datafusion_expr/expr_fn.rs
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965
// Licensed to the Apache Software Foundation (ASF) under one
// or more contributor license agreements. See the NOTICE file
// distributed with this work for additional information
// regarding copyright ownership. The ASF licenses this file
// to you under the Apache License, Version 2.0 (the
// "License"); you may not use this file except in compliance
// with the License. You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing,
// software distributed under the License is distributed on an
// "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
// KIND, either express or implied. See the License for the
// specific language governing permissions and limitations
// under the License.
//! Functions for creating logical expressions
use crate::expr::{
AggregateFunction, BinaryExpr, Cast, Exists, GroupingSet, InList, InSubquery,
Placeholder, TryCast, Unnest, WildcardOptions, WindowFunction,
};
use crate::function::{
AccumulatorArgs, AccumulatorFactoryFunction, PartitionEvaluatorFactory,
StateFieldsArgs,
};
use crate::{
conditional_expressions::CaseBuilder, expr::Sort, logical_plan::Subquery,
AggregateUDF, Expr, LogicalPlan, Operator, PartitionEvaluator,
ScalarFunctionImplementation, ScalarUDF, Signature, Volatility,
};
use crate::{
AggregateUDFImpl, ColumnarValue, ScalarUDFImpl, WindowFrame, WindowUDF, WindowUDFImpl,
};
use arrow::compute::kernels::cast_utils::{
parse_interval_day_time, parse_interval_month_day_nano, parse_interval_year_month,
};
use arrow::datatypes::{DataType, Field};
use datafusion_common::{plan_err, Column, Result, ScalarValue, TableReference};
use datafusion_functions_window_common::field::WindowUDFFieldArgs;
use datafusion_functions_window_common::partition::PartitionEvaluatorArgs;
use sqlparser::ast::NullTreatment;
use std::any::Any;
use std::fmt::Debug;
use std::ops::Not;
use std::sync::Arc;
/// Create a column expression based on a qualified or unqualified column name. Will
/// normalize unquoted identifiers according to SQL rules (identifiers will become lowercase).
///
/// For example:
///
/// ```rust
/// # use datafusion_expr::col;
/// let c1 = col("a");
/// let c2 = col("A");
/// assert_eq!(c1, c2);
///
/// // note how quoting with double quotes preserves the case
/// let c3 = col(r#""A""#);
/// assert_ne!(c1, c3);
/// ```
pub fn col(ident: impl Into<Column>) -> Expr {
Expr::Column(ident.into())
}
/// Create an out reference column which hold a reference that has been resolved to a field
/// outside of the current plan.
pub fn out_ref_col(dt: DataType, ident: impl Into<Column>) -> Expr {
Expr::OuterReferenceColumn(dt, ident.into())
}
/// Create an unqualified column expression from the provided name, without normalizing
/// the column.
///
/// For example:
///
/// ```rust
/// # use datafusion_expr::{col, ident};
/// let c1 = ident("A"); // not normalized staying as column 'A'
/// let c2 = col("A"); // normalized via SQL rules becoming column 'a'
/// assert_ne!(c1, c2);
///
/// let c3 = col(r#""A""#);
/// assert_eq!(c1, c3);
///
/// let c4 = col("t1.a"); // parses as relation 't1' column 'a'
/// let c5 = ident("t1.a"); // parses as column 't1.a'
/// assert_ne!(c4, c5);
/// ```
pub fn ident(name: impl Into<String>) -> Expr {
Expr::Column(Column::from_name(name))
}
/// Create placeholder value that will be filled in (such as `$1`)
