Merge pull request #349 from mindsdb/staging

Release 0.10.4

mindsdb_sql/__about__.py

@@ -1,6 +1,6 @@
 __title__ = 'mindsdb_sql'
 __package_name__ = 'mindsdb_sql'
-__version__ = '0.10.3'
+__version__ = '0.10.4'
 __description__ = "Pure python SQL parser"
 __email__ = "[email protected]"
 __author__ = 'MindsDB Inc'

mindsdb_sql/parser/dialects/mindsdb/create_job.py

@@ -12,6 +12,7 @@ def __init__(self,
                  start_str=None,
                  end_str=None,
                  repeat_str=None,
+                 if_query_str=None,
                  if_not_exists=False,
                  *args, **kwargs):
         super().__init__(*args, **kwargs)
@@ -22,6 +23,7 @@ def __init__(self,
         self.repeat_str = repeat_str
         self.date_format = '%Y-%m-%d %H:%M:%S'
         self.if_not_exists = if_not_exists
+        self.if_query_str = if_query_str
 
     def to_tree(self, *args, level=0, **kwargs):
         ind = indent(level)
@@ -46,13 +48,18 @@ def to_tree(self, *args, level=0, **kwargs):
         if self.if_not_exists:
             if_not_exists_str = f'\n{ind1}if_not_exists=True,'
 
+        if_query_str = ''
+        if self.if_query_str is not None:
+            if_query_str = f"\n{ind1}if_query='{self.if_query_str}'"
+
         out_str = f'{ind}CreateJob(' \
                   f'{if_not_exists_str}' \
                   f'{name_str}' \
                   f'{query_str}' \
                   f'{start_str}' \
                   f'{end_str}' \
                   f'{repeat_str}' \
+                  f'{if_query_str}' \
                   f'\n{ind})'
         return out_str
 
@@ -70,5 +77,9 @@ def get_string(self, *args, **kwargs):
         if self.repeat_str is not None:
             repeat_str = f" EVERY '{self.repeat_str}'"
 
-        out_str = f'CREATE JOB {"IF NOT EXISTS" if self.if_not_exists else ""} {self.name.to_string()} ({self.query_str}){start_str}{end_str}{repeat_str}'
+        if_query_str = ''
+        if self.if_query_str is not None:
+            if_query_str = f" IF '{self.if_query_str}'"
+
+        out_str = f'CREATE JOB {"IF NOT EXISTS" if self.if_not_exists else ""} {self.name.to_string()} ({self.query_str}){start_str}{end_str}{repeat_str}{if_query_str}'
         return out_str

mindsdb_sql/parser/dialects/mindsdb/create_predictor.py

@@ -35,6 +35,7 @@ def __init__(self,
         self.is_replace = is_replace
         self.if_not_exists = if_not_exists
         self.task = task
+        self._action = 'CREATE'
 
     def to_tree(self, *args, level=0, **kwargs):
         ind = indent(level)
@@ -74,8 +75,10 @@ def to_tree(self, *args, level=0, **kwargs):
         using_str = f'\n{ind1}using={repr(self.using)},'
 
         if_not_exists_str = f'\n{ind1}if_not_exists={self.if_not_exists},' if self.if_not_exists else ''
+        or_replace_str = f'\n{ind1}is_replace={self.is_replace},' if self.is_replace else ''
 
         out_str = f'{ind}{self.__class__.__name__}(' \
+                  f'{or_replace_str}' \
                   f'{if_not_exists_str}' \
                   f'{name_str}' \
                   f'{integration_name_str}' \
@@ -126,9 +129,11 @@ def get_string(self, *args, **kwargs):
         if self.integration_name is not None:
             integration_name_str = f'FROM {self.integration_name.to_string()} '
 
+        or_replace_str = ' OR REPLACE' if self.is_replace else ''
         if_not_exists_str = 'IF NOT EXISTS ' if self.if_not_exists else ''
+        object_str = self._object + ' ' if self._object else ''
 
