Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add Apache Drill Dialect to SqlGlot #707

Merged
merged 6 commits into from
Nov 15, 2022
Merged
Show file tree
Hide file tree
Changes from 5 commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions sqlglot/dialects/__init__.py
Original file line number Diff line number Diff line change
Expand Up @@ -2,6 +2,7 @@
from sqlglot.dialects.clickhouse import ClickHouse
from sqlglot.dialects.databricks import Databricks
from sqlglot.dialects.dialect import Dialect, Dialects
from sqlglot.dialects.drill import Drill
from sqlglot.dialects.duckdb import DuckDB
from sqlglot.dialects.hive import Hive
from sqlglot.dialects.mysql import MySQL
Expand Down
1 change: 1 addition & 0 deletions sqlglot/dialects/dialect.py
Original file line number Diff line number Diff line change
Expand Up @@ -32,6 +32,7 @@ class Dialects(str, Enum):
TRINO = "trino"
TSQL = "tsql"
DATABRICKS = "databricks"
DRILL = "drill"


class _Dialect(type):
Expand Down
174 changes: 174 additions & 0 deletions sqlglot/dialects/drill.py
Original file line number Diff line number Diff line change
@@ -0,0 +1,174 @@
from __future__ import annotations

from sqlglot import exp, generator, parser, tokens
from sqlglot.dialects.dialect import (
Dialect,
create_with_partitions_sql,
format_time_lambda,
no_pivot_sql,
no_trycast_sql,
rename_func,
str_position_sql,
)
from sqlglot.dialects.postgres import _lateral_sql
from sqlglot.tokens import TokenType


def _to_timestamp(args):
# TO_TIMESTAMP accepts either a single double argument or (text, text)
if len(args) == 1 and args[0].is_number:
return exp.UnixToTime.from_arg_list(args)
return format_time_lambda(exp.StrToTime, "drill")(args)


def _str_to_time_sql(self, expression):
return f"STRPTIME({self.sql(expression, 'this')}, {self.format_time(expression)})"


def _ts_or_ds_to_date_sql(self, expression):
time_format = self.format_time(expression)
if time_format and time_format not in (Drill.time_format, Drill.date_format):
return f"CAST({_str_to_time_sql(self, expression)} AS DATE)"
return f"CAST({self.sql(expression, 'this')} AS DATE)"


def _date_add_sql(kind):
def func(self, expression):
this = self.sql(expression, "this")
unit = expression.text("unit").upper() or "DAY"
expression = self.sql(expression, "expression")
return f"DATE_{kind}({this}, INTERVAL '{expression}' {unit})"

return func


def if_sql(self, expression):
"""
Drill requires backticks around certain SQL reserved words, IF being one of them, This function
adds the backticks around the keyword IF.
Args:
self: The Drill dialect
expression: The input IF expression

Returns: The expression with IF in backticks.

"""
expressions = self.format_args(
expression.this, expression.args.get("true"), expression.args.get("false")
)
return f"`IF`({expressions})"


def _str_to_date(self, expression):
this = self.sql(expression, "this")
time_format = self.format_time(expression)
if time_format == Drill.date_format:
return f"CAST({this} AS DATE)"
return f"TO_DATE({this}, {time_format})"


class Drill(Dialect):
normalize_functions = None
null_ordering = "nulls_are_last"
date_format = "'yyyy-MM-dd'"
dateint_format = "'yyyyMMdd'"
time_format = "'yyyy-MM-dd HH:mm:ss'"

time_mapping = {
"y": "%Y",
"Y": "%Y",
"YYYY": "%Y",
"yyyy": "%Y",
"YY": "%y",
"yy": "%y",
"MMMM": "%B",
"MMM": "%b",
"MM": "%m",
"M": "%-m",
"dd": "%d",
"d": "%-d",
"HH": "%H",
"H": "%-H",
"hh": "%I",
"h": "%-I",
"mm": "%M",
"m": "%-M",
"ss": "%S",
"s": "%-S",
"SSSSSS": "%f",
"a": "%p",
"DD": "%j",
"D": "%-j",
"E": "%a",
"EE": "%a",
"EEE": "%a",
"EEEE": "%A",
"''T''": "T",
}

class Tokenizer(tokens.Tokenizer):
QUOTES = ["'"]
IDENTIFIERS = ["`"]
ESCAPES = ["\\"]
ENCODE = "utf-8"

KEYWORDS = {**tokens.Tokenizer.KEYWORDS, "VARBINARY": TokenType.BINARY}
Copy link
Owner

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

why did you add varbinary? that's already a type

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Removed. Drill doesn't have a BINARY data type, just a VARBINARY, so that's what I was trying to do here, but I think that is covered elsewhere.

