pyobvector 0.2.21__tar.gz → 0.2.23__tar.gz

{pyobvector-0.2.21 → pyobvector-0.2.23}/.github/workflows/ci.yml

@@ -36,10 +36,10 @@ jobs:
         run: uv sync --dev
 
       - name: Lint
-        run: uv run ruff check
+        run: make check
 
       - name: Package build test
-        run: uv build
+        run: make build
 
       - name: Free disk space
         uses: kfir4444/free-disk-space@main
@@ -64,4 +64,4 @@ jobs:
 
       - name: Run tests
         run: |
-          uv run pytest --log-cli-level=INFO ${{ matrix.test_filter }}
+          make test TEST_FILTER='${{ matrix.test_filter }}'

{pyobvector-0.2.21 → pyobvector-0.2.23}/.github/workflows/python-publish.yml

@@ -22,7 +22,7 @@ jobs:
 
     steps:
     - uses: actions/checkout@v6
-    
+
     - name: Set up Python
       uses: actions/setup-python@v6
       with:

pyobvector-0.2.23/.pre-commit-config.yaml

@@ -0,0 +1,26 @@
+repos:
+  - repo: https://github.com/pre-commit/pre-commit-hooks
+    rev: "v5.0.0"
+    hooks:
+      - id: check-case-conflict
+      - id: check-merge-conflict
+      - id: check-toml
+      - id: check-yaml
+      - id: check-json
+      - id: end-of-file-fixer
+      - id: trailing-whitespace
+
+  - repo: https://github.com/astral-sh/ruff-pre-commit
+    rev: "v0.12.7"
+    hooks:
+      - id: ruff-check
+        args: [ --exit-non-zero-on-fix ]
+      - id: ruff-format
+
+  - repo: local
+    hooks:
+      - id: compileall
+        name: compileall
+        entry: make compileall
+        language: system
+        pass_filenames: false

{pyobvector-0.2.21 → pyobvector-0.2.23}/LICENSE

@@ -199,4 +199,4 @@
    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.
+   limitations under the License.

pyobvector-0.2.23/Makefile

@@ -0,0 +1,45 @@
+# Minimal makefile for Sphinx documentation
+#
+
+# You can set these variables from the command line, and also
+# from the environment for the first two.
+SPHINXOPTS    ?=
+SPHINXBUILD   ?= uv run sphinx-build
+SOURCEDIR     = source
+BUILDDIR      = build
+
+.PHONY: install
+install: ## Install the virtual environment and install the pre-commit hooks
+	@echo "Creating virtual environment using uv"
+	@uv sync --dev
+	@uv run prek install
+
+.PHONY: check
+check: ## Run code quality tools.
+	@echo "Linting code: Running pre-commit via prek"
+	@uv run prek run -a
+
+.PHONY: test
+test: ## Test the code with pytest
+	@echo "Testing code: Running pytest"
+	@uv run python -m pytest $(TEST_FILTER)
+
+.PHONY: compileall
+compileall: ## Byte-compile Python sources.
+	@uv run python -m compileall pyobvector tests
+
+.PHONY: build
+build: ## Build wheel file
+	@echo "Creating wheel file"
+	@uv build
+
+# Put it first so that "make" without argument is like "make help".
+help:
+	@$(SPHINXBUILD) -M help "$(SOURCEDIR)" "$(BUILDDIR)" $(SPHINXOPTS) $(O)
+
+.PHONY: help install check test build compileall Makefile
+
+# Catch-all target: route all unknown targets to Sphinx using the new
+# "make mode" option.  $(O) is meant as a shortcut for $(SPHINXOPTS).
+%: Makefile
+	@$(SPHINXBUILD) -M $@ "$(SOURCEDIR)" "$(BUILDDIR)" $(SPHINXOPTS) $(O)

