chore: merge dev into v2 (#36312)
Co-authored-by: opencode-agent[bot] <219766164+opencode-agent[bot]@users.noreply.github.com> Co-authored-by: LukeParkerDev <10430890+Hona@users.noreply.github.com> Co-authored-by: opencode-agent[bot] <opencode-agent[bot]@users.noreply.github.com> Co-authored-by: Aiden Cline <63023139+rekram1-node@users.noreply.github.com> Co-authored-by: Brendan Allan <14191578+Brendonovich@users.noreply.github.com> Co-authored-by: Aarav Sareen <96787824+arvsrn@users.noreply.github.com> Co-authored-by: Julian Coy <julian@ex-machina.co> Co-authored-by: Brendan Allan <git@brendonovich.dev> Co-authored-by: usrnk1 <7547651+usrnk1@users.noreply.github.com> Co-authored-by: opencode <opencode@sst.dev> Co-authored-by: Vladimir Glafirov <vglafirov@gitlab.com> Co-authored-by: Adam <2363879+adamdotdevin@users.noreply.github.com> Co-authored-by: Frank <frank@anoma.ly> Co-authored-by: Jay <53023+jayair@users.noreply.github.com> Co-authored-by: Dustin Deus <deusdustin@gmail.com> Co-authored-by: Kit Langton <kit.langton@gmail.com> Co-authored-by: James Long <longster@gmail.com> Co-authored-by: Simon Klee <hello@simonklee.dk> Co-authored-by: Jay <air@live.ca> Co-authored-by: Jack <jack@anoma.ly> Co-authored-by: David Hill <1879069+iamdavidhill@users.noreply.github.com> Co-authored-by: Aiden Cline <aidenpcline@gmail.com> Co-authored-by: James Long <jlongster@users.noreply.github.com> Co-authored-by: 冯基魁 <56265583+fengjikui@users.noreply.github.com> Co-authored-by: Aiden Cline <rekram1-node@users.noreply.github.com> Co-authored-by: Victor Navarro <vn4varro@gmail.com>
This commit is contained in:
parent
9028c2d8f8
commit
43ecf3ff1b
82 changed files with 4519 additions and 1231 deletions
|
|
@ -94,10 +94,15 @@ export type StatsModelComparisonEntry = {
|
|||
tokenShare: number
|
||||
tokenChange: number
|
||||
totals: StatsModelData["totals"]
|
||||
usage: ModelUsagePoint[]
|
||||
}
|
||||
export type StatsModelComparisonInput = {
|
||||
provider: string
|
||||
model: string
|
||||
}
|
||||
export type StatsModelComparisonData = {
|
||||
updatedAt: string | null
|
||||
models: [StatsModelComparisonEntry | null, StatsModelComparisonEntry | null]
|
||||
models: (StatsModelComparisonEntry | null)[]
|
||||
}
|
||||
export type StatsHomeData = {
|
||||
updatedAt: string | null
|
||||
|
|
@ -289,27 +294,35 @@ function dateValue(value: unknown) {
|
|||
return value instanceof Date ? value : new Date(stringValue(value))
|
||||
}
|
||||
|
||||
export const getStatsModelComparisonData: (
|
||||
firstProvider: string,
|
||||
firstModel: string,
|
||||
secondProvider: string,
|
||||
secondModel: string,
|
||||
) => Effect.Effect<StatsModelComparisonData, DatabaseError, ModelStatRepo> = Effect.fn("StatsModelComparison.getData")(
|
||||
function* (firstProvider, firstModel, secondProvider, secondModel) {
|
||||
export const getStatsModelsComparisonData: (
|
||||
models: readonly StatsModelComparisonInput[],
|
||||
) => Effect.Effect<StatsModelComparisonData, DatabaseError, ModelStatRepo> = Effect.fn("StatsModelsComparison.getData")(
|
||||
function* (models) {
|
||||
const modelStats = yield* ModelStatRepo
|
||||
const rows = yield* modelStats.listDaily()
|
||||
const first = toComparisonEntry(buildStatsModelData(firstModel, rows, [], firstProvider))
|
||||
const second = toComparisonEntry(buildStatsModelData(secondModel, rows, [], secondProvider))
|
||||
const latest = [first?.updatedAt, second?.updatedAt]
|
||||
const entries = models.map((model) => toComparisonEntry(buildStatsModelData(model.model, rows, [], model.provider)))
|
||||
const latest = entries
|
||||
.map((model) => model?.updatedAt)
|
||||
