Conversation
zzylol
force-pushed
the
feat/exact-composition-planner
branch
from
August 30, 2026 18:21
8d33043 to
5859f35
Compare
zzylol
force-pushed
the
feat/exact-composition-planner
branch
from
August 30, 2026 18:30
5859f35 to
c4135df
Compare
zzylol
force-pushed
the
feat/exact-composition-planner
branch
from
August 30, 2026 18:35
c4135df to
be9b131
Compare
zzylol
force-pushed
the
feat/exact-composition-planner
branch
from
August 30, 2026 18:46
be9b131 to
88bced3
Compare
…ase boundaries (#171) Add phase-explicit post-ASAP nodes SummaryExpr::{ExactTransform, ExactPostProcess} carrying a non-exhaustive ExactOperator::Aggregate payload (never an intact QueryExpr subtree), plus an ExecutionAvailability {UpdateValue, SummaryState, ReadoutValue} derivation/validation (post_asap::phase) returning typed PhaseErrors at construction. construct_summary_agg now validates its edge, so a maintained summary over a query-time readout falls back conservatively instead of producing an unexecutable plan. Add ExactCompositionStrategy (registered in default_strategies) proposing Replacement::ExactComposition candidates that reference the child target rather than selecting a child; PlanSpace::global_selection commits the compatible parent/child pair using the issue's cost-units-per-second formulas (postprocess/pretransform vs raw-recompute baseline), counts shared child state once, and GlobalSelection::materialize links the committed decisions into one validated DAG with shared Rc identity. Cost hooks: CostModel::mixed_execution_capabilities and exact_composition_cost_inputs (unknowns stay None, never zero; missing statistics keep KeepPreAsap). DAG export gains explicit per-node stage, decision provenance, cost unit and child-decision links (additive). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
zzylol
force-pushed
the
feat/exact-composition-planner
branch
from
August 30, 2026 19:03
88bced3 to
aa8353d
Compare
zzylol
force-pushed
the
feat/exact-summary-composition-171
branch
from
August 30, 2026 21:08
1b96726 to
5bcfb69
Compare
Contributor
Author
|
Superseded by the revised PR #300 design. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Stacked on #300.
Why
Exact outer operations should reuse inner summaries when accuracy and execution-data-state contracts permit it.
Before this PR
The planner could not construct and cost an exact operation around a summary result.
After this PR
For example,
max by (zone)over a quantile becomes a read-time post-process, while an exact accumulator beneath a sketch becomes a maintenance-time transform. Invalid transitions fail before costing.What and how
CompositionPlacement.Verification
Stack
Base: #300. Next: #315.