Scope dotted author name handling to author tags

Signed-off-by: Hrithik Sharma <sharmahrithik2806@gmail.com>

Author: Hrithikcrick

src/cluecode/copyrights.py

@@ -440,6 +440,13 @@ def get_tokens(numbered_lines, splitter=re.compile(r'[\t =;]+').split):
             string=line,
         )
 
+        # Normalize dotted names only when they follow an Author tag.
+        line = re.sub(
+            pattern=r'(^|\s)([Aa]uthor)\s+([A-Z][a-z]+)\.([A-Z][a-z]+)(?=\b|[,;])',
+            repl=r'\1\2 \3 \4',
+            string=line,
+        )
+
         for tok in splitter(line):
             # strip trailing quotes+comma
             if tok.endswith("',"):
@@ -1814,9 +1821,6 @@ def build_detection_from_node(
     # Proper Nouns
     ############################################################################
 
-    # Dotted proper names such as Frankie.Chu
-    (r'^[A-Z][a-z]+(?:\.[A-Z][a-z]+)+$', 'NAME'),
-
     # Title case word with a trailing parens is an NNP, including with an optional trailing period
     (r'^[A-Z][a-z]{3,}\)\.?$', 'NNP'),
     # Title case word with a leading parens is an NNP

tests/cluecode/test_copyrights_basic.py

@@ -325,7 +325,7 @@ def test_detect_author_without_space_and_with_dotted_name(self):
             (2, '//  Author:Frankie.Chu'),
         ]
         expected = [
-            copyrights.AuthorDetection('Frankie.Chu', 2, 2),
+            copyrights.AuthorDetection('Frankie Chu', 2, 2),
         ]
         results = list(copyrights.detect_copyrights_from_lines(
             numbered_lines,