.flatMap((value) => (value ? [dateTime(value)] : []))
|
||||
.toSorted((a, b) => b - a)[0]
|
||||
return {
|
||||
updatedAt: latest === undefined ? null : new Date(latest).toISOString(),
|
||||
models: [first, second],
|
||||
models: entries,
|
||||
}
|
||||
},
|
||||
)
|
||||
|
||||
export const getStatsModelComparisonData = (
|
||||
firstProvider: string,
|
||||
firstModel: string,
|
||||
secondProvider: string,
|
||||
secondModel: string,
|
||||
) =>
|
||||
getStatsModelsComparisonData([
|
||||
{ provider: firstProvider, model: firstModel },
|
||||
{ provider: secondProvider, model: secondModel },
|
||||
])
|
||||
|
||||
function buildStatsHomeData(
|
||||
modelRows: ModelStatMetric[],
|
||||
providerRows: ProviderStatMetric[],
|
||||
|
|
@ -501,6 +514,7 @@ function toComparisonEntry(data: StatsModelData | null): StatsModelComparisonEnt
|
|||
tokenShare: data.tokenShare,
|
||||
tokenChange: data.tokenChange,
|
||||
totals: data.totals,
|
||||
usage: data.usage,
|
||||
}
|
||||
}
|
||||
|
||||
|
|
|
|||
|
|
@ -14,7 +14,10 @@ import { normalizeCountry, normalizeTier, type StatBaseAggregate } from "./stat"
|
|||
|
||||
export type StatDimension = "model" | "provider" | "geo" | "geo_model"
|
||||
|
||||
export function buildStatsQuery(periodStart: Date, periodEnd: Date, dimension: StatDimension) {
|
||||
// All stat dimensions and both grains are computed in one query via GROUPING SETS so
|
||||
// the source table is scanned once per sync pass; separate queries per dimension (and
|
||||
// the previous weekly/daily UNION ALL) each re-scanned the same events.
|
||||
export function buildStatsQuery(periodStart: Date, periodEnd: Date) {
|
||||
const periodStartValue = sqlString(periodStart.toISOString())
|
||||
const periodEndValue = sqlString(periodEnd.toISOString())
|
||||
const periodStartDateValue = sqlString(periodStart.toISOString().slice(0, 10))
|
||||
|
|
@ -22,23 +25,6 @@ export function buildStatsQuery(periodStart: Date, periodEnd: Date, dimension: S
|
|||
const sourceTable = [Resource.InferenceEvent.catalog, Resource.InferenceEvent.database, Resource.InferenceEvent.table]
|
||||
.map(sqlIdentifier)
|
||||
.join(".")
|
||||
const dimensionSql = (() => {
|
||||
if (dimension === "model")
|
||||
return {
|
||||
select: "provider, model, COALESCE(MAX(NULLIF(provider_model, '')), '') AS provider_model",
|
||||
groupBy: "provider, model",
|
||||
}
|
||||
if (dimension === "provider") return { select: "provider", groupBy: "provider" }
|
||||
if (dimension === "geo_model")
|
||||
return {
|
||||
select: "provider, model, country, COALESCE(MAX(NULLIF(continent, '')), '') AS continent",
|
||||
groupBy: "provider, model, country",
|
||||
}
|
||||
return {
|
||||
select: "'all' AS provider, 'all' AS model, country, COALESCE(MAX(NULLIF(continent, '')), '') AS continent",
|
||||
groupBy: "country",
|
||||
}
|
||||
})()
|
||||
const aggregateColumns = `
|
||||
COUNT(DISTINCT session) AS sessions,
|
||||
COUNT(*) AS requests,
|
||||
|
|
@ -135,34 +121,41 @@ WITH normalized AS (
|
|||
COALESCE(cost_total_microcents, cost_total * 1000000) AS cost_total_microcents
|
||||
FROM normalized
|
||||
WHERE lower(model) NOT IN (${[...EXCLUDED_MODELS].map(sqlString).join(", ")})
|
||||
), weekly AS (
|
||||
), periods AS (
|
||||
SELECT
|
||||
concat(CAST(year_of_week(event_time) AS varchar), '-W', lpad(CAST(week(event_time) AS varchar), 2, '0')) AS week_key,
|
||||
substr(to_iso8601(date_trunc('day', event_time)), 1, 10) AS day_key,
|
||||
*
|
||||
FROM filtered
|
||||
), daily AS (
|
||||
SELECT substr(to_iso8601(date_trunc('day', event_time)), 1, 10) AS day_key, *
|
||||
FROM filtered
|
||||
)
|
||||
SELECT
|
||||
'week' AS grain,
|
||||
