nucliadb 6.2.1.post2842__py3-none-any.whl → 6.2.1.post2855__py3-none-any.whl

migrations/0029_backfill_field_status.py

@@ -0,0 +1,149 @@
+# 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/>.
+#
+
+"""Migration #29
+
+Backfill field status (from error)
+"""
+
+import logging
+from typing import Optional
+
+from nucliadb.migrator.context import ExecutionContext
+from nucliadb_protos import resources_pb2, writer_pb2
+
+logger = logging.getLogger(__name__)
+
+
+async def migrate(context: ExecutionContext) -> None:
+    start: Optional[str] = ""
+    while True:
+        if start is None:
+            break
+        start = await do_batch(context, start)
+
+
+async def migrate_kb(context: ExecutionContext, kbid: str) -> None: ...
+
+
+async def do_batch(context: ExecutionContext, start: str) -> Optional[str]:
+    logger.info(f"Running batch from {start}")
+    async with context.kv_driver.transaction(read_only=False) as txn:
+        async with txn.connection.cursor() as cur:  # type: ignore
+            # Retrieve a batch of fields
+            await cur.execute(
+                """
+                SELECT key FROM resources
+                WHERE key ~ '^/kbs/[^/]*/r/[^/]*/f/[^/]*/[^/]*$'
+                AND key > %s
+                ORDER BY key
+                LIMIT 500""",
+                (start,),
+            )
+            records = await cur.fetchall()
+            if len(records) == 0:
+                return None
+
+            field_keys = [r[0] for r in records]
+
+            # Retrieve resources basic (to check status)
+            resource_keys = set(["/".join(f.split("/")[:5]) for f in field_keys])
+            await cur.execute(
+                """
+                SELECT key, value FROM resources
+                WHERE key = ANY (%s)
+                ORDER BY key
+                """,
+                (list(resource_keys),),
+            )
+            records = await cur.fetchall()
+            resources_basic = {}
+            for k, v in records:
+                row_basic = resources_pb2.Basic()
+                row_basic.ParseFromString(v)
+                resources_basic[k] = row_basic
+
+            # Retrieve field errors
+            await cur.execute(
+                """
+                SELECT key, value FROM resources
+                WHERE key ~ '^/kbs/[^/]*/r/[^/]*/f/[^/]*/[^/]*/error$'
+                AND key > %s AND key <= %s
+                ORDER BY key
+                """,
+                (start, field_keys[-1] + "/error"),
+            )
+            records = await cur.fetchall()
+            errors = {}
+            for k, v in records:
+                row_error = writer_pb2.Error()
+                row_error.ParseFromString(v)
+                errors[k] = row_error
+
+            # Retrieve existing status keys
+            await cur.execute(
+                """
+                SELECT key FROM resources
+                WHERE key ~ '^/kbs/[^/]*/r/[^/]*/f/[^/]*/[^/]*/status$'
+                AND key > %s AND key <= %s
+                ORDER BY key
+                """,
+                (start, field_keys[-1] + "/status"),
+            )
+            records = await cur.fetchall()
+            has_status = [r[0] for r in records]
+
+            set_batch = []
+            for field_key in field_keys:
+                if field_key + "/status" in has_status:
+                    # Already has status, skip
+                    continue
+
+                resource_key = "/".join(field_key.split("/")[:5])
+                basic = resources_basic.get(resource_key, None)
+                if basic is None:
+                    logger.warn(f"{field_key} resource has no basic, skipped")
+                    continue
+
+                status = writer_pb2.FieldStatus()
+                status.status = writer_pb2.FieldStatus.Status.PROCESSED
+                error = errors.get(field_key + "/error", None)
+                # We only copy errors if they come from data augmentation or if the resource is in error
+                # This way we ensure we do not set an error for resources that were previously not in error
+                # There is no way to do this 100% accurate since the /error key is only cleared on field deletion
+                if error:
+                    if (
+                        error.code == writer_pb2.Error.ErrorCode.DATAAUGMENTATION
+                        or basic.metadata.status == resources_pb2.Metadata.Status.ERROR
+                    ):
+                        field_error = writer_pb2.FieldError(
+                            source_error=error,
+                        )
+                        status.errors.append(field_error)
+                        status.status = writer_pb2.FieldStatus.Status.ERROR
+                set_batch.append((field_key + "/status", status.SerializeToString()))
+
+            # Write everything to the database in batch
+            async with cur.copy("COPY resources (key, value) FROM STDIN") as copy:
+                for row in set_batch:
+                    await copy.write_row(row)
+            await txn.commit()
+
+            return field_keys[-1]

nucliadb/common/datamanagers/fields.py

@@ -23,11 +23,13 @@ from typing import Optional
 from google.protobuf.message import Message
 
 from nucliadb.common.datamanagers.utils import get_kv_pb
+from nucliadb.common.ids import FIELD_TYPE_PB_TO_STR
 from nucliadb.common.maindb.driver import Transaction
 from nucliadb_protos import writer_pb2
 
