Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[python/c++] Remove tiledb-py from test_factory.py #3158

Merged
merged 3 commits into from
Oct 10, 2024
Merged
Show file tree
Hide file tree
Changes from 2 commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
12 changes: 10 additions & 2 deletions apis/python/src/tiledbsoma/common.cc
Original file line number Diff line number Diff line change
Expand Up @@ -209,7 +209,10 @@ py::dict meta(std::map<std::string, MetadataValue> metadata_mapping) {
}

void set_metadata(
SOMAObject& soma_object, const std::string& key, py::array value) {
SOMAObject& soma_object,
const std::string& key,
py::array value,
bool force) {
tiledb_datatype_t value_type = np_to_tdb_dtype(value.dtype());

// For https://github.com/single-cell-data/TileDB-SOMA/pull/2900:
Expand All @@ -228,7 +231,12 @@ void set_metadata(

auto value_num = is_tdb_str(value_type) ? value.nbytes() : value.size();
soma_object.set_metadata(
key, value_type, value_num, value_num > 0 ? value.data() : nullptr);
key,
value_type,
value_num,
value_num > 0 ? value.data() : nullptr,
force); // The force flag is intended to only be toggled for testing in
// test_factory.py
nguyenv marked this conversation as resolved.
Show resolved Hide resolved
}

} // namespace tiledbsoma
5 changes: 4 additions & 1 deletion apis/python/src/tiledbsoma/common.h
Original file line number Diff line number Diff line change
Expand Up @@ -28,7 +28,10 @@ std::optional<py::object> to_table(

py::dict meta(std::map<std::string, MetadataValue> metadata_mapping);
void set_metadata(
SOMAObject& soma_object, const std::string& key, py::array value);
SOMAObject& soma_object,
const std::string& key,
py::array value,
bool force = false);

class PyQueryCondition {
private:
Expand Down
16 changes: 14 additions & 2 deletions apis/python/src/tiledbsoma/soma_array.cc
Original file line number Diff line number Diff line change
Expand Up @@ -242,6 +242,9 @@ void load_soma_array(py::module& m) {
return pa_schema_import(
py::capsule(array.arrow_schema().get()));
})
.def(
"config_options_from_schema",
&SOMAArray::config_options_from_schema)
.def("context", &SOMAArray::ctx)

// After this are short functions expected to be invoked when the coords
Expand Down Expand Up @@ -951,9 +954,18 @@ void load_soma_array(py::module& m) {
return meta(array.get_metadata());
})

.def("set_metadata", set_metadata)
.def(
"set_metadata",
set_metadata,
py::arg("key"),
py::arg("value"),
py::arg("force") = false)

.def("delete_metadata", &SOMAArray::delete_metadata)
.def(
"delete_metadata",
&SOMAArray::delete_metadata,
py::arg("key"),
py::arg("force") = false)

.def(
"get_metadata",
Expand Down
18 changes: 14 additions & 4 deletions apis/python/src/tiledbsoma/soma_group.cc
Original file line number Diff line number Diff line change
Expand Up @@ -94,18 +94,28 @@ void load_soma_group(py::module& m) {
return py::none();
return py::cast(group.timestamp()->second);
})

.def_property_readonly(
"meta",
[](SOMAGroup& group) -> py::dict {
return meta(group.get_metadata());
})

.def(
"set_metadata",
[](SOMAGroup& group, const std::string& key, py::array value) {
set_metadata(group, key, value);
})
.def("delete_metadata", &SOMAGroup::delete_metadata)
set_metadata,
py::arg("key"),
py::arg("value"),
py::arg("force") = false)

.def(
"delete_metadata",
&SOMAGroup::delete_metadata,
py::arg("key"),
py::arg("force") = false)

.def("has_metadata", &SOMAGroup::has_metadata)

.def("metadata_num", &SOMAGroup::metadata_num);
}
} // namespace libtiledbsomacpp
195 changes: 78 additions & 117 deletions apis/python/tests/test_factory.py
Original file line number Diff line number Diff line change
Expand Up @@ -2,140 +2,86 @@
from typing import Type

import numpy as np
import pyarrow as pa
import pytest

import tiledbsoma as soma
from tiledbsoma import _constants
import tiledb

UNKNOWN_ENCODING_VERSION = "3141596"