///
/// Note the parameter type can be inferred using [`Expr::infer_placeholder_types`]
///
/// # Example
///
/// ```rust
/// # use datafusion_expr::{placeholder};
/// let p = placeholder("$0"); // $0, refers to parameter 1
/// assert_eq!(p.to_string(), "$0")
/// ```
pub fn placeholder(id: impl Into<String>) -> Expr {
Expr::Placeholder(Placeholder {
id: id.into(),
data_type: None,
})
}
/// Create an '*' [`Expr::Wildcard`] expression that matches all columns
///
/// # Example
///
/// ```rust
/// # use datafusion_expr::{wildcard};
/// let p = wildcard();
/// assert_eq!(p.to_string(), "*")
/// ```
pub fn wildcard() -> Expr {
Expr::Wildcard {
qualifier: None,
options: WildcardOptions::default(),
}
}
/// Create an '*' [`Expr::Wildcard`] expression with the wildcard options
pub fn wildcard_with_options(options: WildcardOptions) -> Expr {
Expr::Wildcard {
qualifier: None,
options,
}
}
/// Create an 't.*' [`Expr::Wildcard`] expression that matches all columns from a specific table
///
/// # Example
///
/// ```rust
/// # use datafusion_common::TableReference;
/// # use datafusion_expr::{qualified_wildcard};
/// let p = qualified_wildcard(TableReference::bare("t"));
/// assert_eq!(p.to_string(), "t.*")
/// ```
pub fn qualified_wildcard(qualifier: impl Into<TableReference>) -> Expr {
Expr::Wildcard {
qualifier: Some(qualifier.into()),
options: WildcardOptions::default(),
}
}
/// Create an 't.*' [`Expr::Wildcard`] expression with the wildcard options
pub fn qualified_wildcard_with_options(
qualifier: impl Into<TableReference>,
options: WildcardOptions,
) -> Expr {
Expr::Wildcard {
qualifier: Some(qualifier.into()),
options,
}
}
/// Return a new expression `left <op> right`
pub fn binary_expr(left: Expr, op: Operator, right: Expr) -> Expr {
Expr::BinaryExpr(BinaryExpr::new(Box::new(left), op, Box::new(right)))
}
/// Return a new expression with a logical AND
pub fn and(left: Expr, right: Expr) -> Expr {
Expr::BinaryExpr(BinaryExpr::new(
Box::new(left),
Operator::And,
Box::new(right),
))
}
/// Return a new expression with a logical OR
pub fn or(left: Expr, right: Expr) -> Expr {
Expr::BinaryExpr(BinaryExpr::new(
Box::new(left),
Operator::Or,
Box::new(right),
))
}
/// Return a new expression with a logical NOT
pub fn not(expr: Expr) -> Expr {
expr.not()
}
/// Return a new expression with bitwise AND
pub fn bitwise_and(left: Expr, right: Expr) -> Expr {
Expr::BinaryExpr(BinaryExpr::new(
Box::new(left),
Operator::BitwiseAnd,
Box::new(right),
))
}
/// Return a new expression with bitwise OR
pub fn bitwise_or(left: Expr, right: Expr) -> Expr {
Expr::BinaryExpr(BinaryExpr::new(
Box::new(left),
Operator::BitwiseOr,
Box::new(right),
))
}
/// Return a new expression with bitwise XOR
pub fn bitwise_xor(left: Expr, right: Expr) -> Expr {
Expr::BinaryExpr(BinaryExpr::new(
Box::new(left),
Operator::BitwiseXor,
Box::new(right),
))
}
/// Return a new expression with bitwise SHIFT RIGHT
pub fn bitwise_shift_right(left: Expr, right: Expr) -> Expr {
Expr::BinaryExpr(BinaryExpr::new(
Box::new(left),
Operator::BitwiseShiftRight,
Box::new(right),
))
}
/// Return a new expression with bitwise SHIFT LEFT
pub fn bitwise_shift_left(left: Expr, right: Expr) -> Expr {
Expr::BinaryExpr(BinaryExpr::new(
Box::new(left),
Operator::BitwiseShiftLeft,
Box::new(right),
))
}
/// Create an in_list expression
pub fn in_list(expr: Expr, list: Vec<Expr>, negated: bool) -> Expr {
