linkml-store 0.2.5__tar.gz → 0.2.6__tar.gz

{linkml_store-0.2.5 → linkml_store-0.2.6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: linkml-store
-Version: 0.2.5
+Version: 0.2.6
 Summary: linkml-store
 License: MIT
 Author: Author 1
@@ -34,7 +34,6 @@ Requires-Dist: duckdb (>=0.10.1)
 Requires-Dist: duckdb-engine (>=0.11.2)
 Requires-Dist: fastapi ; extra == "fastapi"
 Requires-Dist: frictionless ; extra == "frictionless"
-Requires-Dist: gcsfs
 Requires-Dist: google-cloud-bigquery ; extra == "bigquery"
 Requires-Dist: h5py ; extra == "h5py"
 Requires-Dist: jinja2 (>=3.1.4,<4.0.0)
@@ -54,7 +53,7 @@ Requires-Dist: plotly ; extra == "analytics"
 Requires-Dist: py2neo ; extra == "neo4j"
 Requires-Dist: pyarrow ; extra == "pyarrow"
 Requires-Dist: pydantic (>=2.0.0,<3.0.0)
-Requires-Dist: pymongo ; extra == "mongodb"
+Requires-Dist: pymongo (>=4.11,<5.0) ; extra == "mongodb"
 Requires-Dist: pystow (>=0.5.4,<0.6.0)
 Requires-Dist: python-dotenv (>=1.0.1,<2.0.0)
 Requires-Dist: ruff (>=0.6.2) ; extra == "tests"

{linkml_store-0.2.5 → linkml_store-0.2.6}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "linkml-store"
-version = "0.2.5"
+version = "0.2.6"
 description = "linkml-store"
 authors = ["Author 1 <author@org.org>"]
 license = "MIT"
@@ -24,7 +24,7 @@ black = { version=">=24.0.0", optional = true }
 ruff = { version=">=0.6.2", optional = true }
 llm = { version="*", optional = true }
 tiktoken = { version="*", optional = true }
-pymongo = { version="*", optional = true }
+pymongo = "^4.11"
 neo4j = { version="*", optional = true }
 py2neo = { version="*", optional = true }
 networkx = { version="*", optional = true }

{linkml_store-0.2.5 → linkml_store-0.2.6}/src/linkml_store/api/client.py

@@ -15,6 +15,7 @@ logger = logging.getLogger(__name__)
 
 HANDLE_MAP = {
     "duckdb": "linkml_store.api.stores.duckdb.duckdb_database.DuckDBDatabase",
+    "sqlite": "linkml_store.api.stores.duckdb.duckdb_database.DuckDBDatabase",
     "solr": "linkml_store.api.stores.solr.solr_database.SolrDatabase",
     "mongodb": "linkml_store.api.stores.mongodb.mongodb_database.MongoDBDatabase",
     "chromadb": "linkml_store.api.stores.chromadb.chromadb_database.ChromaDBDatabase",
@@ -24,6 +25,8 @@ HANDLE_MAP = {
 
 SUFFIX_MAP = {
     "ddb": "duckdb:///{path}",
+    "duckdb": "duckdb:///{path}",
+    "db": "duckdb:///{path}",
 }
 
 
@@ -204,9 +207,10 @@ class Client:
         if ":" not in handle:
             if alias is None:
                 alias = handle
-            suffix = handle.split(".")[-1]
-            if suffix in SUFFIX_MAP:
-                handle = SUFFIX_MAP[suffix].format(path=handle)
+            if "." in handle:
+                suffix = handle.split(".")[-1]
+                if suffix in SUFFIX_MAP:
+                    handle = SUFFIX_MAP[suffix].format(path=handle)
         if ":" not in handle:
             scheme = handle
             handle = None

{linkml_store-0.2.5 → linkml_store-0.2.6}/src/linkml_store/api/collection.py

@@ -1,6 +1,7 @@
 """A structure for representing collections of similar objects."""
 
