Pin numpy version (#49)

.github/workflows/ci.yaml

@@ -13,7 +13,7 @@ jobs:
     strategy:
       matrix:
         os: [ubuntu-latest, macos-latest, windows-latest]
-        python-version: ["3.8", "3.9", "3.10"]
+        python-version: ["3.9", "3.10", "3.11"]
 
     steps:
       - uses: actions/checkout@v2

.github/workflows/deploy_public.yml

@@ -35,10 +35,10 @@ jobs:
     steps:
       - uses: actions/checkout@v3
 
-      - name: Set up Python 3.9
+      - name: Set up Python 3.10
         uses: actions/setup-python@v4
         with:
-          python-version: "3.9"
+          python-version: "3.10"
 
       - name: Install dependencies
         run: |

docs/user_guide/concepts_utils/data_source.rst

@@ -92,7 +92,7 @@ next time it is requested.
         drio_client,
         labels,
         cache='.cache'
-        cache_size=pd.to_timedelta("3H")
+        cache_size=pd.to_timedelta("3h")
     )
 
 .. tip::
@@ -137,7 +137,7 @@ the index is datetime-like, fixed-frequency start and end index pairs can be gen
 
 
     start, end = iter_index.date_range(
-        start="2020-01-01 00:00", end="2020-02-01 00:00", freq="1H"
+        start="2020-01-01 00:00", end="2020-02-01 00:00", freq="1h"
     )
 
     for index_i, data_i in source.iter(start, end):

docs/user_guide/concepts_utils/example.rst

@@ -73,7 +73,7 @@ are stored in *Azure Blob Storage* using the :class:`~fourinsight.engineroom.uti
     end = pd.to_datetime("now", utc=True)
 
     # Iterate over the data in 1-hour chunks
-    for index_i, data_i in source.iter(*iter_index.date_range(start, end, freq="1H")):
+    for index_i, data_i in source.iter(*iter_index.date_range(start, end, freq="1h")):
         results.new_row(index_i)
 
         series_a = data_i["A"]

fourinsight/engineroom/utils/_core.py

@@ -458,7 +458,11 @@ def pull(self, raise_on_missing=True, strict=True):
 
         self._handler.seek(0)
         df_source = pd.read_csv(
-            self._handler, index_col=0, parse_dates=True, dtype=self._headers
+            self._handler,
+            index_col=0,
+            parse_dates=True,
+            dtype=self._headers,
+            date_format="ISO8601",
         )
 
         if strict and set(df_source.columns) != set(self._headers.keys()):

fourinsight/engineroom/utils/_datamanage.py

@@ -526,7 +526,7 @@ class DrioDataSource(BaseDataSource):
     cache : str, optional
         Cache folder. If ``None`` (default), caching is disabled.
     cache_size :
-        Cache size as an index partition (see Notes). Defaults to ``'24H'`` if the
+        Cache size as an index partition (see Notes). Defaults to ``'24h'`` if the
         `index_type` is 'datetime', otherwise ``None`` is default.
     **get_kwargs : optional
         Keyword arguments that will be passed on to the ``drio_client.get`` method.
@@ -578,7 +578,7 @@ def __init__(
 
         if index_type == "datetime":
             index_converter = DatetimeIndexConverter()
-            cache_size = cache_size or "24H"
+            cache_size = cache_size or "24h"
         elif index_type == "integer":
             index_converter = IntegerIndexConverter()
         elif isinstance(index_type, BaseIndexConverter):
@@ -798,4 +798,10 @@ def get(self, start, end, refresh_cache=False):
             source_i.get(start_i, end_i, refresh_cache=refresh_cache)
             for (start_i, end_i), source_i in zip(pairwise(index_list), sources_list)
         ]
+
+        data_list = [df.dropna(how="all", axis=1) for df in data_list if not df.empty]
+
+        if not data_list:
+            return pd.DataFrame([], columns=self._labels)
+
         return pd.concat(data_list).infer_objects()

setup.cfg

@@ -10,17 +10,18 @@ url = https://4insight.io/
 classifiers =
     License :: OSI Approved :: MIT License
     Operating System :: OS Independent
-    Programming Language :: Python :: 3.8
     Programming Language :: Python :: 3.9
     Programming Language :: Python :: 3.10
+    Programming Language :: Python :: 3.11
 
 [options]
 packages = find_namespace:
-python_requires = >=3.8
+python_requires = >=3.9
 install_requires =
     pandas
     azure-storage-blob >= 12.4.0
     pyarrow
+    numpy < 2.0.0
 
