py310 (#143)

* py3.10 automated build

* lz4 version that supports py3.10

* py3.10 fix (test only)

* min python version & updated readme to reflect

* update zlib-state version for windows support on py3.10

* Revert "update zlib-state version for windows support on py3.10"

This reverts commit 698f6b8.

.github/workflows/push.yml

@@ -14,7 +14,7 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        python-version: [3.7, 3.8, 3.9]
+        python-version: [3.7, 3.8, 3.9, '3.10']
         os: ['ubuntu-latest', 'windows-latest', 'macOs-latest']
         architecture: ['x64']
 

README.md

@@ -44,7 +44,7 @@ $ python setup.py bdist_wheel
 $ pip install dist/ir_datasets-*.whl
 ```
 
-Tested with python versions 3.6 and 3.7
+Tested with python versions 3.7, 3.8, 3.9, and 3.10. (Mininum python version is 3.7.)
 
 ## Features
 

requirements.txt

@@ -6,7 +6,7 @@ pyyaml>=5.3.1
 requests>=2.22.0
 tqdm>=4.38.0
 trec-car-tools>=2.5.4
-lz4>=3.1.1
+lz4>=3.1.10
 warc3-wet>=0.2.3
 warc3-wet-clueweb09>=0.2.5
 zlib-state>=0.1.3

setup.py

@@ -17,7 +17,7 @@
     packages=setuptools.find_packages(include=['ir_datasets', 'ir_datasets.*']),
     install_requires=list(open('requirements.txt')),
     classifiers=[],
-    python_requires='>=3.6',
+    python_requires='>=3.7',
     entry_points={
         'console_scripts': ['ir_datasets=ir_datasets:main_cli'],
     },

test/formats/test_tsv.py

@@ -1,7 +1,6 @@
 import os
 from typing import NamedTuple, Tuple
 import shutil
-from collections import namedtuple
 import unittest
 from ir_datasets.formats import TsvDocs, TsvQueries, TsvDocPairs
 from ir_datasets.util import StringFile
@@ -10,7 +9,10 @@
 class TestTsv(unittest.TestCase):
 
     def test_core(self):
-        data_type = namedtuple('data_type', ['doc_id', 'field1', 'field2'])
+        class data_type(NamedTuple):
+            doc_id: str
+            field1: str
+            field2: str
         mock_file = StringFile('''
 123\tsome field\tanother field
 123\t  repeated  entry \tshouldn't filter
@@ -37,7 +39,10 @@ def test_core(self):
 
 
     def test_too_many_columns(self):
-        data_type = namedtuple('data_type', ['doc_id', 'field1', 'field2'])
+        class data_type(NamedTuple):
+            doc_id: str
+            field1: str
+            field2: str
         mock_file = StringFile('''
 123\tsome field\tanother field
 123\trepeated entry\tshouldn't filter\ttoo many columns
@@ -59,7 +64,10 @@ def test_too_many_columns(self):
 
 
     def test_too_few_columns(self):
-        data_type = namedtuple('data_type', ['doc_id', 'field1', 'field2'])
+        class data_type(NamedTuple):
+            doc_id: str
+            field1: str
+            field2: str
         mock_file = StringFile('''
 123\tsome field\tanother field
 123\ttoo few fields