nucliadb 6.3.0.post3402__py3-none-any.whl → 6.3.0.post3417__py3-none-any.whl

migrations/pg/0004_catalog_facets.py

@@ -0,0 +1,51 @@
+# Copyright (C) 2021 Bosutech XXI S.L.
+#
+# nucliadb is offered under the AGPL v3.0 and as commercial software.
+# For commercial licensing, contact us at info@nuclia.com.
+#
+# AGPL:
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as
+# published by the Free Software Foundation, either version 3 of the
+# License, or (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU Affero General Public License for more details.
+#
+# You should have received a copy of the GNU Affero General Public License
+# along with this program. If not, see <http://www.gnu.org/licenses/>.
+#
+
+from nucliadb.common.maindb.pg import PGTransaction
+
+
+async def migrate(txn: PGTransaction) -> None:
+    async with txn.connection.cursor() as cur:
+        await cur.execute("""
+            CREATE OR REPLACE FUNCTION extract_facets(facets TEXT[])
+            RETURNS TEXT[] AS $$
+                    SELECT array_agg(prefix) FROM (
+                            SELECT DISTINCT p.prefix
+                            FROM unnest(facets) AS facet,
+                            LATERAL (
+                                SELECT string_agg(part, '/') OVER(ROWS UNBOUNDED PRECEDING) AS prefix
+                                FROM regexp_split_to_table(facet, '/') AS part
+                            ) AS p
+                    ) AS s
+                    WHERE prefix != '';
+            $$ LANGUAGE SQL IMMUTABLE;
+        """)
+
+    # Concurrent index must be created outside of a transaction but psycopg automatically
+    # creates transactions. We temporarily disable this for this single statement.
+    await txn.connection.commit()
+    try:
+        await txn.connection.set_autocommit(True)
+        await txn.connection.execute(
+            "CREATE INDEX CONCURRENTLY ON catalog USING gin (kbid, extract_facets(labels));"
+        )
+        await txn.connection.execute("ANALYZE catalog")
+    finally:
+        await txn.connection.set_autocommit(False)

nucliadb/export_import/datamanager.py

@@ -30,7 +30,7 @@ from nucliadb_telemetry import errors
 from nucliadb_utils.storages.storage import Storage, StorageField
 
 MAINDB_EXPORT_KEY = "/kbs/{kbid}/exports/{id}"
-MAINDB_IMPORT_KY = "/kbs/{kbid}/imports/{id}"
+MAINDB_IMPORT_KEY = "/kbs/{kbid}/imports/{id}"
 STORAGE_EXPORT_KEY = "exports/{export_id}"
 STORAGE_IMPORT_KEY = "imports/{import_id}"
 
@@ -49,7 +49,7 @@ class ExportImportDataManager:
     def _get_maindb_metadata_key(self, type: str, kbid: str, id: str) -> str:
         if type not in ("export", "import"):
             raise ValueError(f"Invalid type: {type}")
-        key = MAINDB_EXPORT_KEY if type == "export" else MAINDB_IMPORT_KY
+        key = MAINDB_EXPORT_KEY if type == "export" else MAINDB_IMPORT_KEY
         return key.format(kbid=kbid, id=id)
 
     async def get_metadata(self, type: str, kbid: str, id: str) -> Metadata:

nucliadb/export_import/utils.py

@@ -134,6 +134,7 @@ async def import_binary(
     bucket_name = context.blob_storage.get_bucket_name(kbid)
     new_cf.bucket_name = bucket_name
 
+    # Replace the source kbid with the destination kbid
     src_kb = cf.uri.split("/")[1]
     new_cf.uri = new_cf.uri.replace(src_kb, kbid, 1)
 
@@ -235,7 +236,8 @@ def _clone_collect_cf(binaries: list[resources_pb2.CloudFile], origin: resources
 async def download_binary(
     context: ApplicationContext, cf: resources_pb2.CloudFile
 ) -> AsyncGenerator[bytes, None]:
-    async for data in context.blob_storage.download(cf.bucket_name, cf.uri):
+    bucket_name = context.blob_storage.get_bucket_name_from_cf(cf)
+    async for data in context.blob_storage.download(bucket_name, cf.uri):
         yield data
 
 

nucliadb/search/search/pgcatalog.py

@@ -61,7 +61,7 @@ def _convert_filter(filter, filter_params):
     if op == "literal":
         param_name = f"param{len(filter_params)}"
         filter_params[param_name] = [operands]
-        return f"labels @> %({param_name})s"
+        return f"extract_facets(labels) @> %({param_name})s"
     elif op in ("and", "or"):
         array_op = "@>" if op == "and" else "&&"
         sql = []
@@ -69,7 +69,7 @@ def _convert_filter(filter, filter_params):
         if literals:
             param_name = f"param{len(filter_params)}"
             filter_params[param_name] = literals
-            sql.append(f"labels {array_op} %({param_name})s")
+            sql.append(f"extract_facets(labels) {array_op} %({param_name})s")
         for nonlit in nonliterals:
             sql.append(_convert_filter(nonlit, filter_params))
         return "(" + f" {op.upper()} ".join(sql) + ")"

