leaves12138 commented on code in PR #10476: URL: https://github.com/apache/paimon/pull/10476#discussion_r4231176134
########## paimon-python/pypaimon/table/source/native_vector_search.py: ########## @@ -0,0 +1,93 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor license agreements. See the NOTICE file +# distributed with this work for additional information +# regarding copyright ownership. The ASF licenses this file +# to you under the Apache License, Version 2.0 (the +# "License"); you may not use this file except in compliance +# with the License. You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, +# software distributed under the License is distributed on an +# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +# KIND, either express or implied. See the License for the +# specific language governing permissions and limitations +# under the License. + +"""Local vector search through the Rust core Scan -> Plan -> Read path.""" + +import logging + +from pypaimon.catalog.table_query_auth import reject_search_under_query_auth +from pypaimon.globalindex.vector_search_result import DictBasedScoredIndexResult +from pypaimon.read.native_plan import ( + _catalog_metastore, _native_table, _partition_fields, _predicate_to_native, + _raise_if_native_fork_safety_error, native_method_available) +from pypaimon.read.split_serializer import deserialize_split_v1 +from pypaimon.table.source.primary_key_scored_result import ( + PrimaryKeyScoredResult, PrimaryKeySearchPosition) + +logger = logging.getLogger(__name__) + + +def try_native_vector_search(builder, queries, batch=False): + """Return existing Python result types, or None to use the Python reader. + + Search, scalar filtering, refinement and physical-position selection all + belong to Rust core. Python only converts the scored result metadata. + """ + table = builder._table + reject_search_under_query_auth(table) + try: + native_enabled = getattr(table.options, 'native_read_enabled', None) + if not callable(native_enabled) or not native_enabled(): + return None + loader = getattr(getattr(table, 'catalog_environment', None), 'catalog_loader', None) + if loader is None or _catalog_metastore(loader) != 'rest': + return None + context_fn = getattr(loader, 'context', None) + if not callable(context_fn): + return None + context = context_fn() + if any(getattr(context, attr, None) is not None for attr in ( + 'hadoop_conf', 'prefer_io_loader', 'fallback_io_loader')): + return None + # Python's batch reader currently supports only global row-ID results. + if batch and builder._is_primary_key_vector_search(): + return None + method = 'new_batch_vector_search_builder' if batch else 'new_vector_search_builder' + if not native_method_available('Table', method): + return None + native = getattr(_native_table(table), method)() + if builder._vector_column is not None: + native.with_vector_column(builder._vector_column.name) + native.with_limit(builder._limit).with_options(builder._options) + if builder._filter is not None: + native.with_filter(_predicate_to_native(builder._filter)) Review Comment: [P2] Preserve global-index.filter.refine-from-data semantics during delegation With an inexact scalar-index answer, Java and this PR's classic reader exclude candidates when `global-index.filter.refine-from-data=false`. The delegated core path instead reads the data to build an exact allow-list, ignoring that switch and changing the existing public API's result and I/O behavior just by enabling Native reading. On a real REST table with a BTREE index on `name`, an IVF-flat vector index, `name LIKE '%zeta%'`, and limit 1, classic returns 0 hits with refinement disabled whereas the new Native path returns 1 (batch: `[0, 0]` versus `[1, 1]`). All FAST/FULL/DETAIL cases reproduce it; enabled-refinement controls agree. Please require the companion core fix, or decline Native delegation for these unsupported refinement semantics. ########## paimon-python/pypaimon/table/source/native_vector_search.py: ########## @@ -0,0 +1,93 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor license agreements. See the NOTICE file +# distributed with this work for additional information +# regarding copyright ownership. The ASF licenses this file +# to you under the Apache License, Version 2.0 (the +# "License"); you may not use this file except in compliance +# with the License. You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, +# software distributed under the License is distributed on an +# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +# KIND, either express or implied. See the License for the +# specific language governing permissions and limitations +# under the License. + +"""Local vector search through the Rust core Scan -> Plan -> Read path.""" + +import logging + +from pypaimon.catalog.table_query_auth import reject_search_under_query_auth +from pypaimon.globalindex.vector_search_result import DictBasedScoredIndexResult +from pypaimon.read.native_plan import ( + _catalog_metastore, _native_table, _partition_fields, _predicate_to_native, + _raise_if_native_fork_safety_error, native_method_available) +from pypaimon.read.split_serializer import deserialize_split_v1 +from pypaimon.table.source.primary_key_scored_result import ( + PrimaryKeyScoredResult, PrimaryKeySearchPosition) + +logger = logging.getLogger(__name__) + + +def try_native_vector_search(builder, queries, batch=False): + """Return existing Python result types, or None to use the Python reader. + + Search, scalar filtering, refinement and physical-position selection all + belong to Rust core. Python only converts the scored result metadata. + """ + table = builder._table + reject_search_under_query_auth(table) + try: + native_enabled = getattr(table.options, 'native_read_enabled', None) + if not callable(native_enabled) or not native_enabled(): + return None + loader = getattr(getattr(table, 'catalog_environment', None), 'catalog_loader', None) + if loader is None or _catalog_metastore(loader) != 'rest': + return None + context_fn = getattr(loader, 'context', None) + if not callable(context_fn): + return None + context = context_fn() + if any(getattr(context, attr, None) is not None for attr in ( + 'hadoop_conf', 'prefer_io_loader', 'fallback_io_loader')): + return None + # Python's batch reader currently supports only global row-ID results. + if batch and builder._is_primary_key_vector_search(): + return None + method = 'new_batch_vector_search_builder' if batch else 'new_vector_search_builder' + if not native_method_available('Table', method): + return None + native = getattr(_native_table(table), method)() + if builder._vector_column is not None: + native.with_vector_column(builder._vector_column.name) + native.with_limit(builder._limit).with_options(builder._options) + if builder._filter is not None: + native.with_filter(_predicate_to_native(builder._filter)) + if builder._partition_filter is not None: + native.with_partition_filter(_predicate_to_native(builder._partition_filter)) + if batch: + native.with_query_vectors(queries) + results = native.execute_batch_local() + return [DictBasedScoredIndexResult(result.row_ids()) for result in results] + native.with_query_vector(queries) + result = native.execute_local() Review Comment: [P1] Do not enable Native DE filtering before matching Java's raw fallback For a scalar predicate that cannot be evaluated by a scalar index, the classic reader added in this PR correctly follows Java `AbstractDataEvolutionVectorRead.prepareSplits`: it reroutes vector-index-covered ranges to raw scoring. This new delegation bypasses that policy and accepts an empty Native ANN result as a successful result, so the exception fallback does not help. Using a real REST IVF-flat index with four clusters, `ivf.nprobe=1`, query `[0, 0]`, limit 3, and `id >= 100` without an `id` scalar index, the classic path returns 3 hits while the new public Native path returns 0. This reproduces for single/batch and FAST/FULL/DETAIL; probing all four clusters recovers the three Native hits. The core issue is detailed on apache/paimon-rust#1087. Please land the Java-equivalent core routing before enabling these filtered searches, or keep unsupported cases on the classic lane until then. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