 KB_RESOURCE_FIELD = "/kbs/{kbid}/r/{uuid}/f/{type}/{field}"
 KB_RESOURCE_FIELD_ERROR = "/kbs/{kbid}/r/{uuid}/f/{type}/{field}/error"
+KB_RESOURCE_FIELD_STATUS = "/kbs/{kbid}/r/{uuid}/f/{type}/{field}/status"
 
 
 async def get_raw(
@@ -52,13 +54,7 @@ async def set(
 
 async def delete(txn: Transaction, *, kbid: str, rid: str, field_type: str, field_id: str):
     base_key = KB_RESOURCE_FIELD.format(kbid=kbid, uuid=rid, type=field_type, field=field_id)
-    # Make sure we explicitly delete the field and any nested key
-    keys_to_delete = []
-    async for key in txn.keys(base_key):
-        keys_to_delete.append(key)
-
-    for key in keys_to_delete:
-        await txn.delete(key)
+    await txn.delete_by_prefix(base_key)
 
 
 # Error
@@ -82,3 +78,46 @@ async def set_error(
 ):
     key = KB_RESOURCE_FIELD_ERROR.format(kbid=kbid, uuid=rid, type=field_type, field=field_id)
     await txn.set(key, error.SerializeToString())
+
+
+# Status, replaces error
+
+
+async def get_status(
+    txn: Transaction, *, kbid: str, rid: str, field_type: str, field_id: str
+) -> Optional[writer_pb2.FieldStatus]:
+    key = KB_RESOURCE_FIELD_STATUS.format(kbid=kbid, uuid=rid, type=field_type, field=field_id)
+    return await get_kv_pb(txn, key, writer_pb2.FieldStatus)
+
+
+async def get_statuses(
+    txn: Transaction, *, kbid: str, rid: str, fields: list[writer_pb2.FieldID]
+) -> list[writer_pb2.FieldStatus]:
+    keys = [
+        KB_RESOURCE_FIELD_STATUS.format(
+            kbid=kbid, uuid=rid, type=FIELD_TYPE_PB_TO_STR[fid.field_type], field=fid.field
+        )
+        for fid in fields
+    ]
+    serialized = await txn.batch_get(keys, for_update=False)
+    statuses = []
+    for serialized_status in serialized:
+        pb = writer_pb2.FieldStatus()
+        if serialized_status is not None:
+            pb.ParseFromString(serialized_status)
+        statuses.append(pb)
+
+    return statuses
+
+
+async def set_status(
+    txn: Transaction,
+    *,
+    kbid: str,
+    rid: str,
+    field_type: str,
+    field_id: str,
+    status: writer_pb2.FieldStatus,
+):
+    key = KB_RESOURCE_FIELD_STATUS.format(kbid=kbid, uuid=rid, type=field_type, field=field_id)
+    await txn.set(key, status.SerializeToString())

nucliadb/ingest/fields/base.py

@@ -41,7 +41,7 @@ from nucliadb_protos.resources_pb2 import (
     QuestionAnswers,
 )
 from nucliadb_protos.utils_pb2 import ExtractedText, VectorObject
-from nucliadb_protos.writer_pb2 import Error
+from nucliadb_protos.writer_pb2 import Error, FieldStatus
 from nucliadb_utils.storages.storage import Storage, StorageField
 
 SUBFIELDFIELDS = ("c",)
@@ -215,6 +215,25 @@ class Field(Generic[PbType]):
             error=error,
         )
 
