Aggregation Functions

Aggregation functions are used within summarize operators to compute values across groups of rows. Most common aggregations — count(), sum(), avg(), min(), max() — work across all dialects. More advanced aggregations like arg_max(), percentile(), dcountif(), and make_bag() are ClickHouse-only or have limited availability.

Legend: ✅ Supported · ⚠️ Approximated · ❌ Not Supported · 🔄 Rewritten

Function	SQLite	MySQL	ClickHouse	PostgreSQL	Notes
`count()`	✅	✅	✅	✅
`countif()`	✅	✅	✅	✅	Non-ClickHouse: `SUM(CASE WHEN ...)`
`sum()`	✅	✅	✅	✅
`sumif()`	✅	✅	✅	✅	Non-ClickHouse: `SUM(CASE WHEN ...)`
`avg()`	✅	✅	✅	✅
`avgif()`	✅	✅	✅	✅	Non-ClickHouse: `AVG(CASE WHEN ...)`
`min()`	✅	✅	✅	✅
`minif()` / `min_if()`	✅	✅	✅	✅
`max()`	✅	✅	✅	✅
`maxif()` / `max_if()`	✅	✅	✅	✅
`dcount()`	✅	✅	✅	✅	ClickHouse: `uniq()`; others: `COUNT(DISTINCT ...)`
`dcountif()`	❌	❌	✅	❌	ClickHouse: `uniqIf()`
`any()` / `take_any()`	⚠️	⚠️	✅	⚠️	KQL returns an arbitrary value. Non-ClickHouse: `MIN()` used as a deterministic placeholder — results may differ since `MIN()` imposes ordering. Use `ANY_VALUE()` on MySQL 8.0+ if available
`anyif()`	✅	✅	✅	✅	Non-ClickHouse: `MIN(CASE WHEN ...)`
`arg_max()`	❌	❌	✅	❌	ClickHouse: `argMax()`
`arg_min()`	❌	❌	✅	❌	ClickHouse: `argMin()`
`stdev()`	✅	✅	✅	✅	ClickHouse: `stddevPop()`
`stdevif()`	❌	❌	✅	❌
`variance()`	✅	✅	✅	✅	ClickHouse: `varPop()`
`varianceif()`	❌	❌	✅	❌
`percentile()`	❌	❌	✅	✅	ClickHouse: `quantile()`; Postgres: `percentile_cont() WITHIN GROUP`
`percentiles()`	❌	❌	✅	⚠️	Postgres: expands to single `percentile_cont` with warning
`hll()`	✅	✅	✅	✅	ClickHouse: `uniqHLL12()`; others: `COUNT(DISTINCT ...)`
`hll_merge()`	❌	❌	✅	❌
`tdigest()` / `tdigest_merge()`	❌	❌	✅	❌	ClickHouse: `quantileTDigest()`
`make_list()`	❌	✅	✅	✅
`make_set()`	❌	✅	✅	✅	ClickHouse: `groupUniqArray`
`make_bag()`	❌	❌	✅	❌	ClickHouse: `groupArray()`
`coalesce()`	✅	✅	✅	✅	`COALESCE()`
`nullif()`	✅	✅	✅	✅	`NULLIF()`
`case()`	✅	✅	✅	✅	`CASE WHEN ... END`
`iif()` / `iff()`	✅	✅	✅	✅	ClickHouse: `if()`; others: `CASE WHEN`