From 89f843d223635907894cfa45a16723721f616cbe Mon Sep 17 00:00:00 2001
From: BoulderBadgeDad <batmanesignal@gmail.com>
Date: Thu, 11 Jun 2026 10:46:36 -0700
Subject: [PATCH] #851: normalize '/' ':' '_' to spaces so slash-titles match
 underscore sources
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The candidate matcher rejected valid downloads of titles with a '/' or ':' (e.g.
Sawano's "You See Big Girl / T:T") because the unified normalize() removed those
chars ("t:t" -> "tt") while keeping the '_' that source filenames substitute for
them ("T_T" -> "t_t") — the asymmetry tanked the similarity score. Now '/ \ : _'
all map to spaces before the strip, so "/ T:T" and "_ T_T" both normalize to "t t".

Verified on the real library: similarity for the Sawano pair 0.927 -> 1.000; only
348/40786 strings (0.85%, all containing those separators) change; worst-case
joined-variant match (e.g. "12:05" vs "1205") stays 0.889, well above the 0.70
title threshold — no match regressions. Fixes the matching half of #851.
---
 core/matching/audio_verification.py        |  5 +++++
 tests/matching/test_normalize_slash_851.py | 24 ++++++++++++++++++++++
 2 files changed, 29 insertions(+)
 create mode 100644 tests/matching/test_normalize_slash_851.py

diff --git a/core/matching/audio_verification.py b/core/matching/audio_verification.py
index 06b90949..a347ce73 100644
--- a/core/matching/audio_verification.py
+++ b/core/matching/audio_verification.py
@@ -67,6 +67,11 @@ def normalize(text: str) -> str:
         '', s, flags=re.IGNORECASE,
     )
     s = re.sub(r'\s*-\s*from\s+.+$', '', s, flags=re.IGNORECASE)
+    # Path/separator punctuation -> space so a title keeps matching a source
+    # filename that substituted '_' for an illegal '/' or ':' (#851): the on-disk
+    # "You See Big Girl _ T_T" must normalize the same as "You See Big Girl / T:T".
+    # Done before the strip below so they become word boundaries, not joins.
+    s = re.sub(r'[\\/:_]+', ' ', s)
     # Drop remaining punctuation but keep word chars (incl. CJK) + spaces.
     s = re.sub(r'[^\w\s]', '', s)
     s = re.sub(r'\s+', ' ', s).strip()
diff --git a/tests/matching/test_normalize_slash_851.py b/tests/matching/test_normalize_slash_851.py
new file mode 100644
index 00000000..9b4b3c8c
--- /dev/null
+++ b/tests/matching/test_normalize_slash_851.py
@@ -0,0 +1,24 @@
+"""#851: a '/' or ':' in a title must normalize the same as a source filename
+that substituted '_' for them, so the candidate matcher stops rejecting valid
+downloads (e.g. Sawano's "You See Big Girl / T:T" vs on-disk "..._ T_T")."""
+
+from __future__ import annotations
+
+from core.matching.audio_verification import normalize, similarity
+
+
+def test_slash_colon_title_matches_underscore_source():
+    assert normalize("You See Big Girl / T:T") == normalize("You See Big Girl _ T_T")
+    assert similarity("You See Big Girl / T:T", "You See Big Girl _ T_T") == 1.0
+
+
+def test_separators_become_word_boundaries():
+    assert normalize("Re:Zero") == "re zero"
+    assert normalize("AC/DC") == "ac dc"
+    assert normalize("T_T") == "t t"
+
+
+def test_joined_variant_stays_above_title_threshold():
+    # Spacing a separator must not drop a joined-variant match below 0.70.
+    assert similarity("AC/DC", "ACDC") >= 0.70
+    assert similarity("12:05", "1205") >= 0.70