Module window 

Window Functions Kernels Module - High-Performance Analytical Window Operations

Advanced window function kernels for sliding window computations, ranking operations, and positional analytics with SIMD acceleration and null-aware semantics. Backbone of time series analysis, analytical SQL window functions, and chunked streaming computations.

Core Operations

• Moving averages: Rolling mean calculations with configurable window sizes
• Cumulative functions: Running sums, products, and statistical aggregations
• Ranking functions: ROW_NUMBER, RANK, DENSE_RANK with tie-handling strategies
• Lead/lag operations: Positional value access with configurable offsets
• Percentile functions: Moving quantile calculations with interpolation support
• Window aggregates: MIN, MAX, SUM operations over sliding windows

Constants

W8

Auto-generated SIMD lane widths from build.rs SIMD lane count for 8-bit elements (u8, i8). Determined at build time based on target architecture capabilities, or overridden via SIMD_LANES_OVERRIDE.

W16

SIMD lane count for 16-bit elements (u16, i16). Determined at build time based on target architecture capabilities, or overridden via SIMD_LANES_OVERRIDE.

W32

SIMD lane count for 32-bit elements (u32, i32, f32). Determined at build time based on target architecture capabilities, or overridden via SIMD_LANES_OVERRIDE.

W64

SIMD lane count for 64-bit elements (u64, i64, f64). Determined at build time based on target architecture capabilities, or overridden via SIMD_LANES_OVERRIDE.

Functions

dense_rank_float

Computes SQL DENSE_RANK() ranking for floating-point data with IEEE 754 compliance.

dense_rank_int

Computes SQL DENSE_RANK() ranking for integer data with consecutive rank assignment.

dense_rank_str

Computes SQL DENSE_RANK() ranking for string data with lexicographic dense ordering.

lag_float

Accesses values from previous positions in floating-point arrays with IEEE 754 compliance.

lag_int

Accesses values from previous positions in integer arrays with configurable offset.

lag_str

Accesses string values from previous positions with UTF-8 string handling.

lead_float

Accesses values from future positions in floating-point arrays with IEEE 754 compliance.

lead_int

Accesses values from future positions in integer arrays with configurable offset.

lead_str

Accesses string values from future positions with efficient UTF-8 processing.

rank_float

Computes standard SQL ROW_NUMBER() ranking for floating-point data with IEEE 754 compliance.

rank_int

Computes standard SQL ROW_NUMBER() ranking for integer data with tie handling.

rank_str

Computes standard SQL ROW_NUMBER() ranking for string data with lexicographic ordering.

rolling_count

Computes rolling counts of elements within sliding windows for positional analysis.

rolling_max_float

Computes rolling maximum values over floating-point data with IEEE 754 compliance.

rolling_max_int

Computes rolling maximum values over integer data within sliding windows.

rolling_mean_float

Computes rolling arithmetic means over floating-point data with IEEE 754 compliance.

rolling_mean_int

Computes rolling arithmetic means over integer data with high-precision floating-point output.

rolling_min_float

Computes rolling minimum values over floating-point data with IEEE 754 compliance.

rolling_min_int

For min, we skip that very first window slot so that the first non-zero result appears one step later. Computes rolling minimum values over integer data within sliding windows.

rolling_product_bool

Computes rolling logical AND operations over boolean data within sliding windows.

rolling_product_float

Computes rolling products over floating-point data with IEEE 754 mathematical semantics.

rolling_product_int

Computes rolling products over a sliding window for integer data with overflow protection.

rolling_sum_bool

Computes rolling sums over boolean data, counting true values within sliding windows.

rolling_sum_float

Computes rolling sums over a sliding window for floating-point data with IEEE 754 compliance.

rolling_sum_int

Computes rolling sums over a sliding window for integer data with null-aware semantics.

shift_float

Shifts floating-point array elements with IEEE 754 compliance and bidirectional support.

shift_int

Shifts integer array elements by specified offset with bidirectional support.

shift_str

Shifts string array elements with UTF-8 integrity and bidirectional offset support.