python/sqlglot
1
0
Fork 0
Code Issues Activity

Merging upstream version 11.7.1.

Browse source 
Signed-off-by: Daniel Baumann <daniel@debian.org>
This commit is contained in:
Daniel Baumann 2025-02-13 15:52:09 +01:00
parent 0c053462ae
commit 8d96084fad
Signed by: daniel
GPG key ID: FBB4F0E80A80222F
144 changed files with 44104 additions and 39367 deletions
Download patch file Download diff file Expand all files Collapse all files

61
sqlglot/dialects/bigquery.py
View file

@ -13,6 +13,7 @@ from sqlglot.dialects.dialect import (
max_or_greatest,
min_or_least,
no_ilike_sql,
parse_date_delta_with_interval,
rename_func,
timestrtotime_sql,
ts_or_ds_to_date_sql,
@ -23,18 +24,6 @@ from sqlglot.tokens import TokenType
E = t.TypeVar("E", bound=exp.Expression)
def _date_add(expression_class: t.Type[E]) -> t.Callable[[t.Sequence], E]:
def func(args):
interval = seq_get(args, 1)
return expression_class(
this=seq_get(args, 0),
expression=interval.this,
unit=interval.args.get("unit"),
)
return func
def _date_add_sql(
data_type: str, kind: str
) -> t.Callable[[generator.Generator, exp.Expression], str]:
@ -142,6 +131,7 @@ class BigQuery(Dialect):
KEYWORDS = {
**tokens.Tokenizer.KEYWORDS,
"ANY TYPE": TokenType.VARIANT,
"BEGIN": TokenType.COMMAND,
"BEGIN TRANSACTION": TokenType.BEGIN,
"CURRENT_DATETIME": TokenType.CURRENT_DATETIME,
@ -155,14 +145,19 @@ class BigQuery(Dialect):
KEYWORDS.pop("DIV")
class Parser(parser.Parser):
PREFIXED_PIVOT_COLUMNS = True
LOG_BASE_FIRST = False
LOG_DEFAULTS_TO_LN = True
FUNCTIONS = {
**parser.Parser.FUNCTIONS, # type: ignore
"DATE_TRUNC": lambda args: exp.DateTrunc(
unit=exp.Literal.string(seq_get(args, 1).name), # type: ignore
this=seq_get(args, 0),
),
"DATE_ADD": _date_add(exp.DateAdd),
"DATETIME_ADD": _date_add(exp.DatetimeAdd),
"DATE_ADD": parse_date_delta_with_interval(exp.DateAdd),
"DATETIME_ADD": parse_date_delta_with_interval(exp.DatetimeAdd),
"DIV": lambda args: exp.IntDiv(this=seq_get(args, 0), expression=seq_get(args, 1)),
"REGEXP_CONTAINS": exp.RegexpLike.from_arg_list,
"REGEXP_EXTRACT": lambda args: exp.RegexpExtract(
@ -174,12 +169,12 @@ class BigQuery(Dialect):
if re.compile(str(seq_get(args, 1))).groups == 1
else None,
),
"TIME_ADD": _date_add(exp.TimeAdd),
"TIMESTAMP_ADD": _date_add(exp.TimestampAdd),
"DATE_SUB": _date_add(exp.DateSub),
"DATETIME_SUB": _date_add(exp.DatetimeSub),
"TIME_SUB": _date_add(exp.TimeSub),
"TIMESTAMP_SUB": _date_add(exp.TimestampSub),
"TIME_ADD": parse_date_delta_with_interval(exp.TimeAdd),
"TIMESTAMP_ADD": parse_date_delta_with_interval(exp.TimestampAdd),
"DATE_SUB": parse_date_delta_with_interval(exp.DateSub),
"DATETIME_SUB": parse_date_delta_with_interval(exp.DatetimeSub),