@pytest.fixture
def tiledb_object_uri(tmp_path, object_type, metadata_typename, encoding_version):
def tiledb_object_uri(tmp_path, metadata_typename, encoding_version, soma_type):
"""Create an object with specified metadata"""
object_uri = f"{tmp_path}/object"

# create object
if object_type == "array":
schema = tiledb.ArraySchema(
domain=tiledb.Domain(
tiledb.Dim(name="rows", domain=(0, 100), dtype=np.int64)
),
attrs=[
tiledb.Attr(name="a", dtype=np.int32),
tiledb.Attr(name="b", dtype=np.float32),
],
kwargs = {}

if issubclass(soma_type, (soma.DenseNDArray, soma.SparseNDArray)):
kwargs["type"] = pa.int64()
kwargs["shape"] = (100,)
elif issubclass(soma_type, soma.DataFrame):
kwargs["schema"] = pa.schema(
[("rows", pa.int64()), ("a", pa.int32()), ("b", pa.float32())]
)
tiledb.Array.create(object_uri, schema)
with tiledb.open(object_uri, mode="w") as A:
_setmetadata(A, metadata_typename, encoding_version)
else:
tiledb.group_create(object_uri)
with tiledb.Group(object_uri, mode="w") as G:
_setmetadata(G, metadata_typename, encoding_version)

with soma_type.create(object_uri, **kwargs) as soma_obj:
_setmetadata(soma_obj, metadata_typename, encoding_version)

return object_uri


@pytest.mark.parametrize(
"object_type,metadata_typename,encoding_version,expected_soma_type",
"metadata_typename, encoding_version, soma_type",
[
("group", "SOMAExperiment", _constants.SOMA_ENCODING_VERSION, soma.Experiment),
(
"group",
"SOMAMeasurement",
_constants.SOMA_ENCODING_VERSION,
soma.Measurement,
),
("group", "SOMACollection", _constants.SOMA_ENCODING_VERSION, soma.Collection),
("array", "SOMADataFrame", _constants.SOMA_ENCODING_VERSION, soma.DataFrame),
(
"array",
"SOMADenseNDArray",
_constants.SOMA_ENCODING_VERSION,
soma.DenseNDArray,
),
(
"array",
"SOMADenseNdArray",
_constants.SOMA_ENCODING_VERSION,
soma.DenseNDArray,
),
(
"array",
"SOMASparseNDArray",
_constants.SOMA_ENCODING_VERSION,
soma.SparseNDArray,
),
(
"array",
"SOMASparseNdArray",
_constants.SOMA_ENCODING_VERSION,
soma.SparseNDArray,
),
("SOMAExperiment", _constants.SOMA_ENCODING_VERSION, soma.Experiment),
("SOMAMeasurement", _constants.SOMA_ENCODING_VERSION, soma.Measurement),
("SOMACollection", _constants.SOMA_ENCODING_VERSION, soma.Collection),
("SOMADataFrame", _constants.SOMA_ENCODING_VERSION, soma.DataFrame),
("SOMADenseNDArray", _constants.SOMA_ENCODING_VERSION, soma.DenseNDArray),
("SOMADenseNdArray", _constants.SOMA_ENCODING_VERSION, soma.DenseNDArray),
("SOMASparseNDArray", _constants.SOMA_ENCODING_VERSION, soma.SparseNDArray),
("SOMASparseNdArray", _constants.SOMA_ENCODING_VERSION, soma.SparseNDArray),
],
)
def test_open(tiledb_object_uri, expected_soma_type: Type):
def test_open(tiledb_object_uri, soma_type: Type):
"""Happy path tests"""
# TODO: Fix Windows test failures without the following.
sleep(0.01)
soma_obj = soma.open(tiledb_object_uri)
assert isinstance(soma_obj, expected_soma_type)
typed_soma_obj = soma.open(tiledb_object_uri, soma_type=expected_soma_type)
assert isinstance(typed_soma_obj, expected_soma_type)
str_typed_soma_obj = soma.open(
tiledb_object_uri, soma_type=expected_soma_type.soma_type
)
assert isinstance(str_typed_soma_obj, expected_soma_type)
assert expected_soma_type.exists(tiledb_object_uri)
assert isinstance(soma_obj, soma_type)
typed_soma_obj = soma.open(tiledb_object_uri, soma_type=soma_type)
assert isinstance(typed_soma_obj, soma_type)
str_typed_soma_obj = soma.open(tiledb_object_uri, soma_type=soma_type.soma_type)
assert isinstance(str_typed_soma_obj, soma_type)
assert soma_type.exists(tiledb_object_uri)


@pytest.mark.parametrize(
("object_type", "metadata_typename", "encoding_version", "wrong_type"),
("metadata_typename", "encoding_version", "soma_type"),
[
("group", "SOMAExperiment", _constants.SOMA_ENCODING_VERSION, soma.Measurement),
("group", "SOMAMeasurement", _constants.SOMA_ENCODING_VERSION, soma.DataFrame),
(
"group",
"SOMAMeasurement",
_constants.SOMA_ENCODING_VERSION,
"SOMACollection",
),
(
"array",
"SOMADenseNDArray",
_constants.SOMA_ENCODING_VERSION,
soma.Collection,
),
(
"array",
"SOMADenseNdArray",
_constants.SOMA_ENCODING_VERSION,
soma.SparseNDArray,
),
(
"array",
"SOMASparseNDArray",
_constants.SOMA_ENCODING_VERSION,
"SOMADenseNDArray",
),
("SOMAExperiment", _constants.SOMA_ENCODING_VERSION, soma.Measurement),
("SOMAMeasurement", _constants.SOMA_ENCODING_VERSION, soma.DataFrame),
("SOMAMeasurement", _constants.SOMA_ENCODING_VERSION, soma.Collection),
("SOMADenseNDArray", _constants.SOMA_ENCODING_VERSION, soma.Collection),
("SOMADenseNdArray", _constants.SOMA_ENCODING_VERSION, soma.SparseNDArray),
("SOMASparseNDArray", _constants.SOMA_ENCODING_VERSION, soma.DenseNDArray),
],
)
def test_open_wrong_type(tiledb_object_uri, wrong_type):
def test_open_wrong_type(tiledb_object_uri, soma_type):
with pytest.raises((soma.SOMAError, TypeError)):
soma.open(tiledb_object_uri, soma_type=wrong_type)
soma.open(tiledb_object_uri, soma_type=soma_type)


@pytest.mark.parametrize(
"object_type,metadata_typename,encoding_version",
"metadata_typename, encoding_version, soma_type",
[
("group", "SOMAExperiment", UNKNOWN_ENCODING_VERSION),
("group", "SOMAMeasurement", UNKNOWN_ENCODING_VERSION),
("group", "SOMACollection", UNKNOWN_ENCODING_VERSION),
("array", "SOMADataFrame", UNKNOWN_ENCODING_VERSION),
("array", "SOMADenseNDArray", UNKNOWN_ENCODING_VERSION),
("array", "SOMASparseNDArray", UNKNOWN_ENCODING_VERSION),
("SOMAExperiment", UNKNOWN_ENCODING_VERSION, soma.Experiment),
("SOMAMeasurement", UNKNOWN_ENCODING_VERSION, soma.Measurement),
("SOMACollection", UNKNOWN_ENCODING_VERSION, soma.Collection),
("SOMADataFrame", UNKNOWN_ENCODING_VERSION, soma.DataFrame),
("SOMADenseNDArray", UNKNOWN_ENCODING_VERSION, soma.DenseNDArray),
("SOMASparseNDArray", UNKNOWN_ENCODING_VERSION, soma.SparseNDArray),
],
)
def test_factory_unsupported_version(tiledb_object_uri):
Expand All @@ -147,25 +93,33 @@ def test_factory_unsupported_version(tiledb_object_uri):


@pytest.mark.parametrize(
"object_type,metadata_typename,encoding_version",
"metadata_typename, encoding_version, soma_type",
[
("array", "AnUnknownTypeName", _constants.SOMA_ENCODING_VERSION),
("group", "AnUnknownTypeName", _constants.SOMA_ENCODING_VERSION),
("array", "AnUnknownTypeName", None),
("group", "AnUnknownTypeName", None),
("array", None, _constants.SOMA_ENCODING_VERSION),
("group", None, _constants.SOMA_ENCODING_VERSION),
("array", None, None),
("group", None, None),
(
"array",
"AnUnknownTypeName",
_constants.SOMA_ENCODING_VERSION,
soma.DataFrame,
), # Invalid type
(
"AnUnknownTypeName",
_constants.SOMA_ENCODING_VERSION,
soma.Collection,
), # Invalid type
("AnUnknownTypeName", None, soma.DataFrame), # Invalid type and no version
("AnUnknownTypeName", None, soma.Collection), # Invalid type and no version
(None, _constants.SOMA_ENCODING_VERSION, soma.DataFrame), # No type given
(None, _constants.SOMA_ENCODING_VERSION, soma.Collection), # No type give
(None, None, soma.DataFrame), # Neither type nor version filled
(None, None, soma.Collection), # Neither type nor version filled
(
"SOMACollection",
_constants.SOMA_ENCODING_VERSION,
), # Collections can't be arrays
soma.DataFrame,
), # Collections can't be an array
(
"group",
"SOMADataFrame",
_constants.SOMA_ENCODING_VERSION,
soma.Collection,
), # DataFrame can't be a group
],
)
Expand All @@ -182,11 +136,18 @@ def test_factory_unknown_files():


def _setmetadata(open_tdb_object, metadata_typename, encoding_version):
"""set only those values which are not None"""
changes = {}
"""force modify the metadata values"""
set_metadata = open_tdb_object._handle._handle.set_metadata
nguyenv marked this conversation as resolved.
Show resolved Hide resolved
del_metadata = open_tdb_object._handle._handle.delete_metadata

if metadata_typename is not None:
changes[_constants.SOMA_OBJECT_TYPE_METADATA_KEY] = metadata_typename
val = np.array([metadata_typename], "S")
set_metadata(_constants.SOMA_OBJECT_TYPE_METADATA_KEY, val, True)
else:
del_metadata(_constants.SOMA_OBJECT_TYPE_METADATA_KEY, True)

if encoding_version is not None:
changes[_constants.SOMA_ENCODING_VERSION_METADATA_KEY] = encoding_version
if changes:
open_tdb_object.meta.update(changes)
val = np.array([encoding_version], "S")
set_metadata(_constants.SOMA_ENCODING_VERSION_METADATA_KEY, val, True)
else:
del_metadata(_constants.SOMA_ENCODING_VERSION_METADATA_KEY, True)
6 changes: 3 additions & 3 deletions libtiledbsoma/src/soma/soma_array.cc
Original file line number Diff line number Diff line change
Expand Up @@ -1094,12 +1094,12 @@ void SOMAArray::set_metadata(
metadata_.insert(mdpair);
}

void SOMAArray::delete_metadata(const std::string& key) {
if (key.compare(SOMA_OBJECT_TYPE_KEY) == 0) {
void SOMAArray::delete_metadata(const std::string& key, bool force) {
if (!force && key.compare(SOMA_OBJECT_TYPE_KEY) == 0) {
throw TileDBSOMAError(SOMA_OBJECT_TYPE_KEY + " cannot be deleted.");
}

if (key.compare(ENCODING_VERSION_KEY) == 0) {
if (!force && key.compare(ENCODING_VERSION_KEY) == 0) {
throw TileDBSOMAError(ENCODING_VERSION_KEY + " cannot be deleted.");
}

Expand Down
2 changes: 1 addition & 1 deletion libtiledbsoma/src/soma/soma_array.h
Original file line number Diff line number Diff line change
Expand Up @@ -664,7 +664,7 @@ class SOMAArray : public SOMAObject {
* @note If the key does not exist, this will take no effect
* (i.e., the function will not error out).
*/
void delete_metadata(const std::string& key);
void delete_metadata(const std::string& key, bool force = false);

/**
* @brief Given a key, get the associated value datatype, number of
Expand Down
Loading
Loading