refactor: rename utf8_literal to string_literal and add alias str_lit

Author: kosiew

python/datafusion/__init__.py

@@ -107,9 +107,16 @@ def literal(value):
     return Expr.literal(value)
 
 
-def utf8_literal(value):
-    """Create a UTF8 literal expression."""
-    return Expr.utf8_literal(value)
+def string_literal(value):
+    """Create a UTF8 literal expression.
+    It differs from `literal`  creates a UTF8view literal.
+    """
+    return Expr.string_literal(value)
+
+
+def str_lit(value):
+    """Alias for `string_literal`"""
+    return string_literal(value)
 
 
 def lit(value):

python/datafusion/expr.py

@@ -381,7 +381,7 @@ def literal(value: Any) -> Expr:
         return Expr(expr_internal.Expr.literal(value))
 
     @staticmethod
-    def utf8_literal(value: str) -> Expr:
+    def string_literal(value: str) -> Expr:
         """Creates a new expression representing a UTF8 literal value.
 
         It is different from `literal` because it is pa.string() instead of

python/tests/test_functions.py

@@ -23,7 +23,7 @@
 
 from datafusion import SessionContext, column
 from datafusion import functions as f
-from datafusion import literal, utf8_literal
+from datafusion import literal, string_literal
 
 np.seterr(invalid="ignore")
 
@@ -909,11 +909,11 @@ def test_temporal_functions(df):
 
 def test_arrow_cast(df):
     df = df.select(
-        # we use `utf8_literal` to return utf8 instead of `literal` which returns
+        # we use `string_literal` to return utf8 instead of `literal` which returns
         # utf8view because datafusion.arrow_cast expects a utf8 instead of utf8view
         # https://github.com/apache/datafusion/blob/86740bfd3d9831d6b7c1d0e1bf4a21d91598a0ac/datafusion/functions/src/core/arrow_cast.rs#L179
-        f.arrow_cast(column("b"), utf8_literal("Float64")).alias("b_as_float"),
-        f.arrow_cast(column("b"), utf8_literal("Int32")).alias("b_as_int"),
+        f.arrow_cast(column("b"), string_literal("Float64")).alias("b_as_float"),
+        f.arrow_cast(column("b"), string_literal("Int32")).alias("b_as_int"),
     )
     result = df.collect()
     assert len(result) == 1