 import hashlib
+import json
 import logging
 from collections import defaultdict
 from pathlib import Path
@@ -210,8 +211,59 @@ class Collection(Generic[DatabaseType]):
         """
         raise NotImplementedError
 
+    def index (
+        self,
+        objs: Union[OBJECT, List[OBJECT]],
+        index_name: Optional[str] = None,
+        replace: bool = False,
+        unique: bool = False,
+        **kwargs,
+    ) -> None:
+        """
+        Index objects in the collection.
+
+        :param objs:
+        :param index_name:
+        :param replace: replace the index, or not
+        :param unique: boolean used to declare the index unique or not
+        :param kwargs:
+        :return:
+        """
+        raise NotImplementedError
+
+    def upsert(self,
+               objs: Union[OBJECT, List[OBJECT]],
+               filter_fields: List[str],
+               update_fields: Union[List[str], None] = None, **kwargs):
+        """
+        Add one or more objects to the collection.
+
+        >>> from linkml_store import Client
+        >>> client = Client()
+        >>> db = client.attach_database("mongodb", alias="test")
+        >>> collection = db.create_collection("Person")
+        >>> objs = [{"id": "P1", "name": "John", "age_in_years": 30}, {"id": "P2", "name": "Alice", "age_in_years": 25}]
+        >>> collection.upsert(objs)
+
+        :param objs:
+        :param filter_fields: List of field names to use as the filter for matching existing collections.
+        :param update_fields: List of field names to include in the update. If None, all fields are updated.
+        :param kwargs:
+
+        :return:
+        """
+        raise NotImplementedError
+
     def _pre_query_hook(self, query: Optional[Query] = None, **kwargs):
-        logger.info(f"Pre-query hook (state: {self._initialized}; Q= {query}")
+        """
+        Pre-query hook.
+
+        This is called before a query is executed. It is used to materialize derivations and indexes.
+        :param query:
+        :param kwargs:
+        :return:
+        """
+        logger.debug(f"Pre-query hook (state: {self._initialized}; Q= {query}")  # if logging.info, this is very noisy.
         if not self._initialized:
             self._materialize_derivations()
             self._initialized = True
@@ -536,7 +588,13 @@ class Collection(Generic[DatabaseType]):
         qr = ix_coll.find(where=where, limit=-1, **kwargs)
         index_col = ix.index_field
         # TODO: optimize this for large indexes
-        vector_pairs = [(row, np.array(row[index_col], dtype=float)) for row in qr.rows]
+        def row2array(row):
+            v = row[index_col]
+            if isinstance(v, str):
+                # sqlite stores arrays as strings
+                v = json.loads(v)
+            return np.array(v, dtype=float)
+        vector_pairs = [(row, row2array(row)) for row in qr.rows]
         results = ix.search(query, vector_pairs, limit=limit, mmr_relevance_factor=mmr_relevance_factor, **kwargs)
         for r in results:
             del r[1][index_col]

{linkml_store-0.2.5 → linkml_store-0.2.6}/src/linkml_store/api/database.py

@@ -276,14 +276,15 @@ class Database(ABC, Generic[CollectionType]):
 
         Examples:
 
-        >>> from linkml_store.api.client import Client
-        >>> client = Client()
-        >>> db = client.attach_database("duckdb", alias="test")
-        >>> collection = db.create_collection("Person", alias="persons")
-        >>> collection.alias
-        'persons'
-        >>> collection.target_class_name
-        'Person'
+            >>> from linkml_store.api.client import Client
+            >>> client = Client()
+            >>> db = client.attach_database("duckdb", alias="test")
+            >>> collection = db.create_collection("Person", alias="persons")
+            >>> collection.alias
+            'persons'
+
+            >>> collection.target_class_name
+            'Person'
 
         If alias is not provided, it defaults to the name of the type.
 