week_key AS period_key,
|
||||
CASE WHEN grouping(week_key) = 0 THEN 'week' ELSE 'day' END AS grain,
|
||||
COALESCE(week_key, day_key) AS period_key,
|
||||
${sqlString(Resource.StatsSyncConfig.dataset)} AS dataset,
|
||||
CASE
|
||||
WHEN grouping(country) = 0 AND grouping(model) = 0 THEN 'geo_model'
|
||||
WHEN grouping(country) = 0 THEN 'geo'
|
||||
WHEN grouping(model) = 0 THEN 'model'
|
||||
ELSE 'provider'
|
||||
END AS dimension,
|
||||
tier,
|
||||
${dimensionSql.select},
|
||||
CASE WHEN grouping(provider) = 0 THEN provider ELSE 'all' END AS provider,
|
||||
CASE WHEN grouping(model) = 0 THEN model WHEN grouping(country) = 0 THEN 'all' END AS model,
|
||||
CASE WHEN grouping(model) = 0 AND grouping(country) = 1 THEN COALESCE(MAX(NULLIF(provider_model, '')), '') END AS provider_model,
|
||||
CASE WHEN grouping(country) = 0 THEN country END AS country,
|
||||
CASE WHEN grouping(country) = 0 THEN COALESCE(MAX(NULLIF(continent, '')), '') END AS continent,
|
||||
${aggregateColumns}
|
||||
FROM weekly
|
||||
GROUP BY week_key, tier, ${dimensionSql.groupBy}
|
||||
UNION ALL
|
||||
SELECT
|
||||
'day' AS grain,
|
||||
day_key AS period_key,
|
||||
${sqlString(Resource.StatsSyncConfig.dataset)} AS dataset,
|
||||
tier,
|
||||
${dimensionSql.select},
|
||||
${aggregateColumns}
|
||||
FROM daily
|
||||
GROUP BY day_key, tier, ${dimensionSql.groupBy}
|
||||
FROM periods
|
||||
GROUP BY GROUPING SETS (
|
||||
(week_key, tier, provider, model),
|
||||
(week_key, tier, provider),
|
||||
(week_key, tier, country),
|
||||
(week_key, tier, provider, model, country),
|
||||
(day_key, tier, provider, model),
|
||||
(day_key, tier, provider),
|
||||
(day_key, tier, country),
|
||||
(day_key, tier, provider, model, country)
|
||||
)
|
||||
ORDER BY grain, period_key, total_tokens DESC
|
||||
`
|
||||
}
|
||||
|
|
|
|||
|
|
@ -1,4 +1,4 @@
|
|||
import { and, asc, eq, inArray, or } from "drizzle-orm"
|
||||
import { and, asc, eq, inArray, max, or } from "drizzle-orm"
|
||||
import { Effect, Layer } from "effect"
|
||||
import * as Context from "effect/Context"
|
||||
import { DatabaseError, DrizzleClient } from "../database"
|
||||
|
|
@ -43,6 +43,7 @@ export type ModelStatMetric = {
|
|||
export declare namespace ModelStatRepo {
|
||||
export interface Service {
|
||||
readonly listDaily: () => Effect.Effect<ModelStatMetric[], DatabaseError>
|
||||
readonly lastSyncedAt: () => Effect.Effect<Date | null, DatabaseError>
|
||||
readonly upsert: (rows: ModelStatRow[]) => Effect.Effect<void, DatabaseError>
|
||||
readonly deleteRetiredDimensions: (rows: ModelStatRow[]) => Effect.Effect<void, DatabaseError>
|
||||
}
|
||||
|
|
@ -111,6 +112,14 @@ export class ModelStatRepo extends Context.Service<ModelStatRepo, ModelStatRepo.
|
|||
})
|
||||
})
|
||||
|
||||
const lastSyncedAt = Effect.fn("ModelStatRepo.lastSyncedAt")(function* () {
|
||||
const result = yield* Effect.tryPromise({
|
||||
try: () => db.select({ value: max(modelStat.updated_at) }).from(modelStat),
|
||||
catch: (cause) => DatabaseError.make({ cause }),
|
||||
})
|
||||
return result[0]?.value ?? null
|
||||
})
|
||||
|
||||
const upsert = Effect.fn("ModelStatRepo.upsert")(function* (rows: ModelStatRow[]) {
|
||||
yield* Effect.forEach(
|
||||
chunks(rows, UPSERT_CHUNK_SIZE),
|
||||
|
|
@ -192,7 +201,7 @@ export class ModelStatRepo extends Context.Service<ModelStatRepo, ModelStatRepo.
|
|||
})
|
||||
})
|
||||
|
||||
return ModelStatRepo.of({ listDaily, upsert, deleteRetiredDimensions })
|
||||
return ModelStatRepo.of({ listDaily, lastSyncedAt, upsert, deleteRetiredDimensions })
|
||||
}),
|
||||
)
|
||||
}
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue