Merge Bugfixes into Main

docs/source/changelog.rst

@@ -1,3 +1,30 @@
+opencosmo 0.9.4 (2025-09-26)
+============================
+
+Bugfixes
+--------
+
+- Fix a bug that could cause opening halos and galaxies only to fail
+
+
+opencosmo 0.9.3 (2025-09-25)
+============================
+
+Bugfixes
+--------
+
+- Fix an issue that could cause opens with multiple files to fail in MPI contexts
+
+
+opencosmo 0.9.2 (2025-09-25)
+============================
+
+Bugfixes
+--------
+
+- Fix a but that could cause opening properties and profiles without particles to fail.
+
+
 opencosmo 0.9.1 (2025-09-16)
 ============================
 
@@ -7,7 +34,7 @@ Bugfixes
 - Re-add license file to package for conda-forge compatability
 
 
-Opencosmo 0.9.0 (2025-09-15)
+opencosmo 0.9.0 (2025-09-15)
 ============================
 
 Features
@@ -50,7 +77,7 @@ Misc
 - Update evaluate to respect default parameters
 
 
-Opencosmo 0.8.0 (2025-07-22)
+opencosmo 0.8.0 (2025-07-22)
 ============================
 
 Features

docs/source/conf.py

@@ -9,7 +9,7 @@
 project = "OpenCosmo"
 copyright = "2025, OpenCosmo Team"
 author = "OpenCosmo Team"
-release = "0.9.1"
+release = "0.9.4"
 
 # -- General configuration ---------------------------------------------------
 # https://www.sphinx-doc.org/en/master/usage/configuration.html#general-configuration

src/opencosmo/collection/io.py

@@ -47,6 +47,7 @@ def get_collection_type(
     2. The files contain a single non-lightcone datatype.
     3. The files are linked together into a structure collection
     """
+
     handles_by_type = {target.data_type: target.group for target in targets}
     is_lightcone = [target.header.file.is_lightcone for target in targets]
     unique_data_types = set(handles_by_type.keys())
@@ -60,9 +61,16 @@ def get_collection_type(
 
     elif len(unique_data_types) == 1 and all(not il for il in is_lightcone):
         return oc.SimulationCollection
-    elif unique_file_types == set([FILE_TYPE.PARTICLES, FILE_TYPE.DATASET]):
+
+    elif FILE_TYPE.HALO_PROPERTIES in unique_file_types and set(
+        [FILE_TYPE.HALO_PARTICLES, FILE_TYPE.SOD_BINS, FILE_TYPE.GALAXY_PROPERTIES]
+    ).intersection(unique_file_types):
         validate_linked_groups(handles_by_type)
         return oc.StructureCollection
+
+    elif unique_file_types == {FILE_TYPE.GALAXY_PROPERTIES, FILE_TYPE.GALAXY_PARTICLES}:
+        return oc.StructureCollection
+
     elif (
         len(unique_file_types) == 1
         and unique_file_types.pop() == FILE_TYPE.STRUCTURE_COLLECTION

src/opencosmo/collection/structure/io.py

@@ -151,6 +151,16 @@ def build_structure_collection(targets: list[io.io.OpenTarget], ignore_empty: bo
         else:
             return collection
 
+    if (
+        link_sources["halo_properties"]
+        and len(link_sources["galaxy_properties"]) == 1
+        and not link_targets["galaxy_targets"]
+    ):
+        galaxy_properties = io.io.open_single_dataset(
+            link_sources["galaxy_properties"][0]
+        )
+        link_targets["halo_targets"]["galaxy_properties"] = galaxy_properties
+
     if len(link_sources["halo_properties"]) == 1 and link_targets["halo_targets"]:
         handlers = get_link_handlers(
             link_sources["halo_properties"][0].group,

src/opencosmo/io/io.py

@@ -59,11 +59,15 @@
 
 
 class FILE_TYPE(Enum):
-    DATASET = 0
-    PARTICLES = 1
-    LIGHTCONE = 2
-    STRUCTURE_COLLECTION = 3
-    SIMULATION_COLLECTION = 4
+    HALO_PROPERTIES = 0
+    HALO_PARTICLES = 1
+    GALAXY_PROPERTIES = 2
+    GALAXY_PARTICLES = 3
+    SOD_BINS = 4
+    LIGHTCONE = 5
+    STRUCTURE_COLLECTION = 6
+    SIMULATION_COLLECTION = 7
+    SYNTHETIC_CATALOG = 8
 
 
 class COLLECTION_TYPE(Enum):
@@ -83,17 +87,22 @@ def data_type(self):
 
 
 def get_file_type(file: h5py.File) -> FILE_TYPE:
-    file_keys = set(file.keys())
-    if len(set(["header", "data"]).intersection(file_keys)) == 2:
-        return FILE_TYPE.DATASET
-
-    elif all("particles" in fk or fk == "header" for fk in file_keys):
-        return FILE_TYPE.PARTICLES
-
-    elif "header" in file.keys():
-        groups = {name: group for name, group in file.items() if name != "header"}
-        collection.structure.io.validate_linked_groups(groups)
-        return FILE_TYPE.STRUCTURE_COLLECTION
+    if "header" in file.keys():
+        dtype = file["header"]["file"].attrs["data_type"]
+        if dtype == "halo_particles":
+            return FILE_TYPE.HALO_PARTICLES
+        elif dtype == "halo_profiles":
+            return FILE_TYPE.SOD_BINS
+        elif dtype == "halo_properties":
+            return FILE_TYPE.HALO_PROPERTIES
+        elif dtype == "galaxy_properties":
+            return FILE_TYPE.GALAXY_PROPERTIES
+        elif dtype == "galaxy_particles":
+            return FILE_TYPE.GALAXY_PARTICLES
+        elif dtype == "diffsky_fits":
+            return FILE_TYPE.SYNTHETIC_CATALOG
+        else:
+            raise ValueError(f"Unknown file type {dtype}")
 
     if not all("header" in group.keys() for group in file.values()):
         for subgroup in file.values():
@@ -206,8 +215,11 @@ def open(
 
