Script 'mail_helper' called by obssrc
Hello community,
here is the log from the commit of package python-dask for openSUSE:Factory
checked in at 2021-09-22 22:13:06
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Comparing /work/SRC/openSUSE:Factory/python-dask (Old)
and /work/SRC/openSUSE:Factory/.python-dask.new.1899 (New)
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Package is "python-dask"
Wed Sep 22 22:13:06 2021 rev:48 rq:921018 version:2021.9.1
Changes:
--------
--- /work/SRC/openSUSE:Factory/python-dask/python-dask.changes 2021-08-12
09:02:17.182091782 +0200
+++ /work/SRC/openSUSE:Factory/.python-dask.new.1899/python-dask.changes
2021-09-22 22:13:44.836362429 +0200
@@ -1,0 +2,108 @@
+Wed Sep 22 12:50:07 UTC 2021 - Ben Greiner <c...@bnavigator.de>
+
+- Update to 2021.09.1
+ * Fix groupby for future pandas
+ * Remove warning filters in tests that are no longer needed
+ * Add link to diagnostic visualize function in local diagnostic
+ docs
+ * Add datetime_is_numeric to dataframe.describe
+ * Remove references to pd.Int64Index in anticipation of
+ deprecation
+ * Use loc if needed for series __get_item__
+ * Specifically ignore warnings on mean for empty slices
+ * Skip groupby nunique test for pandas >= 1.3.3
+ * Implement ascending arg for sort_values
+ * Replace operator.getitem
+ * Deprecate zero_broadcast_dimensions and homogeneous_deepmap
+ * Add error if drop_index is negative
+ * Allow scheduler to be an Executor
+ * Handle asarray/asanyarray cases where like is a dask.Array
+ * Fix index_col duplication if index_col is type str
+ * Add dtype and order to asarray and asanyarray definitions
+ * Deprecate dask.dataframe.Series.__contains__
+ * Fix edge case with like-arrays in _wrapped_qr
+ * Deprecate boundary_slice kwarg: kind for pandas compat
+- Release 2021.09.0
+ * Fewer open files
+ * Add FileNotFound to expected http errors
+ * Add DataFrame.sort_values to API docs
+ * Change to dask.order: be more eager at times
+ * Add pytest color to CI
+ * FIX: make_people works with processes scheduler
+ * Adds deep param to Dataframe copy method and restrict it to
+ False
+ * Fix typo in configuration docs
+ * Update formatting in DataFrame.query docstring
+ * Un-xfail sparse tests for 0.13.0 release
+ * Add axes property to DataFrame and Series
+ * Add CuPy support in da.unique (values only)
+ * Unit tests for sparse.zeros_like (xfailed)
+ * Add explicit like kwarg support to array creation functions
+ * Separate Array and DataFrame mindeps builds
+ * Fork out percentile_dispatch to dask.array
+ * Ensure filepath exists in to_parquet
+ * Update scheduler plugin usage in
+ test_scheduler_highlevel_graph_unpack_import
+ * Add DataFrame.shuffle to API docs
+ * Order requirements alphabetically
+- Release 2021.08.1
+ * Add ignore_metadata_file option to read_parquet
+ (pyarrow-dataset and fastparquet support only)
+ * Add reference to pytest-xdist in dev docs
+ * Include tz in meta from to_datetime
+ * CI Infra Docs
+ * Include invalid DataFrame key in assert_eq check
+ * Use __class__ when creating DataFrames
+ * Use development version of distributed in gpuCI build
+ * Ignore whitespace when gufunc signature
+ * Move pandas import and percentile dispatch refactor
+ * Add colors to represent high level layer types
+ * Upstream instance fix
+ * Add dask.widgets and migrate HTML reprs to jinja2
+ * Remove wrap_func_like_safe, not required with
+ NumPy >= 1.17
+ * Fix threaded scheduler memory backpressure regression
+ * Add percentile dispatch
+ * Use a publicly documented attribute obj in groupby rather than
+ private _selected_obj
+ * Specify module to import rechunk from
+ * Use dict to store data for {nan,}arg{min,max} in certain cases
+ * Fix blocksize description formatting in read_pandas
+ * Fix "point" -> "pointers" typo in docs
+- Release 2021.08.0
+ * Fix to_orc delayed compute behavior
+ * Don't convert to low-level task graph in
+ compute_as_if_collection
+ * Fix multifile read for hdf
+ * Resolve warning in distributed tests
+ * Update to_orc collection name
+ * Resolve skipfooter problem
+ * Raise NotImplementedError for non-indexable arg passed to
+ to_datetime
+ * Ensure we error on warnings from distributed
+ * Added dict format in to_bag accessories of DataFrame
+ * Delayed docs indirect dependencies
+ * Add tooltips to graphviz high-level graphs
+ * Close 2021 User Survey
+ * Reorganize CuPy tests into multiple files
+ * Refactor and Expand Dask-Dataframe ORC API
+ * Don't enforce columns if enforce=False
+ * Fix map_overlap trimming behavior when drop_axis is not None
+ * Mark gpuCI CuPy test as flaky
+ * Avoid using Delayed in to_csv and to_parquet
+ * Removed redundant check_dtypes
+ * Use pytest.warns instead of raises for checking parquet engine
+ deprecation
+ * Bump RAPIDS_VER in gpuCI to 21.10
+ * Add back pyarrow-legacy test coverage for pyarrow>=5
+ * Allow pyarrow>=5 in to_parquet and read_parquet
+ * Skip CuPy tests requiring NEP-35 when NumPy < 1.20 is available
+ * Add tail and head to SeriesGroupby
+ * Update Zoom link for monthly meeting
+ * Add gpuCI build script
+ * Deprecate daily_stock utility
+ * Add distributed.nanny to configuration reference docs
+ * Require NumPy 1.18+ & Pandas 1.0+
+- Add dask-fix8169-pandas13.patch -- gh#dask/dask#8169
+
+-------------------------------------------------------------------
Old:
----
dask-2021.7.2.tar.gz
New:
----
dask-2021.9.1.tar.gz
dask-fix8169-pandas13.patch
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Other differences:
------------------
++++++ python-dask.spec ++++++
--- /var/tmp/diff_new_pack.GL69r1/_old 2021-09-22 22:13:45.420362915 +0200
+++ /var/tmp/diff_new_pack.GL69r1/_new 2021-09-22 22:13:45.424362918 +0200
@@ -29,21 +29,23 @@
%define skip_python36 1
Name: python-dask%{psuffix}
# Note: please always update together with python-distributed!
-Version: 2021.7.2
+Version: 2021.9.1
Release: 0
Summary: Minimal task scheduling abstraction
License: BSD-3-Clause
URL: https://dask.org
Source:
https://files.pythonhosted.org/packages/source/d/dask/dask-%{version}.tar.gz
+# PATCH-FIX-UPSTREAM dask-fix8169-pandas13.patch -- gh#dask/dask#8169
+Patch0: dask-fix8169-pandas13.patch
BuildRequires: %{python_module base >= 3.7}
-BuildRequires: %{python_module packaging}
+BuildRequires: %{python_module packaging >= 20.0}
BuildRequires: %{python_module setuptools}
BuildRequires: fdupes
BuildRequires: python-rpm-macros
Requires: python-PyYAML
Requires: python-cloudpickle >= 1.1.1
Requires: python-fsspec >= 0.6.0
-Requires: python-packaging
+Requires: python-packaging >= 20.0
Requires: python-partd >= 0.3.10
Requires: python-toolz >= 0.8.2
Recommends: %{name}-array = %{version}
@@ -144,7 +146,7 @@
Summary: Numpy-like array data structure for dask
Requires: %{name} = %{version}
Requires: %{name}-delayed = %{version}
-Requires: python-numpy >= 1.16
+Requires: python-numpy >= 1.18
Recommends: python-scipy
%description array
@@ -191,8 +193,9 @@
Summary: Pandas-like DataFrame data structure for dask
Requires: %{name} = %{version}
Requires: %{name}-array = %{version}
-Requires: python-numpy >= 1.16
-Requires: python-pandas >= 0.25.0
+Requires: python-numpy >= 1.18
+# Patch0 requires pandas 1.3+ -- https://github.com/dask/dask/issues/8169
+Requires: python-pandas >= 1.3
%description dataframe
A flexible library for parallel computing in Python.
@@ -236,6 +239,7 @@
%package diagnostics
Summary: Diagnostics for dask
Requires: %{name} = %{version}
+Requires: python-Jinja2
Requires: python-bokeh >= 1.0.0
%description diagnostics
@@ -294,6 +298,8 @@
%prep
%autosetup -p1 -n dask-%{version}
+sed -i '/addopts/ {s/--durations=10//; s/--color=yes//}' setup.cfg
+chmod a-x dask/dataframe/io/orc/utils.py
%build
%python_build
@@ -334,7 +340,7 @@
donttest+="or (test_distributed and test_await)"
donttest+="or (test_threaded and test_interrupt)"
# requires otherwise optional pyarrow (not available on TW)
-donttest+="or (test_parquet and test_chunksize and fastparquet)"
+donttest+="or (test_parquet and test_chunksize)"
%pytest --pyargs dask -rfEs -m "not network" -k "not ($donttest)" -n auto
%endif
++++++ dask-2021.7.2.tar.gz -> dask-2021.9.1.tar.gz ++++++
++++ 13410 lines of diff (skipped)
++++++ dask-fix8169-pandas13.patch ++++++
Index: dask-2021.9.1/dask/dataframe/backends.py
===================================================================
--- dask-2021.9.1.orig/dask/dataframe/backends.py
+++ dask-2021.9.1/dask/dataframe/backends.py
@@ -356,7 +356,7 @@ class ShuffleGroupResult(SimpleSizeof, d
@group_split_dispatch.register((pd.DataFrame, pd.Series, pd.Index))
def group_split_pandas(df, c, k, ignore_index=False):
indexer, locations = pd._libs.algos.groupsort_indexer(
- c.astype(np.int64, copy=False), k
+ c.astype(np.intp, copy=False), k
)
df2 = df.take(indexer)
locations = locations.cumsum()
