Renamed the default heuristic comparator to highlight it's far from p…

…erfect (it sucks actually, lots of false positives).

beangulp/extract_test.py

@@ -63,7 +63,7 @@ def test_mark_duplicate_entries(self):
           1970-01-02 * "Test"
             Assets:Tests  20.00 USD
         '''))
-        compare = similar.comparator()
+        compare = similar.heuristic_comparator()
         extract.mark_duplicate_entries(entries, entries[:1], timedelta(days=2), compare)
         self.assertTrue(entries[0].meta[extract.DUPLICATE])
         self.assertNotIn(extract.DUPLICATE, entries[1].meta)

beangulp/importer.py

@@ -121,7 +121,7 @@ def extract(self, filepath: str, existing: data.Entries) -> data.Entries:
         """
         return []
 
-    cmp = staticmethod(similar.comparator())
+    cmp = staticmethod(similar.heuristic_comparator())
 
     def deduplicate(self, entries: data.Entries, existing: data.Entries) -> None:
         """Mark duplicates in extracted entries.

beangulp/similar.py

@@ -36,7 +36,7 @@ def find_similar_entries(entries, existing_entries, cmp=None, window_days=2):
     Args:
       entries: The list of entries to classify as duplicate or note.
       existing_entries: The list of entries against which to match.
-      comparator: A functor used to establish the similarity of two entries.
+      cmp: A functor used to establish the similarity of two entries.
       window_days: The number of days (inclusive) before or after to scan the
         entries to classify against.
 
@@ -50,7 +50,7 @@ def find_similar_entries(entries, existing_entries, cmp=None, window_days=2):
     window_tail = datetime.timedelta(days=window_days + 1)
 
     if cmp is None:
-        cmp = comparator()
+        cmp = heuristic_comparator()
 
     # For each of the new entries, look at existing entries at a nearby date.
     duplicates = []
@@ -81,7 +81,7 @@ def __getattr__(self, name):
 Comparator = Callable[[data.Directive, data.Directive], bool]
 
 
-def comparator(
+def heuristic_comparator(
     max_date_delta: datetime.timedelta | None = None, epsilon: Decimal | None = None
 ) -> Comparator:
     """Generic comparison function generator.
@@ -103,7 +103,7 @@ def comparator(
       epsilon: A Decimal fraction representing how close the amounts are
         required to be of each other. For example, Decimal("0.01") for 1%.
     Returns:
-      A comparator predicte accepting two directives and returning a bool.
+      A comparator predicate accepting two directives and returning a bool.
     """
 
     if epsilon is None:
@@ -173,6 +173,10 @@ def cmp(entry1: data.Directive, entry2: data.Directive) -> bool:
     return cmp
 
 
+# Old alias to the heuristic comparator kept for backwards compatibility.
+comparator = heuristic_comparator
+
+
 def amounts_map(entry):
     """Compute a mapping of (account, currency) -> Decimal balances.
 

beangulp/similar_test.py

@@ -129,7 +129,7 @@ def test_amounts_map(self, entries, _, __):
 class TestSimilarityComparator(cmptest.TestCase):
 
     def setUp(self):
-        self.comparator = similar.comparator(datetime.timedelta(days=2))
+        self.comparator = similar.heuristic_comparator(datetime.timedelta(days=2))
 
     @loader.load_doc()
     def test_simple(self, entries, _, __):