "foo", id="leading whitespace"),
+        pytest.param(" foo ", "foo", id="trailing and leading whitespaces"),
+        pytest.param(" foo\r\n ", "foo", id="trailing and leading whitespaces and line break"),
+        # hyphen requirements discussed here:
+        # https://github.com/Hochfrequenz/xml-fundamend-python/issues/172#issue-3427724092
+        pytest.param(" Foo-\r\nbar ", "Foobar", id="hyphen with line break"),
+        pytest.param(" Foo\r\n and bar ", "Foo and bar", id="line break w/o hyphen"),
+    ],
+)
+def test_anwendungsfall_beschreibung_normalization(original: str, expected: str) -> None:
+    actual = remove_linebreaks_and_hyphens(original)
+    assert actual == expected