Compare commits
10 Commits
ac0268dbf2
...
1a858dd6e1
Author | SHA1 | Date | |
---|---|---|---|
1a858dd6e1 | |||
|
2b65f28774 | ||
|
7f30dda570 | ||
|
cd8020f346 | ||
|
9934e1ad39 | ||
|
38f2f4e1cd | ||
|
c5e584ede7 | ||
|
c7038e4f1a | ||
|
b17de15cba | ||
|
6fb256b537 |
5
.gitignore
vendored
5
.gitignore
vendored
@ -70,3 +70,8 @@
|
||||
/dask-2023.3.2.tar.gz
|
||||
/dask-2023.4.0.tar.gz
|
||||
/dask-2023.4.1.tar.gz
|
||||
/dask-2023.7.0.tar.gz
|
||||
/dask-2023.7.1.tar.gz
|
||||
/dask-2023.8.1.tar.gz
|
||||
/dask-2023.11.0.tar.gz
|
||||
/dask-2023.12.0.tar.gz
|
||||
|
@ -1,7 +1,7 @@
|
||||
From 5ee683719a19c8865a7b80725a0adf55baa3e9c9 Mon Sep 17 00:00:00 2001
|
||||
From 1fcc2a30f6e5fea9aff448a52a0b4c4bbed908ea Mon Sep 17 00:00:00 2001
|
||||
From: Elliott Sales de Andrade <quantum.analyst@gmail.com>
|
||||
Date: Sun, 7 Mar 2021 04:07:32 -0500
|
||||
Subject: [PATCH 1/2] Skip test_encoding_gh601 on big-endian machines.
|
||||
Subject: [PATCH 1/6] Skip test_encoding_gh601 on big-endian machines.
|
||||
|
||||
Signed-off-by: Elliott Sales de Andrade <quantum.analyst@gmail.com>
|
||||
---
|
||||
@ -9,10 +9,10 @@ Signed-off-by: Elliott Sales de Andrade <quantum.analyst@gmail.com>
|
||||
1 file changed, 2 insertions(+)
|
||||
|
||||
diff --git a/dask/dataframe/io/tests/test_csv.py b/dask/dataframe/io/tests/test_csv.py
|
||||
index b50a727b..0ac1df4c 100644
|
||||
index 1df7202f..54ff1dff 100644
|
||||
--- a/dask/dataframe/io/tests/test_csv.py
|
||||
+++ b/dask/dataframe/io/tests/test_csv.py
|
||||
@@ -2,6 +2,7 @@ import gzip
|
||||
@@ -4,6 +4,7 @@ import gzip
|
||||
import os
|
||||
import warnings
|
||||
from io import BytesIO, StringIO
|
||||
@ -20,7 +20,7 @@ index b50a727b..0ac1df4c 100644
|
||||
from unittest import mock
|
||||
|
||||
import pytest
|
||||
@@ -1146,6 +1147,7 @@ def test_read_csv_with_datetime_index_partitions_n():
|
||||
@@ -1155,6 +1156,7 @@ def test_read_csv_with_datetime_index_partitions_n():
|
||||
xfail_pandas_100 = pytest.mark.xfail(reason="https://github.com/dask/dask/issues/5787")
|
||||
|
||||
|
||||
@ -29,5 +29,5 @@ index b50a727b..0ac1df4c 100644
|
||||
"encoding",
|
||||
[
|
||||
--
|
||||
2.40.0
|
||||
2.42.0
|
||||
|
||||
|
@ -1,7 +1,7 @@
|
||||
From 54487feeef6d7d41e3b6d4e4ece5b6342071068b Mon Sep 17 00:00:00 2001
|
||||
From f99ee06a1c64ff3096e90cbc3a11ed4f172de691 Mon Sep 17 00:00:00 2001
|
||||
From: Elliott Sales de Andrade <quantum.analyst@gmail.com>
|
||||
Date: Sun, 7 May 2023 23:13:59 -0400
|
||||
Subject: [PATCH 2/2] Skip coverage testing
|
||||
Subject: [PATCH 2/6] Skip coverage testing
|
||||
|
||||
Signed-off-by: Elliott Sales de Andrade <quantum.analyst@gmail.com>
|
||||
---
|
||||
@ -9,10 +9,10 @@ Signed-off-by: Elliott Sales de Andrade <quantum.analyst@gmail.com>
|
||||
1 file changed, 1 insertion(+), 1 deletion(-)
|
||||
|
||||
diff --git a/pyproject.toml b/pyproject.toml
|
||||
index a7d504ff..36dfca8e 100644
|
||||
index 0ef22fb6..a11ac921 100644
|
||||
--- a/pyproject.toml
|
||||
+++ b/pyproject.toml
|
||||
@@ -121,7 +121,7 @@ markers = [
|
||||
@@ -127,7 +127,7 @@ markers = [
|
||||
"skip_with_pyarrow_strings: Tests that should be skipped when pyarrow string conversion is turned on",
|
||||
"xfail_with_pyarrow_strings: Tests that should be xfailed when pyarrow string conversion is turned on",
|
||||
]
|
||||
@ -22,5 +22,5 @@ index a7d504ff..36dfca8e 100644
|
||||
# From Cython-1753
|
||||
"ignore:can't resolve:ImportWarning",
|
||||
--
|
||||
2.40.0
|
||||
2.42.0
|
||||
|
||||
|
@ -0,0 +1,29 @@
|
||||
From 691eb92e4c6099309919e3aaa05a3dd6a2f4ddb1 Mon Sep 17 00:00:00 2001
|
||||
From: Elliott Sales de Andrade <quantum.analyst@gmail.com>
|
||||
Date: Thu, 20 Jul 2023 00:05:48 -0400
|
||||
Subject: [PATCH 3/6] TST: Increase maximum for sizeof test to pass 32-bit
|
||||
|
||||
In that case, the result is 1244, which is ~2.6 * `sys.sizeof` (476),
|
||||
slightly over the 2 used in the existing test.
