Conversation
…ontroller Adopts ASAPController's current AggIntent as the base (~40 variants vs control_plane's pre-merge 25), per docs/migration-plan-backend-plan.md Phase 1. Two deliberate deviations from a byte-for-byte port, both documented in agg_intent.rs's module docs: - `window: Duration` stays on the time-series-derivative variants instead of moving to QueryExpr::TimeRange -- that requires verifying query_expr.rs/lower.rs thread it correctly everywhere, which is Phase 2 scope. - `AggIntent::Frequency` is kept, not folded into ASAPController's RankingMeasure::Frequency -- they're unrelated concepts that share a name (standalone point-frequency-via-CMS query vs. a TopK ranking-measure classifier). This corrects the decision recorded in PR #389/#390; RankingMeasure is still adopted additively. Also folds `irate` into `Rate` per the agreed tie-break (ASAPController's approach): `AggIntent::Irate` is removed. This was already a no-op on real query behavior -- the PromQL walk already lowered "rate"|"irate" to the same AggFunc::Rate with no AggFunc::Irate ever existing, so AggIntent::Irate was unreachable from real parsing; only unit tests constructed it directly. New intent families added (histogram accessors, math/trig transforms, time/calendar accessors, presence functions, Group/CountValues, extended range-vector reducers) are all archive-only for now -- none has a Bind* rule yet, consistent with existing policy for intents with no ASAP-tier sketch binding. Renames to match ASAPController: Idelta -> IDelta, HoltWinters -> DoubleExpSmoothing, Absent -> {Absent, AbsentOverTime}, Present -> PresentOverTime. Adds col: Option<ColumnId> to the single-column reducers (Sum/Min/Max/Avg/Quantile/Cardinality/StdDev/Variance), defaulting to None everywhere today (PromQL sample-value convention) -- plumbing for the Phase 0 "control_plane gains SQL support" decision, populated once frontend-sql lands in Phase 2. Verified: full workspace builds clean (cargo build --workspace); control_plane's 837-test suite passes unchanged (1 pre-existing failure, confirmed identical on unmodified main via git stash, unrelated to this change). The analyzer-parity-matrix.md corpus this phase's testing bar was meant to gate on no longer exists under its documented name (data_plane/.../engine.rs::analyzer_parity_tests) -- flagging for Phase 3, since that phase's plan explicitly depends on it as the acceptance test for the capability_for() fix.
zzylol
added a commit
that referenced
this pull request
Jul 18, 2026
zzylol
added a commit
that referenced
this pull request
Jul 18, 2026
Frequency: corrected to "kept, not folded" per #391's finding. Also flags that analyzer-parity-matrix.md's named acceptance test doesn't exist in current data_plane -- Phase 3 needs to locate or rebuild it before starting, not assume it's there.
3 tasks
Contributor
Author
zzylol
added a commit
that referenced
this pull request
Jul 24, 2026
Phase 1 wasn't done as scoped (#391 closed unmerged; #392 "Phase 1b" substituted a bigger structural fix -- direct git-dep on ASAPController's IR crates instead of an in-tree vocabulary copy-merge). Phase 3 is substantially done already (capability_for() routes Sum/Min/Max/Rate/ Increase to exact-agg on main) but not via this plan's sequencing, and its own documented blocker (missing analyzer_parity_tests corpus) is still unresolved. Phases 4-5 haven't started. Also flags an unplanned parallel thread (#407/#408 Step A/B, merged; #409 Step C, open) that adopts asap_plan::bind::implement_tree / asap_sketch::L4Node directly and overlaps with what Phases 4-5 were meant to deliver -- cross-referenced against the RoutingIndex reconciliation just landed on design-backend-plan-wire-format.md (#389) so Phases 4-5 get re-scoped against what that thread actually ships before anyone executes them as originally written. No process/plan changes here beyond recording status -- this is the same kind of staleness correction this doc already applied to ASAPController/docs/migration-plan.md. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
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.
Summary
First implementation PR against #390's migration plan. Replaces
control_plane'sAggIntent(25 variants) with ASAPController's currentversion (~40 variants), per Phase 1.
Two corrections to decisions recorded in #389/#390, found while actually
reading how the code is used (not just diffing enum signatures):
window: Durationstays on the intent (Rate/Increase/Changes/Delta/IDelta/Deriv/Resets/PredictLinear/DoubleExpSmoothing)instead of moving to
QueryExpr::TimeRangeas ASAPController does.Moving it requires verifying
query_expr.rs/lower.rsthread windowinfo correctly everywhere
AggIntentis built or consulted -- that'sPhase 2 (those are Phase 2 files). Reconcile there.
AggIntent::Frequencyis kept, not folded intoRankingMeasure::Frequency.These are unrelated concepts that happen to share a name:
Frequencyis control_plane's standalone point-frequency-via-CMS query(
count(*) WHERE key = k, bound inbind_cms_count.rs);RankingMeasure::Frequencyclassifies what aTopKranks by. Foldingone into the other would have silently deleted a real, currently-used
capability.
RankingMeasureis still adopted, additively -- it's areal gap control_plane had, just not a replacement for
Frequency.irate/ratefold (per the agreed tie-break) turned out to be a behavioralno-op: the PromQL walk already lowers
"rate" | "irate"to the sameAggFunc::Rate, and noAggFunc::Irateever existed, soAggIntent::Iratewas unreachable from real query parsing -- only unit tests constructed it
directly.
New intent families (histogram accessors, math/trig, time/calendar
accessors, presence functions,
Group/CountValues, extendedrange-vector reducers) land as archive-only, consistent with existing
policy for intents with no
Bind*rule yet.Verification
cargo build --workspace-- clean.cargo test -p control_plane --lib-- 837 passed, 1 pre-existingfailure (
invalid_sketch_type_override_falls_back_to_default),confirmed identical on unmodified
mainviagit stash-- not causedby this change.
analyzer-parity-matrix.md's corpus (the doc'snamed acceptance test,
data_plane/.../engine.rs::analyzer_parity_tests)no longer exists under that name in current
data_plane. Phase 3'splan explicitly depends on it as the gate for the
capability_for()fix -- worth locating or rebuilding before that phase starts.
Test plan
cargo build --workspacecargo test -p control_plane --libupdated in docs(control_plane): design doc for BackendPlan wire format #389's decision table)
🤖 Generated with Claude Code