Merge branch 'main' into dev

Author: eric-epsilla

examples/hello_epsilla_enterprise.py

@@ -15,11 +15,16 @@
 
 db_name = "helloepsilla"
 db_id = "helloepsilla-1234567890"
-table_name = "HelloEpsilla"
-
 
 # Create a new db
-status_code, response = client.create_db(db_name, db_id)
+status_code, response = client.create_db(
+    db_name,
+    db_id,
+    min_replicas = 2,
+    max_replicas = 2,
+    sharding_init_number = 3,
+    sharding_capacity = 1000000,
+)
 print(status_code, response)
 
 # Get info of db

pyepsilla/cloud/client.py

@@ -104,7 +104,7 @@ def insert(self, table_name: str, records: list[dict]):
 
 
     ## query data from table
-    def query(self, table_name: str, query_field: str = None, query_vector: list = None, response_fields: Optional[list] = None, limit: int = 2, filter: Optional[str] = None, with_distance: Optional[bool] = False):
+    def query(self, table_name: str, query_field: str = None, query_vector: Union[list,dict] = None, response_fields: Optional[list] = None, limit: int = 2, filter: Optional[str] = None, with_distance: Optional[bool] = False):
         req_url = "{}/data/query".format(self._baseurl)
         req_data = { "table": table_name }
         if query_field != None:

pyepsilla/enterprise/client.py

@@ -77,9 +77,25 @@ def create_db(
         db_name: str = Field(pattern=r"^[a-zA-Z-0-9]{4,32}$", strict=True),
         db_id: Optional[str] = None,
         project_id: Optional[str] = "default",
+        min_replicas: Optional[int] = 0,
+        max_replicas: Optional[int] = 1,
+        sharding_init_number: Optional[int] = 1,
+        sharding_increase_step: Optional[int] = 2,
+        sharding_capacity: Optional[int] = 150000,
+        sharding_increase_threshold: Optional[float] = 0.9,
     ):
         req_url = "{}/vectordb/create".format(self._baseurl)
-        req_data = {"db_name": db_name, "db_uuid": db_id, "project_id": project_id}
+        req_data = {
+            "db_name": db_name,
+            "db_uuid": db_id,
+            "project_id": project_id,
+            "min_replicas": min_replicas,
+            "max_replicas": max_replicas,
+            "sharding_init_number": sharding_init_number,
+            "sharding_increase_step": sharding_increase_step,
+            "sharding_capacity": sharding_capacity,
+            "sharding_increase_threshold": sharding_increase_threshold,
+        }
         resp = requests.post(
             url=req_url,
             data=json.dumps(req_data),
@@ -205,7 +221,7 @@ def query(
         self,
         table_name: str,
         query_field: str = None,
-        query_vector: list = None,
+        query_vector: Union[list,dict] = None,
         response_fields: Optional[list] = None,
         limit: int = 2,
         filter: Optional[str] = None,

pyepsilla/vectordb/client.py

@@ -152,7 +152,7 @@ def rebuild(self, timeout: int = 7200):
     def query(
         self, table_name: str,
         query_field: str = "",
-        query_vector: list = None,
+        query_vector: Union[list,dict] = None,
         response_fields: list = None,
         limit: int = 1,
         filter: str = "",

pyepsilla/vectordb/field.py

@@ -27,6 +27,9 @@ class FieldType(IntEnum):
     DOUBLE = 11
     STRING = 20
     BOOL = 30
+    JSON = 31
     VECTOR_FLOAT = 40
     VECTOR_DOUBLE = 41
+    SPARSE_VECTOR_FLOAT = 50
+    SPARSE_VECTOR_DOUBLE = 51
     UNKNOWN = 999

pyepsilla/vectordb/version.py

@@ -1 +1 @@
-__version__ = "0.1.25"
+__version__ = "0.2.1"