-        out_str = f'{self._command} {if_not_exists_str}{self.name.to_string()} {integration_name_str}{query_str}' \
+        out_str = f'{self._action}{or_replace_str} {object_str}{if_not_exists_str}{self.name.to_string()} {integration_name_str}{query_str}' \
                   f'{datasource_name_str}' \
                   f'{targets_str} ' \
                   f'{order_by_str}' \
@@ -143,12 +148,12 @@ def get_string(self, *args, **kwargs):
 class CreatePredictor(CreatePredictorBase):
     def __init__(self, *args, **kwargs):
         super().__init__(*args, **kwargs)
-        self._command = 'CREATE PREDICTOR'
+        self._object = 'PREDICTOR'
 
 
 # Models by task type
 class CreateAnomalyDetectionModel(CreatePredictorBase):
     def __init__(self, *args, **kwargs):
         super().__init__(*args, **kwargs)
-        self._command = 'CREATE ANOMALY DETECTION MODEL'
+        self._object = 'ANOMALY DETECTION MODEL'
         self.task = Identifier('AnomalyDetection')

mindsdb_sql/parser/dialects/mindsdb/finetune_predictor.py

@@ -4,4 +4,5 @@
 class FinetunePredictor(CreatePredictorBase):
     def __init__(self, *args, **kwargs):
         super().__init__(*args, **kwargs)
-        self._command = 'FINETUNE'
+        self._action = 'FINETUNE'
+        self._object = ''

mindsdb_sql/parser/dialects/mindsdb/lexer.py

@@ -43,7 +43,7 @@ class MindsDBLexer(Lexer):
         GLOBAL, PROCEDURE, FUNCTION, INDEX, WARNINGS,
         ENGINES, CHARSET, COLLATION, PLUGINS, CHARACTER,
         PERSIST, PERSIST_ONLY,
-        IF_EXISTS, IF_NOT_EXISTS, COLUMNS, FIELDS, COLLATE, SEARCH_PATH,
+        IF_EXISTS, IF_NOT_EXISTS, IF, COLUMNS, FIELDS, COLLATE, SEARCH_PATH,
         VARIABLE, SYSTEM_VARIABLE,
 
         # SELECT Keywords
@@ -71,7 +71,7 @@ class MindsDBLexer(Lexer):
         PLUS, MINUS, DIVIDE, MODULO,
         EQUALS, NEQUALS, GREATER, GEQ, LESS, LEQ,
         AND, OR, NOT, IS, IS_NOT,
-        IN, LIKE, CONCAT, BETWEEN, WINDOW, OVER, PARTITION_BY,
+        IN, LIKE, NOT_LIKE, CONCAT, BETWEEN, WINDOW, OVER, PARTITION_BY,
 
         # Data types
         CAST, ID, INTEGER, FLOAT, QUOTE_STRING, DQUOTE_STRING, NULL, TRUE, FALSE,
@@ -174,6 +174,7 @@ class MindsDBLexer(Lexer):
     PERSIST_ONLY = r'\bPERSIST_ONLY\b'
     IF_EXISTS = r'\bIF[\s]+EXISTS\b'
     IF_NOT_EXISTS = r'\bIF[\s]+NOT[\s]+EXISTS\b'
+    IF = r'\bIF\b'
     COLUMNS = r'\bCOLUMNS\b'
     FIELDS = r'\bFIELDS\b'
     EXTENDED = r'\bEXTENDED\b'
@@ -275,6 +276,7 @@ class MindsDBLexer(Lexer):
     AND = r'\bAND\b'
     OR = r'\bOR\b'
     IS_NOT = r'\bIS[\s]+NOT\b'
+    NOT_LIKE = r'\bNOT[\s]+LIKE\b'
     NOT = r'\bNOT\b'
     IS = r'\bIS\b'
     LIKE = r'\bLIKE\b'

