v1.14.4 (#368)

* v1.14.4-rc1

* v1.14.4-rc2

---------

Co-authored-by: rtosholdings-bot <rtosholdings-bot@sig.com>

.github/workflows/python-package.yml

@@ -194,7 +194,7 @@ jobs:
       - name: Test riptable
         run: |
           set -ex
-          python -m riptable.tests.run
+          python -m pytest --pyargs riptable.tests
           # disable tooling integration tests until they work
           # ipython -m pytest riptable/test_tooling_integration
           # disable hypothesis tests until they run faster, are more consistent, and are easier to investigate
@@ -261,7 +261,7 @@ jobs:
           conda list
       - name: Test with pytest
         run: |
-          python -m riptable.tests.run
+          python -m pytest --pyargs riptable.tests
       - name: Tooling integration tests
         run: |
           echo "DISABLED until tooling tests can be updated"

conda_recipe/meta.yaml

@@ -5,8 +5,7 @@ package:
 build:
   number: 0
   noarch: python
-  # Use Python installed in host environment.
-  script: "{{ PYTHON }} -m pip install -v --no-deps --ignore-installed ."
+  script: "pip install -v --no-deps --no-build-isolation ."
 
 source:
   path: ..
@@ -16,13 +15,13 @@ requirements:
     - python {{ python}}
     - setuptools_scm
   run:
-    - python
     - ansi2html >=1.5.2
     - numpy >=1.23
     - numba >=0.56.2
     - pandas >=1.0,<3.0
+    - python
     - python-dateutil
-    - riptide_cpp >=1.16.1,<2 # run with any (compatible) version in this range
+    - riptide_cpp >=1.16.3,<2 # run with any (compatible) version in this range
 
 about:
   home: https://github.com/rtosholdings/riptable

dev_tools/docstring_xfails.txt

@@ -21,7 +21,6 @@ riptable.rt_bin.cut
 riptable.rt_bin.qcut
 riptable.rt_bin.quantile
 riptable.rt_categorical.CatZero
-riptable.rt_categorical.Categorical
 riptable.rt_categorical.Categorical.align
 riptable.rt_categorical.Categorical.apply
 riptable.rt_categorical.Categorical.apply_nonreduce
@@ -41,10 +40,7 @@ riptable.rt_categorical.Categorical.category_replace
 riptable.rt_categorical.Categorical.copy
 riptable.rt_categorical.Categorical.copy_invalid
 riptable.rt_categorical.Categorical.expand_any
-riptable.rt_categorical.Categorical.expand_array
 riptable.rt_categorical.Categorical.expand_dict
-riptable.rt_categorical.Categorical.fill_backward
-riptable.rt_categorical.Categorical.fill_forward
 riptable.rt_categorical.Categorical.fill_invalid
 riptable.rt_categorical.Categorical.filtered_set_name
 riptable.rt_categorical.Categorical.filtered_string
@@ -66,7 +62,6 @@ riptable.rt_categorical.Categorical.isfiltered
 riptable.rt_categorical.Categorical.isin
 riptable.rt_categorical.Categorical.ismultikey
 riptable.rt_categorical.Categorical.isna
-riptable.rt_categorical.Categorical.isnan
 riptable.rt_categorical.Categorical.isnotnan
 riptable.rt_categorical.Categorical.issinglekey
 riptable.rt_categorical.Categorical.lock
@@ -84,7 +79,6 @@ riptable.rt_categorical.Categorical.nunique
 riptable.rt_categorical.Categorical.one_hot_encode
 riptable.rt_categorical.Categorical.set_name
 riptable.rt_categorical.Categorical.set_valid
-riptable.rt_categorical.Categorical.shift
 riptable.rt_categorical.Categorical.shift_cat
 riptable.rt_categorical.Categorical.shrink
 riptable.rt_categorical.Categorical.sort_gb
@@ -644,7 +638,6 @@ riptable.rt_numpy.abs
 riptable.rt_numpy.absolute
 riptable.rt_numpy.all
 riptable.rt_numpy.any
-riptable.rt_numpy.arange
 riptable.rt_numpy.argmax
 riptable.rt_numpy.argmin
 riptable.rt_numpy.argsort
@@ -667,12 +660,9 @@ riptable.rt_numpy.cumprod
 riptable.rt_numpy.cumsum
 riptable.rt_numpy.diff
 riptable.rt_numpy.double
-riptable.rt_numpy.empty
-riptable.rt_numpy.empty_like
 riptable.rt_numpy.float32
 riptable.rt_numpy.float64
 riptable.rt_numpy.floor
-riptable.rt_numpy.full
 riptable.rt_numpy.get_common_dtype
 riptable.rt_numpy.get_dtype
 riptable.rt_numpy.groupby
@@ -688,14 +678,7 @@ riptable.rt_numpy.int64
 riptable.rt_numpy.int8
 riptable.rt_numpy.interp
 riptable.rt_numpy.interp_extrap
-riptable.rt_numpy.isfinite
-riptable.rt_numpy.isinf
 riptable.rt_numpy.ismember
-riptable.rt_numpy.isnan
-riptable.rt_numpy.isnanorzero
-riptable.rt_numpy.isnotfinite
-riptable.rt_numpy.isnotinf
-riptable.rt_numpy.isnotnan
 riptable.rt_numpy.lexsort
 riptable.rt_numpy.log
 riptable.rt_numpy.log10
@@ -715,7 +698,6 @@ riptable.rt_numpy.mask_xor
 riptable.rt_numpy.mask_xori
 riptable.rt_numpy.max
 riptable.rt_numpy.maximum
-riptable.rt_numpy.mean
 riptable.rt_numpy.median
 riptable.rt_numpy.min
 riptable.rt_numpy.min_scalar_type
@@ -725,43 +707,28 @@ riptable.rt_numpy.nan_to_num
 riptable.rt_numpy.nanargmax
 riptable.rt_numpy.nanargmin
 riptable.rt_numpy.nanmax
-riptable.rt_numpy.nanmean
 riptable.rt_numpy.nanmedian
 riptable.rt_numpy.nanmin
 riptable.rt_numpy.nanpercentile
-riptable.rt_numpy.nanstd
-riptable.rt_numpy.nansum
-riptable.rt_numpy.nanvar
-riptable.rt_numpy.ones
-riptable.rt_numpy.ones_like
 riptable.rt_numpy.percentile
 riptable.rt_numpy.power
 riptable.rt_numpy.putmask
 riptable.rt_numpy.reindex_fast
-riptable.rt_numpy.repeat
 riptable.rt_numpy.reshape
 riptable.rt_numpy.round
 riptable.rt_numpy.searchsorted
 riptable.rt_numpy.single
 riptable.rt_numpy.sort
 riptable.rt_numpy.sortinplaceindirect
-riptable.rt_numpy.std
 riptable.rt_numpy.str_
-riptable.rt_numpy.sum
-riptable.rt_numpy.tile
 riptable.rt_numpy.transpose
 riptable.rt_numpy.trunc
 riptable.rt_numpy.uint0
 riptable.rt_numpy.uint16
 riptable.rt_numpy.uint32
 riptable.rt_numpy.uint64
 riptable.rt_numpy.uint8
-riptable.rt_numpy.unique
-riptable.rt_numpy.var
 riptable.rt_numpy.vstack
-riptable.rt_numpy.where
-riptable.rt_numpy.zeros
-riptable.rt_numpy.zeros_like
 riptable.rt_pdataset.PDataset
 riptable.rt_pdataset.PDataset.hstack
 riptable.rt_pdataset.PDataset.igroupby
@@ -864,7 +831,6 @@ riptable.rt_struct.Struct.hstack
 riptable.rt_struct.Struct.info
 riptable.rt_struct.Struct.is_valid_colname
 riptable.rt_struct.Struct.items
-riptable.rt_struct.Struct.key_search
 riptable.rt_struct.Struct.keys
 riptable.rt_struct.Struct.label_as_dict
 riptable.rt_struct.Struct.label_filter

