feat: shorten paths that exceed the device's limit

Rockbox's MAX_PATH is 260, defined in firmware/include/fs_defines.h and used to
size the directory entry buffer in dir.h. It bounds the path as the device sees
it, so the directory the mirror is copied into spends part of the same budget;
--device-prefix accounts for that and defaults to /Music.

Over-budget paths are shortened from the deepest component outward. The track
name carries the least navigational value and the artist directory the most, so
the filename is cut first and the artist only if nothing else will serve. A
shortened component keeps its extension and gains four hex digits of the
original name: two long titles sharing a prefix cut to the same string
otherwise, and a silent collision between two tracks is a worse outcome than an
ugly filename.

The result is stable. The same source always yields the same shortened name, so
one pass does not rename what the last one wrote -- an unstable scheme would
churn the whole mirror every six hours. A path too deeply nested to fit without
reducing every component to nonsense is left alone and reported rather than
mangled.

Migration now tries more than one previous naming, because there is more than
one. A mirror already running with --fat32-safe holds sanitised but unshortened
paths, and matching only the original unsanitised name would have re-encoded
every one of them instead of moving it.

The checker gains the same two options, since it was measuring the
mirror-relative path against a limit that applies to the device-absolute one,
and so under-reported by the length of the destination directory.
This commit is contained in:
Emma Thorpe
2026-08-25 11:45:27 +01:00
parent 3f50577de6
commit 226d375f48
4 changed files with 257 additions and 26 deletions
+92
View File
@@ -613,3 +613,95 @@ def test_renames_are_counted_separately_from_encodes(tmp_path, make_flac, caplog
assert "1 renamed" in caplog.text
assert "0 encoded" in caplog.text
def test_a_long_path_is_shortened_from_the_deepest_component(tmp_path, make_flac):
"""The track name carries the least navigational value and the artist the
most, so the filename is sacrificed before the album."""
artist = "A" * 60
album = "B" * 60
title = "C" * 150
source = tmp_path / "src"
mirror = tmp_path / "dst"
make_flac(source / artist / album / f"{title}.flac")
run(source, mirror, "--fat32-safe", "--max-path", "160", "--device-prefix", "/Music")
written = list(mirror.rglob("*.mp3"))
assert len(written) == 1
relative = written[0].relative_to(mirror)
assert relative.parts[0] == artist, "the artist directory should be untouched"
assert relative.parts[1] == album, "the album directory should be untouched"
assert len(str(relative)) <= 160 - len("Music") - 2
def test_shortening_is_stable_across_passes(tmp_path, make_flac):
"""An unstable name would rename every file on every pass, for ever."""
source = tmp_path / "src"
mirror = tmp_path / "dst"
make_flac(source / ("D" * 80) / ("E" * 80) / f"{'F' * 120}.flac")
run(source, mirror, "--fat32-safe", "--max-path", "180")
first = sorted(str(p.relative_to(mirror)) for p in mirror.rglob("*.mp3"))
stamp = next(mirror.rglob("*.mp3")).stat().st_mtime_ns
run(source, mirror, "--fat32-safe", "--max-path", "180")
assert sorted(str(p.relative_to(mirror)) for p in mirror.rglob("*.mp3")) == first
assert next(mirror.rglob("*.mp3")).stat().st_mtime_ns == stamp
def test_two_long_names_do_not_collide_after_shortening(tmp_path, make_flac):
"""They share a prefix and cut to the same string; the hash is what keeps
them apart."""
source = tmp_path / "src"
mirror = tmp_path / "dst"
shared = "G" * 140
make_flac(source / "Album" / f"{shared}one.flac")
make_flac(source / "Album" / f"{shared}two.flac")
run(source, mirror, "--fat32-safe", "--max-path", "120")
assert len(list(mirror.rglob("*.mp3"))) == 2
def test_a_sanitised_mirror_is_renamed_rather_than_re_encoded_when_shortening(
tmp_path, make_flac
):
"""The previous naming is sanitised-but-not-shortened, not the original."""
source = tmp_path / "src"
mirror = tmp_path / "dst"
make_flac(source / "Album" / f"Where Are You? {'H' * 140}.flac")
run(source, mirror, "--fat32-safe", "--max-path", "400")
before = next(mirror.rglob("*.mp3"))
contents = before.read_bytes()
run(source, mirror, "--fat32-safe", "--max-path", "120")
after = next(mirror.rglob("*.mp3"))
assert after != before
assert after.read_bytes() == contents, "it was re-encoded rather than moved"
def test_shortening_only_applies_when_over_budget(tmp_path, make_flac):
source = tmp_path / "src"
mirror = tmp_path / "dst"
make_flac(source / "Artist" / "Album" / "Short Name.flac")
run(source, mirror, "--fat32-safe")
assert (mirror / "Artist" / "Album" / "Short Name.mp3").is_file()
def test_a_path_that_cannot_be_made_to_fit_is_reported(tmp_path, make_flac, caplog):
"""Too deeply nested to shorten without making every component unreadable."""
deep = Path(*["I" * 20 for _ in range(10)])
source = tmp_path / "src"
mirror = tmp_path / "dst"
make_flac(source / deep / "track.flac")
with caplog.at_level("WARNING"):
run(source, mirror, "--fat32-safe", "--max-path", "80")
assert "too" in caplog.text and "nested" in caplog.text