mindsdb_sql/parser/dialects/mindsdb/parser.py

@@ -47,7 +47,7 @@ class MindsDBParser(Parser):
         ('left', PLUS, MINUS),
         ('left', STAR, DIVIDE),
         ('right', UMINUS),  # Unary minus operator, unary not
-        ('nonassoc', LESS, LEQ, GREATER, GEQ, IN, BETWEEN, IS, IS_NOT, LIKE),
+        ('nonassoc', LESS, LEQ, GREATER, GEQ, IN, BETWEEN, IS, IS_NOT, NOT_LIKE, LIKE),
     )
 
     # Top-level statements
@@ -229,10 +229,20 @@ def drop_trigger(self, p):
     # -- Jobs --
     @_('CREATE JOB if_not_exists_or_empty identifier LPAREN raw_query RPAREN job_schedule',
        'CREATE JOB if_not_exists_or_empty identifier AS LPAREN raw_query RPAREN job_schedule',
+       'CREATE JOB if_not_exists_or_empty identifier LPAREN raw_query RPAREN job_schedule IF LPAREN raw_query RPAREN',
+       'CREATE JOB if_not_exists_or_empty identifier AS LPAREN raw_query RPAREN job_schedule IF LPAREN raw_query RPAREN',
        'CREATE JOB if_not_exists_or_empty identifier LPAREN raw_query RPAREN',
-       'CREATE JOB if_not_exists_or_empty identifier AS LPAREN raw_query RPAREN')
+       'CREATE JOB if_not_exists_or_empty identifier AS LPAREN raw_query RPAREN',
+       'CREATE JOB if_not_exists_or_empty identifier LPAREN raw_query RPAREN IF LPAREN raw_query RPAREN',
+       'CREATE JOB if_not_exists_or_empty identifier AS LPAREN raw_query RPAREN IF LPAREN raw_query RPAREN'
+       )
     def create_job(self, p):
-        query_str = tokens_to_string(p.raw_query)
+        if hasattr(p, 'raw_query0'):
+            query_str = tokens_to_string(p.raw_query0)
+            if_query_str = tokens_to_string(p.raw_query1)
+        else:
+            query_str = tokens_to_string(p.raw_query)
+            if_query_str = None
 
         job_schedule = getattr(p, 'job_schedule', {})
 
@@ -254,6 +264,7 @@ def create_job(self, p):
         return CreateJob(
             name=p.identifier,
             query_str=query_str,
+            if_query_str=if_query_str,
             start_str=start_str,
             end_str=end_str,
             repeat_str=repeat_str,
@@ -776,10 +787,11 @@ def create_predictor(self, p):
         p.create_predictor.order_by = p.ordering_terms
         return p.create_predictor
 
-    @_('CREATE PREDICTOR if_not_exists_or_empty identifier FROM identifier LPAREN raw_query RPAREN PREDICT result_columns',
-       'CREATE PREDICTOR if_not_exists_or_empty identifier PREDICT result_columns',
-       'CREATE MODEL if_not_exists_or_empty identifier FROM identifier LPAREN raw_query RPAREN PREDICT result_columns',
-       'CREATE MODEL if_not_exists_or_empty identifier PREDICT result_columns')
+    @_('CREATE replace_or_empty PREDICTOR if_not_exists_or_empty identifier FROM identifier LPAREN raw_query RPAREN PREDICT result_columns',
+       'CREATE replace_or_empty PREDICTOR if_not_exists_or_empty identifier PREDICT result_columns',
+       'CREATE replace_or_empty MODEL if_not_exists_or_empty identifier FROM identifier LPAREN raw_query RPAREN PREDICT result_columns',
+       'CREATE replace_or_empty MODEL if_not_exists_or_empty identifier PREDICT result_columns'
+       )
     def create_predictor(self, p):
         query_str = None
         if hasattr(p, 'raw_query'):
@@ -796,7 +808,8 @@ def create_predictor(self, p):
             integration_name=getattr(p, 'identifier1', None),
             query_str=query_str,
             targets=p.result_columns,
-            if_not_exists=p.if_not_exists_or_empty
+            if_not_exists=p.if_not_exists_or_empty,
+            is_replace=p.replace_or_empty
         )
 