+    async def get_status(self) -> Optional[FieldStatus]:
+        return await datamanagers.fields.get_status(
+            self.resource.txn,
+            kbid=self.kbid,
+            rid=self.uuid,
+            field_type=self.type,
+            field_id=self.id,
+        )
+
+    async def set_status(self, status: FieldStatus) -> None:
+        await datamanagers.fields.set_status(
+            self.resource.txn,
+            kbid=self.kbid,
+            rid=self.uuid,
+            field_type=self.type,
+            field_id=self.id,
+            status=status,
+        )
+
     async def get_question_answers(self, force=False) -> Optional[FieldQuestionAnswers]:
         if self.question_answers is None or force:
             sf = self.get_storage_field(FieldTypes.QUESTION_ANSWERS)

nucliadb/ingest/orm/resource.py

@@ -21,6 +21,7 @@ from __future__ import annotations
 
 import asyncio
 import logging
+from collections import defaultdict
 from concurrent.futures import ThreadPoolExecutor
 from functools import partial
 from typing import TYPE_CHECKING, Any, AsyncIterator, MutableMapping, Optional, Type
@@ -534,6 +535,86 @@ class Resource:
                 errors=message.errors,  # type: ignore
             )
 
+    @processor_observer.wrap({"type": "apply_fields_status"})
+    async def apply_fields_status(self, message: BrokerMessage, updated_fields: list[FieldID]):
+        # Dictionary of all errors per field (we may have several due to DA tasks)
+        errors_by_field: dict[tuple[FieldType.ValueType, str], list[writer_pb2.Error]] = defaultdict(
+            list
+        )
+
+        # Make sure if a file is updated without errors, it ends up in errors_by_field
+        for field_id in updated_fields:
+            errors_by_field[(field_id.field_type, field_id.field)] = []
+        for fs in message.field_statuses:
+            errors_by_field[(fs.id.field_type, fs.id.field)] = []
+
+        for error in message.errors:
+            errors_by_field[(error.field_type, error.field)].append(error)
+
+        # If this message comes from the processor (not a DA worker), we clear all previous errors
+        # TODO: When generated_by is populated with DA tasks by processor, remove only related errors
+        from_processor = any((x.WhichOneof("generator") == "processor" for x in message.generated_by))
+
+        for (field_type, field), errors in errors_by_field.items():
+            field_obj = await self.get_field(field, field_type, load=False)
+            if from_processor:
+                status = writer_pb2.FieldStatus()
+            else:
+                status = await field_obj.get_status() or writer_pb2.FieldStatus()
+
+            for error in errors:
+                field_error = writer_pb2.FieldError(
+                    source_error=error,
+                )
+                field_error.created.GetCurrentTime()
+                status.errors.append(field_error)
+
+            # We infer the status for processor messages
+            if message.source == BrokerMessage.MessageSource.PROCESSOR:
+                if len(errors) > 0:
+                    status.status = writer_pb2.FieldStatus.Status.ERROR
+                else:
+                    status.status = writer_pb2.FieldStatus.Status.PROCESSED
+            else:
+                field_status = next(
+                    (
+                        fs.status
+                        for fs in message.field_statuses
+                        if fs.id.field_type == field_type and fs.id.field == field
+                    ),
+                    None,
+                )
+                if field_status:
+                    status.status = field_status
+
+            await field_obj.set_status(status)
+
+    async def update_status(self):
+        field_ids = await self.get_all_field_ids(for_update=False)
+        field_statuses = await datamanagers.fields.get_statuses(
+            self.txn, kbid=self.kb.kbid, rid=self.uuid, fields=field_ids.fields
+        )
+        # If any field is processing -> PENDING
+        if any((f.status == writer_pb2.FieldStatus.Status.PENDING for f in field_statuses)):
+            self.basic.metadata.status = PBMetadata.Status.PENDING
+        # If we have any non-DA error -> ERROR
+        elif any(
+            (
+                f.status == writer_pb2.FieldStatus.Status.ERROR
+                and any(
+                    (
+                        e.source_error.code != writer_pb2.Error.ErrorCode.DATAAUGMENTATION
+                        for e in f.errors
+                    )
+                )
+                for f in field_statuses
+            )
+        ):
+            self.basic.metadata.status = PBMetadata.Status.ERROR
+        # Otherwise (everything processed or we only have DA errors) -> PROCESSED
+        else:
+            self.basic.metadata.status = PBMetadata.Status.PROCESSED
+
     @processor_observer.wrap({"type": "apply_extracted"})
     async def apply_extracted(self, message: BrokerMessage):
         errors = False
@@ -563,6 +644,10 @@ class Resource:
         for extracted_text in message.extracted_text:
             await self._apply_extracted_text(extracted_text)
 
