easylink 0.1.21__py3-none-any.whl → 0.1.23__py3-none-any.whl

{easylink-0.1.21.dist-info → easylink-0.1.23.dist-info}/RECORD

@@ -1,41 +1,39 @@
 easylink/__about__.py,sha256=2-oxCfu9t9yUJouLDwqYRZ0eii8kN25SxRzsawjWjho,440
 easylink/__init__.py,sha256=gGMcIVfiVnHtlDw5mZwhevcDb2wt-kuP6F64gnkFack,159
-easylink/_version.py,sha256=qEmNtjnOwhDYQ0cHPPtUkUaghzD2xl0thJEznl4giYw,23
+easylink/_version.py,sha256=0byemO6n6WCv41u9vBG2AIsOkVbxLvok7puvwy8EhfU,23
 easylink/cli.py,sha256=zQO4lOVoZ3eVgPVWT2sCF4zNoKgiDJP9ReTh2Myr9jc,10307
-easylink/configuration.py,sha256=hgmG5SIbYqnHDHfk44Gr3QX7C3yTaEVW6GuKeMqvu6c,12689
+easylink/configuration.py,sha256=TPd3WbqUcJMJDPJuHeo5ZebvZPQrRyfm8-laad2sOFk,12681
 easylink/graph_components.py,sha256=zZDZXg5smReHO3ryQC4pao24wyKXzWDe6jS3C6fM2ak,13892
 easylink/implementation.py,sha256=lSF37g-aQYgdLI0lgFaDrBkp23te9O9B1V-CmmRtB-0,14514
-easylink/implementation_metadata.yaml,sha256=GoU_aWjVryG8-xjUHkC2nCUeznmYD0BwfJYnNrpZ8P4,10670
+easylink/implementation_metadata.yaml,sha256=pKu_H9fLnTsS8E4wCnYRitumW1-zs7mfE3z66BAyO30,10848
 easylink/pipeline.py,sha256=NJyMInbOCjJ_5kRzzuK5AcupvC7ecd5qLOC-f1Gy3Ss,17701
 easylink/pipeline_graph.py,sha256=jtjS7_2IVa189U8cL621600erC_0pa6BKPRRO8VqwiU,23760
-easylink/pipeline_schema.py,sha256=FieJBa3rKgaCIB9QDuQEfWJ9joNBUUp6iHT6xmns-Vk,6886
-easylink/rule.py,sha256=QJPmrvQUZPnqGFD9UmMK8imdJ7VODzGlUOSnpJhb9AU,16677
+easylink/pipeline_schema.py,sha256=sj6YQqMFUS35RRFWdlrhPlud68d3Rb5y7iKfAEZy1qs,6877
+easylink/rule.py,sha256=MM7WyW56J7zT2FVjHlFtjuz62PfdSBGTD3MNcpLfEZM,16598
 easylink/runner.py,sha256=Z9GKgiUAWtp0dW1cyAp86MGthIDeABJtHDXUtzv0-kE,6714
-easylink/step.py,sha256=SqOxinHyRaLCEnB_y5dvhGMaRLyphQDCpVsQ3160c9U,89588
+easylink/step.py,sha256=-vdFhPvwAZ3d69LMQGmSIVdcMG8E8ZtSvTE0UWif7zs,91088
 easylink/devtools/implementation_creator.py,sha256=gZZpfpiOOh912nNjR_d5wR0mBO5c09jWMS0FSYRcE1o,19120
-easylink/images/spark_cluster/Dockerfile,sha256=3PHotbR4jdjVYRHOJ0VQW55b5Qd4tQ1pLLQMrTKWVA0,576
-easylink/images/spark_cluster/README.md,sha256=KdgSttZRplNNWqHn4K1GTsTIab3dTOSG4V99QPLxSp8,569
-easylink/pipeline_schema_constants/__init__.py,sha256=SMNXz49DSwx05PHMKUsunJsgMOqsBJaAHA1fmIOJsUU,1445
+easylink/pipeline_schema_constants/__init__.py,sha256=njUL2v67WFC3dW_FjOXWRLSeOhInox_ZINLEji0_7V8,1523
 easylink/pipeline_schema_constants/development.py,sha256=0x6lWqBmP1K9AebEmeZ4veSnLBcdQcZXzbV6lCU11bc,12622