     """
     if len(files) == 1 and isinstance(files[0], list):
-        return oc.open(*files[0], **open_kwargs)
-    handles = [h5py.File(f) for f in files]
+        file_list = files[0]
+    else:
+        file_list = list(files)
+    file_list.sort()
+    handles = [h5py.File(f) for f in file_list]
     file_types = list(map(get_file_type, handles))
     targets = make_all_targets(handles)
     targets = evaluate_load_conditions(targets, open_kwargs)

test/parallel/test_mpi.py

@@ -7,6 +7,7 @@
 import mpi4py
 import numpy as np
 import pytest
+from mpi4py import MPI
 from pytest_mpi.parallel_assert import parallel_assert
 
 import opencosmo as oc
@@ -22,6 +23,11 @@ def particle_path(snapshot_path):
     return snapshot_path / "haloparticles.hdf5"
 
 
+@pytest.fixture
+def profile_path(snapshot_path):
+    return snapshot_path / "sodproperties.hdf5"
+
+
 @pytest.fixture
 def malformed_header_path(input_path, tmp_path):
     update = {"n_dm": "foo"}
@@ -72,6 +78,20 @@ def test_mpi(input_path):
     parallel_assert(len(data) != 0)
 
 
+@pytest.mark.parallel(nprocs=4)
+def test_structure_collection_open(input_path, profile_path):
+    oc.open(input_path, profile_path)
+
+
+@pytest.mark.parallel(nprocs=4)
+def test_structure_collection_open_2(input_path, profile_path):
+    comm = MPI.COMM_WORLD
+    if comm.Get_rank() in [0, 2]:
+        oc.open(profile_path, input_path)
+    else:
+        oc.open(input_path, profile_path)
+
+
 @pytest.mark.parallel(nprocs=4)
 def test_partitioning_includes_all(input_path):
     with oc.open(input_path) as f:

test/test_collection.py

@@ -19,7 +19,7 @@ def multi_path(snapshot_path):
 
 @pytest.fixture
 def halo_paths(snapshot_path: Path):
-    files = ["haloparticles.hdf5", "haloproperties.hdf5", "sodproperties.hdf5"]
+    files = ["haloproperties.hdf5", "haloparticles.hdf5", "sodproperties.hdf5"]
     hdf_files = [snapshot_path / file for file in files]
     return list(hdf_files)
 
@@ -48,6 +48,27 @@ def conditional_path(multi_path, tmp_path):
     return path
 
 
+def test_open_structures(halo_paths, galaxy_paths):
+    c1 = oc.open(galaxy_paths)
+    assert isinstance(c1, oc.StructureCollection)
+    c2 = oc.open(halo_paths[0], galaxy_paths[1])
+    assert isinstance(c2, oc.StructureCollection)
+    c3 = oc.open(halo_paths[0], *galaxy_paths)
+    assert isinstance(c3, oc.StructureCollection)
+    c3 = oc.open(halo_paths[0], halo_paths[1])
+    assert isinstance(c3, oc.StructureCollection)
+    c3 = oc.open(halo_paths[0], halo_paths[2])
+    assert isinstance(c3, oc.StructureCollection)
+    c3 = oc.open(*halo_paths)
+    assert isinstance(c3, oc.StructureCollection)
+    c3 = oc.open(*halo_paths, galaxy_paths[1])
+    assert isinstance(c3, oc.StructureCollection)
+    c3 = oc.open(*halo_paths, *galaxy_paths)
+    assert isinstance(c3, oc.StructureCollection)
+    c3 = oc.open(halo_paths[0], halo_paths[1], *galaxy_paths)
+    assert isinstance(c3, oc.StructureCollection)
+
+
 def test_multi_filter(multi_path):
     collection = oc.open(multi_path)
     collection = collection.filter(oc.col("sod_halo_mass") > 0)
@@ -56,6 +77,20 @@ def test_multi_filter(multi_path):
         assert all(ds.data["sod_halo_mass"] > 0)
 
 
+def test_link_particles_only(halo_paths):
+    collection = oc.open(halo_paths[0], halo_paths[1])
+    assert isinstance(collection, oc.StructureCollection)
+    for key in collection.keys():
+        assert "particles" in key or key == "halo_properties"
+
+
+def test_link_profiles_only(halo_paths):
+    print(halo_paths)
+    collection = oc.open(halo_paths[0], halo_paths[2])
+    assert isinstance(collection, oc.StructureCollection)
+    assert set(collection.keys()) == {"halo_properties", "halo_profiles"}
+
+
 def test_galaxy_alias_fails_for_halos(halo_paths):
     ds = oc.open(halo_paths)
     with pytest.raises(AttributeError):