sqlglot.dialects.databricks

 1from __future__ import annotations
 2
 3from copy import deepcopy
 4from collections import defaultdict
 5
 6from sqlglot import exp
 7from sqlglot.dialects.spark import Spark
 8from sqlglot.generators.databricks import DatabricksGenerator
 9from sqlglot.parsers.databricks import DatabricksParser
10from sqlglot.tokens import TokenType
11from sqlglot.optimizer.annotate_types import TypeAnnotator
12
13
14class Databricks(Spark):
15    SAFE_DIVISION = False
16    COPY_PARAMS_ARE_CSV = False
17
18    COERCES_TO = defaultdict(set, deepcopy(TypeAnnotator.COERCES_TO))
19    for text_type in exp.DataType.TEXT_TYPES:
20        COERCES_TO[text_type] |= {
21            *exp.DataType.NUMERIC_TYPES,
22            *exp.DataType.TEMPORAL_TYPES,
23            exp.DType.BINARY,
24            exp.DType.BOOLEAN,
25            exp.DType.INTERVAL,
26        }
27
28    class JSONPathTokenizer(Spark.JSONPathTokenizer):
29        IDENTIFIERS = ["`", '"']
30
31    class Tokenizer(Spark.Tokenizer):
32        KEYWORDS = {
33            **Spark.Tokenizer.KEYWORDS,
34            "STREAM": TokenType.STREAM,
35            "VOID": TokenType.VOID,
36        }
37
38    Parser = DatabricksParser
39
40    Generator = DatabricksGenerator