"TIME_SUB": parse_date_delta_with_interval(exp.TimeSub),
"TIMESTAMP_SUB": parse_date_delta_with_interval(exp.TimestampSub),
"PARSE_TIMESTAMP": lambda args: exp.StrToTime(
this=seq_get(args, 1), format=seq_get(args, 0)
),
@ -209,14 +204,17 @@ class BigQuery(Dialect):
PROPERTY_PARSERS = {
**parser.Parser.PROPERTY_PARSERS, # type: ignore
"NOT DETERMINISTIC": lambda self: self.expression(
exp.VolatilityProperty, this=exp.Literal.string("VOLATILE")
exp.StabilityProperty, this=exp.Literal.string("VOLATILE")
),
}
LOG_BASE_FIRST = False
LOG_DEFAULTS_TO_LN = True
class Generator(generator.Generator):
EXPLICIT_UNION = True
INTERVAL_ALLOWS_PLURAL_FORM = False
JOIN_HINTS = False
TABLE_HINTS = False
LIMIT_FETCH = "LIMIT"
TRANSFORMS = {
**generator.Generator.TRANSFORMS, # type: ignore
**transforms.REMOVE_PRECISION_PARAMETERIZED_TYPES, # type: ignore
@ -236,9 +234,7 @@ class BigQuery(Dialect):
exp.IntDiv: rename_func("DIV"),
exp.Max: max_or_greatest,
exp.Min: min_or_least,
exp.Select: transforms.preprocess(
[_unqualify_unnest], transforms.delegate("select_sql")
),
exp.Select: transforms.preprocess([_unqualify_unnest]),
exp.StrToTime: lambda self, e: f"PARSE_TIMESTAMP({self.format_time(e)}, {self.sql(e, 'this')})",
exp.TimeAdd: _date_add_sql("TIME", "ADD"),
exp.TimeSub: _date_add_sql("TIME", "SUB"),
@ -253,7 +249,7 @@ class BigQuery(Dialect):
exp.ReturnsProperty: _returnsproperty_sql,
exp.Create: _create_sql,
exp.Trim: lambda self, e: self.func(f"TRIM", e.this, e.expression),
exp.VolatilityProperty: lambda self, e: f"DETERMINISTIC"
exp.StabilityProperty: lambda self, e: f"DETERMINISTIC"
if e.name == "IMMUTABLE"
else "NOT DETERMINISTIC",
exp.RegexpLike: rename_func("REGEXP_CONTAINS"),
@ -261,6 +257,7 @@ class BigQuery(Dialect):
TYPE_MAPPING = {
**generator.Generator.TYPE_MAPPING, # type: ignore
exp.DataType.Type.BIGDECIMAL: "BIGNUMERIC",
exp.DataType.Type.BIGINT: "INT64",
exp.DataType.Type.BOOLEAN: "BOOL",
exp.DataType.Type.CHAR: "STRING",
@ -272,17 +269,19 @@ class BigQuery(Dialect):
exp.DataType.Type.NVARCHAR: "STRING",
exp.DataType.Type.SMALLINT: "INT64",
exp.DataType.Type.TEXT: "STRING",
exp.DataType.Type.TIMESTAMP: "DATETIME",
exp.DataType.Type.TIMESTAMPTZ: "TIMESTAMP",
exp.DataType.Type.TINYINT: "INT64",
exp.DataType.Type.VARCHAR: "STRING",
exp.DataType.Type.VARIANT: "ANY TYPE",
}
PROPERTIES_LOCATION = {
**generator.Generator.PROPERTIES_LOCATION, # type: ignore
exp.PartitionedByProperty: exp.Properties.Location.POST_SCHEMA,
exp.VolatileProperty: exp.Properties.Location.UNSUPPORTED,
}
EXPLICIT_UNION = True
LIMIT_FETCH = "LIMIT"
def array_sql(self, expression: exp.Array) -> str:
first_arg = seq_get(expression.expressions, 0)
if isinstance(first_arg, exp.Subqueryable):