fix(import): treat cd1/cd2 folders as one audiobook

tests/test_import_flow.py

@@ -1,4 +1,5 @@
 import importlib
+from pathlib import Path
 
 import pytest
 from fastapi.testclient import TestClient
@@ -214,8 +215,9 @@ def test_scan_rel_dir_and_request_update(client):
     for i in range(2):
         (nested / f"t{i}.mp3").write_bytes(b"")
     item = client.get("/api/import/scan").json()["items"][0]
-    assert item["name"] == "CD1"
-    assert item["rel_dir"] == "Sammlung/Sonderfolgen"
+    # a disc subfolder is folded into its parent, which carries the title
+    assert item["name"] == "Sonderfolgen"
+    assert item["rel_dir"] == "Sammlung"
 
     req = client.post("/api/requests", json={"library_id": lib_id, "title": "X"}).json()
     updated = client.put(f"/api/requests/{req['id']}", json={
@@ -264,3 +266,48 @@ def test_retag_after_update(client):
     assert str(tags["TXXX:SERIES"]) == "Rowling Kinderbücher"
     assert str(tags["TXXX:SERIES-PART"]) == "2"
     assert str(tags["TCOM"]) == "Ben Becker"
+
+
+def test_multi_disc_folder_is_one_audiobook(client):
+    root = client.tmp_path / "library" / "dr3i"
+    lib_id = client.post("/api/libraries", json={
+        "name": "DR3i", "media_type": "audiobook", "root_path": str(root),
+        "folder_template": "{Title}", "file_template": "{Title}",
+    }).json()["id"]
+    req_id = client.post("/api/requests", json={
+        "library_id": lib_id, "title": "Die dr3i - Böses Erwachen",
+    }).json()["id"]
+
+    folder = client.downloads / "DiE DR3i - Boeses Erwachen"
+    for disc, tracks in (("CD 1", 3), ("CD2", 2)):
+        d = folder / disc
+        d.mkdir(parents=True)
+        for i in range(1, tracks + 1):
+            (d / f"Track {i}.mp3").write_bytes(b"")
+
+    items = client.get("/api/import/scan").json()["items"]
+    assert len(items) == 1
+    item = items[0]
+    assert item["name"] == "DiE DR3i - Boeses Erwachen"
+    assert item["is_dir"] and len(item["files"]) == 5
+    # discs in order, tracks naturally sorted within a disc
+    assert [Path(f).parent.name for f in item["files"]] == ["CD 1"] * 3 + ["CD2"] * 2
+    assert item["suggested_request_id"] == req_id
+
+    res = client.post("/api/import", json={"items": [{
+        "path": item["path"], "is_dir": True, "files": item["files"],
+        "request_id": req_id,
+    }]}).json()["results"][0]
+    assert res["ok"], res
+    dest = sorted(p.name for p in Path(res["dest"]).iterdir())
+    assert dest == [f"Die dr3i - Böses Erwachen - Part 0{i}.mp3" for i in range(1, 6)]
+    assert not folder.exists()
+
+
+def test_natural_track_order_in_flat_folder(client):
+    folder = client.downloads / "Folge 12"
+    folder.mkdir(parents=True)
+    for i in (1, 2, 10):
+        (folder / f"track{i}.mp3").write_bytes(b"")
+    item = client.get("/api/import/scan").json()["items"][0]
+    assert [Path(f).name for f in item["files"]] == ["track1.mp3", "track2.mp3", "track10.mp3"]

tests/test_matcher.py

@@ -47,3 +47,26 @@ def test_volume_number_must_match():
               "media_type": "audiobook", "is_dir": False, "files": ["/d/f.mp3"]}]
     out = best_matches(items, [r1, r123])
     assert out[0]["suggested_request_id"] == 2
+
+
+def test_leetspeak_title_matches():
+    assert normalize("DiE DR3i - Folge 03") == "die drei folge 03"
+    # digits at a word edge stay put, they carry the episode number
+    assert normalize("Folge03 v2 [m4b]") == "folge03"
+
+    r = req(1, "Die dr3i - Böses Erwachen", media_type="audiobook")
+    r.volume = 2
+    items = [{"path": "/d/x", "name": "Die drei 02 - Boeses Erwachen",
+              "media_type": "audiobook", "is_dir": True, "files": ["/d/x/a.mp3"]}]
+    out = best_matches(items, [r])
+    assert out[0]["suggested_request_id"] == 1
+
+
+def test_digit_inside_word_is_not_a_volume():
+    from wordarr.importer.matcher import score
+    r = req(1, "Die dr3i - Folge 5", media_type="audiobook")
+    r.volume = 5
+    # "dr3i" must not contribute a "3" that satisfies the volume check
+    r3 = req(2, "Die dr3i - Folge 3", media_type="audiobook")
+    r3.volume = 3
+    assert score("DiE DR3i - 05 - Der Fall", r) > score("DiE DR3i - 05 - Der Fall", r3)