Expr::InList(InList::new(Box::new(expr), list, negated))
}
/// Create an EXISTS subquery expression
pub fn exists(subquery: Arc<LogicalPlan>) -> Expr {
let outer_ref_columns = subquery.all_out_ref_exprs();
Expr::Exists(Exists {
subquery: Subquery {
subquery,
outer_ref_columns,
},
negated: false,
})
}
/// Create a NOT EXISTS subquery expression
pub fn not_exists(subquery: Arc<LogicalPlan>) -> Expr {
let outer_ref_columns = subquery.all_out_ref_exprs();
Expr::Exists(Exists {
subquery: Subquery {
subquery,
outer_ref_columns,
},
negated: true,
})
}
/// Create an IN subquery expression
pub fn in_subquery(expr: Expr, subquery: Arc<LogicalPlan>) -> Expr {
let outer_ref_columns = subquery.all_out_ref_exprs();
Expr::InSubquery(InSubquery::new(
Box::new(expr),
Subquery {
subquery,
outer_ref_columns,
},
false,
))
}
/// Create a NOT IN subquery expression
pub fn not_in_subquery(expr: Expr, subquery: Arc<LogicalPlan>) -> Expr {
let outer_ref_columns = subquery.all_out_ref_exprs();
Expr::InSubquery(InSubquery::new(
Box::new(expr),
Subquery {
subquery,
outer_ref_columns,
},
true,
))
}
/// Create a scalar subquery expression
pub fn scalar_subquery(subquery: Arc<LogicalPlan>) -> Expr {
let outer_ref_columns = subquery.all_out_ref_exprs();
Expr::ScalarSubquery(Subquery {
subquery,
outer_ref_columns,
})
}
/// Create a grouping set
pub fn grouping_set(exprs: Vec<Vec<Expr>>) -> Expr {
Expr::GroupingSet(GroupingSet::GroupingSets(exprs))
}
/// Create a grouping set for all combination of `exprs`
pub fn cube(exprs: Vec<Expr>) -> Expr {
Expr::GroupingSet(GroupingSet::Cube(exprs))
}
/// Create a grouping set for rollup
pub fn rollup(exprs: Vec<Expr>) -> Expr {
Expr::GroupingSet(GroupingSet::Rollup(exprs))
}
/// Create a cast expression
pub fn cast(expr: Expr, data_type: DataType) -> Expr {
Expr::Cast(Cast::new(Box::new(expr), data_type))
}
/// Create a try cast expression
pub fn try_cast(expr: Expr, data_type: DataType) -> Expr {
Expr::TryCast(TryCast::new(Box::new(expr), data_type))
}
/// Create is null expression
pub fn is_null(expr: Expr) -> Expr {
Expr::IsNull(Box::new(expr))
}
/// Create is true expression
pub fn is_true(expr: Expr) -> Expr {
Expr::IsTrue(Box::new(expr))
}
/// Create is not true expression
pub fn is_not_true(expr: Expr) -> Expr {
Expr::IsNotTrue(Box::new(expr))
}
/// Create is false expression
pub fn is_false(expr: Expr) -> Expr {
Expr::IsFalse(Box::new(expr))
}
/// Create is not false expression
pub fn is_not_false(expr: Expr) -> Expr {
Expr::IsNotFalse(Box::new(expr))
}
/// Create is unknown expression
pub fn is_unknown(expr: Expr) -> Expr {
Expr::IsUnknown(Box::new(expr))
}
/// Create is not unknown expression
pub fn is_not_unknown(expr: Expr) -> Expr {
Expr::IsNotUnknown(Box::new(expr))
}
/// Create a CASE WHEN statement with literal WHEN expressions for comparison to the base expression.
pub fn case(expr: Expr) -> CaseBuilder {
CaseBuilder::new(Some(Box::new(expr)), vec![], vec![], None)
}
/// Create a CASE WHEN statement with boolean WHEN expressions and no base expression.
pub fn when(when: Expr, then: Expr) -> CaseBuilder {
CaseBuilder::new(None, vec![when], vec![then], None)
}
/// Create a Unnest expression
pub fn unnest(expr: Expr) -> Expr {
Expr::Unnest(Unnest {
expr: Box::new(expr),
})
}
/// Convenience method to create a new user defined scalar function (UDF) with a
/// specific signature and specific return type.
