dbcli/sqlglot
1
0
Fork 0
Code Issues Activity

Merging upstream version 7.1.3.

Browse source 
Signed-off-by: Daniel Baumann <daniel@debian.org>
This commit is contained in:
Daniel Baumann 2025-02-13 14:46:58 +01:00
parent 964bd62de9
commit e6b3d2fe54
Signed by: daniel
GPG key ID: FBB4F0E80A80222F
42 changed files with 1430 additions and 253 deletions
Download patch file Download diff file Expand all files Collapse all files

3
sqlglot/dialects/clickhouse.py
View file

@ -1,7 +1,6 @@
from sqlglot import exp
from sqlglot.dialects.dialect import Dialect, inline_array_sql, var_map_sql
from sqlglot.generator import Generator
from sqlglot.helper import csv
from sqlglot.parser import Parser, parse_var_map
from sqlglot.tokens import Tokenizer, TokenType
@ -66,7 +65,7 @@ class ClickHouse(Dialect):
TRANSFORMS = {
**Generator.TRANSFORMS,
exp.Array: inline_array_sql,
exp.StrPosition: lambda self, e: f"position({csv(self.sql(e, 'this'), self.sql(e, 'substr'), self.sql(e, 'position'))})",
exp.StrPosition: lambda self, e: f"position({self.format_args(e.this, e.args.get('substr'), e.args.get('position'))})",
exp.Final: lambda self, e: f"{self.sql(e, 'this')} FINAL",
exp.Map: lambda self, e: _lower_func(var_map_sql(self, e)),
exp.VarMap: lambda self, e: _lower_func(var_map_sql(self, e)),

25
sqlglot/dialects/dialect.py
View file

@ -2,7 +2,7 @@ from enum import Enum
from sqlglot import exp
from sqlglot.generator import Generator
from sqlglot.helper import csv, list_get
from sqlglot.helper import list_get
from sqlglot.parser import Parser
from sqlglot.time import format_time
from sqlglot.tokens import Tokenizer
@ -177,11 +177,11 @@ class Dialect(metaclass=_Dialect):
def rename_func(name):
def _rename(self, expression):
args = (
self.expressions(expression, flat=True)
expression.expressions
if isinstance(expression, exp.Func) and expression.is_var_len_args
else csv(*[self.sql(e) for e in expression.args.values()])
else expression.args.values()
)
return f"{name}({args})"
return f"{name}({self.format_args(*args)})"
return _rename
@ -189,15 +189,11 @@ def rename_func(name):
def approx_count_distinct_sql(self, expression):
if expression.args.get("accuracy"):
self.unsupported("APPROX_COUNT_DISTINCT does not support accuracy")
return f"APPROX_COUNT_DISTINCT({self.sql(expression, 'this')})"
return f"APPROX_COUNT_DISTINCT({self.format_args(expression.this)})"
def if_sql(self, expression):
expressions = csv(
self.sql(expression, "this"),
self.sql(expression, "true"),
self.sql(expression, "false"),
)
expressions = self.format_args(expression.this, expression.args.get("true"), expression.args.get("false"))
return f"IF({expressions})"
@ -254,6 +250,11 @@ def no_trycast_sql(self, expression):
return self.cast_sql(expression)
def no_properties_sql(self, expression):
self.unsupported("Properties unsupported")
return ""
def str_position_sql(self, expression):
this = self.sql(expression, "this")
substr = self.sql(expression, "substr")
@ -275,13 +276,13 @@ def var_map_sql(self, expression):
if not isinstance(keys, exp.Array) or not isinstance(values, exp.Array):
self.unsupported("Cannot convert array columns into map.")
return f"MAP({self.sql(keys)}, {self.sql(values)})"
return f"MAP({self.format_args(keys, values)})"
args = []
for key, value in zip(keys.expressions, values.expressions):
args.append(self.sql(key))
args.append(self.sql(value))
return f"MAP({csv(*args)})"
return f"MAP({self.format_args(*args)})"
def format_time_lambda(exp_class, dialect, default=None):