 [options.packages.find]
 include =

tests/test_datamanage.py

@@ -689,7 +689,7 @@ def test_partition_start_end_float(self):
     def test_partition_start_end_datetime(self):
         start = pd.to_datetime("2020-01-01 03:00", utc=True)
         end = pd.to_datetime("2020-01-01 09:00", utc=True)
-        partition = pd.to_timedelta("3H")
+        partition = pd.to_timedelta("3h")
         reference = pd.to_datetime("2020-01-01 02:00", utc=True)
         out = BaseDataSourceForTesting._partition_start_end(
             start, end, partition, reference
@@ -705,15 +705,15 @@ def test_partition_start_end_datetime(self):
     def test__is_cached_false(self, tmp_path):
         cache_dir = tmp_path / ".cache"
         source = BaseDataSourceForTesting(
-            DatetimeIndexConverter(), cache=cache_dir, cache_size="1H"
+            DatetimeIndexConverter(), cache=cache_dir, cache_size="1h"
         )
         assert cache_dir.exists()
         assert source._is_cached("filename") is False
 
     def test__is_cached_true(self, tmp_path):
         cache_dir = tmp_path / ".cache"
         source = BaseDataSourceForTesting(
-            DatetimeIndexConverter(), cache=cache_dir, cache_size="1H"
+            DatetimeIndexConverter(), cache=cache_dir, cache_size="1h"
         )
         assert cache_dir.exists()
         (cache_dir / "filename").touch()
@@ -722,7 +722,7 @@ def test__is_cached_true(self, tmp_path):
     def test__cache_read(self, tmp_path):
         cache_dir = tmp_path / ".cache"
         source = BaseDataSourceForTesting(
-            DatetimeIndexConverter(), cache=cache_dir, cache_size="1H"
+            DatetimeIndexConverter(), cache=cache_dir, cache_size="1h"
         )
 
         df = pd.DataFrame(data={"filename": [2, 4, 6], "a": [1, 2, 3]})
@@ -735,7 +735,7 @@ def test__cache_read(self, tmp_path):
     def test__cache_write(self, tmp_path):
         cache_dir = tmp_path / ".cache"
         source = BaseDataSourceForTesting(
-            DatetimeIndexConverter(), cache=cache_dir, cache_size="1H"
+            DatetimeIndexConverter(), cache=cache_dir, cache_size="1h"
         )
 
         df = pd.DataFrame(data={"a": [1, 2, 3]}, index=[2, 4, 6])
@@ -967,7 +967,7 @@ def test__init__(self, tmp_path):
         assert source._get_kwargs == {"convert_date": False, "raise_empty": True}
         assert isinstance(source._index_converter, DatetimeIndexConverter)
         assert source._cache == Path(cache_dir)
-        assert source._cache_size == pd.to_timedelta("24H")
+        assert source._cache_size == pd.to_timedelta("24h")
 
     def test__init__integer(self):
         drio_client = Mock()
@@ -1579,15 +1579,15 @@ def test_to_universal_index_arraylike(self):
         np.testing.assert_array_equal(out, expect)
 
     def test_to_universal_delta(self):
-        out = DatetimeIndexConverter().to_universal_delta("3H")
-        expect = pd.to_timedelta("3H")
+        out = DatetimeIndexConverter().to_universal_delta("3h")
+        expect = pd.to_timedelta("3h")
         assert out == expect
 
     def test_to_universal_delta_arraylike(self):
         out = DatetimeIndexConverter().to_universal_delta(
-            ["3H", "24H", pd.to_timedelta("2D")]
+            ["3h", "24h", pd.to_timedelta("2D")]
         )
-        expect = pd.to_timedelta(["3H", "24H", pd.to_timedelta("2D")])
+        expect = pd.to_timedelta(["3h", "24h", pd.to_timedelta("2D")])
         np.testing.assert_array_equal(out, expect)
 
     def test_to_native_index(self):

tests/test_iter_index.py

@@ -7,7 +7,7 @@ class Test_date_range:
     def test_start_end_freq(self):
         start = "2020-01-01 00:00"
         end = "2020-01-01 05:00"
-        freq = "1H"
+        freq = "1h"
         start_out, end_out = iter_index.date_range(start=start, end=end, freq=freq)
 
         date_range = pd.date_range(start=start, end=end, freq=freq)

tox.ini

@@ -12,7 +12,7 @@ deps =
     pytest-cov
 
 [testenv:docs]
-basepython = python3.9
+basepython = python3.10
 commands = sphinx-build -W -b html -d {toxworkdir}/docs_doctree docs {toxworkdir}/docs_out
 deps =
     sphinx == 5.3.0