///
/// Note this function does not expose all available features of [`ScalarUDF`],
/// such as
///
/// * computing return types based on input types
/// * multiple [`Signature`]s
/// * aliases
///
/// See [`ScalarUDF`] for details and examples on how to use the full
/// functionality.
pub fn create_udf(
name: &str,
input_types: Vec<DataType>,
return_type: DataType,
volatility: Volatility,
fun: ScalarFunctionImplementation,
) -> ScalarUDF {
ScalarUDF::from(SimpleScalarUDF::new(
name,
input_types,
return_type,
volatility,
fun,
))
}
/// Implements [`ScalarUDFImpl`] for functions that have a single signature and
/// return type.
pub struct SimpleScalarUDF {
name: String,
signature: Signature,
return_type: DataType,
fun: ScalarFunctionImplementation,
}
impl Debug for SimpleScalarUDF {
fn fmt(&self, f: &mut std::fmt::Formatter) -> std::fmt::Result {
f.debug_struct("ScalarUDF")
.field("name", &self.name)
.field("signature", &self.signature)
.field("fun", &"<FUNC>")
.finish()
}
}
impl SimpleScalarUDF {
/// Create a new `SimpleScalarUDF` from a name, input types, return type and
/// implementation. Implementing [`ScalarUDFImpl`] allows more flexibility
pub fn new(
name: impl Into<String>,
input_types: Vec<DataType>,
return_type: DataType,
volatility: Volatility,
fun: ScalarFunctionImplementation,
) -> Self {
let name = name.into();
let signature = Signature::exact(input_types, volatility);
Self {
name,
signature,
return_type,
fun,
}
}
}
impl ScalarUDFImpl for SimpleScalarUDF {
fn as_any(&self) -> &dyn Any {
self
}
fn name(&self) -> &str {
&self.name
}
fn signature(&self) -> &Signature {
&self.signature
}
fn return_type(&self, _arg_types: &[DataType]) -> Result<DataType> {
Ok(self.return_type.clone())
}
fn invoke(&self, args: &[ColumnarValue]) -> Result<ColumnarValue> {
(self.fun)(args)
}
}
/// Creates a new UDAF with a specific signature, state type and return type.
/// The signature and state type must match the `Accumulator's implementation`.
pub fn create_udaf(
name: &str,
input_type: Vec<DataType>,
return_type: Arc<DataType>,
volatility: Volatility,
accumulator: AccumulatorFactoryFunction,
state_type: Arc<Vec<DataType>>,
) -> AggregateUDF {
let return_type = Arc::unwrap_or_clone(return_type);
let state_type = Arc::unwrap_or_clone(state_type);
let state_fields = state_type
.into_iter()
.enumerate()
.map(|(i, t)| Field::new(format!("{i}"), t, true))
.collect::<Vec<_>>();
AggregateUDF::from(SimpleAggregateUDF::new(
name,
input_type,
return_type,
volatility,
accumulator,
state_fields,
))
}
/// Implements [`AggregateUDFImpl`] for functions that have a single signature and
/// return type.
pub struct SimpleAggregateUDF {
name: String,
signature: Signature,
return_type: DataType,
accumulator: AccumulatorFactoryFunction,
state_fields: Vec<Field>,
}
impl Debug for SimpleAggregateUDF {
fn fmt(&self, f: &mut std::fmt::Formatter) -> std::fmt::Result {
f.debug_struct("AggregateUDF")
.field("name", &self.name)
.field("signature", &self.signature)
.field("fun", &"<FUNC>")
.finish()
}
}
impl SimpleAggregateUDF {
/// Create a new `AggregateUDFImpl` from a name, input types, return type, state type and
/// implementation. Implementing [`AggregateUDFImpl`] allows more flexibility
pub fn new(
name: impl Into<String>,
input_type: Vec<DataType>,
return_type: DataType,
volatility: Volatility,
accumulator: AccumulatorFactoryFunction,
state_fields: Vec<Field>,
) -> Self {
let name = name.into();
let signature = Signature::exact(input_type, volatility);
Self {
name,
signature,
return_type,
accumulator,
state_fields,
}
}
pub fn new_with_signature(
name: impl Into<String>,
signature: Signature,
return_type: DataType,
accumulator: AccumulatorFactoryFunction,
state_fields: Vec<Field>,
) -> Self {
let name = name.into();
Self {
name,
signature,
return_type,
accumulator,
state_fields,
}
}
}
impl AggregateUDFImpl for SimpleAggregateUDF {
fn as_any(&self) -> &dyn Any {
self
}
fn name(&self) -> &str {
&self.name
}
fn signature(&self) -> &Signature {
&self.signature
}
fn return_type(&self, _arg_types: &[DataType]) -> Result<DataType> {
Ok(self.return_type.clone())
}
fn accumulator(
&self,
acc_args: AccumulatorArgs,
) -> Result<Box<dyn crate::Accumulator>> {
(self.accumulator)(acc_args)
}
fn state_fields(&self, _args: StateFieldsArgs) -> Result<Vec<Field>> {
Ok(self.state_fields.clone())
}
}
/// Creates a new UDWF with a specific signature, state type and return type.
