v3.8.1

What's Changed

• Releases dev guide by @RossKen in #1202
• Fix link only cartesian calc by @ThomasHepworth in #1204

Full Changelog: v3.8.0...v3.8.1

v3.8.0

What's Changed

• Make the example notebooks run faster by @RobinL in #1160
• Add tags by @RossKen in #1165
• Benchmark timeseries commit workflow to run only in upstream repo by @ADBond in #1152
• Create _register_input_tables method in our main linker class by @ThomasHepworth in #1172
• Documentation examples by @RossKen in #1159
• Fix autoblack checkout step by @ADBond in #1169
• Add emojis rather than bullets by @RobinL in #1180
• Add option to pass seed into estimate_u_using_random_sampling by @RossKen in #1161
• Adjust the outputs of truth_space_table_from_labels_with_predictions_sqls to be lowercase by @ThomasHepworth in #1183
• Improve Logging by @NickCrews in #1084
• Improve logging by @ThomasHepworth in #1186
• Add docs for Feature Engineering by @RossKen in #1178
• Add UDFs dev guide by @RossKen in #1182
• [BUG] Fix source dataset issue when running link jobs by @ThomasHepworth in #1193
• 1107 add jaro similarity by @RossKen in #1167
• migrate ComparisonProperties by @ThomasHepworth in #1195
• revert to old comparison script structure by @RossKen in #1197
• 1030 option for auto typecasting datediff by @aliceoleary0 in #1162
• Athena updates by @ThomasHepworth in #1187
• Release 3.8.0 by @RossKen in #1201

New Contributors

• @aliceoleary0 made their first contribution in #1162

Full Changelog: v3.7.3...v3.8.0

v3.7.3

What's Changed

• Linting update by @ADBond in #1131
• Fix autoblack workflow for forks by @ADBond in #1133
• remove invalid comma by @wilko77 in #1143
• Improve readme what does splink do by @RobinL in #1129
• Improve copy writing on readme by @RobinL in #1144
• Improve readme images for clarity by @RobinL in #1145
• Attempt to make examples notebooks action faster by @RobinL in #1147
• ComparisonLevel composition v2 by @NickCrews in #1114
• Update run_demos_examples.yml by @RobinL in #1154
• 1151 fix term frequencies for cols reversed by @afua-moj in #1156
• Add previously breaking tf case to tests by @RossKen in #1157
• Version 3.7.3 by @RossKen in #1158

New Contributors

• @wilko77 made their first contribution in #1143
• @afua-moj made their first contribution in #1156

Full Changelog: v3.7.2...v3.7.3

v3.7.2

What's Changed

• Add comparison template library functions for simple name column by @RossKen in #1125
• Modify defaults for repartitioning by @RobinL in #1138

Full Changelog: v3.7.1...v3.7.2

v3.7.1

What's Changed

• Fix a couple of typos by @ADBond in #1123
• Fix athena linker invalid reference by @davidschrooten in #1135
• Fix clustering in issue 1136 by @RobinL in #1137

New Contributors

• @davidschrooten made their first contribution in #1135

Full Changelog: v3.7.0...v3.7.1

v3.7.0

What's Changed

• Adjust caching for our concat tables by @ThomasHepworth in #1013
• _initialise_df_concat optionally returns list by @RobinL in #1023
• Df concat and df concat with tf return SplinkDataframe or None by @RobinL in #1033
• [Docs] Add a dev guide for creating new ComparisonLevels and Comparisons to Splink libraries by @ADBond in #1041
• correct module name duckb_base -> duckdb_base by @ADBond in #1046
• Some cache tests by @ADBond in #1050
• Improving the cache and make cache invalidation easier and more robust by @RobinL in #987
• Bump version to 3.7.0 by @RobinL in #1056
• Release 3.7.0 as dev version by @RobinL in #1057
• Adds the ability to read directly from a settings filepath by @ThomasHepworth in #1062
• Add code to produce tf cols from concat_with_tf by @ThomasHepworth in #1065
• Use Ruff as a linter by @NickCrews in #1004
• cast all value values to varchar by @ThomasHepworth in #1049
• Automatically add tables of comparison (level)s compatible with each dialect to docs by @ADBond in #1035
• add the ability to pass pandas df into the SparkLinker by @ThomasHepworth in #1068
• Ruff by @ThomasHepworth in #1070
• Replace flake8 with ruff as our main linter by @ThomasHepworth in #1071
• Loosen dependency ranges by @NickCrews in #1080
• add new award by @RossKen in #1081
• Merge load settings methods by @ThomasHepworth in #1078
• Fix docs workflow by @ADBond in #1073
• Add docs for testing and creating a venv by @ThomasHepworth in #1086
• Added ability to profile nested lists by @zslade in #1074
• Workflow test multiple python versions by @ADBond in #1090
• WIP: Update new_library_comparisons_and_levels.md by @RossKen in #1082
• Added error message to catch pandas null casting issue when read into duckdb by @zslade in #1098
• add a bash script for linting by @ThomasHepworth in #1100
• parametrize datediff tests to clean them up by @ThomasHepworth in #1101
• update with parametrize to test more file loading options by @ThomasHepworth in #1105
• Improve citation by @RobinL in #1108
• Simplify specific implementations of SplinkDataframe by @RobinL in #1116
• Create Distance in KM Comparison library function by @RossKen in #1117
• Rename target_rows argument to max_pairs in estimate_u_using_random_sampling() by @NickCrews in #1087
• Create wrapper function for date comparisons by @RossKen in #1094
• Rename target rows as max_pairs by @RossKen in #1119
• Small Fixes by @NickCrews in #1115
• Fix benchmark comment action to work better with forks by @ADBond in #1122
• Release 3.7.0 proper by @ADBond in #1126

Full Changelog: v3.6.0...v3.7.0

v3.7.0.dev01

What's Changed

• Adjust caching for our concat tables by @ThomasHepworth in #1013
• _initialise_df_concat optionally returns list by @RobinL in #1023
• Df concat and df concat with tf return SplinkDataframe or None by @RobinL in #1033
• [Docs] Add a dev guide for creating new ComparisonLevels and Comparisons to Splink libraries by @ADBond in #1041
• correct module name duckb_base -> duckdb_base by @ADBond in #1046
• Some cache tests by @ADBond in #1050
• Improving the cache and make cache invalidation easier and more robust by @RobinL in #987
• Bump version to 3.7.0 by @RobinL in #1056
• Release 3.7.0 as dev version by @RobinL in #1057

Full Changelog: v3.6.0...v3.7.0.dev01

v3.6.0

What's Changed

• Conda install in readme by @ADBond in #1002
• fix: Safeguard against rounding/overflow errors in great_circle_distance_km_sql() by @NickCrews in #1006
• Create FEATURE_REQUEST.md by @OlivierBinette in #1010
• Drop python 3.6 support by @NickCrews in #1003
• Pin Black to v22 temporariliy, since v23 was removing lines by @RobinL in #1015
• Fixing tooltip for count and sum_matches in comparison viewer dashboard tooltip by @James-Osmond in #1016
• feat: Support sqlglot versions >=5.1.0 by @NickCrews in #1018
• Bump version to 3.6.0 ready for release by @RobinL in #1020

New Contributors

• @OlivierBinette made their first contribution in #1010

Full Changelog: v3.5.4...v3.6.0

v3.5.4

What's Changed

• threshold_actual parameter in precision-recall charts by @James-Osmond in #968
• upating enable_splink method to take spark as input parameter by @robertwhiffin in #989
• James osmond pr chart threshold by @RossKen in #971
• Udf register fix by @robertwhiffin in #993
• Update spark udf jars by @ThomasHepworth in #998

New Contributors

• @RossKen made their first contribution in #971

Full Changelog: v3.5.3...v3.5.4

v3.5.3

What's Changed

• Lint SparkLinker by @RobinL in #955
• Fix __splink_df_concat_with_tf_left name by @RobinL in #963
• Cumulative BRs - minor tweaks to fix documentation and rr error by @ThomasHepworth in #974
• Fix distance function comparison typo by @ADBond in #980
• add_l_or_r_to_identifier now has case for type exp.Lambda by @James-Osmond in #979
• Edit unlinkables table name when logged in db by @ThomasHepworth in #978
• Add datediff comparison levels by @ThomasHepworth in #972
• docs updates by @mamonu in #976
• Update find_matches_to_new_records to automatically generate our tf tables by @ThomasHepworth in #983

Full Changelog: v3.5.2...v3.5.3