diff --git a/datacontract/data_contract.py b/datacontract/data_contract.py
index 3ef90cf0..806810db 100644
--- a/datacontract/data_contract.py
+++ b/datacontract/data_contract.py
@@ -4,6 +4,7 @@
 import typing
 
 import yaml
+from pyspark.sql import SparkSession
 
 from datacontract.breaking.breaking import models_breaking_changes, \
     quality_breaking_changes
@@ -53,25 +54,6 @@
 from datacontract.model.run import Run, Check
 
 
-def _determine_sql_server_type(data_contract, sql_server_type):
-    if sql_server_type == "auto":
-        if data_contract.servers is None or len(data_contract.servers) == 0:
-            raise RuntimeError("Export with server_type='auto' requires servers in the data contract.")
-
-        server_types = set([server.type for server in data_contract.servers.values()])
-        if "snowflake" in server_types:
-            return "snowflake"
-        elif "postgres" in server_types:
-            return "postgres"
-        elif "databricks" in server_types:
-            return "databricks"
-        else:
-            # default to snowflake dialect
-            return "snowflake"
-    else:
-        return sql_server_type
-
-
 class DataContract:
     def __init__(
         self,
@@ -83,7 +65,7 @@ def __init__(
         examples: bool = False,
         publish_url: str = None,
         publish_to_opentelemetry: bool = False,
-        spark: str = None,
+        spark: SparkSession = None,
         inline_definitions: bool = False,
     ):
         self._data_contract_file = data_contract_file
@@ -385,13 +367,13 @@ def export(self, export_format, model: str = "all", rdf_base: str = None, sql_se
         if export_format == "terraform":
             return to_terraform(data_contract)
         if export_format == "sql":
-            server_type = _determine_sql_server_type(data_contract, sql_server_type)
+            server_type = self._determine_sql_server_type(data_contract, sql_server_type)
             return to_sql_ddl(data_contract, server_type=server_type)
         if export_format == "sql-query":
             if data_contract.models is None:
                 raise RuntimeError(f"Export to {export_format} requires models in the data contract.")
 
-            server_type = _determine_sql_server_type(data_contract, sql_server_type)
+            server_type = self._determine_sql_server_type(data_contract, sql_server_type)
 
             model_names = list(data_contract.models.keys())
 
@@ -443,6 +425,24 @@ def export(self, export_format, model: str = "all", rdf_base: str = None, sql_se
             print(f"Export format {export_format} not supported.")
             return ""
 
+    def _determine_sql_server_type(data_contract, sql_server_type):
+        if sql_server_type == "auto":
+            if data_contract.servers is None or len(data_contract.servers) == 0:
+                raise RuntimeError("Export with server_type='auto' requires servers in the data contract.")
+
+            server_types = set([server.type for server in data_contract.servers.values()])
+            if "snowflake" in server_types:
+                return "snowflake"
+            elif "postgres" in server_types:
+                return "postgres"
+            elif "databricks" in server_types:
+                return "databricks"
+            else:
+                # default to snowflake dialect
+                return "snowflake"
+        else:
+            return sql_server_type
+
     def _get_examples_server(self, data_contract, run, tmp_dir):
         run.log_info(f"Copying examples to files in temporary directory {tmp_dir}")
         format = "json"
diff --git a/datacontract/engines/soda/check_soda_execute.py b/datacontract/engines/soda/check_soda_execute.py
index 16299988..23651a9c 100644
--- a/datacontract/engines/soda/check_soda_execute.py
+++ b/datacontract/engines/soda/check_soda_execute.py
@@ -1,5 +1,6 @@
 import logging
 
+from pyspark.sql import SparkSession
 from soda.scan import Scan
 
 from datacontract.engines.soda.connections.bigquery import \
@@ -19,7 +20,7 @@
 from datacontract.model.run import Run, Check, Log
 
 
-def check_soda_execute(run: Run, data_contract: DataContractSpecification, server: Server, spark, tmp_dir):
+def check_soda_execute(run: Run, data_contract: DataContractSpecification, server: Server, spark: SparkSession, tmp_dir):
     if data_contract is None:
         run.log_warn("Cannot run engine soda-core, as data contract is invalid")
         return