Struct SqlDataFrame 

pub struct SqlDataFrame { /* private fields */ }

Krishiv-owned wrapper around a DataFusion DataFrame.

Implementations

impl SqlDataFrame

pub fn query(&self) -> Option<&str>

Original SQL query when created from SqlEngine::sql.

pub fn krishiv_logical_plan(&self) -> LogicalPlan

Create a Krishiv logical plan wrapper for this DataFrame.

Walks the DataFusion logical plan tree, creating Krishiv PlanNode entries for each operator. Table-scan nodes are annotated with estimated_rows from the engine’s table-row-count registry, allowing BroadcastAutoRule to identify small tables for broadcast join promotion. The plan is then run through the logical optimizer before being returned.

pub fn explain_logical(&self) -> String

Explain the logical plan without executing it.

pub async fn explain(&self) -> SqlResult<String>

Explain logical and physical plan details through DataFusion.

pub async fn collect(&self) -> SqlResult<Vec<RecordBatch>>

Execute and collect this DataFrame.

pub async fn execute_stream(&self) -> SqlResult<SqlStream>

Execute and return a record batch stream.

pub async fn collect_with_stats( &self, ) -> SqlResult<(Vec<RecordBatch>, SqlExecutionStats)>

Execute and collect this DataFrame, also returning lightweight runtime statistics.

Collects output_rows from DataFusion’s execution metrics. cpu_nanos is approximated from elapsed_compute when available. spill_bytes and spill_count are aggregated across every operator in the physical plan tree (sorts, hash joins, and aggregations report spills when the memory pool forces them to disk); other fields default to 0.

Trait Implementations

impl Clone for SqlDataFrame

fn clone(&self) -> SqlDataFrame

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for SqlDataFrame

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl KrishivDataFrameOps for SqlDataFrame

fn collect<'life0, 'async_trait>( &'life0 self, ) -> Pin<Box<dyn Future<Output = SqlResult<Vec<RecordBatch>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Execute and collect all result batches.

fn collect_with_stats<'life0, 'async_trait>( &'life0 self, ) -> Pin<Box<dyn Future<Output = SqlResult<(Vec<RecordBatch>, SqlExecutionStats)>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Execute, collect results, and return lightweight runtime statistics.

fn explain<'life0, 'async_trait>( &'life0 self, ) -> Pin<Box<dyn Future<Output = SqlResult<String>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Explain the physical and logical plan text (does not execute).

fn explain_logical(&self) -> String

Explain the logical plan text without executing.

fn krishiv_logical_plan(&self) -> LogicalPlan

Build a Krishiv LogicalPlan wrapper for this DataFrame.

fn query(&self) -> Option<&str>

The original SQL query string, if any.

fn execute_stream<'life0, 'async_trait>( &'life0 self, ) -> Pin<Box<dyn Future<Output = SqlResult<SqlStream>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Execute and return a record batch stream.

fn schema(&self) -> SchemaRef

Return the Arrow schema of this DataFrame.

fn select<'life0, 'life1, 'life2, 'async_trait>( &'life0 self, columns: &'life1 [&'life2 str], ) -> Pin<Box<dyn Future<Output = SqlResult<Box<dyn KrishivDataFrameOps>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait, 'life2: 'async_trait,

Select columns by name.

fn select_exprs<'life0, 'life1, 'life2, 'async_trait>( &'life0 self, expressions: &'life1 [&'life2 Expr], ) -> Pin<Box<dyn Future<Output = SqlResult<Box<dyn KrishivDataFrameOps>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait, 'life2: 'async_trait,

Select arbitrary SQL expressions.

fn aggregate<'life0, 'life1, 'life2, 'life3, 'life4, 'async_trait>( &'life0 self, group_exprs: &'life1 [&'life2 Expr], aggregate_exprs: &'life3 [&'life4 Expr], ) -> Pin<Box<dyn Future<Output = SqlResult<Box<dyn KrishivDataFrameOps>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait, 'life2: 'async_trait, 'life3: 'async_trait, 'life4: 'async_trait,