|
||||
|
||||
Signed-off-by: Elliott Sales de Andrade <quantum.analyst@gmail.com>
|
||||
---
|
||||
dask/tests/test_sizeof.py | 2 +-
|
||||
1 file changed, 1 insertion(+), 1 deletion(-)
|
||||
|
||||
diff --git a/dask/tests/test_sizeof.py b/dask/tests/test_sizeof.py
|
||||
index 9b43a7ae..866d119b 100644
|
||||
--- a/dask/tests/test_sizeof.py
|
||||
+++ b/dask/tests/test_sizeof.py
|
||||
@@ -81,7 +81,7 @@ def test_pandas_multiindex():
|
||||
index = pd.MultiIndex.from_product([range(5), ["a", "b", "c", "d", "e"]])
|
||||
actual_size = sys.getsizeof(index)
|
||||
|
||||
- assert 0.5 * actual_size < sizeof(index) < 2 * actual_size
|
||||
+ assert 0.5 * actual_size < sizeof(index) < 3 * actual_size
|
||||
assert isinstance(sizeof(index), int)
|
||||
|
||||
|
||||
--
|
||||
2.42.0
|
||||
|
@ -0,0 +1,36 @@
|
||||
From e5c291fbcd69a80aef7d8b01f7621d736ce497ed Mon Sep 17 00:00:00 2001
|
||||
From: Elliott Sales de Andrade <quantum.analyst@gmail.com>
|
||||
Date: Sat, 19 Aug 2023 16:49:33 -0400
|
||||
Subject: [PATCH 4/6] Fix test_pandas_timestamp_overflow_pyarrow condition
|
||||
|
||||
The new behavior in pyarrow only occurs with Pandas 2.0 as well.
|
||||
|
||||
Signed-off-by: Elliott Sales de Andrade <quantum.analyst@gmail.com>
|
||||
---
|
||||
dask/dataframe/io/tests/test_parquet.py | 6 +++---
|
||||
1 file changed, 3 insertions(+), 3 deletions(-)
|
||||
|
||||
diff --git a/dask/dataframe/io/tests/test_parquet.py b/dask/dataframe/io/tests/test_parquet.py
|
||||
index 0d4dd39d..d5ad8b5f 100644
|
||||
--- a/dask/dataframe/io/tests/test_parquet.py
|
||||
+++ b/dask/dataframe/io/tests/test_parquet.py
|
||||
@@ -3381,13 +3381,13 @@ def test_pandas_timestamp_overflow_pyarrow(tmpdir):
|
||||
table, f"{tmpdir}/file.parquet", use_deprecated_int96_timestamps=False
|
||||
)
|
||||
|
||||
- if pyarrow_version < parse_version("13.0.0.dev"):
|
||||
+ if pyarrow_version.major >= 13 and PANDAS_GE_200:
|
||||
+ dd.read_parquet(str(tmpdir), engine="pyarrow").compute()
|
||||
+ else:
|
||||
# This will raise by default due to overflow
|
||||
with pytest.raises(pa.lib.ArrowInvalid) as e:
|
||||
dd.read_parquet(str(tmpdir), engine="pyarrow").compute()
|
||||
assert "out of bounds" in str(e.value)
|
||||
- else:
|
||||
- dd.read_parquet(str(tmpdir), engine="pyarrow").compute()
|
||||
|
||||
from dask.dataframe.io.parquet.arrow import ArrowDatasetEngine as ArrowEngine
|
||||
|
||||
--
|
||||
2.42.0
|
||||
|
24
0005-Allow-older-versioneer.patch
Normal file
24
0005-Allow-older-versioneer.patch
Normal file
@ -0,0 +1,24 @@
|
||||
From 7803706e5acf1e1edce34de0b0d5321872bde299 Mon Sep 17 00:00:00 2001
|
||||
From: Elliott Sales de Andrade <quantum.analyst@gmail.com>
|
||||
Date: Sat, 25 Nov 2023 20:22:50 -0500
|
||||
Subject: [PATCH 5/6] Allow older versioneer
|
||||
|
||||
Signed-off-by: Elliott Sales de Andrade <quantum.analyst@gmail.com>
