native/spark-expr/src/lib.rs

// 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.

// The clippy throws an error if the reference clone not wrapped into `Arc::clone`
// The lint makes easier for code reader/reviewer separate references clones from more heavyweight ones
#![deny(clippy::clone_on_ref_ptr)]

mod error;
mod query_context;

pub mod kernels;
pub use kernels::temporal::date_trunc_dyn;
mod static_invoke;
pub use static_invoke::*;

mod struct_funcs;
pub use struct_funcs::{CreateNamedStruct, GetStructField};

mod csv_funcs;
mod json_funcs;
pub mod test_common;
mod unbound;
pub use unbound::UnboundColumn;
mod predicate_funcs;
pub mod utils;
pub use predicate_funcs::{spark_isnan, RLike};

mod agg_funcs;
mod array_funcs;
mod comet_scalar_funcs;
pub mod hash_funcs;

mod string_funcs;

mod datetime_funcs;
mod iceberg_funcs;
pub use agg_funcs::*;
pub use iceberg_funcs::{SparkIcebergBucket, SparkIcebergTemporalTransform, SparkIcebergTruncate};

pub use cast::{spark_cast, Cast, SparkCastOptions};

mod bloom_filter;
pub use bloom_filter::{BloomFilterAgg, BloomFilterMightContain, SparkBloomFilterVersion};

pub mod jvm_udf;
mod subquery;
pub use subquery::Subquery;

mod conditional_funcs;
mod conversion_funcs;
mod float_semantics;
pub use float_semantics::{
    canonicalize_nan, normalize_floats, NormalizeNaNAndZero, NormalizeNestedFloats,
};
mod hll_scalar;
pub use hll_scalar::spark_hll_sketch_estimate;
pub use hll_scalar::spark_hll_union;
mod map_funcs;
pub use map_funcs::{spark_map_sort, SparkMapExtract, SparkMapFromArrays};
mod math_funcs;
mod nondetermenistic_funcs;
pub mod url_funcs;

pub use array_funcs::*;
pub use conditional_funcs::*;
pub use conversion_funcs::*;
pub use nondetermenistic_funcs::*;

pub use comet_scalar_funcs::{
    create_comet_physical_fun, create_comet_physical_fun_with_eval_mode,
    register_all_comet_functions,
};
pub use csv_funcs::*;
pub use datetime_funcs::{
    spark_day_name, spark_month_name, spark_seconds_of_time, spark_to_time, SparkDateDiff,
    SparkDateFromUnixDate, SparkDateTrunc, SparkDayOfWeek, SparkHour, SparkHoursTransform,
    SparkMakeDate, SparkMakeInterval, SparkMakeTime, SparkMinute, SparkNextDay, SparkSecond,
    SparkSecondsToTimestamp, SparkUnixTimestamp, SparkWeekDay, TimestampTruncExpr,
};
pub use error::{decimal_overflow_error, SparkError, SparkErrorWithContext, SparkResult};
pub use hash_funcs::*;
pub use json_funcs::{FromJson, JsonArrayLength, ToJson};
pub use math_funcs::{
    abs, checked_add, checked_div, checked_mul, checked_sub, create_modulo_expr,
    create_negate_expr, spark_ceil, spark_decimal_div, spark_decimal_integral_div, spark_floor,
    spark_log, spark_make_decimal, spark_modulo, spark_pow, spark_round, spark_signum, spark_sqrt,
    spark_unhex, spark_unscaled_value, CheckOverflow, CheckedBinaryExpr,
    DecimalRescaleCheckOverflow, NegativeExpr, SparkGreatestLeast, WideDecimalBinaryExpr,
    WideDecimalOp,
};
pub use query_context::{create_query_context_map, QueryContext, QueryContextMap};
pub use string_funcs::*;

/// Spark supports three evaluation modes when evaluating expressions, which affect
/// the behavior when processing input values that are invalid or would result in an
/// error, such as divide by zero errors, and also affects behavior when converting
/// between types.
#[derive(Debug, Hash, PartialEq, Eq, Clone, Copy)]
pub enum EvalMode {
    /// Legacy is the default behavior in Spark prior to Spark 4.0. This mode silently ignores
    /// or replaces errors during SQL operations. Operations resulting in errors (like
    /// division by zero) will produce NULL values instead of failing. Legacy mode also
    /// enables implicit type conversions.
    Legacy,
    /// Adheres to the ANSI SQL standard for error handling by throwing exceptions for
    /// operations that result in errors. Does not perform implicit type conversions.
    Ansi,
    /// Same as Ansi mode, except that it converts errors to NULL values without
    /// failing the entire query.
    Try,
}

#[derive(Debug, Hash, PartialEq, Eq, Clone, Copy)]
pub enum BinaryOutputStyle {
    Utf8,
    Basic,
    Base64,
    Hex,
    HexDiscrete,
}

pub(crate) fn arithmetic_overflow_error(from_type: &str) -> SparkError {
    SparkError::ArithmeticOverflow {
        from_type: from_type.to_string(),
        function_name: String::new(),
    }
}

pub(crate) fn decimal_sum_overflow_error(function_name: &str) -> SparkError {
    SparkError::DecimalSumOverflow {
        function_name: function_name.to_string(),
    }
}

pub(crate) fn divide_by_zero_error() -> SparkError {
    SparkError::DivideByZero
}

pub(crate) fn integral_divide_overflow_error() -> SparkError {
    SparkError::IntegralDivideOverflow
}

pub(crate) fn remainder_by_zero_error() -> SparkError {
    SparkError::RemainderByZero
}