Copy link
Owner

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

then it should instead be in the generator binary mapping to varbinary, look at what presto does

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Gotcha.... and it's already there. 👍

exp.DataType.Type.SMALLINT: "INTEGER",
exp.DataType.Type.TINYINT: "INTEGER",
exp.DataType.Type.BINARY: "VARBINARY",
exp.DataType.Type.TEXT: "VARCHAR",


normalize_functions = None

class Parser(parser.Parser):
STRICT_CAST = False

FUNCTIONS = {
**parser.Parser.FUNCTIONS,
"TO_TIMESTAMP": exp.TimeStrToTime.from_arg_list,
"TO_CHAR": format_time_lambda(exp.TimeToStr, "drill"),
}

class Generator(generator.Generator):
TYPE_MAPPING = {
**generator.Generator.TYPE_MAPPING,
exp.DataType.Type.INT: "INTEGER",
exp.DataType.Type.SMALLINT: "INTEGER",
exp.DataType.Type.TINYINT: "INTEGER",
exp.DataType.Type.BINARY: "VARBINARY",
exp.DataType.Type.TEXT: "VARCHAR",
exp.DataType.Type.NCHAR: "VARCHAR",
exp.DataType.Type.TIMESTAMPLTZ: "TIMESTAMP",
exp.DataType.Type.TIMESTAMPTZ: "TIMESTAMP",
exp.DataType.Type.DATETIME: "TIMESTAMP",
}

ROOT_PROPERTIES = {exp.PartitionedByProperty}

TRANSFORMS = {
**generator.Generator.TRANSFORMS,
exp.CurrentTimestamp: lambda *_: "CURRENT_TIMESTAMP",
exp.Lateral: _lateral_sql,
exp.ArrayContains: rename_func("REPEATED_CONTAINS"),
exp.ArraySize: rename_func("REPEATED_COUNT"),
exp.Create: create_with_partitions_sql,
exp.DateAdd: _date_add_sql("ADD"),
exp.DateStrToDate: lambda self, e: f"CAST({self.sql(e, 'this')} AS DATE)",
exp.DateSub: _date_add_sql("SUB"),
exp.DateToDi: lambda self, e: f"CAST(TO_DATE({self.sql(e, 'this')}, {Drill.dateint_format}) AS INT)",
exp.DiToDate: lambda self, e: f"TO_DATE(CAST({self.sql(e, 'this')} AS VARCHAR), {Drill.dateint_format})",
exp.If: if_sql,
exp.ILike: lambda self, e: f" {self.sql(e, 'this')} `ILIKE` {self.sql(e, 'expression')}",
exp.Levenshtein: rename_func("LEVENSHTEIN_DISTANCE"),
exp.PartitionedByProperty: lambda self, e: f"PARTITION BY {self.sql(e, 'value')}",
exp.Pivot: no_pivot_sql,
exp.RegexpLike: rename_func("REGEXP_MATCHES"),
exp.StrPosition: str_position_sql,
exp.StrToDate: _str_to_date,
exp.StrToTime: lambda self, e: f"TO_TIMESTAMP({self.sql(e, 'this')}, {self.format_time(e)})",
exp.TimeStrToDate: lambda self, e: f"CAST({self.sql(e, 'this')} AS DATE)",
exp.TimeStrToTime: lambda self, e: f"CAST({self.sql(e, 'this')} AS TIMESTAMP)",
exp.TimeStrToUnix: rename_func("UNIX_TIMESTAMP"),
exp.TimeToStr: lambda self, e: f"TO_CHAR({self.sql(e, 'this')}, {self.format_time(e)})",
exp.TimeToUnix: rename_func("UNIX_TIMESTAMP"),
exp.TryCast: no_trycast_sql,
exp.TsOrDsAdd: lambda self, e: f"DATE_ADD(CAST({self.sql(e,'this')} AS DATE), INTERVAL '{self.sql(e,'expression')}' DAY)",
exp.TsOrDsToDate: _ts_or_ds_to_date_sql,
exp.TsOrDiToDi: lambda self, e: f"CAST(SUBSTR(REPLACE(CAST({self.sql(e, 'this')} AS VARCHAR), '-', ''), 1, 8) AS INT)",
}
Loading