+        # TODO: Update field and resource status depending on processing results
+        await self.apply_fields_status(message, self._modified_extracted_text)
+        # await self.update_status()
+
         extracted_languages = []
 
         for link_extracted_data in message.link_extracted_data:

nucliadb/search/search/graph_strategy.py

@@ -311,11 +311,12 @@ async def get_graph_results(
 ) -> tuple[KnowledgeboxFindResults, QueryParser]:
     relations = Relations(entities={})
     explored_entities: set[str] = set()
+    scores: dict[str, list[float]] = {}
     predict = get_predict()
 
     for hop in range(graph_strategy.hops):
         entities_to_explore: Iterable[RelationNode] = []
-        scores: dict[str, list[float]] = {}
+
         if hop == 0:
             # Get the entities from the query
             with metrics.time("graph_strat_query_entities"):
@@ -415,6 +416,7 @@ async def get_graph_results(
                 capture_exception(e)
                 logger.exception("Error in ranking relations for graph strategy")
                 relations = Relations(entities={})
+                scores = {}
                 break
 
     # Get the text blocks of the paragraphs that contain the top relations

nucliadb/writer/api/v1/field.py

@@ -55,7 +55,7 @@ from nucliadb_models.utils import FieldIdString
 from nucliadb_models.writer import ResourceFieldAdded, ResourceUpdated
 from nucliadb_protos import resources_pb2
 from nucliadb_protos.resources_pb2 import FieldID, Metadata
-from nucliadb_protos.writer_pb2 import BrokerMessage
+from nucliadb_protos.writer_pb2 import BrokerMessage, FieldIDStatus, FieldStatus
 from nucliadb_utils.authentication import requires
 from nucliadb_utils.exceptions import LimitsExceededError, SendToProcessError
 from nucliadb_utils.utilities import (
@@ -553,6 +553,12 @@ async def reprocess_file_field(
     writer.source = BrokerMessage.MessageSource.WRITER
     writer.basic.metadata.useful = True
     writer.basic.metadata.status = Metadata.Status.PENDING
+    writer.field_statuses.append(
+        FieldIDStatus(
+            id=FieldID(field_type=resources_pb2.FieldType.FILE, field=field_id),
+            status=FieldStatus.Status.PENDING,
+        )
+    )
     await transaction.commit(writer, partition, wait=False)
     # Send current resource to reprocess.
     try:

nucliadb/writer/api/v1/resource.py

@@ -63,8 +63,8 @@ from nucliadb_models.writer import (
     ResourceUpdated,
     UpdateResourcePayload,
 )
-from nucliadb_protos.resources_pb2 import Metadata
-from nucliadb_protos.writer_pb2 import BrokerMessage, IndexResource
+from nucliadb_protos.resources_pb2 import FieldID, Metadata
+from nucliadb_protos.writer_pb2 import BrokerMessage, FieldIDStatus, FieldStatus, IndexResource
 from nucliadb_telemetry.errors import capture_exception
 from nucliadb_utils.authentication import requires
 from nucliadb_utils.exceptions import LimitsExceededError, SendToProcessError
@@ -422,6 +422,7 @@ async def _reprocess_resource(
     storage = await get_storage(service_name=SERVICE_NAME)
     driver = get_driver()
 
+    writer = BrokerMessage()
     async with driver.transaction() as txn:
         kb = KnowledgeBox(txn, storage, kbid)
 
@@ -430,8 +431,14 @@ async def _reprocess_resource(
             raise HTTPException(status_code=404, detail="Resource does not exist")
 
         await extract_fields(resource=resource, toprocess=toprocess)
+        for field_type, field_id in resource.fields.keys():
+            writer.field_statuses.append(
+                FieldIDStatus(
+                    id=FieldID(field_type=field_type, field=field_id),
+                    status=FieldStatus.Status.PENDING,
+                )
+            )
 
-    writer = BrokerMessage()
     writer.kbid = kbid
     writer.uuid = rid
     writer.source = BrokerMessage.MessageSource.WRITER

nucliadb/writer/api/v1/upload.py

@@ -74,7 +74,7 @@ from nucliadb_utils.utilities import (
     get_storage,
 )
 
-from .router import KB_PREFIX, RESOURCE_PREFIX, RSLUG_PREFIX, api
+from .router import KB_PREFIX, RESOURCE_PREFIX, RESOURCES_PREFIX, RSLUG_PREFIX, api
 
 TUS_HEADERS = {
     "Tus-Resumable": "1.0.0",
@@ -544,8 +544,8 @@ async def _tus_patch(
             raise AttributeError()
         path = await storage_manager.finish(dm)
         headers["Tus-Upload-Finished"] = "1"
-        headers["NDB-Resource"] = f"/{KB_PREFIX}/{kbid}/resources/{rid}"
-        headers["NDB-Field"] = f"/{KB_PREFIX}/{kbid}/resources/{rid}/field/{field}"
+        headers["NDB-Resource"] = f"/{KB_PREFIX}/{kbid}/{RESOURCES_PREFIX}/{rid}"
+        headers["NDB-Field"] = f"/{KB_PREFIX}/{kbid}/{RESOURCES_PREFIX}/{rid}/field/{field}"
 
         item_payload = dm.get("item")
         creation_payload = None

nucliadb/writer/resource/field.py

@@ -37,7 +37,7 @@ from nucliadb_models.writer import (
     UpdateResourcePayload,
 )
 from nucliadb_protos import resources_pb2
-from nucliadb_protos.writer_pb2 import BrokerMessage
+from nucliadb_protos.writer_pb2 import BrokerMessage, FieldIDStatus, FieldStatus
 from nucliadb_utils.storages.storage import StorageField
 from nucliadb_utils.utilities import get_storage
 
@@ -186,6 +186,12 @@ def parse_text_field(
         format=getattr(models.PushTextFormat, text_field.format.value),
         extract_strategy=text_field.extract_strategy,
     )
+    writer.field_statuses.append(
+        FieldIDStatus(
+            id=resources_pb2.FieldID(field_type=resources_pb2.FieldType.TEXT, field=key),
+            status=FieldStatus.Status.PENDING,
+        )
+    )
 
 
 async def parse_file_field(
@@ -204,6 +210,13 @@ async def parse_file_field(
             key, file_field, writer, toprocess, kbid, uuid, skip_store=skip_store
         )
 
+    writer.field_statuses.append(
+        FieldIDStatus(
+            id=resources_pb2.FieldID(field_type=resources_pb2.FieldType.FILE, field=key),
+            status=FieldStatus.Status.PENDING,
+        )
+    )
+
 
 async def parse_internal_file_field(
     key: str,
@@ -310,6 +323,12 @@ def parse_link_field(
         xpath=link_field.xpath,
         extract_strategy=link_field.extract_strategy,
     )
+    writer.field_statuses.append(
+        FieldIDStatus(
+            id=resources_pb2.FieldID(field_type=resources_pb2.FieldType.LINK, field=key),
+            status=FieldStatus.Status.PENDING,
+        )
+    )
 
 
 async def parse_conversation_field(
@@ -385,3 +404,9 @@ async def parse_conversation_field(
 
     toprocess.conversationfield[key] = convs
     writer.conversations[key].CopyFrom(field_value)
+    writer.field_statuses.append(
+        FieldIDStatus(
+            id=resources_pb2.FieldID(field_type=resources_pb2.FieldType.CONVERSATION, field=key),
+            status=FieldStatus.Status.PENDING,
+        )
+    )

{nucliadb-6.2.1.post2842.dist-info → nucliadb-6.2.1.post2855.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: nucliadb
-Version: 6.2.1.post2842
+Version: 6.2.1.post2855
 Home-page: https://docs.nuclia.dev/docs/management/nucliadb/intro
 Author: NucliaDB Community
 Author-email: nucliadb@nuclia.com
@@ -22,10 +22,10 @@ Classifier: Programming Language :: Python :: 3.12
 Classifier: Programming Language :: Python :: 3 :: Only
 Requires-Python: >=3.9, <4
 Description-Content-Type: text/markdown
-Requires-Dist: nucliadb-telemetry[all]>=6.2.1.post2842
-Requires-Dist: nucliadb-utils[cache,fastapi,storages]>=6.2.1.post2842
-Requires-Dist: nucliadb-protos>=6.2.1.post2842
-Requires-Dist: nucliadb-models>=6.2.1.post2842
+Requires-Dist: nucliadb-telemetry[all]>=6.2.1.post2855
+Requires-Dist: nucliadb-utils[cache,fastapi,storages]>=6.2.1.post2855
+Requires-Dist: nucliadb-protos>=6.2.1.post2855
+Requires-Dist: nucliadb-models>=6.2.1.post2855
 Requires-Dist: nucliadb-admin-assets>=1.0.0.post1224
 Requires-Dist: nucliadb-node-binding>=2.26.0
 Requires-Dist: nuclia-models>=0.24.2

