feat: sanitize Anwendungsfall.beschreibung by removing hyphens inside words (#233)

seems like they're writing the XMLs in Word

Co-authored-by: Konstantin <[email protected]>

Author: hf-kklein

src/fundamend/reader/ahbreader.py

@@ -31,7 +31,7 @@
     _is_segment_group,
     _is_uebertragungsdatei,
 )
-from fundamend.utils import lstrip, remove_linebreaks_and_hyphens, strip
+from fundamend.utils import lstrip, remove_linebreaks_and_hyphens, remove_unnecessary_hyphens, strip
 
 # pylint:disable=duplicate-code
 # yes, it's very similar to the MigReader
@@ -252,7 +252,9 @@ 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=remove_linebreaks_and_hyphens(original_element.attrib["Beschreibung"]),
+            beschreibung=remove_unnecessary_hyphens(
+                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

@@ -3,10 +3,13 @@
 """
 
 import re
-from typing import Optional
+from typing import Optional, overload
 
 from fundamend.models.kommunikationsrichtung import Kommunikationsrichtung
 
+_unnecessary_hyphen_pattern = re.compile(r"(?<=[a-zäüöß])-(?=[a-zäüöß])")
+"""if before AND after a hyphen there are only lower case letters, then we can probably remove it"""
+
 
 def lstrip(prefix: str, text: str) -> str:
     """Strip the given prefix from the given text. If the text does not start with the prefix, return the text as is.
@@ -143,4 +146,27 @@ def parse_kommunikation_von(kommunikation_von: Optional[str]) -> list[Kommunikat
     return result
 
 
-__all__ = ["lstrip", "rstrip", "strip", "parse_kommunikation_von", "remove_linebreaks_and_hyphens"]
+@overload
+def remove_unnecessary_hyphens(candidate: str) -> str: ...
+@overload
+def remove_unnecessary_hyphens(candidate: None) -> None: ...
+def remove_unnecessary_hyphens(candidate: Optional[str]) -> Optional[str]:
+    """
+    removes hyphens from the middle of words that are likely unnecessary
+    Example: "Ausfallarbeits-summenzeitreihe" returns "Ausfallarbeitssummenzeitreihe"
+    or "Bestäti-gung" returns "Bestätigung". But "Sperr-/Entsperrauftrag" stays untouched.
+    Handles multiple occurrences: "Bestäti-gung der Stornier-ung" returns "Bestätigung der Stornierung".
+    """
+    if candidate is None:
+        return None
+    return _unnecessary_hyphen_pattern.sub("", candidate)
+
+
+__all__ = [
+    "lstrip",
+    "rstrip",
+    "strip",
+    "parse_kommunikation_von",
+    "remove_linebreaks_and_hyphens",
+    "remove_unnecessary_hyphens",
+]