@@ -419,7 +420,7 @@ class Database(ABC, Generic[CollectionType]):
         >>> from linkml_store.api.client import Client
         >>> from linkml_store.api.queries import Query
         >>> client = Client()
-        >>> db = client.attach_database("duckdb", alias="test")
+        >>> db = client.attach_database("duckdb", alias="test", recreate_if_exists=True)
         >>> collection = db.create_collection("Person")
         >>> collection.insert([{"id": "P1", "name": "John"}, {"id": "P2", "name": "Alice"}])
         >>> query = Query(from_table="Person", where_clause={"name": "John"})
@@ -451,7 +452,7 @@ class Database(ABC, Generic[CollectionType]):
 
         >>> from linkml_store.api.client import Client
         >>> client = Client()
-        >>> db = client.attach_database("duckdb", alias="test")
+        >>> db = client.attach_database("duckdb", alias="test", recreate_if_exists=True)
         >>> collection = db.create_collection("Person", alias="persons")
         >>> collection.insert([{"id": "P1", "name": "John", "age_in_years": 25}])
         >>> schema_view = db.schema_view
@@ -721,7 +722,7 @@ class Database(ABC, Generic[CollectionType]):
 
         >>> from linkml_store.api.client import Client
         >>> client = Client()
-        >>> db = client.attach_database("duckdb", alias="test")
+        >>> db = client.attach_database("duckdb", alias="test", recreate_if_exists=True)
         >>> db.import_database("tests/input/iris.csv", Format.CSV, collection_name="iris")
         >>> db.list_collection_names()
         ['iris']
@@ -741,7 +742,9 @@ class Database(ABC, Generic[CollectionType]):
                 # import into a test instance
                 tmp_handle = source_format.value
                 client = self.parent
-                tmp_db = client.attach_database(tmp_handle, alias="tmp")
+                tmp_alias = "tmp"
+                client.drop_database(tmp_alias, missing_ok=True)
+                tmp_db = client.attach_database(tmp_handle, alias=tmp_alias, recreate_if_exists=True)
                 # TODO: check for infinite recursion
                 tmp_db.import_database(location, source_format=source_format)
                 obj = {}

{linkml_store-0.2.5 → linkml_store-0.2.6}/src/linkml_store/api/stores/duckdb/duckdb_collection.py

@@ -147,16 +147,22 @@ class DuckDBCollection(Collection):
         if self._table_created or self.metadata.is_prepopulated:
             logger.info(f"Already have table for: {cd.name}")
             return
-        query = Query(
-            from_table="information_schema.tables", where_clause={"table_type": "BASE TABLE", "table_name": self.alias}
-        )
-        qr = self.parent.query(query)
-        if qr.num_rows > 0:
+        if self.parent._table_exists(self.alias):
             logger.info(f"Table already exists for {cd.name}")
             self._table_created = True
             self._initialized = True
             self.metadata.is_prepopulated = True
             return
+        # query = Query(
+        #     from_table="information_schema.tables", where_clause={"table_type": "BASE TABLE", "table_name": self.alias}
+        # )
+        # qr = self.parent.query(query)
+        # if qr.num_rows > 0:
+        #     logger.info(f"Table already exists for {cd.name}")
+        #     self._table_created = True
+        #     self._initialized = True
+        #     self.metadata.is_prepopulated = True
+        #     return
         logger.info(f"Creating table for {cd.name}")
         t = self._sqla_table(cd)
         ct = CreateTable(t)

{linkml_store-0.2.5 → linkml_store-0.2.6}/src/linkml_store/api/stores/duckdb/duckdb_database.py

@@ -1,7 +1,7 @@
 import json
 import logging
 from pathlib import Path
-from typing import Optional, Union
+from typing import Optional, Union, List
 
 import pandas as pd
 import sqlalchemy
@@ -14,7 +14,7 @@ from linkml_store.api import Database
 from linkml_store.api.queries import Query, QueryResult
 from linkml_store.api.stores.duckdb.duckdb_collection import DuckDBCollection
 from linkml_store.utils.format_utils import Format
-from linkml_store.utils.sql_utils import introspect_schema, query_to_sql
+from linkml_store.utils.sql_utils import introspect_schema, query_to_sql, where_clause_to_sql
 
