Struct DataFusionQueryExecutor 

pub struct DataFusionQueryExecutor;

DataFusion-powered query executor for repository operations.

This executor converts repository data into Apache Arrow columnar format and leverages DataFusion’s query optimizer for efficient filtering, sorting, and aggregation operations.

Performance Benefits

• Vectorized operations: SIMD optimizations for filtering and sorting
• Pushdown predicates: Filters applied at the storage layer
• Memory efficiency: Columnar format reduces memory overhead
• Query optimization: Cost-based optimization for complex queries

Example

use term_guard::repository::datafusion_executor::DataFusionQueryExecutor;

let results = DataFusionQueryExecutor::execute_optimized_query(
    repository_data,
    before,
    after,
    tag_filters,
    analyzer_filters,
    limit,
    offset,
    ascending
).await?;

Implementations

impl DataFusionQueryExecutor

pub async fn execute_optimized_query( data: Vec<(ResultKey, AnalyzerContext)>, before: Option<i64>, after: Option<i64>, tags: &HashMap<String, String>, analyzers: &Option<Vec<String>>, limit: Option<usize>, offset: Option<usize>, ascending: bool, ) -> Result<Vec<(ResultKey, AnalyzerContext)>>

Executes an optimized query using DataFusion’s query engine.

This method converts repository data into Arrow format, registers it as a table, constructs an SQL query with filters, and executes it using DataFusion’s vectorized query engine.

Arguments

• data - Repository data as (ResultKey, AnalyzerContext) pairs
• before - Optional timestamp filter (exclusive)
• after - Optional timestamp filter (inclusive)
• tags - Tag filters to apply
• analyzers - Optional analyzer name filters
• limit - Maximum number of results to return
• offset - Number of results to skip
• ascending - Sort order (true = ascending, false = descending)

Trait Implementations

impl Default for DataFusionQueryExecutor

fn default() -> Self

Returns the “default value” for a type.