chore: Fix compatibility tests (#2685)

* chore: Fix compatibility tests

* DEBUGGGG

* fix: update deps, fix unit tests

* fix: FE deps

* chore: update compatibility matrix

---------

Co-authored-by: kaloster <rkalo@contractor.chanzuckerberg.com>

.github/workflows/compatibility_tests.yml

@@ -14,7 +14,7 @@ jobs:
   docker-build:
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v2
+      - uses: actions/checkout@v3
       - name: Set up Python ${{ matrix.python-version }}
         uses: actions/setup-python@v4
         with:
@@ -29,21 +29,21 @@ jobs:
       fail-fast: false
       matrix:
         # note: The `macos-latest` is latest Catalina version, and not Big Sur. So we explicitly ask for Big Sur (`macos-11`)
-        os: [ubuntu-latest, macos-latest, macos-11]
+        os: [ubuntu-latest, macos-latest, macos-13]
-        python-version: [3.8, 3.9, 3.10, 3.11]
+        python-version: ["3.10", "3.11"]
         cellxgene_build: [main, latest]
         # add anndata pinned version test for subset of matrix configurations,
         # in order to reduce matrix cross-product explosion
         include:
-          - python-version: 3.9
+          - python-version: 3.10
             cellxgene_build: latest
             # TODO: dynamically use the literal version in requirements.txt,
             # to avoid having to update this in manually in the future
             # TODO: Do not bother running this if anndata latest version
             # matches this pinned version, to avoid a redundant test
-            anndata_version: "==0.10.3"
+            anndata_version: "==0.10.9"
     steps:
-      - uses: actions/checkout@v2
+      - uses: actions/checkout@v3
       - name: Set up Python ${{ matrix.python-version }}
         uses: actions/setup-python@v4
         with:
@@ -55,7 +55,7 @@ jobs:
         run: echo "BREW_CACHE=`brew --cache`" >> $GITHUB_ENV
       # FIXME: Only working for Linux
       - name: Python cache
-        uses: actions/cache@v1
+        uses: actions/cache@v2
         with:
           path: ${{ env.PIP_CACHE }}
           key: ${{ runner.os }}-pip-${{ hashFiles('**/requirements*.txt') }}
@@ -96,6 +96,7 @@ jobs:
           # keep same pip pkg versions as in the cxg release
           sed -i'' -e 's/-r requirements.txt//' server/requirements-dev.txt
           pip install -r server/requirements-dev.txt
+          pip install --force-reinstall numpy==2.0.1 numba>=0.60.0 pandas
       - name: Install anndata version per matrix variable
         run: pip install anndata${{ matrix.anndata_version }}
       - name: Install node

client/.nvmrc

@@ -0,0 +1 @@
+16.20.0

client/__tests__/e2e/puppeteer.setup.js

@@ -13,7 +13,7 @@ import * as ENV_DEFAULT from "../../../environment.default.json";
 // a test can take more time to finish, so we don't want
 // jest to shut off the test too soon
 jest.setTimeout(2 * 60 * 1000);
-setDefaultOptions({ timeout: 20 * 1000 });
+setDefaultOptions({ timeout: 60 * 1000 });
 
 jest.retryTimes(ENV_DEFAULT.RETRY_ATTEMPTS);
 

client/package.json

@@ -18,7 +18,8 @@
   },
   "engineStrict": true,
   "engines": {
-    "npm": ">=3.0.0"
+    "npm": ">=3.0.0",
+    "node": "^16.0.0"
   },
   "eslintConfig": {
     "extends": "./configuration/eslint/eslint.js"
@@ -123,7 +124,7 @@
     "jest-circus": "^27.0.6",
     "jest-environment-puppeteer": "^5.0.1",
     "jest-fetch-mock": "^3.0.3",
-    "jest-puppeteer": "^5.0.1",
+    "jest-puppeteer": "^6.2.0",
     "json-loader": "^0.5.7",
     "lint-staged": "^10.2.11",
     "lodash": "^4.17.21",
@@ -134,7 +135,7 @@
     "lodash.zip": "^4.2.0",
     "mini-css-extract-plugin": "^1.5.0",
     "prettier": "^2.0.5",
-    "puppeteer": "^8.0.0",
+    "puppeteer": "^10.4.0",
     "rimraf": "^3.0.2",
     "serve-favicon": "^2.5.0",
     "terser-webpack-plugin": "^5.1.1",

server/common/utils/type_conversion_utils.py

@@ -116,7 +116,7 @@ def _get_type_info(array: Union[np.ndarray, pd.Series, pd.Index]) -> Tuple[np.dt
         raise TypeError("Unsupported data type.")
 