 TYPE_MAP = {
     "VARCHAR": "string",
@@ -62,7 +62,7 @@ class DuckDBDatabase(Database):
     def engine(self) -> sqlalchemy.Engine:
         if not self._engine:
             handle = self.handle
-            if not handle.startswith("duckdb://") and not handle.startswith(":"):
+            if not handle.startswith("duckdb://") and not handle.startswith(":") and "://" not in handle:
                 handle = f"duckdb:///{handle}"
             if ":memory:" not in handle:
                 # TODO: investigate this; duckdb appears to be prematurely caching
@@ -71,6 +71,10 @@ class DuckDBDatabase(Database):
                 self._engine = sqlalchemy.create_engine(handle)
         return self._engine
 
+    @property
+    def _is_sqlite(self) -> bool:
+        return self.handle and self.handle.startswith("sqlite:")
+
     def commit(self, **kwargs):
         with self.engine.connect() as conn:
             conn.commit()
@@ -89,34 +93,60 @@ class DuckDBDatabase(Database):
             if not missing_ok:
                 raise FileNotFoundError(f"Database file not found: {path}")
 
-    def query(self, query: Query, **kwargs) -> QueryResult:
+    def _table_exists(self, table: str) -> bool:
+        if self._is_sqlite:
+            if table == "sqlite_master":
+                return True
+            meta_query = Query(
+                from_table="sqlite_master",
+                where_clause={
+                    #"type": "table",
+                    "name": table,
+                }
+            )
+        else:
+            if table.startswith("information_schema"):
+                return True
+            meta_query = Query(
+                from_table="information_schema.tables",
+                where_clause={
+                    "table_type": "BASE TABLE",
+                    "table_name": table,
+                }
+            )
+
+        qr = self.query(meta_query)
+        if qr.num_rows == 0:
+            logger.debug(f"Table {self.alias} not created yet")
+            return False
+        return True
+
+    def _json_encoded_cols(self, table_name: str) -> Optional[List[str]]:
         json_encoded_cols = []
-        if query.from_table:
-            if not query.from_table.startswith("information_schema"):
-                meta_query = Query(
-                    from_table="information_schema.tables", where_clause={"table_name": query.from_table}
-                )
-                qr = self.query(meta_query)
-                if qr.num_rows == 0:
-                    logger.debug(f"Table {query.from_table} not created yet")
-                    return QueryResult(query=query, num_rows=0, rows=[])
-            if not query.from_table.startswith("information_schema"):
-                sv = self.schema_view
-            else:
-                sv = None
+        if table_name:
+            if table_name.startswith("information_schema") or table_name.startswith("sqlite"):
+                return []
+            sv = self.schema_view
             if sv:
                 cd = None
                 for c in self._collections.values():
-                    # if c.name == query.from_table or c.metadata.alias == query.from_table:
-                    if c.alias == query.from_table or c.target_class_name == query.from_table:
+                    if c.alias == table_name or c.target_class_name == table_name:
                         cd = c.class_definition()
                         break
                 if cd:
                     for att in sv.class_induced_slots(cd.name):
                         if att.inlined or att.inlined_as_list:
                             json_encoded_cols.append(att.name)
+        return json_encoded_cols
+
+    def query(self, query: Query, **kwargs) -> QueryResult:
+        if not self._table_exists(query.from_table):
+            return QueryResult(query=query, num_rows=0, rows=[])
+        json_encoded_cols = self._json_encoded_cols(query.from_table)
+
         with self.engine.connect() as conn:
             count_query_str = text(query_to_sql(query, count=True))
+            logger.debug(f"count_query_str: {count_query_str}")
             num_rows = list(conn.execute(count_query_str))[0][0]
             logger.debug(f"num_rows: {num_rows}")
             query_str = query_to_sql(query, **kwargs)  # include offset, limit
@@ -167,6 +197,9 @@ class DuckDBDatabase(Database):
         logger.info(f"Inducing schema view for {self.metadata.handle} // {self}")
         sb = SchemaBuilder()
         schema = sb.schema
+        logger.info(f"Checking if {self.metadata.handle} is sqlite: {self._is_sqlite}")
+        if self._is_sqlite:
+            return SchemaView(schema)
         query = Query(from_table="information_schema.tables", where_clause={"table_type": "BASE TABLE"})
         qr = self.query(query)
         logger.info(f"Found {qr.num_rows} information_schema.tables // {qr.rows}")

