The Postgres development platform. Supabase gives you a dedicated Postgres database to build your web, mobile, and AI applications.
TypeScript aialternativeauthdatabasedenoembeddingsexamplefirebasenextjsoauth2pgvectorpostgispostgrespostgresqlpostgrestrealtimesupabasevectorswebsockets
import { ident, literal, safeSql } from '@supabase/pg-meta/src/pg-format'
import { useQuery } from '@tanstack/react-query'
import { databaseQueuesKeys } from './keys'
import {
isQueueNameValid,
pgmqQueueTable,
} from '@/components/interfaces/Integrations/Queues/Queues.utils'
import { executeSql } from '@/data/sql/execute-sql-mutation'
import type { ResponseError, UseCustomQueryOptions } from '@/types'
export type DatabaseQueuesMetricsVariables = {
projectRef?: string
connectionString?: string | null
queueName: string
}
export type PostgresQueueMetric = {
queue_name: string
queue_length: number
method: 'estimated' | 'precise'
}
const preciseMetricsSqlQuery = (queueName: string) =>
safeSql`
set local statement_timeout = '1s';
SELECT
COUNT(*) AS row_count
FROM
${ident('pgmq')}.${ident(pgmqQueueTable(queueName))};
`
const estimateMetricsSqlQuery = (queueName: string) =>
safeSql`
select
reltuples::bigint as estimated_rows
from
pg_class
where
relname = ${literal(pgmqQueueTable(queueName))}
and relnamespace = 'pgmq'::regnamespace;
`
export async function getDatabaseQueuesMetrics({
projectRef,
connectionString,
queueName,
}: DatabaseQueuesMetricsVariables) {
if (!projectRef) throw new Error('Project ref is required')
if (!isQueueNameValid(queueName)) {
throw new Error(
'Invalid queue name: must contain only alphanumeric characters, underscores, and hyphens'
)
}
try {
const { result } = await executeSql({
projectRef,
connectionString,
sql: preciseMetricsSqlQuery(queueName),
})
return {
queue_name: queueName,
queue_length: result[0].row_count,
method: 'precise',
} as PostgresQueueMetric
} catch (error: any) {
// if the error is caused because the count timeouted, try to fetch an approximate count
if (error?.message === 'canceling statement due to statement timeout') {
const { result } = await executeSql({
projectRef,
connectionString,
sql: estimateMetricsSqlQuery(queueName),
})
return {
queue_name: queueName,
queue_length: result[0].estimated_rows,
method: 'estimated',
} as PostgresQueueMetric
}
throw error
}
}
export type DatabaseQueuesMetricsData = PostgresQueueMetric
export type DatabaseQueuesMetricsError = ResponseError
export const useQueuesMetricsQuery = <TData = DatabaseQueuesMetricsData>(
{ projectRef, connectionString, queueName }: DatabaseQueuesMetricsVariables,
{
enabled = true,
...options
}: UseCustomQueryOptions<DatabaseQueuesMetricsData, DatabaseQueuesMetricsError, TData> = {}
) =>
useQuery<DatabaseQueuesMetricsData, DatabaseQueuesMetricsError, TData>({
queryKey: databaseQueuesKeys.metrics(projectRef, queueName),
queryFn: () => getDatabaseQueuesMetrics({ projectRef, connectionString, queueName }),
enabled: enabled && typeof projectRef !== 'undefined',
...options,
})