     # Typed models
@@ -1422,14 +1435,19 @@ def expr(self, p):
        'expr NOT expr',
        'expr IS expr',
        'expr LIKE expr',
+       'expr NOT_LIKE expr',
        'expr CONCAT expr',
        'expr IN expr')
     def expr(self, p):
         if hasattr(p, 'LAST'):
             arg1 = Last()
         else:
             arg1 = p.expr1
-        return BinaryOperation(op=p[1], args=(p[0], arg1))
+        if len(p) > 3:
+            op = ' '.join([p[i] for i in range(1, len(p)-1)])
+        else:
+            op = p[1]
+        return BinaryOperation(op=op, args=(p[0], arg1))
 
     @_('MINUS expr %prec UMINUS',
        'NOT expr %prec UNOT', )
@@ -1729,6 +1747,15 @@ def variable(self, p):
     def variable(self, p):
         return Variable(value=p.VARIABLE)
 
+    @_(
+        'OR REPLACE',
+        'empty'
+    )
+    def replace_or_empty(self, p):
+        if hasattr(p, 'REPLACE'):
+            return True
+        return False
+
     @_(
         'IF_NOT_EXISTS',
         'empty'

mindsdb_sql/parser/dialects/mindsdb/retrain_predictor.py

@@ -1,8 +1,8 @@
-from mindsdb_sql.parser.ast.base import ASTNode
-from mindsdb_sql.parser.utils import indent
 from .create_predictor import CreatePredictorBase
 
+
 class RetrainPredictor(CreatePredictorBase):
     def __init__(self, *args, **kwargs):
         super().__init__(*args, **kwargs)
-        self._command = 'RETRAIN'
+        self._action = 'RETRAIN'
+        self._object = ''

mindsdb_sql/planner/plan_join.py

@@ -407,16 +407,27 @@ def process_predictor(self, item, query_in):
             raise NotImplementedError("TS predictor is not supported here yet")
         data_step = self.step_stack[-1]
         row_dict = None
+
+        predict_target = item.predictor_info.get('to_predict')
+        if isinstance(predict_target, list) and len(predict_target) > 0:
+            predict_target = predict_target[0]
+        if predict_target is not None:
+            predict_target = predict_target.lower()
+
         if item.conditions:
             row_dict = {}
-            for el in item.conditions:
+            for i, el in enumerate(item.conditions):
                 if isinstance(el.args[0], Identifier) and el.op == '=':
+                    col_name = el.args[0].parts[-1]
+                    if col_name.lower() == predict_target:
+                        # don't add predict target to parameters
+                        continue
 
                     if isinstance(el.args[1], (Constant, Parameter)):
                         row_dict[el.args[0].parts[-1]] = el.args[1].value
 
                     # exclude condition
-                    item.conditions[0]._orig_node.args = [Constant(0), Constant(0)]
+                    el._orig_node.args = [Constant(0), Constant(0)]
 
         predictor_step = self.planner.plan.add_step(ApplyPredictorStep(
             namespace=item.integration,

mindsdb_sql/planner/query_planner.py

@@ -289,9 +289,10 @@ def plan_api_db_select(self, query):
         #     keep only limit and where
         #     the rest goes to outer select
         query2 = Select(
-            targets=[Star()],
+            targets=query.targets,
             from_table=query.from_table,
             where=query.where,
+            order_by=query.order_by,
             limit=query.limit,
         )
         prev_step = self.plan_integration_select(query2)

mindsdb_sql/render/sqlalchemy_render.py

@@ -133,6 +133,7 @@ def to_expression(self, t):
                 "is": "is_",
                 "is not": "is_not",
                 "like": "like",
+                "not like": "notlike",
                 "in": "in_",
                 "not in": "notin_",
                 "||": "concat",

tests/test_parser/test_base_sql/test_select_operations.py

@@ -564,3 +564,21 @@ def test_function_with_from(self, dialect):
         assert str(ast) == str(expected_ast)
         assert ast.to_tree() == expected_ast.to_tree()
 