{linkml_store-0.2.5 → linkml_store-0.2.6}/src/linkml_store/api/stores/mongodb/mongodb_collection.py

@@ -41,6 +41,89 @@ class MongoDBCollection(Collection):
             del obj["_id"]
         self._post_insert_hook(objs)
 
+
+    def index(self,
+              objs: Union[OBJECT, List[OBJECT]],
+              index_name: Optional[str] = None,
+              replace: bool = False,
+              unique: bool = False,
+              **kwargs):
+        """
+        Create indexes on the collection.
+
+        :param objs: Field(s) to index.
+        :param index_name: Optional name for the index.
+        :param replace: If True, the index will be dropped and recreated.
+        :param unique: If True, creates a unique index (default: False).
+        """
+
+        if not isinstance(objs, list):
+            objs = [objs]
+
+        existing_indexes = self.mongo_collection.index_information()
+
+        for obj in objs:
+            field_exists = False
+            index_to_drop = None
+
+            # Extract existing index details
+            for index_name_existing, index_details in existing_indexes.items():
+                indexed_fields = [field[0] for field in index_details.get("key", [])]  # Extract field names
+
+                if obj in indexed_fields:  # If this field is already indexed
+                    field_exists = True
+                    index_to_drop = index_name_existing if replace else None
+
+            # Drop the index if replace=True and index_to_drop is valid
+            if index_to_drop:
+                self.mongo_collection.drop_index(index_to_drop)
+                logging.debug(f"Dropped existing index: {index_to_drop}")
+
+            # Create the new index only if it doesn't exist or was dropped
+            if not field_exists or replace:
+                self.mongo_collection.create_index(obj, name=index_name, unique=unique)
+                logging.debug(f"Created new index: {index_name} on field {obj}, unique={unique}")
+            else:
+                logging.debug(f"Index already exists for field {obj}, skipping creation.")
+
+    def upsert(self,
+               objs: Union[OBJECT, List[OBJECT]],
+               filter_fields: List[str],
+               update_fields: Optional[List[str]] = None,
+               **kwargs):
+        """
+        Upsert one or more documents into the MongoDB collection.
+
+        :param objs: The document(s) to insert or update.
+        :param filter_fields: List of field names to use as the filter for matching existing documents.
+        :param update_fields: List of field names to include in the update. If None, all fields are updated.
+        """
+        if not isinstance(objs, list):
+            objs = [objs]
+
+        for obj in objs:
+            # Ensure filter fields exist in the object
+            filter_criteria = {field: obj[field] for field in filter_fields if field in obj}
+            if not filter_criteria:
+                raise ValueError("At least one valid filter field must be present in each object.")
+
+            # Check if a document already exists
+            existing_doc = self.mongo_collection.find_one(filter_criteria)
+
+            if existing_doc:
+                # Update only changed fields
+                updates = {key: obj[key] for key in update_fields if key in obj and obj[key] != existing_doc.get(key)}
+
+                if updates:
+                    self.mongo_collection.update_one(filter_criteria, {"$set": updates})
+                    logging.debug(f"Updated existing document: {filter_criteria} with {updates}")
+                else:
+                    logging.debug(f"No changes detected for document: {filter_criteria}. Skipping update.")
+            else:
+                # Insert a new document
+                self.mongo_collection.insert_one(obj)
+                logging.debug(f"Inserted new document: {obj}")
+
     def query(self, query: Query, limit: Optional[int] = None, offset: Optional[int] = None, **kwargs) -> QueryResult:
         mongo_filter = self._build_mongo_filter(query.where_clause)
         limit = limit or query.limit