{nucliadb-6.2.1.post2842.dist-info → nucliadb-6.2.1.post2855.dist-info}/RECORD

@@ -24,6 +24,7 @@ migrations/0025_assign_models_to_kbs_v2.py,sha256=QC6nDF2Wyc6zQMqNoKzvz-3507UpDy
 migrations/0026_fix_high_cardinality_content_types.py,sha256=BsbBkvZDzjRHQfoouZNNtHA1xMxTKm8wOVnp_WAS9j4,2322
 migrations/0027_rollover_texts3.py,sha256=UQDaMOayVuqDisf82NDrPStoEVveHvdjkSmzbIcU9o4,2730
 migrations/0028_extracted_vectors_reference.py,sha256=49DHCIlBpjofU8cYVHTdWv0EBIlnPTWV2WCezf0rJUo,2392
+migrations/0029_backfill_field_status.py,sha256=QWF69n1da9lpRnbEpgbqPjSQ-Wfn6rMC7Enz6bBYGt4,5663
 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
@@ -70,7 +71,7 @@ nucliadb/common/datamanagers/atomic.py,sha256=DU7RihO8WaGNuh_GTEpQ-8hkoinY5GSpNS
 nucliadb/common/datamanagers/cluster.py,sha256=psTwAWSLj83vhFnC1iJJ6holrolAI4nKos9PuEWspYY,1500
 nucliadb/common/datamanagers/entities.py,sha256=hqw4YcEOumGK_1vgNNfxP-WafHvWN5jf61n4U01WJtc,5311
 nucliadb/common/datamanagers/exceptions.py,sha256=Atz_PP_GGq4jgJaWcAkcRbHBoBaGcC9yJvFteylKtTE,883
-nucliadb/common/datamanagers/fields.py,sha256=fgEP3SM_8asi3eSq2rMxJw2qgvNw0oZ_LoM5zjW27w4,2716
+nucliadb/common/datamanagers/fields.py,sha256=QqWVFqLp58Ib9fadXkSBuaYU-Mo6VJHZWg-taUqreNM,3915
 nucliadb/common/datamanagers/kb.py,sha256=P7EhF4tApIUG2jw_HH1oMufTKG9__kuOLKnrCNGbDM4,6156
 nucliadb/common/datamanagers/labels.py,sha256=Zm0GQpSPoGXEEysUY7VsDIcyKSIIQsMVphj23IyM9_c,4502
 nucliadb/common/datamanagers/processing.py,sha256=ByxdZzdbAfJGqC6__mY-zryjk040TyQfcUq3rxujeoY,1587
@@ -127,7 +128,7 @@ nucliadb/ingest/consumer/service.py,sha256=EZM1sABW_7bj6j2UgKUHUuK-EGIEYnLdtPAn8
 nucliadb/ingest/consumer/shard_creator.py,sha256=19wf-Bu_9hb_muCDVblamWuvLr09e5dMu9Id5I4-rGw,4324
 nucliadb/ingest/consumer/utils.py,sha256=jpX8D4lKzuPCpArQLZeX_Zczq3pfen_zAf8sPJfOEZU,2642
 nucliadb/ingest/fields/__init__.py,sha256=cp15ZcFnHvpcu_5-aK2A4uUyvuZVV_MJn4bIXMa20ks,835
-nucliadb/ingest/fields/base.py,sha256=EdWrNWG7-W5ehyPyJrNULLtoApT_AEsKCcfXwzxYMhU,19257
+nucliadb/ingest/fields/base.py,sha256=GuyZaumtaaAzoy86-mewBMhNX9DaSDBIK4pCqrxZlDc,19836
 nucliadb/ingest/fields/conversation.py,sha256=OcQOHvi72Pm0OyNGwxLo9gONo8f1NhwASq0_gS-E64A,7021
 nucliadb/ingest/fields/exceptions.py,sha256=LBZ-lw11f42Pk-ck-NSN9mSJ2kOw-NeRwb-UE31ILTQ,1171
 nucliadb/ingest/fields/file.py,sha256=1v4jLg3balUua2VmSV8hHkAwPFShTUCOzufZvIUQcQw,4740
@@ -141,7 +142,7 @@ nucliadb/ingest/orm/entities.py,sha256=2PslT1FZ6yCvJtjR0UpKTSzxJrtS-C_gZx4ZTWHun
 nucliadb/ingest/orm/exceptions.py,sha256=k4Esv4NtL4TrGTcsQpwrSfDhPQpiYcRbB1SpYmBX5MY,1432
 nucliadb/ingest/orm/knowledgebox.py,sha256=UpWJrVaVfCtk8R4qfSR6h6vzwOKXa8Teuwkna5QSljE,24508
 nucliadb/ingest/orm/metrics.py,sha256=OkwMSPKLZcKba0ZTwtTiIxwBgaLMX5ydhGieKvi2y7E,1096
-nucliadb/ingest/orm/resource.py,sha256=FIC7Tzef_0hoN92tFwN8PUi8IHlUkyifPJxsA1_XumM,55727
+nucliadb/ingest/orm/resource.py,sha256=jvaKLsTlHtmIWUjjWCu8XBF7qQl5hoUihAa8sHDpLV8,59540
 nucliadb/ingest/orm/utils.py,sha256=vCe_9UxHu26JDFGLwQ0wH-XyzJIpQCTK-Ow9dtZR5Vg,2716
 nucliadb/ingest/orm/processor/__init__.py,sha256=Aqd9wCNTvggkMkCY3WvoI8spdr94Jnqk-0iq9XpLs18,922
 nucliadb/ingest/orm/processor/auditing.py,sha256=TeYhXGJRyQ7ROytbb2u8R0fIh_FYi3HgTu3S1ribY3U,4623
@@ -219,7 +220,7 @@ nucliadb/search/search/fetch.py,sha256=XJHIFnZmXM_8Kb37lb4lg1GYG7cZ1plT-qAIb_Qzi
 nucliadb/search/search/filters.py,sha256=1MkHlJjAQqoRCj7e5cEzK2HvBxGLE17I_omsjiklbtw,6476
 nucliadb/search/search/find.py,sha256=yQbttt85wQFc4NEaj2RNGgozP7IQx_bjAOhHke3fXY0,9890
 nucliadb/search/search/find_merge.py,sha256=_R_YpHAZv5BHh3XABQ8MRd1Ci0seclGYf26yJHJ7H0I,17178
-nucliadb/search/search/graph_strategy.py,sha256=Yw6gPiIcKSj6HtbBcIT_5Poypc7AHkqKhcf8RRzJMnI,33513
+nucliadb/search/search/graph_strategy.py,sha256=6d-KjGDbOnaXQzEwyBpA-iQM0rkveVTiK3A3m2UJq8Q,33538
 nucliadb/search/search/hydrator.py,sha256=-R37gCrGxkyaiHQalnTWHNG_FCx11Zucd7qA1vQCxuw,6985
 nucliadb/search/search/merge.py,sha256=g0PxUejWtYIYWG-VroArMCgwB6AOp3lZMkzoHAPYBKE,22183
 nucliadb/search/search/metrics.py,sha256=81X-tahGW4n2CLvUzCPdNxNClmZqUWZjcVOGCUHoiUM,2872
@@ -314,19 +315,19 @@ nucliadb/writer/api/constants.py,sha256=qWEDjFUycrEZnSJyLnNK4PQNodU2oVmkO4NycaEZ
 nucliadb/writer/api/utils.py,sha256=wIQHlU8RQiIGVLI72suvyVIKlCU44Unh0Ae0IiN6Qwo,1313
 nucliadb/writer/api/v1/__init__.py,sha256=FVn7N9VJ6bsEoy4TRnkclr4Umd5hECiwPXVqRnJ8BME,1095
 nucliadb/writer/api/v1/export_import.py,sha256=6_gn0-emCjmK6bCUX5kgMvG0qkZr4HlfGmBXhhngsxo,8243
-nucliadb/writer/api/v1/field.py,sha256=gYa2SMYGSr9ORi_lrN9EQpek7nj4-PW1JEoJ_hJzMGc,17274
+nucliadb/writer/api/v1/field.py,sha256=OsWOYA0WQ6onE5Rkl20QIEdtrSi7Jgnu62fUt90Ziy8,17503
 nucliadb/writer/api/v1/knowledgebox.py,sha256=Mr1vJSWOtiraDdtoTqQ1V2rSirMdojL4wN0Q3cOiX4k,10929
 nucliadb/writer/api/v1/learning_config.py,sha256=GaYaagjBrVG9ZxrWQyVQfqGMQV3tAJjqJ5CStaKhktU,2058
-nucliadb/writer/api/v1/resource.py,sha256=bXaksvUoyYcHMrLmv3Ufrc_nA9hywuonfV6U6kL5kdE,18429
+nucliadb/writer/api/v1/resource.py,sha256=A8fAHlN5XFsg6XFYKhfWJS8czgNH6yXr-PsnUqz2WUE,18757
 nucliadb/writer/api/v1/router.py,sha256=RjuoWLpZer6Kl2BW_wznpNo6XL3BOpdTGqXZCn3QrrQ,1034
 nucliadb/writer/api/v1/services.py,sha256=U8OGxhA1tdt-wxw2uDAjFpwFXFEXSDTfBe1iV5nfmx8,9897
 nucliadb/writer/api/v1/slug.py,sha256=xlVBDBpRi9bNulpBHZwhyftVvulfE0zFm1XZIWl-AKY,2389
 nucliadb/writer/api/v1/transaction.py,sha256=d2Vbgnkk_-FLGSTt3vfldwiJIUf0XoyD0wP1jQNz_DY,2430
-nucliadb/writer/api/v1/upload.py,sha256=Lb5yTh8MfpyhFYT4UCn26NfaYohu44mK_em3v4NUp6Q,32869
+nucliadb/writer/api/v1/upload.py,sha256=yfVbIDHZEmFUDLi0Fenv9i-oj1m0JZPyTK7UPlq8Wws,32905
 nucliadb/writer/resource/__init__.py,sha256=cp15ZcFnHvpcu_5-aK2A4uUyvuZVV_MJn4bIXMa20ks,835
 nucliadb/writer/resource/audit.py,sha256=FvxMZPzrNHtd31HgpZEvxzwAkbxJTZRhPLqRYYJi3tA,1426
 nucliadb/writer/resource/basic.py,sha256=l9zD-Qiq4eUkHezMf0w1Ksx2izKYLYuNoMIlXcNxxpM,11163
-nucliadb/writer/resource/field.py,sha256=sh4BrewJrPKAajA7jx3dcnm3UB0L7FKbLueBWnpxq9I,14601
+nucliadb/writer/resource/field.py,sha256=HsOERELyAsb9e0dx2IkSQ9lk0SThALFRcDKCVBw8ifU,15478
 nucliadb/writer/resource/origin.py,sha256=pvhUDdU0mlWPUcpoQi4LDUJaRtfjzVVrA8XcGVI_N8k,2021
 nucliadb/writer/tus/__init__.py,sha256=huWpKnDnjsrKlBBJk30ta5vamlA-4x0TbPs_2Up8hyM,5443
 nucliadb/writer/tus/azure.py,sha256=XhWAlWTM0vmXcXtuEPYjjeEhuZjiZXZu8q9WsJ7omFE,4107
@@ -337,9 +338,9 @@ 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-6.2.1.post2842.dist-info/METADATA,sha256=CmYjCtJ5NGjNjopTPvSGiJNDrhzypADVZGncMsY834k,4689
-nucliadb-6.2.1.post2842.dist-info/WHEEL,sha256=In9FTNxeP60KnTkGw7wk6mJPYd_dQSjEZmXdBdMCI-8,91
-nucliadb-6.2.1.post2842.dist-info/entry_points.txt,sha256=XqGfgFDuY3zXQc8ewXM2TRVjTModIq851zOsgrmaXx4,1268
-nucliadb-6.2.1.post2842.dist-info/top_level.txt,sha256=hwYhTVnX7jkQ9gJCkVrbqEG1M4lT2F_iPQND1fCzF80,20
-nucliadb-6.2.1.post2842.dist-info/zip-safe,sha256=AbpHGcgLb-kRsJGnwFEktk7uzpZOCcBY74-YBdrKVGs,1
-nucliadb-6.2.1.post2842.dist-info/RECORD,,
+nucliadb-6.2.1.post2855.dist-info/METADATA,sha256=F8qPYGVhRM6117pUwuEC5Sk8k8EAjk5jcJ79wjawFxw,4689
+nucliadb-6.2.1.post2855.dist-info/WHEEL,sha256=In9FTNxeP60KnTkGw7wk6mJPYd_dQSjEZmXdBdMCI-8,91
+nucliadb-6.2.1.post2855.dist-info/entry_points.txt,sha256=XqGfgFDuY3zXQc8ewXM2TRVjTModIq851zOsgrmaXx4,1268
+nucliadb-6.2.1.post2855.dist-info/top_level.txt,sha256=hwYhTVnX7jkQ9gJCkVrbqEG1M4lT2F_iPQND1fCzF80,20
+nucliadb-6.2.1.post2855.dist-info/zip-safe,sha256=AbpHGcgLb-kRsJGnwFEktk7uzpZOCcBY74-YBdrKVGs,1
+nucliadb-6.2.1.post2855.dist-info/RECORD,,