dev_tools/gen_requirements.py

@@ -52,7 +52,8 @@ def is_python(major: int, minor: int) -> bool:
 ] + toolchain_reqs
 
 # PyPI setup build requirements.
-# Most everything else will be specified in setup.py.
+# Most everything *should* be in pyproject.toml, but some packages
+# need to be set up manually here.
 pypi_reqs = [
     "build",  # PEP-517 py build frontend
     _BENCHMARK_REQ,  # PyPI package doesn't exist

dev_tools/validate_docstrings.py

@@ -51,7 +51,7 @@
     Validator,
     validate,
 )
-import pandas
+
 import riptable
 from riptable.Utils.common import cached_weakref_property
 
@@ -98,7 +98,6 @@
 
 IMPORT_CONTEXT = {
     "np": numpy,
-    "pd": pandas,
     "rt": riptable,
 }
 

docs/source/conf.py

@@ -87,8 +87,16 @@ def parse_filters(filterpath: str) -> typing.Tuple[typing.Optional[list[str]], t
     "sphinx_design",
 ]
 
+# Put the list of valid time zones into docstrings using rst_prolog.
+from riptable import TimeZone
+
+tz_list = 'Supported timezones: "' + '", "'.join(TimeZone.valid_timezones) + '"'
+current_tzs_note = f"{tz_list}. To see supported timezones, use ``rt.TimeZone.valid_timezones``."
+
+
 rst_prolog = f"""
 .. |rtosholdings_docs| replace:: {os.getenv("RTOSHOLDINGS_DOCS", "rtosholdings-docs@sig.com")}
+.. |To see supported timezones, use ``rt.TimeZone.valid_timezones``.| replace:: {current_tzs_note}
 """
 
 

docs/source/tutorial/tutorial_cat_adv_instantiation.rst

@@ -1,9 +1,9 @@
 
-A Useful Way to Instantiate a Categorical
-*****************************************
+Build a Categorical Iteratively
+*******************************
 
 It can sometimes be useful to instantiate a Categorical with only one
-category, then fill it in as needed.
+category, then fill it in iteratively as needed.
 