{linkml_store-0.2.5 → linkml_store-0.2.6}/src/linkml_store/api/stores/mongodb/mongodb_database.py

@@ -3,6 +3,7 @@
 import logging
 from pathlib import Path
 from typing import Optional, Union
+from urllib.parse import urlparse
 
 from pymongo import MongoClient
 from pymongo.database import Database as NativeDatabase
@@ -38,10 +39,13 @@ class MongoDBDatabase(Database):
     @property
     def _db_name(self) -> str:
         if self.handle:
-            db = self.handle.split("/")[-1]
+            parsed_url = urlparse(self.handle)
+            path_parts = parsed_url.path.lstrip("/").split("?")[0].split("/")
+            print(path_parts)
+            db_name = path_parts[0] if path_parts else "default"
         else:
-            db = "default"
-        return db
+            db_name = "default"
+        return db_name
 
     @property
     def native_client(self) -> MongoClient:

{linkml_store-0.2.5 → linkml_store-0.2.6}/src/linkml_store/cli.py

@@ -186,7 +186,7 @@ def cli(ctx, verbose: int, quiet: bool, stacktrace: bool, database, collection,
 
 
 @cli.command()
-@click.argument("files", type=click.Path(exists=True), nargs=-1)
+@click.argument("files", type=click.Path(), nargs=-1)
 @click.option("--replace/--no-replace", default=False, show_default=True, help="Replace existing objects")
 @click.option("--format", "-f", type=format_choice, help="Input format")
 @click.option("--object", "-i", multiple=True, help="Input object as YAML")

{linkml_store-0.2.5 → linkml_store-0.2.6}/src/linkml_store/utils/format_utils.py

@@ -3,6 +3,7 @@ import gzip
 import io
 import json
 import logging
+import re
 import sys
 import tarfile
 from enum import Enum
@@ -31,10 +32,13 @@ class Format(Enum):
     TSV = "tsv"
     CSV = "csv"
     XML = "xml"
+    OBO = "obo"
+    PKL = "pkl"
     PYTHON = "python"
     PARQUET = "parquet"
     FORMATTED = "formatted"
     TABLE = "table"
+    XLSX = "xlsx"
     SQLDUMP_DUCKDB = "duckdb"
     SQLDUMP_POSTGRES = "postgres"
     DUMP_MONGODB = "mongodb"
@@ -67,6 +71,9 @@ class Format(Enum):
     def is_dump_format(self):
         return self in [Format.SQLDUMP_DUCKDB, Format.SQLDUMP_POSTGRES, Format.DUMP_MONGODB]
 
+    def is_binary_format(self):
+        return self in [Format.PARQUET, Format.XLSX]
+
     def is_xsv(self):
         return self in [Format.TSV, Format.CSV]
 
@@ -95,6 +102,26 @@ def load_objects_from_url(
     return objs
 
 
+def clean_pandas_value(v):
+    """Clean a single value from pandas."""
+    import math
+
+    if isinstance(v, float):
+        if math.isnan(v) or math.isinf(v):
+            return None
+        return float(v)  # Ensures proper float type
+    return v
+
+
+def clean_nested_structure(obj):
+    """Recursively clean a nested structure of dicts/lists from pandas."""
+    if isinstance(obj, dict):
+        return {k: clean_nested_structure(v) for k, v in obj.items()}
+    elif isinstance(obj, list):
+        return [clean_nested_structure(item) for item in obj]  # Fixed: using 'item' instead of 'v'
+    else:
+        return clean_pandas_value(obj)
+
 def process_file(
     f: IO, format: Format, expected_type: Optional[Type] = None, header_comment_token: Optional[str] = None
 ) -> List[Dict[str, Any]]:
@@ -128,6 +155,19 @@ def process_file(
         objs = list(reader)
     elif format == Format.XML:
         objs = xmltodict.parse(f.read())
+    elif format == Format.PKL:
+        objs = pd.read_pickle(f).to_dict(orient="records")
+    elif format == Format.XLSX:
+        xls = pd.ExcelFile(f)
+        objs = {sheet: clean_nested_structure(xls.parse(sheet).to_dict(orient="records")) for sheet in xls.sheet_names}
+    elif format == Format.OBO:
+        blocks = split_document(f.read(), "\n\n")
+        id_pattern = re.compile(r"id: (\S+)")
+        def get_id(block):
+            m = id_pattern.search(block)
+            return m.group(1) if m else None
+        objs = [{"id": get_id(block), "content": block} for block in blocks]
+        objs = [obj for obj in objs if obj["id"]]
     elif format == Format.PARQUET:
         import pyarrow.parquet as pq
 
