DataFusion’s optimizer detects that it has reached a fixed point by hashing each pass’s plan into a signature of node count plus content hash and checking a set of previous …
DataFusion’s decorrelator pulls a correlated filter out of a subquery and turns it into a join predicate, adding a __always_true indicator column so the empty-set semantics of …
Instead of hard-coding shuffles, each ExecutionPlan states the distribution it needs from its children; a single optimizer rule turns unmet requirements into RepartitionExec nodes.
DataFusion’s external sorter has two modes: sort the whole buffered input in one concatenated batch when it is small, and otherwise sort per-batch runs and merge them — coalescing …
DataFusion constant-folds expressions by running them against a one-row dummy batch of NULLs, and guards the trick with checks for volatility, leaf nodes, and regexes that would …