Ignore case for queries in the parser configuration

dask_sql/context.py

@@ -800,8 +800,16 @@ def _get_ral(self, sql):
         # get the schema of what we currently have registered
         schemas = self._prepare_schemas()
 
-        # Now create a relational algebra from that
-        generator_builder = RelationalAlgebraGeneratorBuilder(self.schema_name)
+        # True if the SQL query should be case sensitive and False otherwise
+        case_sensitive = (
+            self.schema[self.schema_name]
+            .config.get_config_by_prefix("dask.sql.identifier.case.sensitive")
+            .get("dask.sql.identifier.case.sensitive", True)
+        )
+
+        generator_builder = RelationalAlgebraGeneratorBuilder(
+            self.schema_name, case_sensitive
+        )
         for schema in schemas:
             generator_builder.addSchema(schema)
         generator = generator_builder.build()

planner/src/main/java/com/dask/sql/application/RelationalAlgebraGenerator.java

@@ -63,14 +63,14 @@ public class RelationalAlgebraGenerator {
 	final HepPlanner hepPlanner;
 
 	/// Create a new relational algebra generator from a schema
-	public RelationalAlgebraGenerator(final String rootSchemaName, final List<DaskSchema> schemas) throws ClassNotFoundException, SQLException {
+	public RelationalAlgebraGenerator(final String rootSchemaName, final List<DaskSchema> schemas, final boolean case_sensitive) throws ClassNotFoundException, SQLException {
 		// Taken from https://calcite.apache.org/docs/ and blazingSQL
 		final SchemaPlus rootSchema = createRootSchema(rootSchemaName, schemas);
 
 		final JavaTypeFactoryImpl typeFactory = createTypeFactory();
 		final CalciteCatalogReader calciteCatalogReader = createCatalogReader(rootSchemaName, rootSchema, typeFactory);
 		final SqlOperatorTable operatorTable = createOperatorTable(calciteCatalogReader);
-		final SqlParser.Config parserConfig = createParserConfig();
+		final SqlParser.Config parserConfig = createParserConfig(case_sensitive);
 		final SchemaPlus schemaPlus = rootSchema.getSubSchema(rootSchemaName);
 		final FrameworkConfig frameworkConfig = createFrameworkConfig(schemaPlus, operatorTable, parserConfig);
 
@@ -180,8 +180,9 @@ private SqlOperatorTable createOperatorTable(final CalciteCatalogReader calciteC
 		return operatorTable;
 	}
 
-	private Config createParserConfig() {
+	private Config createParserConfig(boolean case_sensitive) {
 		return getDialect().configureParser(SqlParser.Config.DEFAULT).withConformance(SqlConformanceEnum.DEFAULT)
+				.withCaseSensitive(case_sensitive)
 				.withParserFactory(new DaskSqlParserImplFactory());
 	}
 

planner/src/main/java/com/dask/sql/application/RelationalAlgebraGeneratorBuilder.java

@@ -9,10 +9,12 @@
 public class RelationalAlgebraGeneratorBuilder {
 	private final String rootSchemaName;
 	private final List<DaskSchema> schemas;
+	private final boolean case_sensitive; // True if case should be ignored when comparing SQLNode(s)
 
-	public RelationalAlgebraGeneratorBuilder(final String rootSchemaName) {
+	public RelationalAlgebraGeneratorBuilder(final String rootSchemaName, final boolean case_sensitive) {
 		this.rootSchemaName = rootSchemaName;
 		this.schemas = new ArrayList<>();
+		this.case_sensitive = case_sensitive;
 	}
 
 	public RelationalAlgebraGeneratorBuilder addSchema(final DaskSchema schema) {
@@ -21,6 +23,6 @@ public RelationalAlgebraGeneratorBuilder addSchema(final DaskSchema schema) {
 	}
 
 	public RelationalAlgebraGenerator build() throws ClassNotFoundException, SQLException {
-		return new RelationalAlgebraGenerator(rootSchemaName, schemas);
+		return new RelationalAlgebraGenerator(rootSchemaName, schemas, case_sensitive);
 	}
 }

tests/integration/test_compatibility.py

@@ -17,6 +17,7 @@
 from pandas.testing import assert_frame_equal
 
 from dask_sql import Context
+from dask_sql.utils import ParsingException
 
 
 def cast_datetime_to_string(df):
@@ -934,3 +935,16 @@ def test_integration_1():
         """,
         a=a,
     )
+
+
+def test_query_case_sensitivity():
+    c = Context()
+    c.set_config(("dask.sql.identifier.case.sensitive", False))
+    df = pd.DataFrame({"id": [0, 1]})
+
+    c.create_table("test", df)
+
+    try:
+        c.sql("select ID from test")
+    except ParsingException as pe:
+        assert False, f"Queries should be case insensitve but raised exception {pe}"

tests/integration/test_select.py

@@ -95,6 +95,7 @@ def test_select_of_select_with_casing(c, df):
     expected_df = pd.DataFrame(
         {"AAA": df["a"] + df["b"], "aaa": 2 * df["b"], "aAa": df["a"] - 1}
     )
+
     assert_frame_equal(result_df, expected_df)