{pyobvector-0.2.21 → pyobvector-0.2.23}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: pyobvector
-Version: 0.2.21
+Version: 0.2.23
 Summary: A python SDK for OceanBase Vector Store, based on SQLAlchemy, compatible with Milvus API.
 Project-URL: Homepage, https://github.com/oceanbase/pyobvector
 Project-URL: Repository, https://github.com/oceanbase/pyobvector.git
@@ -34,7 +34,7 @@ uv sync
 - install with pip:
 
 ```shell
-pip install pyobvector==0.2.21
+pip install pyobvector==0.2.23
 ```
 
 ## Build Doc
@@ -154,8 +154,8 @@ client.create_table(test_collection_name, columns=cols, partitions=range_part)
 
 # create vector index
 client.create_index(
-    test_collection_name, 
-    is_vec_index=True, 
+    test_collection_name,
+    is_vec_index=True,
     index_name='vidx',
     column_names=['embedding'],
     vidx_params='distance=l2, type=hnsw, lib=vsag',
@@ -179,8 +179,8 @@ client.insert(test_collection_name, data=data1)
 ```python
 # perform ann search with basic column selection
 res = self.client.ann_search(
-    test_collection_name, 
-    vec_data=[0,0,0], 
+    test_collection_name,
+    vec_data=[0,0,0],
     vec_column_name='embedding',
     distance_func=l2_distance,
     topk=5,
@@ -194,8 +194,8 @@ from sqlalchemy import Table, text, func
 
 table = Table(test_collection_name, client.metadata_obj, autoload_with=client.engine)
 res = self.client.ann_search(
-    test_collection_name, 
-    vec_data=[0,0,0], 
+    test_collection_name,
+    vec_data=[0,0,0],
     vec_column_name='embedding',
     distance_func=l2_distance,
     topk=5,
@@ -211,8 +211,8 @@ res = self.client.ann_search(
 
 # perform ann search with distance threshold (filter results by distance)
 res = self.client.ann_search(
-    test_collection_name, 
-    vec_data=[0,0,0], 
+    test_collection_name,
+    vec_data=[0,0,0],
     vec_column_name='embedding',
     distance_func=l2_distance,
     with_dist=True,
@@ -230,17 +230,17 @@ res = self.client.ann_search(
 The `ann_search` method supports flexible output column selection through the `output_columns` parameter:
 
 - **`output_columns`** (recommended): Accepts SQLAlchemy Column objects, expressions, or a mix of both
-  
+
   - Column objects: `table.c.id`, `table.c.name`
   - Expressions: `(table.c.age + 10).label('age_plus_10')`
   - JSON queries: `text("JSON_EXTRACT(meta, '$.key') as extracted_key")`
   - String functions: `func.concat(table.c.name, ' (', table.c.age, ')').label('name_age')`
 - **`output_column_names`** (legacy): Accepts list of column name strings
-  
+
   - Example: `['id', 'name', 'meta']`
 - **Parameter Priority**: `output_columns` takes precedence over `output_column_names` when both are provided
 - **`distance_threshold`** (optional): Filter results by distance threshold
-  
+
   - Type: `Optional[float]`
   - Only returns results where `distance <= threshold`
   - Example: `distance_threshold=0.5` returns only results with distance <= 0.5
@@ -449,4 +449,3 @@ You can also get the actual SQL that will be executed:
 sql = client.get_sql(index=test_table_name, body=body)
 print(sql)  # prints the SQL query
 ```
-

{pyobvector-0.2.21 → pyobvector-0.2.23}/README.md

@@ -15,7 +15,7 @@ uv sync
 - install with pip:
 
 ```shell
-pip install pyobvector==0.2.21
+pip install pyobvector==0.2.23
 ```
 