     dtype = array.dtype
-
+    
     res = _get_type_info_from_dtype(dtype)
     if res is not None:
         return res
@@ -140,7 +140,6 @@ def _get_type_info(array: Union[np.ndarray, pd.Series, pd.Index]) -> Tuple[np.dt
 
     if dtype.kind in ["i", "u"] and _can_cast_array_values_to_int32(array):
         return (np.int32, {"type": "int32"})
-
     if dtype.kind == "f":
         _float64_warning(array.dtype)
         return (np.float32, {"type": "float32"})

server/requirements.txt

@@ -12,10 +12,10 @@ flatten-dict>=0.2.0
 fsspec>0.8.0
 gunicorn>=20.0.4
 h5py>=3.0.0
-numba>=0.51.2
+numba>=0.60.0
-numpy>1.22
+numpy==2.0.1
 packaging>=20.0
-pandas<2.0.0
+pandas>=2.2.2
 PyYAML>=5.4  # CVE-2020-14343
 requests>=2.22.0
 s3fs==0.4.2

test/unit/compute/test_est_dist.py

@@ -65,13 +65,13 @@ class EstDistTest(unittest.TestCase):
 
         # non-finites
         self.assertEqual(estimate_approximate_distribution(np.array([np.nan])), XApproximateDistribution.NORMAL)
-        self.assertEqual(estimate_approximate_distribution(np.array([np.PINF])), XApproximateDistribution.NORMAL)
+        self.assertEqual(estimate_approximate_distribution(np.array([np.inf])), XApproximateDistribution.NORMAL)
-        self.assertEqual(estimate_approximate_distribution(np.array([np.NINF])), XApproximateDistribution.NORMAL)
+        self.assertEqual(estimate_approximate_distribution(np.array([np.inf])), XApproximateDistribution.NORMAL)
         self.assertEqual(
-            estimate_approximate_distribution(np.array([np.PINF, np.NINF, 0])), XApproximateDistribution.NORMAL
+            estimate_approximate_distribution(np.array([np.inf, np.inf, 0])), XApproximateDistribution.NORMAL
         )
         self.assertEqual(
-            estimate_approximate_distribution(np.array([np.nan, np.PINF, np.NINF])), XApproximateDistribution.NORMAL
+            estimate_approximate_distribution(np.array([np.nan, np.inf, np.inf])), XApproximateDistribution.NORMAL
         )
 
         raw = np.random.exponential(scale=1000, size=(50, 3))
@@ -82,15 +82,15 @@ class EstDistTest(unittest.TestCase):
             XApproximateDistribution.COUNT,
         )
         self.assertEqual(
-            estimate_approximate_distribution(put(raw, [1], [np.PINF])),
+            estimate_approximate_distribution(put(raw, [1], [np.inf])),
             XApproximateDistribution.COUNT,
         )
         self.assertEqual(
-            estimate_approximate_distribution(put(raw, [1], [np.NINF])),
+            estimate_approximate_distribution(put(raw, [1], [np.inf])),
             XApproximateDistribution.COUNT,
         )
         self.assertEqual(
-            estimate_approximate_distribution(put(raw, [1, 3, 88], [np.nan, np.PINF, np.NINF])),
+            estimate_approximate_distribution(put(raw, [1, 3, 88], [np.nan, np.inf, np.inf])),
             XApproximateDistribution.COUNT,
         )
         self.assertEqual(
@@ -103,15 +103,15 @@ class EstDistTest(unittest.TestCase):
             XApproximateDistribution.NORMAL,
         )
         self.assertEqual(
-            estimate_approximate_distribution(put(logged, [1], [np.PINF])),
+            estimate_approximate_distribution(put(logged, [1], [np.inf])),
             XApproximateDistribution.NORMAL,
         )
         self.assertEqual(
-            estimate_approximate_distribution(put(logged, [1], [np.NINF])),
+            estimate_approximate_distribution(put(logged, [1], [np.inf])),
             XApproximateDistribution.NORMAL,
         )
         self.assertEqual(
-            estimate_approximate_distribution(put(logged, [1, 3, 88], [np.nan, np.PINF, np.NINF])),
+            estimate_approximate_distribution(put(logged, [1, 3, 88], [np.nan, np.inf, np.inf])),
             XApproximateDistribution.NORMAL,
         )
         self.assertEqual(

