From dbf9528fea12961567d6d674d8be334c259eada4 Mon Sep 17 00:00:00 2001
From: rmsitz <rmsitz@gmail.com>
Date: Sat, 22 Aug 2026 09:22:54 -0400
Subject: [PATCH] Speed up library search on large libraries

search_library_courses was walking via list_library_directory, which
computes a full recursive file count (rglob) and thumbnail lookup for
every course at every level - turning a search into an O(every file in
every course) scan regardless of how few results actually match. Give
search its own lightweight directory-only walk and defer the expensive
per-course lookups until after a name match is confirmed, so cost now
scales with matches found rather than total library size.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
---
 offlineu_core.py | 53 +++++++++++++++++++++++++++++++++++++-----------
 1 file changed, 41 insertions(+), 12 deletions(-)

diff --git a/offlineu_core.py b/offlineu_core.py
index e0c0b74..13aa3fb 100644
--- a/offlineu_core.py
+++ b/offlineu_core.py
@@ -735,23 +735,52 @@ def get_library_root() -> str:
 def search_library_courses(dir_path: str, query: str) -> List[Dict[str, Any]]:
     """
     Recursively search the library for courses whose name contains `query`
-    (case-insensitive). Walks via list_library_directory, so it reuses the
-    exact same course/directory detection and hidden-path filtering as
-    normal browsing - a hidden course or folder never shows up in results.
+    (case-insensitive), respecting hidden paths exactly like normal browsing.
+
+    Deliberately doesn't reuse list_library_directory for the walk itself:
+    that function computes media_count (a full recursive file count via
+    rglob) and thumbnail presence for *every* course at each level, which is
+    fine for showing one directory's worth of courses but turns a full-
+    library search into an O(every file in every course) scan. Here, the
+    directory walk only touches directory entries (cheap - iterdir, no file
+    stats), and the expensive per-course lookups only run for the handful of
+    courses whose name actually matches.
     """
     query_lower = query.lower()
+    hidden_set = set(get_hidden_paths())
     results: List[Dict[str, Any]] = []
 
-    def walk(path: str):
-        level = list_library_directory(path)
-        for item in level['items']:
-            if item['type'] == 'course':
-                if query_lower in item['name'].lower():
-                    results.append(item)
-            else:
-                walk(item['path'])
+    def walk(directory: Path):
+        try:
+            entries = sorted(
+                (p for p in directory.iterdir() if p.is_dir() and not p.name.startswith('.')),
+                key=lambda p: p.name.lower()
+            )
+        except (PermissionError, OSError):
+            return
 
-    walk(dir_path)
+        for entry in entries:
+            if os.path.abspath(str(entry)) in hidden_set:
+                continue
+
+            if _looks_like_course(entry):
+                if query_lower in entry.name.lower():
+                    media_count = len([
+                        f for f in entry.rglob('*')
+                        if f.is_file() and f.suffix.lower() in VIDEO_EXTENSIONS | AUDIO_EXTENSIONS
+                    ])
+                    results.append({
+                        'type': 'course',
+                        'name': entry.name,
+                        'path': str(entry),
+                        'media_files': media_count,
+                        'hidden': False,
+                        'has_thumbnail': find_course_thumbnail(str(entry)) is not None
+                    })
+            else:
+                walk(entry)
+
+    walk(Path(dir_path))
     return results