///
/// The signature and state type must match the [`PartitionEvaluator`]'s implementation`.
///
/// [`PartitionEvaluator`]: crate::PartitionEvaluator
pub fn create_udwf(
name: &str,
input_type: DataType,
return_type: Arc<DataType>,
volatility: Volatility,
partition_evaluator_factory: PartitionEvaluatorFactory,
) -> WindowUDF {
let return_type = Arc::unwrap_or_clone(return_type);
WindowUDF::from(SimpleWindowUDF::new(
name,
input_type,
return_type,
volatility,
partition_evaluator_factory,
))
}
/// Implements [`WindowUDFImpl`] for functions that have a single signature and
/// return type.
pub struct SimpleWindowUDF {
name: String,
signature: Signature,
return_type: DataType,
partition_evaluator_factory: PartitionEvaluatorFactory,
}
impl Debug for SimpleWindowUDF {
fn fmt(&self, f: &mut std::fmt::Formatter) -> std::fmt::Result {
f.debug_struct("WindowUDF")
.field("name", &self.name)
.field("signature", &self.signature)
.field("return_type", &"<func>")
.field("partition_evaluator_factory", &"<FUNC>")
.finish()
}
}
impl SimpleWindowUDF {
/// Create a new `SimpleWindowUDF` from a name, input types, return type and
/// implementation. Implementing [`WindowUDFImpl`] allows more flexibility
pub fn new(
name: impl Into<String>,
input_type: DataType,
return_type: DataType,
volatility: Volatility,
partition_evaluator_factory: PartitionEvaluatorFactory,
) -> Self {
let name = name.into();
let signature = Signature::exact([input_type].to_vec(), volatility);
Self {
name,
signature,
return_type,
partition_evaluator_factory,
}
}
}
impl WindowUDFImpl for SimpleWindowUDF {
fn as_any(&self) -> &dyn Any {
self
}
fn name(&self) -> &str {
&self.name
}
fn signature(&self) -> &Signature {
&self.signature
}
fn partition_evaluator(
&self,
_partition_evaluator_args: PartitionEvaluatorArgs,
) -> Result<Box<dyn PartitionEvaluator>> {
(self.partition_evaluator_factory)()
}
fn field(&self, field_args: WindowUDFFieldArgs) -> Result<Field> {
Ok(Field::new(
field_args.name(),
self.return_type.clone(),
true,
))
}
}
pub fn interval_year_month_lit(value: &str) -> Expr {
let interval = parse_interval_year_month(value).ok();
Expr::Literal(ScalarValue::IntervalYearMonth(interval))
}
pub fn interval_datetime_lit(value: &str) -> Expr {
let interval = parse_interval_day_time(value).ok();
Expr::Literal(ScalarValue::IntervalDayTime(interval))
}
pub fn interval_month_day_nano_lit(value: &str) -> Expr {
let interval = parse_interval_month_day_nano(value).ok();
Expr::Literal(ScalarValue::IntervalMonthDayNano(interval))
}
/// Extensions for configuring [`Expr::AggregateFunction`] or [`Expr::WindowFunction`]
///
/// Adds methods to [`Expr`] that make it easy to set optional options
/// such as `ORDER BY`, `FILTER` and `DISTINCT`
///
/// # Example
/// ```no_run
/// # use datafusion_common::Result;
/// # use datafusion_expr::test::function_stub::count;
/// # use sqlparser::ast::NullTreatment;
/// # use datafusion_expr::{ExprFunctionExt, lit, Expr, col};
/// # // first_value is an aggregate function in another crate
/// # fn first_value(_arg: Expr) -> Expr {
/// unimplemented!() }
/// # fn main() -> Result<()> {
/// // Create an aggregate count, filtering on column y > 5