Group by expressions and compute aggregate expressions.

fn aggregate_grouping<'life0, 'life1, 'life2, 'life3, 'async_trait>( &'life0 self, grouping: GroupingMode<'life1>, aggregate_exprs: &'life2 [&'life3 Expr], ) -> Pin<Box<dyn Future<Output = SqlResult<Box<dyn KrishivDataFrameOps>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait, 'life2: 'async_trait, 'life3: 'async_trait,

Aggregate using GROUPING SETS, CUBE, or ROLLUP.

fn pivot<'life0, 'life1, 'life2, 'life3, 'life4, 'life5, 'async_trait>( &'life0 self, group_exprs: &'life1 [&'life2 Expr], pivot_column: &'life3 Expr, aggregate_expr: &'life4 Expr, values: &'life5 [(ScalarValue, String)], ) -> Pin<Box<dyn Future<Output = SqlResult<Box<dyn KrishivDataFrameOps>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait, 'life2: 'async_trait, 'life3: 'async_trait, 'life4: 'async_trait, 'life5: 'async_trait,

Pivot known values into aggregate columns.

fn unpivot<'life0, 'life1, 'life2, 'life3, 'life4, 'async_trait>( &'life0 self, columns: &'life1 [&'life2 str], name_column: &'life3 str, value_column: &'life4 str, ) -> Pin<Box<dyn Future<Output = SqlResult<Box<dyn KrishivDataFrameOps>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait, 'life2: 'async_trait, 'life3: 'async_trait, 'life4: 'async_trait,

Unpivot columns into name/value rows while preserving other columns.

fn filter<'life0, 'life1, 'async_trait>( &'life0 self, predicate: &'life1 str, ) -> Pin<Box<dyn Future<Output = SqlResult<Box<dyn KrishivDataFrameOps>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait,

Filter rows by a SQL predicate expression.

fn filter_expr<'life0, 'life1, 'async_trait>( &'life0 self, predicate: &'life1 Expr, ) -> Pin<Box<dyn Future<Output = SqlResult<Box<dyn KrishivDataFrameOps>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait,

Filter rows using the engine-owned typed expression AST.

fn limit<'life0, 'async_trait>( &'life0 self, n: usize, ) -> Pin<Box<dyn Future<Output = SqlResult<Box<dyn KrishivDataFrameOps>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Limit the number of rows.

fn distinct<'life0, 'async_trait>( &'life0 self, ) -> Pin<Box<dyn Future<Output = SqlResult<Box<dyn KrishivDataFrameOps>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Remove duplicate rows.

fn drop_nulls<'life0, 'life1, 'life2, 'async_trait>( &'life0 self, columns: &'life1 [&'life2 str], ) -> Pin<Box<dyn Future<Output = SqlResult<Box<dyn KrishivDataFrameOps>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait, 'life2: 'async_trait,

Drop rows with nulls in selected columns; an empty list checks all columns.

fn sample<'life0, 'async_trait>( &'life0 self, fraction: f64, ) -> Pin<Box<dyn Future<Output = SqlResult<Box<dyn KrishivDataFrameOps>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Bernoulli-sample rows.

fn sort<'life0, 'life1, 'life2, 'life3, 'async_trait>( &'life0 self, columns: &'life1 [&'life2 str], descending: &'life3 [bool], ) -> Pin<Box<dyn Future<Output = SqlResult<Box<dyn KrishivDataFrameOps>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait, 'life2: 'async_trait, 'life3: 'async_trait,

Sort by columns with optional descending flags.

fn alias<'life0, 'life1, 'async_trait>( &'life0 self, alias: &'life1 str, ) -> Pin<Box<dyn Future<Output = SqlResult<Box<dyn KrishivDataFrameOps>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait,

Assign an alias (table name) to this DataFrame.

fn drop_columns<'life0, 'life1, 'life2, 'async_trait>( &'life0 self, columns: &'life1 [&'life2 str], ) -> Pin<Box<dyn Future<Output = SqlResult<Box<dyn KrishivDataFrameOps>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait, 'life2: 'async_trait,

Drop columns by name.

fn rename_column<'life0, 'life1, 'life2, 'async_trait>( &'life0 self, old: &'life1 str, new: &'life2 str, ) -> Pin<Box<dyn Future<Output = SqlResult<Box<dyn KrishivDataFrameOps>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait, 'life2: 'async_trait,

Rename a column from old to new.

fn with_column<'life0, 'life1, 'life2, 'async_trait>( &'life0 self, name: &'life1 str, expr: &'life2 str, ) -> Pin<Box<dyn Future<Output = SqlResult<Box<dyn KrishivDataFrameOps>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait, 'life2: 'async_trait,

Add or replace a column with a computed expression.

fn as_any(&self) -> &dyn Any

Return the underlying concrete type for downcasting.

fn describe<'life0, 'async_trait>( &'life0 self, ) -> Pin<Box<dyn Future<Output = SqlResult<Box<dyn KrishivDataFrameOps>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Compute summary statistics (delegates to DataFusion’s describe).

fn fill_null<'life0, 'life1, 'life2, 'async_trait>( &'life0 self, column: &'life1 str, value: &'life2 str, ) -> Pin<Box<dyn Future<Output = SqlResult<Box<dyn KrishivDataFrameOps>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait, 'life2: 'async_trait,

Fill null values in column with the literal SQL value.

fn join<'life0, 'life1, 'life2, 'life3, 'life4, 'life5, 'life6, 'async_trait>( &'life0 self, right: &'life1 dyn KrishivDataFrameOps, how: &'life2 str, left_on: &'life3 [&'life4 str], right_on: &'life5 [&'life6 str], ) -> Pin<Box<dyn Future<Output = SqlResult<Box<dyn KrishivDataFrameOps>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait, 'life2: 'async_trait, 'life3: 'async_trait, 'life4: 'async_trait, 'life5: 'async_trait, 'life6: 'async_trait,

Join with another DataFrame using a join type and equi-join keys.

fn union<'life0, 'life1, 'async_trait>( &'life0 self, right: &'life1 dyn KrishivDataFrameOps, ) -> Pin<Box<dyn Future<Output = SqlResult<Box<dyn KrishivDataFrameOps>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait,

Union this DataFrame with another (UNION ALL semantics).

fn union_distinct<'life0, 'life1, 'async_trait>( &'life0 self, right: &'life1 dyn KrishivDataFrameOps, ) -> Pin<Box<dyn Future<Output = SqlResult<Box<dyn KrishivDataFrameOps>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait,

fn intersect<'life0, 'life1, 'async_trait>( &'life0 self, right: &'life1 dyn KrishivDataFrameOps, distinct: bool, ) -> Pin<Box<dyn Future<Output = SqlResult<Box<dyn KrishivDataFrameOps>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait,

fn except<'life0, 'life1, 'async_trait>( &'life0 self, right: &'life1 dyn KrishivDataFrameOps, distinct: bool, ) -> Pin<Box<dyn Future<Output = SqlResult<Box<dyn KrishivDataFrameOps>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait,

fn register_batches<'life0, 'life1, 'async_trait>( &'life0 self, name: &'life1 str, batches: Vec<RecordBatch>, ) -> Pin<Box<dyn Future<Output = SqlResult<()>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait,

Register a list of record batches as a named in-memory table in the same session context that backs this DataFrame. Used by cache().

fn deregister_table<'life0, 'life1, 'async_trait>( &'life0 self, name: &'life1 str, ) -> Pin<Box<dyn Future<Output = SqlResult<()>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait,

Deregister a named table from the session context. Used by unpersist().

fn create_view<'life0, 'life1, 'async_trait>( &'life0 self, name: &'life1 str, replace: bool, ) -> Pin<Box<dyn Future<Output = SqlResult<()>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait,

Create (or replace) a SQL view named name backed by this DataFrame’s query. Used by create_or_replace_temp_view().