@@ -167,6 +207,14 @@ def load_objects(
     if isinstance(file_path, Path):
         file_path = str(file_path)
 
+    for url_scheme in ["http", "https", "ftp"]:
+        if file_path.startswith(f"{url_scheme}://"):
+            return load_objects_from_url(
+                file_path,
+                format=format,
+                expected_type=expected_type,
+            )
+
     if isinstance(format, str):
         format = Format(format)
 
@@ -185,9 +233,9 @@ def load_objects(
     else:
         if Path(file_path).is_dir():
             raise ValueError(f"{file_path} is a dir, which is invalid for {format}")
-        mode = "rb" if format == Format.PARQUET or compression == "gz" else "r"
         open_func = gzip.open if compression == "gz" else open
         format = Format.guess_format(file_path) if not format else format
+        mode = "rb" if (format and format.is_binary_format()) or compression == "gz" else "r"
         with open_func(file_path, mode) if file_path != "-" else sys.stdin as f:
             if compression == "gz" and mode == "r":
                 f = io.TextIOWrapper(f)
@@ -343,3 +391,14 @@ def guess_format(path: str) -> Optional[Format]:
     :return: The guessed format.
     """
     return Format.guess_format(path)
+
+
+def split_document(doc: str, delimiter: str):
+    """
+    Split a document into parts based on a delimiter.
+
+    :param doc: The document to split.
+    :param delimiter: The delimiter.
+    :return: The parts of the document.
+    """
+    return doc.split(delimiter)

{linkml_store-0.2.5 → linkml_store-0.2.6}/src/linkml_store/utils/sql_utils.py

@@ -5,7 +5,7 @@ import sqlalchemy
 import sqlalchemy.sql.sqltypes as sqlt
 from linkml_runtime.linkml_model import SchemaDefinition, SlotDefinition
 from linkml_runtime.utils.schema_builder import SchemaBuilder
-from sqlalchemy import MetaData
+from sqlalchemy import MetaData, quoted_name
 
 from linkml_store.api.queries import Query
 
@@ -115,7 +115,13 @@ def facet_count_sql(query: Query, facet_column: Union[str, Tuple[str, ...]], mul
         conditions = [cond for cond in where_clause_sql.split(" AND ") if not cond.startswith(f"{facet_column} ")]
         modified_where = " AND ".join(conditions)
 
+    def make_col_safe(col):
+        return '"' + quoted_name(col, True) + '"' if ' ' in col else col
+
+    if isinstance(facet_column, str):
+        facet_column = make_col_safe(facet_column)
     if isinstance(facet_column, tuple):
+        facet_column = [make_col_safe(col) for col in facet_column]
         if multivalued:
             raise NotImplementedError("Multivalued facets are not supported for multiple columns")
         facet_column = ", ".join(facet_column)

{linkml_store-0.2.5 → linkml_store-0.2.6}/src/linkml_store/utils/vector_utils.py

@@ -34,7 +34,7 @@ def pairwise_cosine_similarity(vector1: np.array, vector2: np.array) -> float:
     dot_product = np.dot(vector1, vector2)
     norm1 = np.linalg.norm(vector1)
     norm2 = np.linalg.norm(vector2)
-    return dot_product / (norm1 * norm2)
+    return float(dot_product / (norm1 * norm2))
 
 
 def compute_cosine_similarity_matrix(list1: LOL, list2: LOL) -> np.ndarray: