Hello community, here is the log from the commit of package python-dask for openSUSE:Factory checked in at 2019-11-13 13:26:36 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Comparing /work/SRC/openSUSE:Factory/python-dask (Old) and /work/SRC/openSUSE:Factory/.python-dask.new.2990 (New) ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Package is "python-dask" Wed Nov 13 13:26:36 2019 rev:22 rq:747781 version:2.7.0 Changes: -------- --- /work/SRC/openSUSE:Factory/python-dask/python-dask.changes 2019-06-20 18:54:37.792782485 +0200 +++ /work/SRC/openSUSE:Factory/.python-dask.new.2990/python-dask.changes 2019-11-13 13:26:41.603595513 +0100 @@ -1,0 +2,367 @@ +Mon Nov 11 18:24:07 UTC 2019 - Todd R <[email protected]> + +- Update to 2.7.0 + + Array + * Reuse code for assert_eq util method + * Update da.array to always return a dask array + * Skip transpose on trivial inputs + * Avoid NumPy scalar string representation in tokenize + * Remove unnecessary tiledb shape constraint + * Removes bytes from sparse array HTML repr + + Core + * Drop Python 3.5 + * Update the use of fixtures in distributed tests + * Changed deprecated bokeh-port to dashboard-address + * Avoid updating with identical dicts in ensure_dict + * Test Upstream + * Accelerate reverse_dict + * Update test_imports.sh + * Support cgroups limits on cpu count in multiprocess and threaded schedulers + * Update minimum pyarrow version on CI + * Make cloudpickle optional + + DataFrame + * Add an example of index_col usage + * Explicitly use iloc for row indexing + * Accept dask arrays on columns assignemnt + * Implement unique and value_counts for SeriesGroupBy + * Add sizeof definition for pyarrow tables and columns + * Enable row-group task partitioning in pyarrow-based read_parquet + * Removes npartitions='auto' from dd.merge docstring + * Apply enforce error message shows non-overlapping columns. + * Optimize meta_nonempty for repetitive dtypes + * Remove import of dask_cudf, which is now a part of cudf + + Documentation + * Make capitalization more consistent in FAQ docs + * Add CONTRIBUTING.md + * Document optional dependencies + * Update helm chart docs to reflect new chart repo + * Add Resampler to API docs + * Fix typo in read_sql_table + * Add adaptive deployments screencast +- Update to 2.6.0 + + Core + * Call ``ensure_dict`` on graphs before entering ``toolz.merge`` + * Consolidating hash dispatch functions + + DataFrame + * Support Python 3.5 in Parquet code + * Avoid identity check in ``warn_dtype_mismatch`` + * Enable unused groupby tests + * Remove old parquet and bcolz dataframe optimizations + * Add getitem optimization for ``read_parquet`` + * Use ``_constructor_sliced`` method to determine Series type + * Fix map(series) for unsorted base series index + * Fix ``KeyError`` with Groupby label + + Documentation + * Use Zoom meeting instead of appear.in + * Added curated list of resources + * Update SSH docs to include ``SSHCluster`` + * Update "Why Dask?" page + * Fix typos in docstrings +- Update to 2.5.2 + + Array + * Correct chunk size logic for asymmetric overlaps + * Make da.unify_chunks public API + + DataFrame + * Fix dask.dataframe.fillna handling of Scalar object + + Documentation + * Remove boxes in Spark comparison page + * Add latest presentations + * Update cloud documentation +- Update to 2.5.0 + + Core + * Add sentinel no_default to get_dependencies task + * Update fsspec version + * Remove PY2 checks + + DataFrame + * Add option to not check meta in dd.from_delayed + * Fix test_timeseries_nulls_in_schema failures with pyarrow master + * Reduce read_metadata output size in pyarrow/parquet + * Test numeric edge case for repartition with npartitions. + * Unxfail pandas-datareader test + * Add DataFrame.pop implementation + * Enable merge/set_index for cudf-based dataframes with cupy ``values`` + * drop_duplicates support for positional subset parameter + + Documentation + * Add screencasts to array, bag, dataframe, delayed, futures and setup + * Fix delimeter parsing documentation + * Update overview image +- Update to 2.4.0 + + Array + * Adds explicit ``h5py.File`` mode + * Provides method to compute unknown array chunks sizes + * Ignore runtime warning in Array ``compute_meta`` + * Add ``_meta`` to ``Array.__dask_postpersist__`` + * Fixup ``da.asarray`` and ``da.asanyarray`` for datetime64 dtype and xarray objects + * Add shape implementation + * Add chunktype to array text repr + * Array.random.choice: handle array-like non-arrays + + Core + * Remove deprecated code + * Fix ``funcname`` when vectorized func has no ``__name__`` + * Truncate ``funcname`` to avoid long key names + * Add support for ``numpy.vectorize`` in ``funcname`` + * Fixed HDFS upstream test + * Support numbers and None in ``parse_bytes``/``timedelta`` + * Fix tokenizing of subindexes on memmapped numpy arrays + * Upstream fixups + + DataFrame + * Allow pandas to cast type of statistics + * Preserve index dtype after applying ``dd.pivot_table`` + * Implement explode for Series and DataFrame + * ``set_index`` on categorical fails with less categories than partitions + * Support output to a single CSV file + * Add ``groupby().transform()`` + * Adding filter kwarg to pyarrow dataset call + * Implement and check compression defaults for parquet + * Pass sqlalchemy params to delayed objects + * Fixing schema handling in arrow-parquet + * Add support for DF and Series ``groupby().idxmin/max()`` + * Add correlation calculation and add test + + Documentation + * Numpy docstring standard has moved + * Reference correct NumPy array name + * Minor edits to Array chunk documentation + * Add methods to API docs + * Add namespacing to configuration example + * Add get_task_stream and profile to the diagnostics page + * Add best practice to load data with Dask + * Update ``institutional-faq.rst`` + * Add threads and processes note to the best practices + * Update cuDF links + * Fixed small typo with parentheses placement + * Update link in reshape docstring +- Update to 2.3.0 + + Array + * Raise exception when ``from_array`` is given a dask array + * Avoid adjusting gufunc's meta dtype twice + * Add ``meta=`` keyword to map_blocks and add test with sparse + * Add rollaxis and moveaxis + * Always increment old chunk index + * Shuffle dask array + * Fix ordering when indexing a dask array with a bool dask array + + Bag + * Add workaround for memory leaks in bag generators + + Core + * Set strict xfail option + * test-upstream + * Fixed HDFS CI failure + * Error nicely if no file size inferred + * A few changes to ``config.set`` + * Fixup black string normalization + * Pin NumPy in windows tests + * Ensure parquet tests are skipped if fastparquet and pyarrow not installed + * Add fsspec to readthedocs + * Bump NumPy and Pandas to 1.17 and 0.25 in CI test + + DataFrame + * Fix ``DataFrame.query`` docstring (incorrect numexpr API) + * Parquet metadata-handling improvements + * Improve messaging around sorted parquet columns for index + * Add ``rearrange_by_divisions`` and ``set_index`` support for cudf + * Fix ``groupby.std()`` with integer colum names + * Add ``Series.__iter__`` + * Generalize ``hash_pandas_object`` to work for non-pandas backends + * Add rolling cov + * Add columns argument in drop function + + Documentation + * Update institutional FAQ doc + * Add draft of institutional FAQ + * Make boxes for dask-spark page + * Add motivation for shuffle docs + * Fix links and API entries for best-practices + * Remove "bytes" (internal data ingestion) doc page + * Redirect from our local distributed page to distributed.dask.org + * Cleanup API page + * Remove excess endlines from install docs + * Remove item list in phases of computation doc + * Remove custom graphs from the TOC sidebar + * Remove experimental status of custom collections + * Adds table of contents to Why Dask? + * Moves bag overview to top-level bag page + * Remove use-cases in favor of stories.dask.org + * Removes redundant TOC information in index.rst + * Elevate dashboard in distributed diagnostics documentation + * Updates "add" layer in HLG docs example + * Update GUFunc documentation +- Update to 2.2.0 + + Array + * Use da.from_array(..., asarray=False) if input follows NEP-18 + * Add missing attributes to from_array documentation + * Fix meta computation for some reduction functions + * Raise informative error in to_zarr if unknown chunks + * Remove invalid pad tests + * Ignore NumPy warnings in compute_meta + * Fix kurtosis calc for single dimension input array + * Support Numpy 1.17 in tests + + Bag + * Supply pool to bag test to resolve intermittent failure + + Core ++++ 170 more lines (skipped) ++++ between /work/SRC/openSUSE:Factory/python-dask/python-dask.changes ++++ and /work/SRC/openSUSE:Factory/.python-dask.new.2990/python-dask.changes Old: ---- dask-1.2.2.tar.gz New: ---- dask-2.7.0.tar.gz ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Other differences: ------------------ ++++++ python-dask.spec ++++++ --- /var/tmp/diff_new_pack.loxd5y/_old 2019-11-13 13:26:42.683596636 +0100 +++ /var/tmp/diff_new_pack.loxd5y/_new 2019-11-13 13:26:42.687596641 +0100 @@ -25,8 +25,9 @@ %define psuffix %{nil} %bcond_with test %endif +%define skip_python2 1 Name: python-dask%{psuffix} -Version: 1.2.2 +Version: 2.7.0 Release: 0 Summary: Minimal task scheduling abstraction License: BSD-3-Clause @@ -37,6 +38,7 @@ BuildRequires: fdupes BuildRequires: python-rpm-macros Requires: python-toolz >= 0.7.3 +Requires: python-tornado >= 5 Recommends: %{name}-array = %{version} Recommends: %{name}-bag = %{version} Recommends: %{name}-dataframe = %{version} @@ -62,6 +64,7 @@ BuildRequires: %{python_module chest} BuildRequires: %{python_module cloudpickle >= 0.2.1} BuildRequires: %{python_module distributed} +BuildRequires: %{python_module fsspec >= 0.5.1} BuildRequires: %{python_module graphviz} BuildRequires: %{python_module h5py} BuildRequires: %{python_module jupyter_ipython} @@ -80,7 +83,7 @@ BuildRequires: %{python_module scipy} BuildRequires: %{python_module six} BuildRequires: %{python_module tables} -BuildRequires: %{python_module tornado} +BuildRequires: %{python_module tornado >= 5} BuildRequires: graphviz BuildRequires: graphviz-gd BuildRequires: graphviz-gnome @@ -124,7 +127,7 @@ Summary: Numpy-like array data structure for dask Group: Development/Languages/Python Requires: %{name} = %{version} -Requires: python-numpy +Requires: python-numpy >= 1.13.0 Recommends: python-chest Recommends: python-h5py Recommends: python-pandas @@ -149,7 +152,9 @@ Group: Development/Languages/Python Requires: %{name} = %{version} Requires: %{name}-multiprocessing = %{version} -Recommends: python-partd >= 0.3.7 +Requires: python-cloudpickle >= 0.2.1 +Requires: python-fsspec >= 0.5.1 +Requires: python-partd >= 0.3.10 %description bag A minimal task scheduling abstraction and parallel arrays. @@ -172,8 +177,10 @@ Requires: %{name} = %{version} Requires: %{name}-array = %{version} Requires: %{name}-multiprocessing = %{version} -Requires: python-numpy -Requires: python-pandas +Requires: python-fsspec >= 0.5.1 +Requires: python-numpy >= 1.13.0 +Requires: python-pandas >= 0.21.0 +Requires: python-partd >= 0.3.10 Requires: python-six Recommends: %{name}-bag = %{version} Recommends: python-SQLAlchemy @@ -181,7 +188,6 @@ Recommends: python-chest Recommends: python-fastparquet Recommends: python-pandas-datareader -Recommends: python-partd >= 0.3.7 Recommends: python-psutil Recommends: python-pyarrow @@ -204,8 +210,7 @@ Summary: Interface with the distributed task scheduler in dask Group: Development/Languages/Python Requires: %{name} = %{version} -Requires: python-distributed >= 1.16 -Requires: python-s3fs >= 0.0.8 +Requires: python-distributed >= 2.0 %description distributed A minimal task scheduling abstraction and parallel arrays. @@ -272,10 +277,12 @@ %if %{with test} %check # Tests need network: +# test_await # test_serializable_groupby_agg # test_persist # test_local_get_with_distributed_active -%python_expand PYTHONPATH=%{buildroot}%{$python_sitelib} py.test-%{python_bin_suffix} -v dask/tests -k 'not (test_serializable_groupby_agg or test_persist or test_local_get_with_distributed_active)' +# test_local_scheduler +%python_expand PYTHONPATH=%{buildroot}%{$python_sitelib} py.test-%{python_bin_suffix} -v dask/tests -k 'not (test_serializable_groupby_agg or test_persist or test_local_get_with_distributed_active or test_await or test_local_scheduler)' %endif %if !%{with test} ++++++ dask-1.2.2.tar.gz -> dask-2.7.0.tar.gz ++++++ ++++ 97003 lines of diff (skipped)
