fix(impala): fix lstrip, rstrip, strip

Author: gforsyth

ibis/backends/impala/tests/snapshots/test_string_builtins/test_string_builtins/lstrip/out.sql

@@ -1,3 +1,3 @@
 SELECT
-  LTRIM(`t0`.`string_col`) AS `LStrip(string_col)`
+  LTRIM(`t0`.`string_col`, ' \t\n\r\v\f') AS `LStrip(string_col)`
 FROM `functional_alltypes` AS `t0`

ibis/backends/impala/tests/snapshots/test_string_builtins/test_string_builtins/rstrip/out.sql

@@ -1,3 +1,3 @@
 SELECT
-  RTRIM(`t0`.`string_col`) AS `RStrip(string_col)`
+  RTRIM(`t0`.`string_col`, ' \t\n\r\v\f') AS `RStrip(string_col)`
 FROM `functional_alltypes` AS `t0`

ibis/backends/impala/tests/snapshots/test_string_builtins/test_string_builtins/strip/out.sql

@@ -1,3 +1,3 @@
 SELECT
-  TRIM(`t0`.`string_col`) AS `Strip(string_col)`
+  RTRIM(LTRIM(`t0`.`string_col`, ' \t\n\r\v\f'), ' \t\n\r\v\f') AS `Strip(string_col)`
 FROM `functional_alltypes` AS `t0`

ibis/backends/sql/compilers/impala.py

@@ -60,10 +60,7 @@ class ImpalaCompiler(SQLGlotCompiler):
         ops.DayOfWeekName: "dayname",
         ops.ExtractEpochSeconds: "unix_timestamp",
         ops.Hash: "fnv_hash",
-        ops.LStrip: "ltrim",
         ops.Ln: "ln",
-        ops.RStrip: "rtrim",
-        ops.Strip: "trim",
         ops.TypeOf: "typeof",
     }
 
@@ -324,5 +321,11 @@ def visit_DateDelta(self, op, *, left, right, part):
             )
         return self.f.datediff(left, right)
 
+    def visit_Strip(self, op, *, arg):
+        # Impala's `TRIM` doesn't allow specifying characters to trim off, unlike
+        # Impala's `RTRIM` and `LTRIM` which accept a set of characters to
+        # remove.
+        return self.visit_RStrip(op, arg=self.visit_LStrip(op, arg=arg))
+
 
 compiler = ImpalaCompiler()

ibis/backends/tests/test_string.py

@@ -1265,7 +1265,7 @@ def string_temp_table(backend, con):
             id="lstrip",
             marks=[
                 pytest.mark.notimpl(
-                    ["impala", "pyspark"],
+                    ["pyspark"],
                     raises=AssertionError,
                     reason="doesn't strip newline or tabs",
                 ),
@@ -1282,7 +1282,7 @@ def string_temp_table(backend, con):
             id="rstrip",
             marks=[
                 pytest.mark.notimpl(
-                    ["impala", "pyspark"],
+                    ["pyspark"],
                     raises=AssertionError,
                     reason="doesn't strip newline or tabs",
                 ),
@@ -1298,16 +1298,6 @@ def string_temp_table(backend, con):
             lambda t: t.str.strip(),
             id="strip",
             marks=[
-                pytest.mark.notimpl(
-                    ["impala"],
-                    raises=AssertionError,
-                    reason="""
-                    not stripping anything but space
-                    can use
-                    TRIM(TRAILING '\t\n\r ' FROM string_col)
-                    TRIM(LEADING '\t\n\r ' FROM string_col)
-                    """,
-                ),
                 pytest.mark.notimpl(
                     ["flink"],
                     raises=AssertionError,