auchann-compatible error detection method

deregularise.py

@@ -34,7 +34,7 @@
 import csv
 import re
 import os
-from typing import Dict, List, Tuple
+from typing import cast, Dict, List, Optional, Tuple
 from config import SD_DIR
 
 tab = '\t'
@@ -50,6 +50,14 @@
 wrongen = 'Wrong -en suffix'
 nolabel = 'Correct'
 
+chat_errors = {
+    'Overgeneralisation': 'm',
+    'Lacking ge prefix': 'm',
+    'Prefix ge without onset': 'm',
+    'Wrong Overgeneralisation': 'm',
+    'Wrong -en suffix': 'm'
+}
+
 metaarr = {}
 metaarr['ge'] = ''
 metaarr[''] = noge
@@ -98,6 +106,34 @@ def correctinflection(word: str) -> List[Tuple[str, str]]:
     return result
 
 
+def map_error(error_type: str) -> str:
+    try:
+        return chat_errors[error_type]
+    except KeyError:
+        return error_type
+
+
+def detect_error(original: str, correction: str) -> Tuple[int, Optional[str]]:
+    """Detects an error comparing a text with a correction and returns
+    the desired editing distance and the CHAT error code
+
+    Args:
+        original (str): transcribed text
+        correction (str): correction
+
+    Returns:
+        Tuple[int, Optional[str]]: editing distance and error code
+    """
+    error = None
+    for candidate, candidate_error in correctinflection(original):
+        if candidate == correction:
+            error = map_error(candidate_error)
+    if error is not None:
+        return 1, cast(str, error)
+    else:
+        return 0, None
+
+
 def alt(thestr):
     result = '[' + thestr + ']'
     return result