/// let agg = count(col("x")).filter(col("y").gt(lit(5))).build()?;
///
/// // Find the first value in an aggregate sorted by column y
/// // equivalent to:
/// // `FIRST_VALUE(x ORDER BY y ASC IGNORE NULLS)`
/// let sort_expr = col("y").sort(true, true);
/// let agg = first_value(col("x"))
/// .order_by(vec![sort_expr])
/// .null_treatment(NullTreatment::IgnoreNulls)
/// .build()?;
///
/// // Create a window expression for percent rank partitioned on column a
/// // equivalent to:
/// // `PERCENT_RANK() OVER (PARTITION BY a ORDER BY b ASC NULLS LAST IGNORE NULLS)`
/// // percent_rank is an udwf function in another crate
/// # fn percent_rank() -> Expr {
/// unimplemented!() }
/// let window = percent_rank()
/// .partition_by(vec![col("a")])
/// .order_by(vec![col("b").sort(true, true)])
/// .null_treatment(NullTreatment::IgnoreNulls)
/// .build()?;
/// # Ok(())
/// # }
/// ```
pub trait ExprFunctionExt {
/// Add `ORDER BY <order_by>`
fn order_by(self, order_by: Vec<Sort>) -> ExprFuncBuilder;
/// Add `FILTER <filter>`
fn filter(self, filter: Expr) -> ExprFuncBuilder;
/// Add `DISTINCT`
fn distinct(self) -> ExprFuncBuilder;
/// Add `RESPECT NULLS` or `IGNORE NULLS`
fn null_treatment(
self,
null_treatment: impl Into<Option<NullTreatment>>,
) -> ExprFuncBuilder;
/// Add `PARTITION BY`
fn partition_by(self, partition_by: Vec<Expr>) -> ExprFuncBuilder;
/// Add appropriate window frame conditions
fn window_frame(self, window_frame: WindowFrame) -> ExprFuncBuilder;
}
#[derive(Debug, Clone)]
pub enum ExprFuncKind {
Aggregate(AggregateFunction),
Window(WindowFunction),
}
/// Implementation of [`ExprFunctionExt`].
///
/// See [`ExprFunctionExt`] for usage and examples
#[derive(Debug, Clone)]
pub struct ExprFuncBuilder {
fun: Option<ExprFuncKind>,
order_by: Option<Vec<Sort>>,
filter: Option<Expr>,
distinct: bool,
null_treatment: Option<NullTreatment>,
partition_by: Option<Vec<Expr>>,
window_frame: Option<WindowFrame>,
}
impl ExprFuncBuilder {
/// Create a new `ExprFuncBuilder`, see [`ExprFunctionExt`]
fn new(fun: Option<ExprFuncKind>) -> Self {
Self {
fun,
order_by: None,
filter: None,
distinct: false,
null_treatment: None,
partition_by: None,
window_frame: None,
}
}
/// Updates and returns the in progress [`Expr::AggregateFunction`] or [`Expr::WindowFunction`]
///
/// # Errors:
///
/// Returns an error if this builder [`ExprFunctionExt`] was used with an
/// `Expr` variant other than [`Expr::AggregateFunction`] or [`Expr::WindowFunction`]
pub fn build(self) -> Result<Expr> {
let Self {
fun,
order_by,
filter,
distinct,
null_treatment,
partition_by,
window_frame,
} = self;
let Some(fun) = fun else {
return plan_err!(
"ExprFunctionExt can only be used with Expr::AggregateFunction or Expr::WindowFunction"
);
};
let fun_expr = match fun {
ExprFuncKind::Aggregate(mut udaf) => {
udaf.order_by = order_by;
udaf.filter = filter.map(Box::new);
udaf.distinct = distinct;
udaf.null_treatment = null_treatment;
Expr::AggregateFunction(udaf)
}
ExprFuncKind::Window(mut udwf) => {
let has_order_by = order_by.as_ref().map(|o| !o.is_empty());
udwf.order_by = order_by.unwrap_or_default();
udwf.partition_by = partition_by.unwrap_or_default();