 ## Build Doc
@@ -135,8 +135,8 @@ client.create_table(test_collection_name, columns=cols, partitions=range_part)
 
 # create vector index
 client.create_index(
-    test_collection_name, 
-    is_vec_index=True, 
+    test_collection_name,
+    is_vec_index=True,
     index_name='vidx',
     column_names=['embedding'],
     vidx_params='distance=l2, type=hnsw, lib=vsag',
@@ -160,8 +160,8 @@ client.insert(test_collection_name, data=data1)
 ```python
 # perform ann search with basic column selection
 res = self.client.ann_search(
-    test_collection_name, 
-    vec_data=[0,0,0], 
+    test_collection_name,
+    vec_data=[0,0,0],
     vec_column_name='embedding',
     distance_func=l2_distance,
     topk=5,
@@ -175,8 +175,8 @@ from sqlalchemy import Table, text, func
 
 table = Table(test_collection_name, client.metadata_obj, autoload_with=client.engine)
 res = self.client.ann_search(
-    test_collection_name, 
-    vec_data=[0,0,0], 
+    test_collection_name,
+    vec_data=[0,0,0],
     vec_column_name='embedding',
     distance_func=l2_distance,
     topk=5,
@@ -192,8 +192,8 @@ res = self.client.ann_search(
 
 # perform ann search with distance threshold (filter results by distance)
 res = self.client.ann_search(
-    test_collection_name, 
-    vec_data=[0,0,0], 
+    test_collection_name,
+    vec_data=[0,0,0],
     vec_column_name='embedding',
     distance_func=l2_distance,
     with_dist=True,
@@ -211,17 +211,17 @@ res = self.client.ann_search(
 The `ann_search` method supports flexible output column selection through the `output_columns` parameter:
 
 - **`output_columns`** (recommended): Accepts SQLAlchemy Column objects, expressions, or a mix of both
-  
+
   - Column objects: `table.c.id`, `table.c.name`
   - Expressions: `(table.c.age + 10).label('age_plus_10')`
   - JSON queries: `text("JSON_EXTRACT(meta, '$.key') as extracted_key")`
   - String functions: `func.concat(table.c.name, ' (', table.c.age, ')').label('name_age')`
 - **`output_column_names`** (legacy): Accepts list of column name strings
-  
+
   - Example: `['id', 'name', 'meta']`
 - **Parameter Priority**: `output_columns` takes precedence over `output_column_names` when both are provided
 - **`distance_threshold`** (optional): Filter results by distance threshold
-  
+
   - Type: `Optional[float]`
   - Only returns results where `distance <= threshold`
   - Example: `distance_threshold=0.5` returns only results with distance <= 0.5
@@ -430,4 +430,3 @@ You can also get the actual SQL that will be executed:
 sql = client.get_sql(index=test_table_name, body=body)
 print(sql)  # prints the SQL query
 ```
-

{pyobvector-0.2.21 → pyobvector-0.2.23}/RELEASE_NOTES.md

@@ -2,6 +2,16 @@
 
 This file documents all released versions and their notable changes for the pyobvector project. Changes are grouped by version and categorized as Added (new features), Changed (modifications), Fixed (bug fixes), and Security (security updates).
 
+## [0.2.23](https://github.com/oceanbase/pyobvector/compare/release-v0.2.22...release-v0.2.23) - 2026-01-29
+
+- Cchore: intro pre commit
+- Fix: make SeekDB version check case-insensitive
+
+## [0.2.22](https://github.com/oceanbase/pyobvector/compare/release-v0.2.21...release-v0.2.22) - 2026-01-15
+
+- Fix: HybridSearch.search() crashes when OceanBase returns NULL for empty results
+- Feat: add **kwargs support to create_table_with_index_params for heap organization
+
 ## [0.2.21](https://github.com/oceanbase/pyobvector/compare/release-v0.2.20...release-v0.2.21) - 2026-01-13
 
 - Migrate tool.poetry section to project section following PEP 518
@@ -352,4 +362,3 @@ This file documents all released versions and their notable changes for the pyob
 ### Fixed
 
 - Fix upsert JSON column: no literal value renderer
-

{pyobvector-0.2.21 → pyobvector-0.2.23}/pyobvector/__init__.py

@@ -1,10 +1,10 @@
 """A python SDK for OceanBase Vector Store, based on SQLAlchemy, compatible with Milvus API.
 
-`pyobvector` supports two modes: 
-1. `Milvus compatible mode`: You can use the `MilvusLikeClient` class to use vector storage 
+`pyobvector` supports two modes:
+1. `Milvus compatible mode`: You can use the `MilvusLikeClient` class to use vector storage
 in a way similar to the Milvus API.
-2. `SQLAlchemy hybrid mode`: You can use the vector storage function provided by the 
-`ObVecClient` class and execute the relational database statement with the SQLAlchemy library. 
+2. `SQLAlchemy hybrid mode`: You can use the vector storage function provided by the
+`ObVecClient` class and execute the relational database statement with the SQLAlchemy library.
 In this mode, you can regard `pyobvector` as an extension of SQLAlchemy.
 
 * ObVecClient           MySQL client in SQLAlchemy hybrid mode
@@ -19,7 +19,7 @@ In this mode, you can regard `pyobvector` as an extension of SQLAlchemy.
 * FtsIndex              Full Text Search Index
 * FieldSchema           Clas to define field schema in collection for MilvusLikeClient
 * CollectionSchema      Class to define collection schema for MilvusLikeClient
-* PartType              Specify partition type of table or collection 
+* PartType              Specify partition type of table or collection
                         for both ObVecClient and MilvusLikeClient
 * ObPartition           Abstract type class of all kind of Partition strategy
 * RangeListPartInfo     Specify Range/RangeColumns/List/ListColumns partition info
@@ -40,6 +40,7 @@ In this mode, you can regard `pyobvector` as an extension of SQLAlchemy.
 * FtsIndexParam         Full Text Search index parameter
 * MatchAgainst          Full Text Search clause
 """
+
 from .client import *
 from .schema import (
     ARRAY,

{pyobvector-0.2.21 → pyobvector-0.2.23}/pyobvector/client/__init__.py

@@ -1,9 +1,9 @@
 """Multi-type Vector Store Client:
 
-1. `Milvus compatible mode`: You can use the `MilvusLikeClient` class to use vector storage 
+1. `Milvus compatible mode`: You can use the `MilvusLikeClient` class to use vector storage
 in a way similar to the Milvus API.
-2. `SQLAlchemy hybrid mode`: You can use the vector storage function provided by the 
-`ObVecClient` class and execute the relational database statement with the SQLAlchemy library. 
+2. `SQLAlchemy hybrid mode`: You can use the vector storage function provided by the
+`ObVecClient` class and execute the relational database statement with the SQLAlchemy library.
 In this mode, you can regard `pyobvector` as an extension of SQLAlchemy.
 
 * ObVecClient           MySQL client in SQLAlchemy hybrid mode
@@ -14,7 +14,7 @@ In this mode, you can regard `pyobvector` as an extension of SQLAlchemy.
 * DataType              Specify field type in collection schema for MilvusLikeClient
 * FieldSchema           Clas to define field schema in collection for MilvusLikeClient
 * CollectionSchema      Class to define collection schema for MilvusLikeClient
-* PartType              Specify partition type of table or collection 
+* PartType              Specify partition type of table or collection
                         for both ObVecClient and MilvusLikeClient
 * ObPartition           Abstract type class of all kind of Partition strategy
 * RangeListPartInfo     Specify Range/RangeColumns/List/ListColumns partition info
@@ -30,6 +30,7 @@ In this mode, you can regard `pyobvector` as an extension of SQLAlchemy.
 * FtsParser             Text Parser Type for Full Text Search
 * FtsIndexParam         Full Text Search index parameter
 """
+
 from .ob_vec_client import ObVecClient
 from .milvus_like_client import MilvusLikeClient
 from .ob_vec_json_table_client import ObVecJsonTableClient