 easylink/pipeline_schema_constants/main.py,sha256=kcAhdbK_BhS79LUMhKkn-Uttl2vA-MHVX4M1StTloJQ,22934
-easylink/pipeline_schema_constants/testing.py,sha256=G7szRMyY48dL8kUHWq2MeMaV2G0F-AdAPsQxFzdUnFI,20567
+easylink/pipeline_schema_constants/testing.py,sha256=ti08DeUuF-eWrGKMj4BMyOFFJnEYooDaWX0DGiferbk,24579
 easylink/steps/cascading/exclude_clustered.def,sha256=GfoDqO2Vtsh7VI8SwGaJtv_KtKjs-UmBcivqQ7OPkjk,503
-easylink/steps/cascading/exclude_clustered.py,sha256=NSA6GZBzGa7e6CH4tacCGfr0Y9sUM29g9Nf8NquHB44,2612
+easylink/steps/cascading/exclude_clustered.py,sha256=Bpsyf9vAZ431Fh96RVzHkF7fy77NQjo1Cl6bHCIy69c,2580
 easylink/steps/cascading/exclude_none.def,sha256=iFUhUMocxtkA0NErkjVrBxY0MUdS3DIPNsbCpTJRP0k,488
-easylink/steps/cascading/exclude_none.py,sha256=KntBX3q-V47d96ztOlPNRY_kCFJNi1LNYQ7UNs5wB4c,2507
+easylink/steps/cascading/exclude_none.py,sha256=5DK5bNG4TneMwUKE49Kmz7VDnKBNZWjOERkuSJU3BmA,2475
 easylink/steps/cascading/update_clusters_by_connected_components.def,sha256=sAAAWOod8EuAnotR1cayaGAvs7x6xoMVlwmLso_a9Cc,578
-easylink/steps/cascading/update_clusters_by_connected_components.py,sha256=43D5GBmPXSgxcjgbJTvEoGFvPzBCGqYgBaT42pncNNw,3661
+easylink/steps/cascading/update_clusters_by_connected_components.py,sha256=sFZXMGXl17jcGt8Fu5hgQz1KW5bFvPYdCoQGZ9Erc0I,3629
 easylink/steps/default/default_clusters_to_links.def,sha256=9PjUygLvsoYMUZDznceuuv55t8fPs473P57J_RMl3U0,527
-easylink/steps/default/default_clusters_to_links.py,sha256=EIYeP0lj0plBl2OpTRuv3iDEQl-zNVJONUg0kgKSEF0,2848
+easylink/steps/default/default_clusters_to_links.py,sha256=Ckm53d3W-ETNlTvQJPOkpHmSqCmxSWknMPQrEAIoTBY,2816
 easylink/steps/default/default_determining_exclusions.def,sha256=zZUEHDdrpLxzx3gTm-dki2ge5ivCCg4ziIwTErqCII0,542
-easylink/steps/default/default_determining_exclusions.py,sha256=tF2lcga-6n99shgYEmhpNuqok33u7dcW9q5wV3xgp5w,2661
+easylink/steps/default/default_determining_exclusions.py,sha256=4diLfuqYm_Koj7gwifjwe_7mLZ6xb6RQiEdk-RRtB94,2629
 easylink/steps/default/default_removing_records.def,sha256=QqacmOu6myxFSULHRKeKsVD8l73KDm4VEkPkPlovwqs,524
-easylink/steps/default/default_removing_records.py,sha256=LIlFS8EvJ6h5XqEfgWZYyIAjcKj7Oo8_I5a-vXHOozs,1938
+easylink/steps/default/default_removing_records.py,sha256=P4mmX2D4mhSoWd_S5CaNT4hlHOMAeZiMhCScWQiR_fQ,1906
 easylink/steps/default/default_schema_alignment.def,sha256=hFHJkos0Fhe4LvpjLOCd6klIaIqOKqECDDSTVu3G03Y,524
-easylink/steps/default/default_schema_alignment.py,sha256=Uxi6uTFveFKSiiRZG9MnTXOklQngSKGMafqnvKDc0rY,1459