|
||||
---
|
||||
pyproject.toml | 2 +-
|
||||
1 file changed, 1 insertion(+), 1 deletion(-)
|
||||
|
||||
diff --git a/pyproject.toml b/pyproject.toml
|
||||
index a11ac921..d74f0fa5 100644
|
||||
--- a/pyproject.toml
|
||||
+++ b/pyproject.toml
|
||||
@@ -1,5 +1,5 @@
|
||||
[build-system]
|
||||
-requires = ["setuptools>=62.6", "versioneer[toml]==0.29"]
|
||||
+requires = ["setuptools>=62.6", "versioneer[toml]>=0.28"]
|
||||
build-backend = "setuptools.build_meta"
|
||||
|
||||
[project]
|
||||
--
|
||||
2.42.0
|
||||
|
29
0006-Ignore-NumPy-warnings-from-Pandas.patch
Normal file
29
0006-Ignore-NumPy-warnings-from-Pandas.patch
Normal file
@ -0,0 +1,29 @@
|
||||
From 7db95f23f20cfd48eb6352c9ac0cba7780e925f6 Mon Sep 17 00:00:00 2001
|
||||
From: James Bourbeau <jrbourbeau@gmail.com>
|
||||
Date: Mon, 22 May 2023 12:53:20 -0500
|
||||
Subject: [PATCH 6/6] Ignore NumPy warnings from Pandas
|
||||
|
||||
Signed-off-by: Elliott Sales de Andrade <quantum.analyst@gmail.com>
|
||||
---
|
||||
pyproject.toml | 5 +++++
|
||||
1 file changed, 5 insertions(+)
|
||||
|
||||
diff --git a/pyproject.toml b/pyproject.toml
|
||||
index d74f0fa5..b93c2187 100644
|
||||
--- a/pyproject.toml
|
||||
+++ b/pyproject.toml
|
||||
@@ -143,6 +143,11 @@ filterwarnings = [
|
||||
# https://pandas.pydata.org/docs/dev/whatsnew/v1.5.0.html#using-group-keys-with-transformers-in-groupby-apply
|
||||
"ignore:Not prepending group keys:FutureWarning",
|
||||
"ignore:.*:dask.tests.warning_aliases.RemovedIn20Warning",
|
||||
+ # This is coming from pandas use of np.find_common_type
|
||||
+ # See https://github.com/pandas-dev/pandas/issues/53236
|
||||
+ "ignore:np.find_common_type is deprecated:DeprecationWarning",
|
||||
+ "ignore:`cumproduct` is deprecated:DeprecationWarning",
|
||||
+ "ignore:`product` is deprecated:DeprecationWarning",
|
||||
"ignore:When grouping with a length-1 list-like, you will need to pass a length-1 tuple to get_group in a future version of pandas:FutureWarning",
|
||||
'ignore:DataFrameGroupBy\.apply operated on the grouping columns\. This behavior is deprecated, and in a future version of pandas the grouping columns will be excluded from the operation\. Either pass `include_groups=False` to exclude the groupings or explicitly select the grouping columns after groupby to silence this warning\.:FutureWarning',
|
||||
'ignore:Passing a BlockManager to DataFrame is deprecated and will raise in a future version. Use public APIs instead:DeprecationWarning', # https://github.com/apache/arrow/issues/35081
|
||||
--
|
||||
2.42.0
|
||||
|
19
force-little-endian-random.patch
Normal file
19
force-little-endian-random.patch
Normal file
@ -0,0 +1,19 @@
|
||||
Author: Diane Trout <diane@ghic.org>
|
||||
Description: Force initializing the random seed with the same
|
||||
byte order interpretation as on x86.