+class TestOperationsMindsdb:
+    def test_select_binary_operations(self):
+        for op in ['not like',]:
+            sql = f'SELECT column1 {op.upper()} column2 FROM tab'
+            ast = parse_sql(sql)
+
+            expected_ast = Select(
+                targets=[BinaryOperation(op=op,
+                                         args=(
+                                             Identifier.from_path_str('column1'), Identifier.from_path_str('column2')
+                                         )),
+                         ],
+                from_table=Identifier.from_path_str('tab')
+            )
+
+            assert str(ast).lower() == sql.lower()
+            assert str(ast) == str(expected_ast)
+            assert ast.to_tree() == expected_ast.to_tree()

tests/test_parser/test_mindsdb/test_create_predictor.py

@@ -104,6 +104,17 @@ def test_create_predictor_no_with(self):
 
         assert ast.to_tree() == expected_ast.to_tree()
 
+        # test or replace
+        sql = """CREATE or replace model pred
+                FROM integration_name 
+                (select * FROM table_name)
+                PREDICT f1 as f1_alias, f2
+                """
+        ast = parse_sql(sql, dialect='mindsdb')
+        expected_ast.is_replace = True
+
+        assert ast.to_tree() == expected_ast.to_tree()
+
     def test_create_predictor_quotes(self):
         sql = """CREATE PREDICTOR xxx 
                  FROM `yyy` 

tests/test_parser/test_mindsdb/test_jobs.py

@@ -56,6 +56,23 @@ def test_create_job(self):
         assert str(ast) == str(expected_ast)
         assert ast.to_tree() == expected_ast.to_tree()
 
+        # 4
+
+        sql = '''
+            create job j1 ( retrain p1 )
+            every 2 hours
+            if (select a from table)
+        '''
+        ast = parse_sql(sql, dialect='mindsdb')
+        expected_ast = CreateJob(
+            name=Identifier('j1'),
+            query_str="retrain p1",
+            if_query_str="select a from table",
+            repeat_str="2 hours"
+        )
+        assert str(ast) == str(expected_ast)
+        assert ast.to_tree() == expected_ast.to_tree()
+
     def test_create_job_minimal_with_if_not_exists(self):
         sql = '''
             create job if not exists proj2.j1 ( 

tests/test_planner/test_integration_select.py

@@ -524,7 +524,7 @@ def test_select_from_table_subselect_api_integration(self):
             steps=[
                 FetchDataframeStep(
                     integration='int1',
-                    query=parse_sql('select * from tab1'),
+                    query=parse_sql('select tab1.`id` AS `id` from tab1'),
                 ),
                 SubSelectStep(
                     dataframe=Result(0),
@@ -534,7 +534,7 @@ def test_select_from_table_subselect_api_integration(self):
                 FetchDataframeStep(
                     integration='int1',
                     query=Select(
-                        targets=[Star()],
+                        targets=[Identifier('tab2.x', alias=Identifier('x'))],
                         from_table=Identifier('tab2'),
                         where=BinaryOperation(
                             op='in',
@@ -597,7 +597,7 @@ def test_delete_from_table_subselect_api_integration(self):
             steps=[
                 FetchDataframeStep(
                     integration='int1',
-                    query=parse_sql('select * from tab1'),
+                    query=parse_sql('select tab1.`id` AS `id` from tab1'),
                 ),
                 SubSelectStep(
                     dataframe=Result(0),