{pyobvector-0.2.21 → pyobvector-0.2.23}/pyobvector/client/collection_schema.py

@@ -1,4 +1,5 @@
 """FieldSchema & CollectionSchema definition module to be compatible with Milvus."""
+
 import copy
 from typing import Optional
 from sqlalchemy import Column
@@ -6,6 +7,7 @@ from .schema_type import DataType, convert_datatype_to_sqltype
 from .exceptions import *
 from .partitions import *
 
+
 class FieldSchema:
     """FieldSchema definition.
 
@@ -18,6 +20,7 @@ class FieldSchema:
     nullable (bool) : whether the field can be null
     type_params (dict) : different parameters for different data type
     """
+
     def __init__(
         self,
         name: str,
@@ -117,12 +120,13 @@ class FieldSchema:
 
 class CollectionSchema:
     """CollectionSchema definition.
-    
+
     Attributes:
     fields (List[FieldSchema]) : a list of FieldSchema
     description (string) : collection description (not used in OceanBase)
     partitions (ObPartition) : partition strategy of this collection
     """
+
     def __init__(
         self,
         fields: Optional[list[FieldSchema]] = None,

{pyobvector-0.2.21 → pyobvector-0.2.23}/pyobvector/client/enum.py

@@ -1,7 +1,7 @@
 """Common module for int type enumerate."""
+
 from enum import Enum
 
 
 class IntEnum(int, Enum):
     """Int type enumerate definition."""
-    

{pyobvector-0.2.21 → pyobvector-0.2.23}/pyobvector/client/exceptions.py

@@ -1,9 +1,11 @@
 """Exception for MilvusLikeClient."""
+
 from .enum import IntEnum
 
 
 class ErrorCode(IntEnum):
     """Error codes for MilvusLikeClient."""
+
     SUCCESS = 0
     UNEXPECTED_ERROR = 1
     INVALID_ARGUMENT = 2
@@ -14,6 +16,7 @@ class ErrorCode(IntEnum):
 
 class ObException(Exception):
     """Base class for MilvusLikeClient exception."""
+
     def __init__(
         self,
         code: int = ErrorCode.UNEXPECTED_ERROR,
@@ -75,6 +78,7 @@ class ClusterVersionException(ObException):
 
 class ExceptionsMessage:
     """Exception Messages definition."""
+
     PartitionExprNotExists = "Partition expression string does not exist."
     PartitionMultiField = "Multi-Partition Field is not supported."
     PartitionLevelMoreThanTwo = "Partition Level should less than or equal to 2."
@@ -93,12 +97,8 @@ class ExceptionsMessage:
     PartitionListColNameListMissing = (
         "Column name list is necessary when partition type is ListColumns"
     )
-    PartitionHashNameListAndPartCntMissing = (
-        "One of hash_part_name_list and part_count must be set when partition type is Hash"
-    )
-    PartitionKeyNameListAndPartCntMissing = (
-        "One of key_part_name_list and part_count must be set when partition type is Key"
-    )
+    PartitionHashNameListAndPartCntMissing = "One of hash_part_name_list and part_count must be set when partition type is Hash"
+    PartitionKeyNameListAndPartCntMissing = "One of key_part_name_list and part_count must be set when partition type is Key"
     PrimaryFieldType = "Param primary_field must be int or str type."
     VectorFieldMissingDimParam = "Param 'dim' must be set for vector field."
     VarcharFieldMissingLengthParam = "Param 'max_length' must be set for varchar field."
@@ -108,7 +108,9 @@ class ExceptionsMessage:
     )
     CollectionNotExists = "Collection does not exist."
     MetricTypeParamTypeInvalid = "MetricType param type should be string."
-    MetricTypeValueInvalid = "MetricType should be 'l2'/'ip'/'neg_ip'/'cosine' in ann search."
+    MetricTypeValueInvalid = (
+        "MetricType should be 'l2'/'ip'/'neg_ip'/'cosine' in ann search."
+    )
     UsingInIDsWhenMultiPrimaryKey = "Using 'ids' when table has multi primary key."
     ClusterVersionIsLow = (
         "OceanBase %s feature is not supported because cluster version is below %s."

