unsloth/studio/backend/tests/test_cached_gguf_routes.py

# SPDX-License-Identifier: AGPL-3.0-only
# Copyright 2026-present the Unsloth AI Inc. team. All rights reserved. See /studio/LICENSE.AGPL-3.0

import asyncio
import sys
import types
from pathlib import Path
from types import SimpleNamespace

# Keep this test runnable without optional logging deps.
if "structlog" not in sys.modules:

    class _DummyLogger:
        def __getattr__(self, _name):
            return lambda *args, **kwargs: None

    sys.modules["structlog"] = types.SimpleNamespace(
        BoundLogger = _DummyLogger,
        get_logger = lambda *args, **kwargs: _DummyLogger(),
    )

import routes.models as models_route


def _repo(
    repo_id: str,
    files: list[SimpleNamespace],
    repo_path: Path,
    *,
    revisions: list[SimpleNamespace] | None = None,
) -> SimpleNamespace:
    return SimpleNamespace(
        repo_id = repo_id,
        repo_type = "model",
        repo_path = repo_path,
        revisions = revisions or [SimpleNamespace(files = files)],
    )


def _file(
    name: str,
    size_on_disk: int,
    *,
    blob_path: str | None = None,
) -> SimpleNamespace:
    return SimpleNamespace(
        file_name = name,
        size_on_disk = size_on_disk,
        blob_path = blob_path,
    )


def test_iter_gguf_paths_matches_extension_case_insensitively(tmp_path):
    nested = tmp_path / "snapshots" / "rev"
    nested.mkdir(parents = True)
    lower = nested / "Q4_K_M.gguf"
    upper = nested / "Q8_0.GGUF"
    other = nested / "README.md"
    lower.write_text("a")
    upper.write_text("b")
    other.write_text("c")

    result = sorted(path.name for path in models_route._iter_gguf_paths(tmp_path))

    assert result == ["Q4_K_M.gguf", "Q8_0.GGUF"]


def test_list_cached_gguf_includes_non_suffix_repo_when_cache_contains_gguf(monkeypatch, tmp_path):
    repo = _repo(
        "HauhauCS/Gemma-4-E4B-Uncensored-HauhauCS-Aggressive",
        [_file("Q4_K_M.gguf", 5_000), _file("README.md", 10)],
        tmp_path / "models--HauhauCS--Gemma",
    )
    scan = SimpleNamespace(repos = [repo])

    monkeypatch.setattr(models_route, "_all_hf_cache_scans", lambda: [scan])

    result = asyncio.run(models_route.list_cached_gguf(current_subject = "test-user"))

    assert result["cached"] == [
        {
            "repo_id": "HauhauCS/Gemma-4-E4B-Uncensored-HauhauCS-Aggressive",
            "size_bytes": 5_000,
            "cache_path": str(repo.repo_path),
        }
    ]


def test_list_cached_gguf_matches_extension_case_insensitively(monkeypatch, tmp_path):
    repo = _repo(
        "Org/Model-Without-Suffix",
        [_file("Q8_0.GGUF", 7_000)],
        tmp_path / "models--Org--Model-Without-Suffix",
    )
    scan = SimpleNamespace(repos = [repo])

    monkeypatch.setattr(models_route, "_all_hf_cache_scans", lambda: [scan])

    result = asyncio.run(models_route.list_cached_gguf(current_subject = "test-user"))

    assert result["cached"] == [
        {
            "repo_id": "Org/Model-Without-Suffix",
            "size_bytes": 7_000,
            "cache_path": str(repo.repo_path),
        }
    ]


def test_list_cached_gguf_skips_repos_without_positive_gguf_size(monkeypatch, tmp_path):
    missing = _repo(
        "Org/ReadmeOnly",
        [_file("README.md", 10)],
        tmp_path / "models--Org--ReadmeOnly",
    )
    zero = _repo(
        "Org/ZeroSize",
        [_file("Q4_K_M.gguf", 0)],
        tmp_path / "models--Org--ZeroSize",
    )
    scan = SimpleNamespace(repos = [missing, zero])

    monkeypatch.setattr(models_route, "_all_hf_cache_scans", lambda: [scan])

    result = asyncio.run(models_route.list_cached_gguf(current_subject = "test-user"))

    assert result["cached"] == []


def test_list_cached_gguf_keeps_largest_duplicate_repo_across_scans(monkeypatch, tmp_path):
    smaller = _repo(
        "Org/Dupe",
        [_file("Q4_K_M.gguf", 2_000)],
        tmp_path / "models--Org--Dupe-a",
    )
    larger = _repo(
        "org/dupe",
        [_file("Q4_K_M.gguf", 5_000), _file("Q6_K.gguf", 1_000)],
        tmp_path / "models--Org--Dupe-b",
    )