13
sqlglot/dialects/duckdb.py
View file

@ -6,6 +6,7 @@ from sqlglot.dialects.dialect import (
arrow_json_extract_sql,
format_time_lambda,
no_pivot_sql,
no_properties_sql,
no_safe_divide_sql,
no_tablesample_sql,
rename_func,
@ -68,6 +69,12 @@ def _struct_pack_sql(self, expression):
return f"STRUCT_PACK({', '.join(args)})"
def _datatype_sql(self, expression):
if expression.this == exp.DataType.Type.ARRAY:
return f"{self.expressions(expression, flat=True)}[]"
return self.datatype_sql(expression)
class DuckDB(Dialect):
class Tokenizer(Tokenizer):
KEYWORDS = {
@ -106,6 +113,8 @@ class DuckDB(Dialect):
}
class Generator(Generator):
STRUCT_DELIMITER = ("(", ")")
TRANSFORMS = {
**Generator.TRANSFORMS,
exp.ApproxDistinct: approx_count_distinct_sql,
@ -113,8 +122,9 @@ class DuckDB(Dialect):
exp.ArraySize: rename_func("ARRAY_LENGTH"),
exp.ArraySort: _array_sort_sql,
exp.ArraySum: rename_func("LIST_SUM"),
exp.DataType: _datatype_sql,
exp.DateAdd: _date_add,
exp.DateDiff: lambda self, e: f"""DATE_DIFF({self.sql(e, 'unit') or "'day'"}, {self.sql(e, 'expression')}, {self.sql(e, 'this')})""",
exp.DateDiff: lambda self, e: f"""DATE_DIFF({self.format_args(e.args.get("unit") or "'day'", e.expression, e.this)})""",
exp.DateStrToDate: lambda self, e: f"CAST({self.sql(e, 'this')} AS DATE)",
exp.DateToDi: lambda self, e: f"CAST(STRFTIME({self.sql(e, 'this')}, {DuckDB.dateint_format}) AS INT)",
exp.DiToDate: lambda self, e: f"CAST(STRPTIME(CAST({self.sql(e, 'this')} AS TEXT), {DuckDB.dateint_format}) AS DATE)",
@ -124,6 +134,7 @@ class DuckDB(Dialect):
exp.JSONBExtract: arrow_json_extract_sql,
exp.JSONBExtractScalar: arrow_json_extract_scalar_sql,
exp.Pivot: no_pivot_sql,
exp.Properties: no_properties_sql,
exp.RegexpLike: rename_func("REGEXP_MATCHES"),
exp.RegexpSplit: rename_func("STR_SPLIT_REGEX"),
exp.SafeDivide: no_safe_divide_sql,