udwf.window_frame =
window_frame.unwrap_or(WindowFrame::new(has_order_by));
udwf.null_treatment = null_treatment;
Expr::WindowFunction(udwf)
}
};
Ok(fun_expr)
}
}
impl ExprFunctionExt for ExprFuncBuilder {
/// Add `ORDER BY <order_by>`
fn order_by(mut self, order_by: Vec<Sort>) -> ExprFuncBuilder {
self.order_by = Some(order_by);
self
}
/// Add `FILTER <filter>`
fn filter(mut self, filter: Expr) -> ExprFuncBuilder {
self.filter = Some(filter);
self
}
/// Add `DISTINCT`
fn distinct(mut self) -> ExprFuncBuilder {
self.distinct = true;
self
}
/// Add `RESPECT NULLS` or `IGNORE NULLS`
fn null_treatment(
mut self,
null_treatment: impl Into<Option<NullTreatment>>,
) -> ExprFuncBuilder {
self.null_treatment = null_treatment.into();
self
}
fn partition_by(mut self, partition_by: Vec<Expr>) -> ExprFuncBuilder {
self.partition_by = Some(partition_by);
self
}
fn window_frame(mut self, window_frame: WindowFrame) -> ExprFuncBuilder {
self.window_frame = Some(window_frame);
self
}
}
impl ExprFunctionExt for Expr {
fn order_by(self, order_by: Vec<Sort>) -> ExprFuncBuilder {
let mut builder = match self {
Expr::AggregateFunction(udaf) => {
ExprFuncBuilder::new(Some(ExprFuncKind::Aggregate(udaf)))
}
Expr::WindowFunction(udwf) => {
ExprFuncBuilder::new(Some(ExprFuncKind::Window(udwf)))
}
_ => ExprFuncBuilder::new(None),
};
if builder.fun.is_some() {
builder.order_by = Some(order_by);
}
builder
}
fn filter(self, filter: Expr) -> ExprFuncBuilder {
match self {
Expr::AggregateFunction(udaf) => {
let mut builder =
ExprFuncBuilder::new(Some(ExprFuncKind::Aggregate(udaf)));
builder.filter = Some(filter);
builder
}
_ => ExprFuncBuilder::new(None),
}
}
fn distinct(self) -> ExprFuncBuilder {
match self {
Expr::AggregateFunction(udaf) => {
let mut builder =
ExprFuncBuilder::new(Some(ExprFuncKind::Aggregate(udaf)));
builder.distinct = true;
builder
}
_ => ExprFuncBuilder::new(None),
}
}
fn null_treatment(
self,
null_treatment: impl Into<Option<NullTreatment>>,
) -> ExprFuncBuilder {
let mut builder = match self {
Expr::AggregateFunction(udaf) => {
ExprFuncBuilder::new(Some(ExprFuncKind::Aggregate(udaf)))
}
Expr::WindowFunction(udwf) => {
ExprFuncBuilder::new(Some(ExprFuncKind::Window(udwf)))
}
_ => ExprFuncBuilder::new(None),
};
if builder.fun.is_some() {
builder.null_treatment = null_treatment.into();
}
builder
}
fn partition_by(self, partition_by: Vec<Expr>) -> ExprFuncBuilder {
match self {
Expr::WindowFunction(udwf) => {
let mut builder = ExprFuncBuilder::new(Some(ExprFuncKind::Window(udwf)));
builder.partition_by = Some(partition_by);
builder
}
_ => ExprFuncBuilder::new(None),
}
}
fn window_frame(self, window_frame: WindowFrame) -> ExprFuncBuilder {
match self {
Expr::WindowFunction(udwf) => {
let mut builder = ExprFuncBuilder::new(Some(ExprFuncKind::Window(udwf)));
builder.window_frame = Some(window_frame);
builder
}
_ => ExprFuncBuilder::new(None),
}
}
}
#[cfg(test)]
mod test {
use super::*;
#[test]
fn filter_is_null_and_is_not_null() {
let col_null = col("col1");
let col_not_null = ident("col2");
assert_eq!(format!("{}", col_null.is_null()), "col1 IS NULL");
assert_eq!(
format!("{}", col_not_null.is_not_null()),
"col2 IS NOT NULL"
);
}
}