 For example, let’s say we have a Dataset with a column that has a lot of
 categories, and we want to create a new Categorical column that keeps

docs/source/tutorial/tutorial_categoricals.rst

@@ -1172,7 +1172,11 @@ resulting Dataset is expanded.
 
 Note that until a reported bug is fixed, column names might not persist through grouping operations.
 
-For more in-depth information about Categoricals, see the `Categoricals User Guide <categoricals_user_guide>`.
+For more in-depth information about Categoricals, see:
+
+- Appendix: :doc:`Build a Categorical Iteratively <tutorial_cat_adv_instantiation>` 
+- :doc:`Categoricals User Guide <categoricals_user_guide>`
+- API Reference: :py:class:`~.rt_categorical.Categorical` class
 
 In the next section, `Accums <tutorial_accums.rst>`__, we look at
 another way to do multi-key groupings with fancier output.

docs/source/tutorial/tutorial_datetimes.rst

@@ -149,19 +149,13 @@ strings. Strings are common when the data is from, say, a CSV file.
 
 Unlike ``Date`` objects, ``DateTimeNano``\ s are time-zone-aware. When
 you create a ``DateTimeNano``, you need to specify the time zone of
-origin with the ``from_tz`` argument. Since Riptable is mainly used for
-financial market data, its time zone options are limited to NYC, DUBLIN,
-and (as of Riptable 1.3.6) Australia/Sydney, plus GMT and UTC (which is
-an alias for GMT).
-
-(If you’re wondering why ‘Australia/Sydney’ isn’t abbreviated, it’s
-because Riptable uses the standard time zone name from the `tz
-database <https://en.wikipedia.org/wiki/Tz_database>`__. In the future,
-Riptable will support only the `standard
-names <https://en.wikipedia.org/wiki/List_of_tz_database_time_zones>`__
-in the tz database.)
+origin with the ``from_tz`` argument. Riptable supports time zones in 
+certain areas of interest, such as New York, Dublin, Sydney, Tokyo, and 
+Hong Kong (more are added as needed). It also supports GMT and UTC.
 
-::
+To see the list of valid time zones, use ``rt.TimeZone.valid_timezones``.
+
+Here, we create a ``DateTimeNano`` with 'GMT' as the time zone of origin::
 
     >>> rt.DateTimeNano(['20210101 09:31:15', '20210519 05:21:17'], from_tz='GMT')
     DateTimeNano(['20210101 04:31:15.000000000', '20210519 01:21:17.000000000'], to_tz='NYC')

docs/source/tutorial/tutorial_visualize.rst

@@ -6,7 +6,7 @@ Matplotlib, to create visualizations of your data. You can also take
 advantage of the plotting and HTML styling tools offered by Pandas.
 
 In this section we’ll look at a couple of simple examples using
-Matplotlib, Pandas, and Playa.
+Matplotlib and Pandas.
 
 ::
 
@@ -202,59 +202,9 @@ your Dataset for the rendering::
         </tr>
       </tbody>
     </table>
-
-Groupscatter Plots with Playa
------------------------------
 
-Playa’s ``GroupScatter()`` method groups data into buckets based on
-x-values and returns a Matplotlib plot summarizing the data.
-
-::
-
-    from playa.plot import GroupScatter
-
-Make a noisier price signal
-^^^^^^^^^^^^^^^^^^^^^^^^^^^
-
-::
-
-    >>> ds.NoisyPrice = ds.Price + rng.normal(0, 10, ds.shape[0])
-
-A regular Matplotlib scatter plot, for comparison
-^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-
-::
-
-    >>> num_rows = int(rt.ceil(len(symbols)/2))
-    >>> fig, axes = plt.subplots(num_rows, 2, figsize=(20, 5 * num_rows))
-    >>> for (ax, symbol) in zip(axes.flatten(), symbols):
-    ...     f = ds.Symbol==symbol
-    ...     ax.scatter(ds.Time[f], ds.NoisyPrice[f])
-    ...     ax.grid()
-    ...     ax.set_xlabel('Time')
-    ...     ax.set_ylabel('Price')
-    ...     ax.set_title(f'{symbol} Noisy Stock Price by Time')
-    >>> plt.show()
-
-.. image:: output_25_0.png
-
-
-Now a GroupScatter for each one, you can see how it clarifies the point cloud and reveals the shape.
-^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-
-::
-
-    >>> fig, axes = plt.subplots(num_rows, 2, figsize=(20, 5 * num_rows))
-    >>> for (ax, symbol) in zip(axes.flatten(), symbols):
-    ...     f = ds.Symbol==symbol
-    ...     gs = GroupScatter(ds.Time[f].hour, ds.NoisyPrice[f])
-    ...     gs.plot(title=f'{symbol} Noisy Stock Price Over Time', x_label='Hour of the Day', y_label='Price', ax=ax)
-    >>> plt.show()
-
-.. image:: output_27_0.png
-
-This was just a brief introduction – check out the Matpotlib, Pandas,
-and Playa documentation for more details and possibilities.
+This was a brief introduction – check out the Matpotlib and Pandas documentation for 
+more details and possibilities.
 
 Next we cover useful tools for working with ``NaN``\ s and other missing
 values: `Working with Missing Data <tutorial_missing_data.rst>`__.