From 7869839aa7a6419931d2385817686d478c81be48 Mon Sep 17 00:00:00 2001
From: Jayaram Kancherla <jayaram.kancherla@gmail.com>
Date: Wed, 29 May 2024 07:40:21 -0700
Subject: [PATCH] Register `save_dataset` for `SummarizedExperiment` class
 (#11)

Update tests
---
 docs/conf.py                 |  1 +
 src/scrnaseq/save_dataset.py | 23 +++++++++++++++++------
 tests/test_fetch_dataset.py  |  1 +
 tests/test_polish_dataset.py |  1 +
 tests/test_save_dataset.py   |  4 ++--
 5 files changed, 22 insertions(+), 8 deletions(-)

diff --git a/docs/conf.py b/docs/conf.py
index 52b4089..16b0213 100644
--- a/docs/conf.py
+++ b/docs/conf.py
@@ -316,6 +316,7 @@
     "gypsum_client": ("https://artifactdb.github.io/gypsum-py", None),
     "delayedarray": ("https://biocpy.github.io/DelayedArray", None),
     "dolomite_base": ("https://artifactdb.github.io/dolomite-base", None),
+    "anndata": ("https://anndata.readthedocs.io/en/latest/", None),
 }
 
 print(f"loading configurations for {project} {version} ...", file=sys.stderr)
\ No newline at end of file
diff --git a/src/scrnaseq/save_dataset.py b/src/scrnaseq/save_dataset.py
index edb78c3..8d9a10c 100644
--- a/src/scrnaseq/save_dataset.py
+++ b/src/scrnaseq/save_dataset.py
@@ -7,6 +7,7 @@
 import dolomite_base as dl
 from gypsum_client import fetch_metadata_schema, validate_metadata
 from singlecellexperiment import SingleCellExperiment
+from summarizedexperiment import SummarizedExperiment
 
 from .utils import format_object_metadata
 
@@ -22,7 +23,9 @@ def save_dataset(x: Any, path, metadata):
     Args:
         x:
             An object containing single-cell data.
-            May be a derivative of SummarizedExperiment or AnnData.
+            May be a derivative of
+            :py:class:`~summarizedexperiment.SummarizedExperiment.SummarizedExperiment`
+            or :py:class:`~anndata.AnnData`.
 
         path:
             Path to a new directory to save the dataset.
@@ -80,13 +83,11 @@ def save_dataset(x: Any, path, metadata):
     )
 
 
-@save_dataset.register
-def save_dataset_sce(x: SingleCellExperiment, path: str, metadata: dict):
-    """Save :py:class:`~singlecellexperiment.SingleCellExperiment.SingleCellExperiment` to disk."""
+def _save_se(x, path, metadata):
     schema = fetch_metadata_schema()
 
     if "bioconductor_version" not in metadata:
-        metadata["bioconductor_version"] = "3.14"  # current release
+        metadata["bioconductor_version"] = "3.19"  # current release
 
     validate_metadata(metadata, schema)
 
@@ -109,7 +110,17 @@ def save_dataset_sce(x: SingleCellExperiment, path: str, metadata: dict):
     with open(os.path.join(path, "_bioconductor.json"), "w") as f:
         f.write(contents)
 
-    return
+
+@save_dataset.register
+def save_dataset_sce(x: SingleCellExperiment, path: str, metadata: dict):
+    """Save :py:class:`~singlecellexperiment.SingleCellExperiment.SingleCellExperiment` to disk."""
+    return _save_se(x, path, metadata)
+
+
+@save_dataset.register
+def save_dataset_se(x: SummarizedExperiment, path: str, metadata: dict):
+    """Save :py:class:`~summarizedexperiment.SummarizedExperiment.SummarizedExperiment` to disk."""
+    return _save_se(x, path, metadata)
 
 
 has_anndata = False
diff --git a/tests/test_fetch_dataset.py b/tests/test_fetch_dataset.py
index f9ce31b..07acfca 100644
--- a/tests/test_fetch_dataset.py
+++ b/tests/test_fetch_dataset.py
@@ -35,6 +35,7 @@ def test_fetch_dataset():
         assert all(isinstance(a, (np.ndarray)) for _, a in alt_exp_ass.items())
 
 
+@pytest.mark.skip("takes too long")
 def test_fetch_dataset_realizes_reduced_dimensions():
     sce = fetch_dataset("aztekin-tail-2019", "2023-12-14", realize_reduced_dims=False)
     red_dim = sce.get_reduced_dims()
diff --git a/tests/test_polish_dataset.py b/tests/test_polish_dataset.py
index 9a310ff..415518f 100644
--- a/tests/test_polish_dataset.py
+++ b/tests/test_polish_dataset.py
@@ -168,6 +168,7 @@ def test_polish_dataset_forbids_highly_nested_altexps():
     )
 
 
+@pytest.mark.skip("takes too long")
 def test_polish_existing_dataset():
     sce = fetch_dataset("zeisel-brain-2015", "2023-12-14")
 
diff --git a/tests/test_save_dataset.py b/tests/test_save_dataset.py
index 08433a4..97be19b 100644
--- a/tests/test_save_dataset.py
+++ b/tests/test_save_dataset.py
@@ -44,7 +44,7 @@ def test_save_dataset_sce():
     with open(os.path.join(tmp, "_bioconductor.json")) as f:
         saved_meta = json.load(f)
 
-    assert saved_meta["bioconductor_version"] == "3.14"  # Placeholder version
+    assert saved_meta["bioconductor_version"] == "3.19"
 
     # Test validation failure
     meta["title"] = 1234
@@ -90,7 +90,7 @@ def test_save_dataset_anndata():
     with open(os.path.join(tmp, "_bioconductor.json")) as f:
         saved_meta = json.load(f)
 
-    assert saved_meta["bioconductor_version"] == "3.14"  # Placeholder version
+    assert saved_meta["bioconductor_version"] == "3.19"
 
     # Test validation failure
     meta["title"] = 1234