{nucliadb-6.3.0.post3402.dist-info → nucliadb-6.3.0.post3417.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: nucliadb
-Version: 6.3.0.post3402
+Version: 6.3.0.post3417
 Summary: NucliaDB
 Author-email: Nuclia <nucliadb@nuclia.com>
 License: AGPL
@@ -20,11 +20,11 @@ Classifier: Programming Language :: Python :: 3.12
 Classifier: Programming Language :: Python :: 3 :: Only
 Requires-Python: <4,>=3.9
 Description-Content-Type: text/markdown
-Requires-Dist: nucliadb-telemetry[all]>=6.3.0.post3402
-Requires-Dist: nucliadb-utils[cache,fastapi,storages]>=6.3.0.post3402
-Requires-Dist: nucliadb-protos>=6.3.0.post3402
-Requires-Dist: nucliadb-models>=6.3.0.post3402
-Requires-Dist: nidx-protos>=6.3.0.post3402
+Requires-Dist: nucliadb-telemetry[all]>=6.3.0.post3417
+Requires-Dist: nucliadb-utils[cache,fastapi,storages]>=6.3.0.post3417
+Requires-Dist: nucliadb-protos>=6.3.0.post3417
+Requires-Dist: nucliadb-models>=6.3.0.post3417
+Requires-Dist: nidx-protos>=6.3.0.post3417
 Requires-Dist: nucliadb-admin-assets>=1.0.0.post1224
 Requires-Dist: nuclia-models>=0.24.2
 Requires-Dist: uvicorn