    monkeypatch.setattr(
        models_route,
        "_all_hf_cache_scans",
        lambda: [
            SimpleNamespace(repos = [smaller]),
            SimpleNamespace(repos = [larger]),
        ],
    )

    result = asyncio.run(models_route.list_cached_gguf(current_subject = "test-user"))

    assert result["cached"] == [
        {
            "repo_id": "org/dupe",
            "size_bytes": 6_000,
            "cache_path": str(larger.repo_path),
        }
    ]


def test_list_cached_gguf_dedupes_shared_blobs_across_revisions(monkeypatch, tmp_path):
    shared = "blobs/shared-q4"
    repo = _repo(
        "Org/SharedBlobRepo",
        [],
        tmp_path / "models--Org--SharedBlobRepo",
        revisions = [
            SimpleNamespace(files = [_file("Q4_K_M.gguf", 5_000, blob_path = shared)]),
            SimpleNamespace(files = [_file("Q4_K_M.gguf", 5_000, blob_path = shared)]),
        ],
    )

    monkeypatch.setattr(
        models_route,
        "_all_hf_cache_scans",
        lambda: [SimpleNamespace(repos = [repo])],
    )

    result = asyncio.run(models_route.list_cached_gguf(current_subject = "test-user"))

    assert result["cached"] == [
        {
            "repo_id": "Org/SharedBlobRepo",
            "size_bytes": 5_000,
            "cache_path": str(repo.repo_path),
        }
    ]


def test_list_cached_models_skips_non_suffix_repo_when_gguf_files_exist(monkeypatch, tmp_path):
    mixed = _repo(
        "Org/MixedRepo",
        [
            _file("Q4_K_M.gguf", 5_000),
            _file("model.safetensors", 10_000),
        ],
        tmp_path / "models--Org--MixedRepo",
    )

    monkeypatch.setattr(
        models_route,
        "_all_hf_cache_scans",
        lambda: [SimpleNamespace(repos = [mixed])],
    )

    result = asyncio.run(models_route.list_cached_models(current_subject = "test-user"))

    assert result["cached"] == []


def test_list_cached_gguf_includes_mixed_repo_with_gguf_and_safetensors(monkeypatch, tmp_path):
    """Mixed repo still surfaces in cached-gguf as a GGUF download."""
    mixed = _repo(
        "Org/MixedRepo",
        [
            _file("Q4_K_M.gguf", 5_000),
            _file("model.safetensors", 10_000),
        ],
        tmp_path / "models--Org--MixedRepo",
    )

    monkeypatch.setattr(
        models_route,
        "_all_hf_cache_scans",
        lambda: [SimpleNamespace(repos = [mixed])],
    )

    result = asyncio.run(models_route.list_cached_gguf(current_subject = "test-user"))

    assert result["cached"] == [
        {
            "repo_id": "Org/MixedRepo",
            "size_bytes": 5_000,
            "cache_path": str(mixed.repo_path),
        }
    ]


def test_list_cached_gguf_handles_none_size_on_disk(monkeypatch, tmp_path):
    """``size_on_disk = None`` (partial download) is treated as zero, not a
    TypeError from ``sum()`` that wipes the response."""
    partial = _repo(
        "Org/PartialDownload",
        [_file("Q4_K_M.gguf", None), _file("Q6_K.gguf", 5_000)],
        tmp_path / "models--Org--PartialDownload",
    )

    monkeypatch.setattr(
        models_route,
        "_all_hf_cache_scans",
        lambda: [SimpleNamespace(repos = [partial])],
    )

    result = asyncio.run(models_route.list_cached_gguf(current_subject = "test-user"))

    assert result["cached"] == [
        {
            "repo_id": "Org/PartialDownload",
            "size_bytes": 5_000,
            "cache_path": str(partial.repo_path),
        }
    ]


def test_list_cached_gguf_skips_malformed_repo_without_wiping_response(monkeypatch, tmp_path):
    """One repo raising during classification must not poison the response."""

    class _ExplodingRepo:
        repo_id = "Org/Broken"
        repo_type = "model"
        repo_path = tmp_path / "models--Org--Broken"

