fix(hive): Enclose exp.Split with \E (#4163)

VaggelisD · web-flow · commit 9c172643aa3f · 2024-09-25T20:23:56.000+03:00
diff --git a/sqlglot/dialects/hive.py b/sqlglot/dialects/hive.py
@@ -546,7 +546,7 @@ class Generator(generator.Generator):
             exp.SchemaCommentProperty: lambda self, e: self.naked_property(e),
             exp.ArrayUniqueAgg: rename_func("COLLECT_SET"),
             exp.Split: lambda self, e: self.func(
-                "SPLIT", e.this, self.func("CONCAT", "'\\\\Q'", e.expression)
+                "SPLIT", e.this, self.func("CONCAT", "'\\\\Q'", e.expression, "'\\\\E'")
             ),
             exp.Select: transforms.preprocess(
                 [
diff --git a/tests/dialects/test_duckdb.py b/tests/dialects/test_duckdb.py
@@ -535,17 +535,17 @@ def test_duckdb(self):
             write={
                 "duckdb": "STR_SPLIT(x, 'a')",
                 "presto": "SPLIT(x, 'a')",
-                "hive": "SPLIT(x, CONCAT('\\\\Q', 'a'))",
-                "spark": "SPLIT(x, CONCAT('\\\\Q', 'a'))",
+                "hive": "SPLIT(x, CONCAT('\\\\Q', 'a', '\\\\E'))",
+                "spark": "SPLIT(x, CONCAT('\\\\Q', 'a', '\\\\E'))",
             },
         )
         self.validate_all(
             "STRING_TO_ARRAY(x, 'a')",
             write={
                 "duckdb": "STR_SPLIT(x, 'a')",
                 "presto": "SPLIT(x, 'a')",
-                "hive": "SPLIT(x, CONCAT('\\\\Q', 'a'))",
-                "spark": "SPLIT(x, CONCAT('\\\\Q', 'a'))",
+                "hive": "SPLIT(x, CONCAT('\\\\Q', 'a', '\\\\E'))",
+                "spark": "SPLIT(x, CONCAT('\\\\Q', 'a', '\\\\E'))",
             },
         )
         self.validate_all(
diff --git a/tests/dialects/test_presto.py b/tests/dialects/test_presto.py
@@ -158,8 +158,8 @@ def test_regex(self):
             write={
                 "duckdb": "STR_SPLIT(x, 'a.')",
                 "presto": "SPLIT(x, 'a.')",
-                "hive": "SPLIT(x, CONCAT('\\\\Q', 'a.'))",
-                "spark": "SPLIT(x, CONCAT('\\\\Q', 'a.'))",
+                "hive": "SPLIT(x, CONCAT('\\\\Q', 'a.', '\\\\E'))",
+                "spark": "SPLIT(x, CONCAT('\\\\Q', 'a.', '\\\\E'))",
             },
         )
         self.validate_all(
diff --git a/tests/dialects/test_starrocks.py b/tests/dialects/test_starrocks.py
@@ -85,8 +85,8 @@ def test_unnest(self):
             r"""SELECT id, t.type, t.scores FROM example_table, unnest(split(type, ";"), scores) AS t(type,scores)""",
             write={
                 "postgres": "SELECT id, t.type, t.scores FROM example_table, UNNEST(SPLIT(type, ';'), scores) AS t(type, scores)",
-                "spark": r"""SELECT id, t.type, t.scores FROM example_table LATERAL VIEW INLINE(ARRAYS_ZIP(SPLIT(type, CONCAT('\\Q', ';')), scores)) t AS type, scores""",
-                "databricks": r"""SELECT id, t.type, t.scores FROM example_table LATERAL VIEW INLINE(ARRAYS_ZIP(SPLIT(type, CONCAT('\\Q', ';')), scores)) t AS type, scores""",
+                "spark": r"""SELECT id, t.type, t.scores FROM example_table LATERAL VIEW INLINE(ARRAYS_ZIP(SPLIT(type, CONCAT('\\Q', ';', '\\E')), scores)) t AS type, scores""",
+                "databricks": r"""SELECT id, t.type, t.scores FROM example_table LATERAL VIEW INLINE(ARRAYS_ZIP(SPLIT(type, CONCAT('\\Q', ';', '\\E')), scores)) t AS type, scores""",
                 "starrocks": r"""SELECT id, t.type, t.scores FROM example_table, UNNEST(SPLIT(type, ';'), scores) AS t(type, scores)""",
                 "hive": UnsupportedError,
             },
@@ -95,7 +95,7 @@ def test_unnest(self):
         self.validate_all(
             r"""SELECT id, t.type, t.scores FROM example_table_2 CROSS JOIN LATERAL unnest(split(type, ";"), scores) AS t(type,scores)""",
             write={
-                "spark": r"""SELECT id, t.type, t.scores FROM example_table_2 LATERAL VIEW INLINE(ARRAYS_ZIP(SPLIT(type, CONCAT('\\Q', ';')), scores)) t AS type, scores""",
+                "spark": r"""SELECT id, t.type, t.scores FROM example_table_2 LATERAL VIEW INLINE(ARRAYS_ZIP(SPLIT(type, CONCAT('\\Q', ';', '\\E')), scores)) t AS type, scores""",
                 "starrocks": r"""SELECT id, t.type, t.scores FROM example_table_2 CROSS JOIN LATERAL UNNEST(SPLIT(type, ';'), scores) AS t(type, scores)""",
                 "hive": UnsupportedError,
             },

Original file line number	Diff line number	Diff line change
`@@ -546,7 +546,7 @@ class Generator(generator.Generator):`
`546`	`546`	`exp.SchemaCommentProperty: lambda self, e: self.naked_property(e),`
`547`	`547`	`exp.ArrayUniqueAgg: rename_func("COLLECT_SET"),`
`548`	`548`	`exp.Split: lambda self, e: self.func(`
`549`		`- "SPLIT", e.this, self.func("CONCAT", "'\\\\Q'", e.expression)`
	`549`	`+ "SPLIT", e.this, self.func("CONCAT", "'\\\\Q'", e.expression, "'\\\\E'")`
`550`	`550`	`),`
`551`	`551`	`exp.Select: transforms.preprocess(`
`552`	`552`	`[`