{pyobvector-0.2.21 → pyobvector-0.2.23}/pyobvector/client/fts_index_param.py

@@ -1,9 +1,12 @@
 """A module to specify fts index parameters"""
+
 from enum import Enum
 from typing import Optional, Union
 
+
 class FtsParser(Enum):
     """Built-in full-text search parser types supported by OceanBase"""
+
     IK = 0
     NGRAM = 1
     NGRAM2 = 2  # NGRAM2 parser (supported from V4.3.5 BP2+)
@@ -13,13 +16,14 @@ class FtsParser(Enum):
 
 class FtsIndexParam:
     """Full-text search index parameter.
-    
+
     Args:
         index_name: Index name
         field_names: List of field names to create full-text index on
         parser_type: Parser type, can be FtsParser enum or string (for custom parsers)
                     If None, uses default Space parser
     """
+
     def __init__(
         self,
         index_name: str,
@@ -34,11 +38,11 @@ class FtsIndexParam:
         """Convert parser type to string format for SQL."""
         if self.parser_type is None:
             return None  # Default Space parser, no need to specify
-        
+
         if isinstance(self.parser_type, str):
             # Custom parser name (e.g., "thai_ftparser")
             return self.parser_type.lower()
-        
+
         if isinstance(self.parser_type, FtsParser):
             if self.parser_type == FtsParser.IK:
                 return "ik"
@@ -52,7 +56,7 @@ class FtsIndexParam:
                 return "jieba"
             # Raise exception for unrecognized FtsParser enum values
             raise ValueError(f"Unrecognized FtsParser enum value: {self.parser_type}")
-        
+
         return None
 
     def __iter__(self):

{pyobvector-0.2.21 → pyobvector-0.2.23}/pyobvector/client/hybrid_search.py

@@ -1,4 +1,5 @@
 """OceanBase Hybrid Search Client."""
+
 import json
 import logging
 from typing import Any
@@ -27,7 +28,7 @@ class HybridSearch(Client):
         super().__init__(uri, user, password, db_name, **kwargs)
 
         min_required_version = ObVersion.from_db_version_nums(4, 4, 1, 0)
-        
+
         if self.ob_version < min_required_version:
             # For versions < 4.4.1.0, check if it's SeekDB
             if self._is_seekdb():
@@ -35,7 +36,8 @@ class HybridSearch(Client):
                 return
             raise ClusterVersionException(
                 code=ErrorCode.NOT_SUPPORTED,
-                message=ExceptionsMessage.ClusterVersionIsLow % ("Hybrid Search", "4.4.1.0"),
+                message=ExceptionsMessage.ClusterVersionIsLow
+                % ("Hybrid Search", "4.4.1.0"),
             )
 
     def search(
@@ -60,7 +62,11 @@ class HybridSearch(Client):
 
         with self.engine.connect() as conn:
             with conn.begin():
-                res = conn.execute(sql, {"index": index, "body_str": body_str}).fetchone()
+                res = conn.execute(
+                    sql, {"index": index, "body_str": body_str}
+                ).fetchone()
+                if res[0] is None:
+                    return []
                 return json.loads(res[0])
 
     def get_sql(
@@ -83,5 +89,9 @@ class HybridSearch(Client):
 
         with self.engine.connect() as conn:
             with conn.begin():
-                res = conn.execute(sql, {"index": index, "body_str": body_str}).fetchone()
+                res = conn.execute(
+                    sql, {"index": index, "body_str": body_str}
+                ).fetchone()
+                if res[0] is None:
+                    return ""
                 return res[0]