8
sqlglot/dialects/hive.py
View file

@ -14,7 +14,7 @@ from sqlglot.dialects.dialect import (
var_map_sql,
)
from sqlglot.generator import Generator
from sqlglot.helper import csv, list_get
from sqlglot.helper import list_get
from sqlglot.parser import Parser, parse_var_map
from sqlglot.tokens import Tokenizer
@ -32,7 +32,7 @@ def _property_sql(self, expression):
def _str_to_unix(self, expression):
return f"UNIX_TIMESTAMP({csv(self.sql(expression, 'this'), _time_format(self, expression))})"
return f"UNIX_TIMESTAMP({self.format_args(expression.this, _time_format(self, expression))})"
def _str_to_date(self, expression):
@ -226,7 +226,7 @@ class Hive(Dialect):
exp.SchemaCommentProperty: lambda self, e: self.naked_property(e),
exp.SetAgg: rename_func("COLLECT_SET"),
exp.Split: lambda self, e: f"SPLIT({self.sql(e, 'this')}, CONCAT('\\\\Q', {self.sql(e, 'expression')}))",
exp.StrPosition: lambda self, e: f"LOCATE({csv(self.sql(e, 'substr'), self.sql(e, 'this'), self.sql(e, 'position'))})",
exp.StrPosition: lambda self, e: f"LOCATE({self.format_args(e.args.get('substr'), e.this, e.args.get('position'))})",
exp.StrToDate: _str_to_date,
exp.StrToTime: _str_to_time,
exp.StrToUnix: _str_to_unix,
@ -241,7 +241,7 @@ class Hive(Dialect):
exp.TsOrDsAdd: lambda self, e: f"DATE_ADD({self.sql(e, 'this')}, {self.sql(e, 'expression')})",
exp.TsOrDsToDate: _to_date_sql,
exp.TryCast: no_trycast_sql,
exp.UnixToStr: lambda self, e: f"FROM_UNIXTIME({csv(self.sql(e, 'this'), _time_format(self, e))})",
exp.UnixToStr: lambda self, e: f"FROM_UNIXTIME({self.format_args(e.this, _time_format(self, e))})",
exp.UnixToTime: rename_func("FROM_UNIXTIME"),
exp.UnixToTimeStr: rename_func("FROM_UNIXTIME"),
exp.PartitionedByProperty: lambda self, e: f"PARTITIONED BY {self.sql(e, 'value')}",

1
sqlglot/dialects/postgres.py
View file

@ -167,6 +167,7 @@ class Postgres(Dialect):
**Tokenizer.KEYWORDS,
"ALWAYS": TokenType.ALWAYS,
"BY DEFAULT": TokenType.BY_DEFAULT,
"COMMENT ON": TokenType.COMMENT_ON,
"IDENTITY": TokenType.IDENTITY,
"GENERATED": TokenType.GENERATED,
"DOUBLE PRECISION": TokenType.DOUBLE,

6
sqlglot/dialects/presto.py
View file

@ -11,7 +11,7 @@ from sqlglot.dialects.dialect import (
)
from sqlglot.dialects.mysql import MySQL
from sqlglot.generator import Generator
from sqlglot.helper import csv, list_get
from sqlglot.helper import list_get
from sqlglot.parser import Parser
from sqlglot.tokens import Tokenizer, TokenType
@ -26,7 +26,7 @@ def _concat_ws_sql(self, expression):
sep, *args = expression.expressions
sep = self.sql(sep)
if len(args) > 1:
return f"ARRAY_JOIN(ARRAY[{csv(*(self.sql(e) for e in args))}], {sep})"
return f"ARRAY_JOIN(ARRAY[{self.format_args(*args)}], {sep})"
return f"ARRAY_JOIN({self.sql(args[0])}, {sep})"
@ -66,7 +66,7 @@ def _no_sort_array(self, expression):
comparator = "(a, b) -> CASE WHEN a < b THEN 1 WHEN a > b THEN -1 ELSE 0 END"
else:
comparator = None
args = csv(self.sql(expression, "this"), comparator)
args = self.format_args(expression.this, comparator)
return f"ARRAY_SORT({args})"

5
sqlglot/dialects/tableau.py
View file

@ -1,7 +1,6 @@
from sqlglot import exp
from sqlglot.dialects.dialect import Dialect
from sqlglot.generator import Generator
from sqlglot.helper import list_get
from sqlglot.parser import Parser
@ -16,7 +15,7 @@ def _coalesce_sql(self, expression):
def _count_sql(self, expression):
this = expression.this
if isinstance(this, exp.Distinct):
return f"COUNTD({self.sql(this, 'this')})"
return f"COUNTD({self.expressions(this, flat=True)})"
return f"COUNT({self.sql(expression, 'this')})"
@ -33,5 +32,5 @@ class Tableau(Dialect):
FUNCTIONS = {
**Parser.FUNCTIONS,
"IFNULL": exp.Coalesce.from_arg_list,
"COUNTD": lambda args: exp.Count(this=exp.Distinct(this=list_get(args, 0))),
"COUNTD": lambda args: exp.Count(this=exp.Distinct(expressions=args)),
}