|
||||
|
||||
Index: dask-2023.8.0+dfsg/dask/utils.py
|
||||
===================================================================
|
||||
--- dask-2023.8.0+dfsg.orig/dask/utils.py
|
||||
+++ dask-2023.8.0+dfsg/dask/utils.py
|
||||
@@ -426,7 +426,9 @@ def random_state_data(n: int, random_sta
|
||||
random_state = np.random.RandomState(random_state)
|
||||
|
||||
random_data = random_state.bytes(624 * n * 4) # `n * 624` 32-bit integers
|
||||
- l = list(np.frombuffer(random_data, dtype=np.uint32).reshape((n, -1)))
|
||||
+ dt = np.dtype(np.uint32)
|
||||
+ dt = dt.newbyteorder("<")
|
||||
+ l = list(np.frombuffer(random_data, dtype=dt).reshape((n, -1)))
|
||||
assert len(l) == n
|
||||
return l
|
||||
|
@ -10,9 +10,9 @@
|
||||
%global debug_package %{nil}
|
||||
|
||||
Name: python-%{srcname}
|
||||
Version: 2023.4.1
|
||||
%global tag 2023.4.1
|
||||
Release: %autorelease
|
||||
Version: 2023.12.0
|
||||
%global tag 2023.12.0
|
||||
Release: %autorelease -e 0.riscv64
|
||||
Summary: Parallel PyData with Task Scheduling
|
||||
|
||||
License: BSD-3-Clause
|
||||
@ -22,6 +22,17 @@ Source0: %{pypi_source %{srcname}}
|
||||
Patch: 0001-Skip-test_encoding_gh601-on-big-endian-machines.patch
|
||||
# Fedora-specific patch.
|
||||
Patch: 0002-Skip-coverage-testing.patch
|
||||
# Drop after dropping 32-bit support.
|
||||
Patch: 0003-TST-Increase-maximum-for-sizeof-test-to-pass-32-bit.patch
|
||||
# https://github.com/dask/dask/pull/10451
|
||||
Patch: 0004-Fix-test_pandas_timestamp_overflow_pyarrow-condition.patch
|
||||
# https://github.com/dask/dask/issues/10423
|
||||
Patch: https://salsa.debian.org/python-team/packages/dask/-/raw/08ffea1b7b53e9c71c9a926d5786288c2e6c1b5b/debian/patches/force-little-endian-random.patch
|
||||
Patch: 0005-Allow-older-versioneer.patch
|
||||
# Ignore warnings from Pandas.
|
||||
# Upstream had https://github.com/dask/dask/pull/10307 but reverted it because
|
||||
# a new Pandas was released that fixed the warning, but we don't have it yet.
|
||||
Patch: 0006-Ignore-NumPy-warnings-from-Pandas.patch
|
||||
|
||||
%description
|
||||
Dask is a flexible parallel computing library for analytics.
|
||||
@ -54,7 +65,10 @@ BuildRequires: python3dist(pyarrow)
|
||||
%endif
|
||||
BuildRequires: python3dist(requests)
|
||||
BuildRequires: python3dist(sqlalchemy)
|
||||
# tables does not support 32 bit architectures and is ExcludeArch.
|
||||
%ifnarch %{ix86}
|
||||
BuildRequires: python3dist(tables)
|
||||
%endif
|
||||
BuildRequires: python3dist(zarr)
|
||||
|
||||
Recommends: python3-%{srcname}+array = %{version}-%{release}
|
||||
@ -208,6 +222,12 @@ k="${k-}${k+ and }not test_development_guidelines_matches_ci"
|
||||
# https://github.com/dask/dask/issues/10418
|
||||
k="${k-}${k+ and }not test_division_or_partition"
|
||||
|
||||
%ifarch riscv64
|
||||
k="${k-}${k+ and }not test_datetime_std_with_larger_dataset[False-False-1]"
|
||||
k="${k-}${k+ and }not test_datetime_std_with_larger_dataset[None-True-1]"
|
||||
k="${k-}${k+ and }not test_cov_corr_stable"
|
||||
%endif
|
||||
|
||||
pytest_args=(
|
||||
-m 'not network'
|
||||
|
||||
|
2
sources
2
sources
@ -1 +1 @@
|
||||
SHA512 (dask-2023.4.1.tar.gz) = 2fad1317aad845f7b11d8efceafb11eb9e945aca649bcf7ccd71e8692c3ca18c6f595c82b4e2930cba4e9622dd384965b3cf0ea3f68bdbc2f6f18c7a2d806b26
|
||||
SHA512 (dask-2023.12.0.tar.gz) = b4e81f7e0ff23862824b71c98f36293da2d11bc49b5f82df556be0023d001aa85181db336e8333bb4c589259d18435ef04289219706479e3982b2c2bb398e7c6
|
||||
|
Loading…
Reference in New Issue
Block a user