+easylink/steps/default/default_schema_alignment.py,sha256=oT5jbYQ3C3ocLgqqOnvH0SIJ6NeTtPBWWmCqr_frnAI,1479
 easylink/steps/default/default_updating_clusters.def,sha256=vDzSkTknDfeiXeHREpw4BkUxFcTWamxr81c3rZ7_enY,527
-easylink/steps/default/default_updating_clusters.py,sha256=A-lO3ussM1Ntffp-ZyPQGbbxZg4QNiZ8AvSOGVJDXnA,2139
-easylink/steps/dev/README.md,sha256=u9dZUggpY2Lf2qb-xkDLWWgHjcmi4osbQtzSNo4uklE,4549
+easylink/steps/default/default_updating_clusters.py,sha256=uwblSM9w4PoqPgWA_5cwmnU7ARSruickWxTfr2EZCM0,2107
+easylink/steps/dev/README.md,sha256=lEHtM48SkFTV1FL-B5bbrEGjSVa_mb690Ed08nnheww,4533
 easylink/steps/dev/build-containers-local.sh,sha256=Wy3pfcyt7I-BNvHcr7ZXDe0g5Ihd00BIPqt9YuRbLeA,259
 easylink/steps/dev/build-containers-remote.sh,sha256=Hy-kaaXf-ta6n8SzOz_ahByjMY5T7J71MvzXRXDvQw8,271
 easylink/steps/dev/test.py,sha256=4iudKBD6CFz2CxbjSBUkc8LCWlMl-Nmw_rB35ZN6TrQ,6835
@@ -45,36 +43,38 @@ easylink/steps/dev/input_data/input_file_1.parquet,sha256=Km8jRyfGNdq0MFdz_-bewl
 easylink/steps/dev/input_data/input_file_2.csv,sha256=YqKLZDC4d-aYN8Dh9OB6iQWWUKmvueu5CszckH1AApU,100016
 easylink/steps/dev/input_data/input_file_2.parquet,sha256=Vpo0sUqQ78qlWLRk8p303Nh89BVcK4uvXJljRGHmsWk,60392
 easylink/steps/dev/python_pandas/README.md,sha256=c_FbtkKKOTjt2R_LfHUo5lBga1qHiYkxLdQeewRr45g,977
-easylink/steps/dev/python_pandas/dummy_step.py,sha256=NvhLUZu40B3Xbj_S-chQ6IkYUPr6X2aGBxYUa3DqwmY,4362
-easylink/steps/dev/python_pandas/python_pandas.def,sha256=24cxwGF8Cqkv2a1zVsu94MfC_bAXBqAINLwfW2zyB_0,769
+easylink/steps/dev/python_pandas/dummy_step.py,sha256=ASZIxk_d46zNz0xUtZ37OuTJUpzfi98EwQduZXDiwK0,4330
+easylink/steps/dev/python_pandas/python_pandas.def,sha256=umAUJL3RQZsCLIvFvsaif_-kCGfWzsnIH8d6Zwx_qYQ,482
 easylink/steps/dev/python_pyspark/README.md,sha256=di29SAfcdTTpar7gdoJRLqKrL8DEfNeayYUyaywdhUg,1563
-easylink/steps/dev/python_pyspark/dummy_step.py,sha256=wxHHI3Uv8MTipKG2ffHbT_eL4JkoNpx49bJoErXumdc,5003
-easylink/steps/dev/python_pyspark/python_pyspark.def,sha256=j_RmVjspmXGOhJTr10ED13RYfbimgxRU3WVTL7VOIUQ,915
+easylink/steps/dev/python_pyspark/dummy_step.py,sha256=6BJi-L6tlXGc6GJnGTI06AnzuRYh3KYBgU09tvId954,4949
+easylink/steps/dev/python_pyspark/python_pyspark.def,sha256=e-OXax5t96OEgAJdsKDqjGXPuCk12GpPYbwIdrUpBOU,666
 easylink/steps/dev/r/README.md,sha256=dPjZdDTqcJsZCiwhddzlOj1ob0P7YocZUNFrLIGM1-0,1201
-easylink/steps/dev/r/dummy_step.R,sha256=1TWZY8CEkT6gavrulBxFsKbDSKJJjk0NtJrGH7TIikE,4975
-easylink/steps/dev/r/r-image.def,sha256=LrhXlt0C3k7d_VJWopRPEVARnFWSuq_oILlwo7g03bE,627
-easylink/steps/fastLink/fastLink_evaluating_pairs.R,sha256=fQRrTPrgb1t5hrQi0V5H55J-PHdWjsATrVRYdXNYtdU,4603
+easylink/steps/dev/r/dummy_step.R,sha256=4eFZgmKaagydjYOVnrOB3W2vjHHcGDhJ8LXsWDrDNyI,4943
+easylink/steps/dev/r/r-image.def,sha256=gBNCfMb_HtgrGPhTt8qEws5tWVfjQIS_GCIHRGAHG9c,391
+easylink/steps/example/middle_name_to_initial.def,sha256=UmD3FCuK8CMD0gQRUqg1BFnGq5Mucu7x8eU19jq7pZ0,518
+easylink/steps/example/middle_name_to_initial.py,sha256=1Q7xaXIxkIvPN6jW98WKTvVhWB9qMC23mRIoO7NYRa8,1901
+easylink/steps/fastLink/fastLink_evaluating_pairs.R,sha256=ucbHibtoYJ4-GDg1mWv-dtv0r_1XomhdT-KC3Zkat2E,4539
 easylink/steps/fastLink/fastLink_evaluating_pairs.def,sha256=5rDi-cmWhyuFEsiGFPpTxtySMqq5TpgJG-y8g_MtEvA,509
-easylink/steps/fastLink/fastLink_links_to_clusters.R,sha256=exVzJl4r7k7cRlMCHSmigOqTlxShqzK-FO3EDhlPksg,4087
+easylink/steps/fastLink/fastLink_links_to_clusters.R,sha256=iM6bi27bHNJRxfOIBOCvZcFUuKT3VauOWbM6d0Ws5dk,4055
 easylink/steps/fastLink/fastLink_links_to_clusters.def,sha256=1xYjOMsHtSS2-AI4EC2r6kL8ZX5F2JhmvESefEKeJVY,512
 easylink/steps/output_dir/dummy_step_1_for_output_dir_example.def,sha256=CkQVG-uDRQ9spAavdkZbhx2GD_fRsKZGELPrr8yltsc,550
-easylink/steps/output_dir/dummy_step_1_for_output_dir_example.py,sha256=dI0OWugE35ABLcSwsI-T3C4dvuPTKXwjE52dtSsCo8Y,428
+easylink/steps/output_dir/dummy_step_1_for_output_dir_example.py,sha256=sBmF-wMgTpcqeM9gVWbHZkcy_w0OzfDYo98-3P2WMaM,412
 easylink/steps/output_dir/dummy_step_2_for_output_dir_example.def,sha256=9gShg1EDJEHZcz7Z5VfZ1A4Gpm9XQes8ezn6rAZDgDM,550
-easylink/steps/output_dir/dummy_step_2_for_output_dir_example.py,sha256=DMJW5TXjhELxhY4U9q2RpLjqxlS1YSosTGL2AfRnaZM,521
+easylink/steps/output_dir/dummy_step_2_for_output_dir_example.py,sha256=ExFljptolMiidU7LiOfQtH13ChbDUGIF3r5qM5paKsA,489
 easylink/steps/rl-dummy/canonicalizing_and_downstream_analysis/dummy_canonicalizing_and_downstream_analysis.def,sha256=YOWtJZxoe-kHFeEyrgGcVGfdqcbD_Fg17A9shOaK-yc,584
-easylink/steps/rl-dummy/canonicalizing_and_downstream_analysis/dummy_canonicalizing_and_downstream_analysis.py,sha256=skZUiZWcSXAOqq8TAlN5I0wztXgCWHQYA_xkuiL5s28,1202
+easylink/steps/rl-dummy/canonicalizing_and_downstream_analysis/dummy_canonicalizing_and_downstream_analysis.py,sha256=zctz6LktX1BsVbeSR2gWuNCzRU7WkmWDGg68wlzZw0Q,1170
 easylink/steps/rl-dummy/input_data/create_input_files.ipynb,sha256=uXvJ8zTTPg0077HgA7MhQ_9jD-aeISFLeMeEBbSnOu8,54498
 easylink/steps/rl-dummy/input_data/input_file_1.parquet,sha256=GQ_7v7ucwdJn-9mTgKVcvqkJ5gTkwb0B7y38mfOYbic,15200
 easylink/steps/rl-dummy/input_data/input_file_2.parquet,sha256=Y4eseBm0HmFroksQr_VApgozRL8h8u7nQO6x_Utyns8,14902
 easylink/steps/rl-dummy/input_data/known_clusters.parquet,sha256=Ysodu65toHZN4AgjVJsm0ueUxPIZAJjbtRm9SVM08JE,2598
 easylink/steps/rl-dummy/pre-processing/dummy_pre-processing.def,sha256=HeUSv2QvMOQzsyVktYR1xYoEqwiNpDo-p7IRcGSMspE,512
-easylink/steps/rl-dummy/pre-processing/dummy_pre-processing.py,sha256=aeDgn9z2um0oTPNSwPcTkBou3-1ajud_MWhkuRoHdOU,1884
-easylink/steps/splink/splink_blocking_and_filtering.def,sha256=foAQAPvhDEkXkevpghS-uftsTbIQnQy9PvTkyldQeAA,539
-easylink/steps/splink/splink_blocking_and_filtering.py,sha256=8-_a9PkOmKSa-8TJ9YMjqI7gLo-YD9JCAO1f8uBhdoE,4469
+easylink/steps/rl-dummy/pre-processing/dummy_pre-processing.py,sha256=I6kqG4e_H2yFW5MpsMXdpoY_NjHcBvVVAHWv89LUgXE,1852
+easylink/steps/splink/splink_blocking_and_filtering.def,sha256=umWzxJhsfdi8g3TD-r2mKpjC-FPAMDk6-IERiWigdQc,557
+easylink/steps/splink/splink_blocking_and_filtering.py,sha256=FO8YJ2_KgCLpQoq5xsM339bTSN1DhCXCL8XT1pb5STY,5259
 easylink/steps/splink/splink_evaluating_pairs.def,sha256=DN3Ohy9qJOAyK58v164neP23HDVYpedMqzCu4eQh4Hg,521
-easylink/steps/splink/splink_evaluating_pairs.py,sha256=JR2qVgb14cNZKozDyOrN11nr1mXOwWv69E6WP0pRlMw,5713
+easylink/steps/splink/splink_evaluating_pairs.py,sha256=m-j1QMRSvPCiSoWVSV1kzzzsK1c_xG8nqYKMd3cj7kM,6195
 easylink/steps/splink/splink_links_to_clusters.def,sha256=RurvOYyGjNs9tx64DTXwI-GSgHD4T7SzDfhAH18pTEM,524
-easylink/steps/splink/splink_links_to_clusters.py,sha256=z5ymdYl9ytp1e5MA6vn8wpGRFWVuhh23LqGq8NJJxZQ,1936
+easylink/steps/splink/splink_links_to_clusters.py,sha256=5Sw8yi0dVLuRB-trN2kXmxbHBR0VJBxYee6u4_usg2Y,1920
 easylink/utilities/__init__.py,sha256=0U33kbv4hoMfFQ_lh5hLwifxRPzOgkLkjKLYxmaK10g,196
 easylink/utilities/aggregator_utils.py,sha256=_DAHRAf9janbDsuj_jnAn5Dzz2s4R5Ni3YeytDpN9UE,954
 easylink/utilities/data_utils.py,sha256=XPRjq3qW_fN0xQ23Jms_xBzpTHbRwqZWDP1AW0nYkP0,6926
@@ -83,9 +83,9 @@ easylink/utilities/paths.py,sha256=9inDKMPorAaWh5vhjc3j1Tj_aXVKhLdodiJO9H1nNes,9
 easylink/utilities/spark.smk,sha256=kGtpem7LfQc71tMh5WAYaqKnHQKFvcdhPQSdumOP70k,5799
 easylink/utilities/splitter_utils.py,sha256=KXiVUYJ9TGxCQmrwos18pB1sxG_0Ay67qoDJT6vem2o,3828
 easylink/utilities/validation_utils.py,sha256=DBJB2TLVklgYw1WaaPse9vqtwPLMGmZNYM2cbCZsoHI,18417
-easylink-0.1.21.dist-info/licenses/LICENSE,sha256=z6NBo3hY2Pbh2Wjm9IcHLyQoG-kd816uASh8PbwhTcQ,1530
-easylink-0.1.21.dist-info/METADATA,sha256=wdHGbqg2d4yte9ep9mO_GAr2EbUmEAVHHjPg6LsvMLE,3565
-easylink-0.1.21.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-easylink-0.1.21.dist-info/entry_points.txt,sha256=OGMZDFltg3yMboT7XjJt3joiPhRfV_7jnREVtrAIQNU,51
-easylink-0.1.21.dist-info/top_level.txt,sha256=oHcOpcF_jDMWFiJRzfGQvuskENGDjSPC_Agu9Z_Xvik,9
-easylink-0.1.21.dist-info/RECORD,,
+easylink-0.1.23.dist-info/licenses/LICENSE,sha256=z6NBo3hY2Pbh2Wjm9IcHLyQoG-kd816uASh8PbwhTcQ,1530
+easylink-0.1.23.dist-info/METADATA,sha256=u-oRyBse4M0AsFkMjTuy0JCpul-BwHJ1JaD9fIALrHU,3565
+easylink-0.1.23.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+easylink-0.1.23.dist-info/entry_points.txt,sha256=OGMZDFltg3yMboT7XjJt3joiPhRfV_7jnREVtrAIQNU,51
+easylink-0.1.23.dist-info/top_level.txt,sha256=oHcOpcF_jDMWFiJRzfGQvuskENGDjSPC_Agu9Z_Xvik,9
+easylink-0.1.23.dist-info/RECORD,,

easylink/images/spark_cluster/Dockerfile

@@ -1,16 +0,0 @@
-# Stage 1: Start with the miniconda3 base image
-FROM continuumio/miniconda3 as conda-base
-
-# Create a new conda environment
-SHELL ["/bin/bash", "--login", "-c"]
-RUN conda init bash \
-    && . ~/.bashrc \
-    && conda create -n spark_cluster python=3.10
-
-# Stage 2: Start with the Apache Spark base image
-FROM apache/spark@sha256:a1dd2487a97fb5e35c5a5b409e830b501a92919029c62f9a559b13c4f5c50f63 as spark-base
-
-COPY --from=conda-base /opt/conda /opt/conda
-
-# Set PATH for conda environment and conda itself
-ENV PATH=/opt/conda/envs/spark_cluster/bin:/opt/conda/condabin:${PATH}

easylink/images/spark_cluster/README.md

@@ -1,15 +0,0 @@
-# spark_cluster container
-NOTE: Spinning up a spark cluster using `easylink` currently requires building an image from this directory.
-
-This is done by running the following commands from this directory:
-
-```
-# build the image
-$ sudo docker build -t easylink:sparkbuilder .
-# save as compressed tarball
-$ sudo docker save easylink:sparkbuilder | gzip > spark_cluster.tar.gz
-# remove the image
-$ sudo docker rmi easylink:sparkbuilder
-# convert the image from the docker image
-$ singularity build --force spark_cluster.sif docker-archive://$(pwd)/spark_cluster.tar.gz
-```