diff options
Diffstat (limited to 'sqlglot/dialects/hive.py')
| -rw-r--r-- | sqlglot/dialects/hive.py | 38 |
1 files changed, 21 insertions, 17 deletions
diff --git a/sqlglot/dialects/hive.py b/sqlglot/dialects/hive.py index 871a180..fbd626a 100644 --- a/sqlglot/dialects/hive.py +++ b/sqlglot/dialects/hive.py @@ -147,13 +147,6 @@ def _to_date_sql(self: generator.Generator, expression: exp.TsOrDsToDate) -> str return f"TO_DATE({this})" -def _index_sql(self: generator.Generator, expression: exp.Index) -> str: - this = self.sql(expression, "this") - table = self.sql(expression, "table") - columns = self.sql(expression, "columns") - return f"{this} ON TABLE {table} {columns}" - - class Hive(Dialect): alias_post_tablesample = True @@ -225,8 +218,7 @@ class Hive(Dialect): STRICT_CAST = False FUNCTIONS = { - **parser.Parser.FUNCTIONS, # type: ignore - "APPROX_COUNT_DISTINCT": exp.ApproxDistinct.from_arg_list, + **parser.Parser.FUNCTIONS, "BASE64": exp.ToBase64.from_arg_list, "COLLECT_LIST": exp.ArrayAgg.from_arg_list, "DATE_ADD": lambda args: exp.TsOrDsAdd( @@ -271,21 +263,29 @@ class Hive(Dialect): } PROPERTY_PARSERS = { - **parser.Parser.PROPERTY_PARSERS, # type: ignore + **parser.Parser.PROPERTY_PARSERS, "WITH SERDEPROPERTIES": lambda self: exp.SerdeProperties( expressions=self._parse_wrapped_csv(self._parse_property) ), } + QUERY_MODIFIER_PARSERS = { + **parser.Parser.QUERY_MODIFIER_PARSERS, + "distribute": lambda self: self._parse_sort(exp.Distribute, "DISTRIBUTE", "BY"), + "sort": lambda self: self._parse_sort(exp.Sort, "SORT", "BY"), + "cluster": lambda self: self._parse_sort(exp.Cluster, "CLUSTER", "BY"), + } + class Generator(generator.Generator): LIMIT_FETCH = "LIMIT" TABLESAMPLE_WITH_METHOD = False TABLESAMPLE_SIZE_IS_PERCENT = True JOIN_HINTS = False TABLE_HINTS = False + INDEX_ON = "ON TABLE" TYPE_MAPPING = { - **generator.Generator.TYPE_MAPPING, # type: ignore + **generator.Generator.TYPE_MAPPING, exp.DataType.Type.TEXT: "STRING", exp.DataType.Type.DATETIME: "TIMESTAMP", exp.DataType.Type.VARBINARY: "BINARY", @@ -294,7 +294,7 @@ class Hive(Dialect): } TRANSFORMS = { - **generator.Generator.TRANSFORMS, # type: ignore + **generator.Generator.TRANSFORMS, exp.Group: transforms.preprocess([transforms.unalias_group]), exp.Select: transforms.preprocess( [ @@ -319,7 +319,6 @@ class Hive(Dialect): exp.FileFormatProperty: lambda self, e: f"STORED AS {self.sql(e, 'this') if isinstance(e.this, exp.InputOutputFormat) else e.name.upper()}", exp.FromBase64: rename_func("UNBASE64"), exp.If: if_sql, - exp.Index: _index_sql, exp.ILike: no_ilike_sql, exp.JSONExtract: rename_func("GET_JSON_OBJECT"), exp.JSONExtractScalar: rename_func("GET_JSON_OBJECT"), @@ -342,7 +341,6 @@ class Hive(Dialect): exp.StrToTime: _str_to_time_sql, exp.StrToUnix: _str_to_unix_sql, exp.StructExtract: struct_extract_sql, - exp.TableFormatProperty: lambda self, e: f"USING {self.sql(e, 'this')}", exp.TimeStrToDate: rename_func("TO_DATE"), exp.TimeStrToTime: timestrtotime_sql, exp.TimeStrToUnix: rename_func("UNIX_TIMESTAMP"), @@ -363,14 +361,13 @@ class Hive(Dialect): exp.SerdeProperties: lambda self, e: self.properties(e, prefix="WITH SERDEPROPERTIES"), exp.NumberToStr: rename_func("FORMAT_NUMBER"), exp.LastDateOfMonth: rename_func("LAST_DAY"), - exp.National: lambda self, e: self.sql(e, "this"), + exp.National: lambda self, e: self.national_sql(e, prefix=""), } PROPERTIES_LOCATION = { - **generator.Generator.PROPERTIES_LOCATION, # type: ignore + **generator.Generator.PROPERTIES_LOCATION, exp.FileFormatProperty: exp.Properties.Location.POST_SCHEMA, exp.PartitionedByProperty: exp.Properties.Location.POST_SCHEMA, - exp.TableFormatProperty: exp.Properties.Location.POST_SCHEMA, exp.VolatileProperty: exp.Properties.Location.UNSUPPORTED, } @@ -396,3 +393,10 @@ class Hive(Dialect): expression = exp.DataType.build(expression.this) return super().datatype_sql(expression) + + def after_having_modifiers(self, expression: exp.Expression) -> t.List[str]: + return super().after_having_modifiers(expression) + [ + self.sql(expression, "distribute"), + self.sql(expression, "sort"), + self.sql(expression, "cluster"), + ] |