{nucliadb-6.3.0.post3402.dist-info → nucliadb-6.3.0.post3417.dist-info}/RECORD

@@ -31,6 +31,7 @@ migrations/__init__.py,sha256=cp15ZcFnHvpcu_5-aK2A4uUyvuZVV_MJn4bIXMa20ks,835
 migrations/pg/0001_bootstrap.py,sha256=Fsqkeof50m7fKiJN05kmNEMwiKDlOrAgcAS5sLLkutA,1256
 migrations/pg/0002_catalog.py,sha256=Rsleecu351Ty19kYZgOpqX5G3MEAY8nMxCJrAeuS2Mw,1690
 migrations/pg/0003_catalog_kbid_index.py,sha256=uKq_vtnuf73GVf0mtl2rhzdk_czAoEU1UdiVKVZpA0M,1044
+migrations/pg/0004_catalog_facets.py,sha256=FJFASHjfEHG3sNve9BP2HnnLO4xr7dnR6Qpctnmt4LE,2180
 migrations/pg/__init__.py,sha256=cp15ZcFnHvpcu_5-aK2A4uUyvuZVV_MJn4bIXMa20ks,835
 nucliadb/__init__.py,sha256=_abCmDJ_0ku483Os4UAjPX7Nywm39cQgAV_DiyjsKeQ,891
 nucliadb/health.py,sha256=UIxxA4oms4HIsCRZM_SZsdkIZIlgzmOxw-qSHLlWuak,3465
@@ -94,13 +95,13 @@ nucliadb/common/models_utils/__init__.py,sha256=cp15ZcFnHvpcu_5-aK2A4uUyvuZVV_MJ
 nucliadb/common/models_utils/from_proto.py,sha256=d-q6-JWvjVVxtI57itQVKRIY8Hzv-tBhDt-tr4aIKNM,15979
 nucliadb/common/models_utils/to_proto.py,sha256=97JvOR_3odu50YvzLa2CERfEN3w_QPmAVcCJwJB5m5A,2438
 nucliadb/export_import/__init__.py,sha256=y-Is0Bxa8TMV6UiOW0deC_D3U465P65CQ5RjBjIWnow,932
-nucliadb/export_import/datamanager.py,sha256=b9Vhf-WqJ8HosTdNpKXlGj-Vi7MHyMoPxL0VpfPlYlE,6720
+nucliadb/export_import/datamanager.py,sha256=ZL96M6WuyrFLcGdoM6Cb6hQwrGFRtSeRlsKuvDjV3sY,6722
 nucliadb/export_import/exceptions.py,sha256=Dw8WqfG4r6MPJc5TPfbjMvCgXXWTcTOecGHRVU1h3kM,1949
 nucliadb/export_import/exporter.py,sha256=kgbW-B7FNW7mlc9rBVEfwkkFTqD58TWSTDe9zkmEnBc,7098
 nucliadb/export_import/importer.py,sha256=v5cq9Nn8c2zrY_K_00mydR52f8mdFxR7tLdtNLQ0qvk,4229
 nucliadb/export_import/models.py,sha256=dbjScNkiMRv4X3Ktudy1JRliD25bfoDTy3JmEZgQSCc,2121
 nucliadb/export_import/tasks.py,sha256=fpCBeFYPReyLIdk38LDM9Tpnw_VczeMrobT4n1RAIp4,2507
-nucliadb/export_import/utils.py,sha256=zrNrkkc9i3uT-R6Ju4J_0WNrzayln3KuQFCz-_qIaIA,19613
+nucliadb/export_import/utils.py,sha256=t7xLA3f5W3zGq3HNXe3noOQnY7gRO8TAoe8S8BG3_so,19733
 nucliadb/ingest/__init__.py,sha256=fsw3C38VP50km3R-nHL775LNGPpJ4JxqXJ2Ib1f5SqE,1011
 nucliadb/ingest/app.py,sha256=n5PlZyrf-bJsrc6jMRSO4DTjGyZLb4q6EA6Y4x6TYfQ,7028
 nucliadb/ingest/cache.py,sha256=w7jMMzamOmQ7gwXna6Dqm6isRNBVv6l5BTBlTxaYWjE,1005
@@ -222,7 +223,7 @@ nucliadb/search/search/ingestion_agents.py,sha256=NeJr4EEX-bvFFMGvXOOwLv8uU7NuQ-
 nucliadb/search/search/merge.py,sha256=i_PTBFRqC5iTTziOMEltxLIlmokIou5hjjgR4BnoLBE,22635
 nucliadb/search/search/metrics.py,sha256=81X-tahGW4n2CLvUzCPdNxNClmZqUWZjcVOGCUHoiUM,2872
 nucliadb/search/search/paragraphs.py,sha256=pNAEiYqJGGUVcEf7xf-PFMVqz0PX4Qb-WNG-_zPGN2o,7799
-nucliadb/search/search/pgcatalog.py,sha256=IaNK4dAxdXs38PoIkTdgqMDuZDjeiOtcXn3LeaT-OMw,8855
+nucliadb/search/search/pgcatalog.py,sha256=hNZJtbke0kFz9ygz70gMMJz9NxhRPznefCad83yQPd4,8887
 nucliadb/search/search/predict_proxy.py,sha256=IFI3v_ODz2_UU1XZnyaD391fE7-2C0npSmj_HmDvzS4,3123
 nucliadb/search/search/query.py,sha256=j7NdAXQXSytWvf_tyqpG-Fz4AhVGI-Y5i7NPm3s7-PI,29980
 nucliadb/search/search/rank_fusion.py,sha256=tRGo_KlsFsVx1CQEy1iqQ6f0T1Dq1kf0axDXHuuzvvM,6946
@@ -338,8 +339,8 @@ nucliadb/writer/tus/local.py,sha256=7jYa_w9b-N90jWgN2sQKkNcomqn6JMVBOVeDOVYJHto,
 nucliadb/writer/tus/s3.py,sha256=vF0NkFTXiXhXq3bCVXXVV-ED38ECVoUeeYViP8uMqcU,8357
 nucliadb/writer/tus/storage.py,sha256=ToqwjoYnjI4oIcwzkhha_MPxi-k4Jk3Lt55zRwaC1SM,2903
 nucliadb/writer/tus/utils.py,sha256=MSdVbRsRSZVdkaum69_0wku7X3p5wlZf4nr6E0GMKbw,2556
-nucliadb-6.3.0.post3402.dist-info/METADATA,sha256=ffldOKngG5BDfwbQ3N430iCHKoBEh8KkL2iM2nAr7_s,4291
-nucliadb-6.3.0.post3402.dist-info/WHEEL,sha256=In9FTNxeP60KnTkGw7wk6mJPYd_dQSjEZmXdBdMCI-8,91
-nucliadb-6.3.0.post3402.dist-info/entry_points.txt,sha256=XqGfgFDuY3zXQc8ewXM2TRVjTModIq851zOsgrmaXx4,1268
-nucliadb-6.3.0.post3402.dist-info/top_level.txt,sha256=hwYhTVnX7jkQ9gJCkVrbqEG1M4lT2F_iPQND1fCzF80,20
-nucliadb-6.3.0.post3402.dist-info/RECORD,,
+nucliadb-6.3.0.post3417.dist-info/METADATA,sha256=Nya_zwsp_AAX_HD3Qf6w7Hwd5UpeppIBUeCi9qqCnro,4291
+nucliadb-6.3.0.post3417.dist-info/WHEEL,sha256=nn6H5-ilmfVryoAQl3ZQ2l8SH5imPWFpm1A5FgEuFV4,91
+nucliadb-6.3.0.post3417.dist-info/entry_points.txt,sha256=XqGfgFDuY3zXQc8ewXM2TRVjTModIq851zOsgrmaXx4,1268
+nucliadb-6.3.0.post3417.dist-info/top_level.txt,sha256=hwYhTVnX7jkQ9gJCkVrbqEG1M4lT2F_iPQND1fCzF80,20
+nucliadb-6.3.0.post3417.dist-info/RECORD,,

{nucliadb-6.3.0.post3402.dist-info → nucliadb-6.3.0.post3417.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (75.8.0)
+Generator: setuptools (75.8.1)
 Root-Is-Purelib: true
 Tag: py3-none-any