test/unit/utils/test_jsonify_strict.py

@@ -16,10 +16,10 @@ class TestJsonifyStrict(unittest.TestCase):
             jsonify_strict({"nan": [np.nan]})
 
         with self.assertRaises(ValueError):
-            jsonify_strict({"pinf": [np.PINF]})
+            jsonify_strict({"pinf": [np.inf]})
 
         with self.assertRaises(ValueError):
-            jsonify_strict({"ninf": [np.NINF]})
+            jsonify_strict({"ninf": [np.inf]})
 
     def test_jsonify_numpy_ndarray(self):
         values = {

test/unit/utils/test_type_conversion_utils.py

@@ -42,7 +42,7 @@ class TestTypeConversionUtils(unittest.TestCase):
             with self.assertRaises(TypeError):
                 get_schema_type_hint_from_dtype(np.dtype(dtype))
 
-        for dtype in [np.float16, np.float32, np.float64]:
+        for dtype in [np.float32, np.float64]:
             self.assertEqual(get_schema_type_hint_from_dtype(np.dtype(dtype)), {"type": "float32"})
 
         for dtype in [np.dtype(object), np.dtype(str)]:
@@ -123,17 +123,18 @@ int_OK_cases = [
 
 float_OK_cases = [
     {
+        "test_case": "float_OK_cases",
         "data": data,
         "expected_encoding_dtype": np.float32,
         "expected_schema_hint": {"type": "float32"},
-        "logs": None if data.dtype != np.float64 else {"level": logging.WARNING, "output": "may lose precision"},
+        "logs": None if dtype == np.float32 else {"level": logging.WARNING, "output": "may lose precision"},
     }
-    for dtype in [np.float16, np.float32, np.float64]
+    for dtype in [np.float32, np.float64]
     for data in [
         np.arange(-128, 1000, dtype=dtype),
         pd.Series(np.arange(-128, 1000, dtype=dtype)),
         pd.Index(np.arange(-129, 1000, dtype=dtype)),
-        np.array([-np.nan, np.NINF, -1, np.NZERO, 0, np.PZERO, 1, np.PINF, np.nan], dtype=dtype),
+        np.array([-np.nan, np.inf, -1, 0.0, 0, 0.0, 1, np.inf, np.nan], dtype=dtype),
         np.array([np.finfo(dtype).min, 0, np.finfo(dtype).max], dtype=dtype),
         sparse.csr_matrix((10, 100), dtype=dtype),
     ]
@@ -198,12 +199,13 @@ category_numeric_OK_cases = [
     # numeric, no NA/NaN, float
     *[
         {
+            "test_case": "numeric, no NA/NaN, float",
             "data": data,
             "expected_encoding_dtype": np.float32,
             "expected_schema_hint": {"type": "categorical"},
-            "logs": {"level": logging.WARNING, "output": "may lose precision"},
+            "logs": None if dtype == np.float32 else {"level": logging.WARNING, "output": "may lose precision"},
         }
-        for dtype in [np.float16, np.float32, np.float64]
+        for dtype in [np.float32, np.float64]
         for data in [
             pd.Series(np.array([0, 1, 2], dtype=dtype), dtype="category"),
             pd.Series(np.array([0, 1, 2], dtype=dtype), dtype="category").cat.remove_categories([1]),
@@ -213,10 +215,11 @@ category_numeric_OK_cases = [
     # numeric, has NA-induced cast to float32
     *[
         {
+            "test_case": "numeric, has NA-induced cast to float32",
             "data": data,
             "expected_encoding_dtype": np.float32,
             "expected_schema_hint": {"type": "categorical"},
-            "logs": {"level": logging.WARNING, "output": "may lose precision"},
+            "logs": None if dtype == np.float32 else {"level": logging.WARNING, "output": "may lose precision"},
         }
         for dtype in [
             np.int8,
@@ -227,7 +230,6 @@ category_numeric_OK_cases = [
             np.uint32,
             np.int64,
             np.uint64,
-            np.float16,
             np.float32,
             np.float64,
         ]
@@ -312,7 +314,6 @@ class TestTypeInference(unittest.TestCase, AssertNoLog):
                     self.assertEqual(encoding_dtype, self.expected_encoding_dtype)
                     self.assertEqual(schema_hint, self.expected_schema_hint)
                     self.assertIn(logs["output"], logger.output[0])
-
             else:
                 with self.assertNoLogs(logging.getLogger(), logging.WARNING):
                     encoding_dtype, schema_hint = get_dtype_and_schema_of_array(self.data)