        @property
        def revisions(self):
            raise RuntimeError("boom")

    healthy = _repo(
        "Org/Healthy",
        [_file("Q4_K_M.gguf", 5_000)],
        tmp_path / "models--Org--Healthy",
    )

    monkeypatch.setattr(
        models_route,
        "_all_hf_cache_scans",
        lambda: [SimpleNamespace(repos = [_ExplodingRepo(), healthy])],
    )

    result = asyncio.run(models_route.list_cached_gguf(current_subject = "test-user"))

    assert result["cached"] == [
        {
            "repo_id": "Org/Healthy",
            "size_bytes": 5_000,
            "cache_path": str(healthy.repo_path),
        }
    ]


def test_list_cached_gguf_skips_repo_with_only_mmproj_gguf(monkeypatch, tmp_path):
    """A repo whose only ``.gguf`` is an mmproj vision adapter is not a GGUF
    repo: mmproj is filtered out, leaving zero variants."""
    mmproj_only = _repo(
        "Org/MmprojOnly",
        [
            _file("mmproj-Q8_0.gguf", 5_000),
            _file("model.safetensors", 10_000),
        ],
        tmp_path / "models--Org--MmprojOnly",
    )

    monkeypatch.setattr(
        models_route,
        "_all_hf_cache_scans",
        lambda: [SimpleNamespace(repos = [mmproj_only])],
    )

    result = asyncio.run(models_route.list_cached_gguf(current_subject = "test-user"))

    assert result["cached"] == []


def test_list_cached_models_includes_repo_with_only_mmproj_gguf(monkeypatch, tmp_path):
    """A safetensors repo with an auxiliary mmproj adapter still surfaces in
    cached-models as a normal model."""
    mmproj_aux = _repo(
        "Org/MmprojAux",
        [
            _file("mmproj-Q8_0.gguf", 5_000),
            _file("model.safetensors", 10_000),
        ],
        tmp_path / "models--Org--MmprojAux",
    )

    monkeypatch.setattr(
        models_route,
        "_all_hf_cache_scans",
        lambda: [SimpleNamespace(repos = [mmproj_aux])],
    )

    result = asyncio.run(models_route.list_cached_models(current_subject = "test-user"))

    assert result["cached"] == [{"repo_id": "Org/MmprojAux", "size_bytes": 15_000}]


def test_list_cached_gguf_includes_vision_repo_with_main_gguf_and_mmproj(monkeypatch, tmp_path):
    """A vision GGUF repo (main weight + mmproj) is a GGUF repo; reported size
    is the main weight only, since mmproj is filtered at classification."""
    vision_repo = _repo(
        "Org/VisionGguf",
        [
            _file("Q4_K_M.gguf", 5_000),
            _file("mmproj-Q8_0.gguf", 1_000),
        ],
        tmp_path / "models--Org--VisionGguf",
    )

    monkeypatch.setattr(
        models_route,
        "_all_hf_cache_scans",
        lambda: [SimpleNamespace(repos = [vision_repo])],
    )

    result = asyncio.run(models_route.list_cached_gguf(current_subject = "test-user"))

    assert result["cached"] == [
        {
            "repo_id": "Org/VisionGguf",
            "size_bytes": 5_000,
            "cache_path": str(vision_repo.repo_path),
        }
    ]


def _gfile(name: str, size: int, mtime: float) -> SimpleNamespace:
    """A cached file carrying a Hugging Face ``blob_last_modified`` timestamp."""
    return SimpleNamespace(
        file_name = name,
        size_on_disk = size,
        blob_path = None,
        blob_last_modified = mtime,
    )


def test_all_hf_cache_scans_survives_inaccessible_aux_cache(monkeypatch, tmp_path):
    """An unreadable auxiliary cache (e.g. an inaccessible
    ``~/.cache/huggingface/hub``) must be skipped, not abort the scan.
    Regression guard for ``extra.is_dir()`` raising and wiping the response.
    """
    import huggingface_hub
    import utils.paths as paths_mod

    active = SimpleNamespace(
        repos = [_repo("Org/Active", [_file("Q4_K_M.gguf", 5_000)], tmp_path / "active")]
    )

    def _fake_scan(cache_dir = None):
        if cache_dir is None:
            return active
        raise AssertionError("auxiliary scan should have been skipped")

    class _Boom:
        def is_dir(self):
            raise PermissionError(13, "Permission denied")

        def resolve(self):
            raise PermissionError(13, "Permission denied")

    monkeypatch.setattr(huggingface_hub, "scan_cache_dir", _fake_scan)
    monkeypatch.setattr(paths_mod, "legacy_hf_cache_dir", lambda: _Boom())
    monkeypatch.setattr(paths_mod, "hf_default_cache_dir", lambda: _Boom())

    scans = models_route._all_hf_cache_scans()
    assert scans == [active]

    # End-to-end: the endpoint still returns the active cache's repo.
    monkeypatch.setattr(models_route, "_all_hf_cache_scans", lambda: [active])
    result = asyncio.run(models_route.list_cached_gguf(current_subject = "test-user"))
    assert result["cached"] == [
        {
            "repo_id": "Org/Active",
            "size_bytes": 5_000,
            "cache_path": str(tmp_path / "active"),
        }
    ]


def test_list_cached_gguf_sorts_newest_first_grouping_by_latest_quant(monkeypatch, tmp_path):
    """Downloaded is ordered newest-first, and a multi-quant repo is placed by
    its most recently downloaded quant (``last_modified`` = newest quant)."""
    older = _repo(
        "Org/Older",
        [_gfile("Older-Q4_K_M.gguf", 5_000, 1_000.0)],
        tmp_path / "models--Org--Older",
    )
    newer = _repo(
        "Org/Newer",
        [
            _gfile("Newer-Q4_K_M.gguf", 5_000, 2_000.0),
            _gfile("Newer-Q8_0.gguf", 9_000, 3_000.0),  # newest quant in the repo
        ],
        tmp_path / "models--Org--Newer",
    )

    monkeypatch.setattr(
        models_route,
        "_all_hf_cache_scans",
        lambda: [SimpleNamespace(repos = [older, newer])],
    )

    result = asyncio.run(models_route.list_cached_gguf(current_subject = "test-user"))

    assert [c["repo_id"] for c in result["cached"]] == ["Org/Newer", "Org/Older"]
    assert result["cached"][0]["last_modified"] == 3_000.0
    assert result["cached"][1]["last_modified"] == 1_000.0


def test_list_cached_gguf_dedupe_keeps_newest_timestamp(monkeypatch, tmp_path):
    """Same repo in two caches with equal size keeps the newest last_modified,
    regardless of scan order."""
    older = _repo("org/dupe", [_gfile("dupe-Q4_K_M.gguf", 5_000, 1_000.0)], tmp_path / "a")
    newer = _repo("org/dupe", [_gfile("dupe-Q4_K_M.gguf", 5_000, 9_000.0)], tmp_path / "b")
    for scans in ([older, newer], [newer, older]):  # both orders
        monkeypatch.setattr(
            models_route,
            "_all_hf_cache_scans",
            lambda s = scans: [SimpleNamespace(repos = [s[0]]), SimpleNamespace(repos = [s[1]])],
        )
        result = asyncio.run(models_route.list_cached_gguf(current_subject = "t"))
        assert len(result["cached"]) == 1
        assert result["cached"][0]["last_modified"] == 9_000.0


def test_gguf_variants_mmproj_does_not_mark_quant_downloaded(monkeypatch, tmp_path):
    """The per-quant 'downloaded' flag is driven by the real weight file in a
    single snapshot; an mmproj vision adapter (matching a quant label) must
    not make that quant appear downloaded."""
    import huggingface_hub.constants as hf_constants

    variants = [
        SimpleNamespace(filename = "model-Q4_K_M.gguf", quant = "Q4_K_M", size_bytes = 10_000),
        SimpleNamespace(filename = "model-F16.gguf", quant = "F16", size_bytes = 20_000),
    ]
    monkeypatch.setattr(
        models_route, "list_gguf_variants", lambda repo_id, hf_token = None: (variants, True)
    )
    monkeypatch.setattr(hf_constants, "HF_HUB_CACHE", str(tmp_path))

    snap = tmp_path / "models--org--repo" / "snapshots" / "rev"
    snap.mkdir(parents = True)
    (snap / "model-Q4_K_M.gguf").write_bytes(b"x" * 10_000)  # real weight, fully present
    (snap / "mmproj-F16.gguf").write_bytes(b"y" * 20_000)  # mmproj adapter, label "F16"

    result = asyncio.run(
        models_route.get_gguf_variants(
            repo_id = "org/repo", hf_token = None, current_subject = "test-user"
        )
    )

    flags = {v.quant: v.downloaded for v in result.variants}
    assert flags["Q4_K_M"] is True
    assert flags["F16"] is False


def test_gguf_download_progress_excludes_mmproj(monkeypatch, tmp_path):
    """A cached mmproj adapter must not count toward a same-label main
    variant's download progress (mmproj-F16 vs an F16 weight)."""
    import huggingface_hub.constants as hf_constants

    monkeypatch.setattr(hf_constants, "HF_HUB_CACHE", str(tmp_path))
    snap = tmp_path / "models--org--repo" / "snapshots" / "rev"
    snap.mkdir(parents = True)
    (snap / "mmproj-F16.gguf").write_bytes(b"y" * 20_000)  # only the adapter on disk

    result = asyncio.run(
        models_route.get_gguf_download_progress(
            repo_id = "org/repo",
            variant = "F16",
            expected_bytes = 20_000,
            current_subject = "test-user",
        )
    )

    assert result["downloaded_bytes"] == 0
    assert result["progress"] == 0