Skip to main content

AQL Functions Overview

AQL functions are the main building blocks of an AQL expression. They transform an input into an output based on specified arguments, and are typically combined using the pipe operator.

This page is the entry point for all AQL functions. Each category below links to its own reference page with the full list of functions, their signatures, and examples.

CategoryWhat it does
Table FunctionsTransform a table expression into another table. select, group, filter, unique, top, bottom.
Metric FunctionsModify the context of a metric expression. Includes Condition, Relationship, LOD, Time-based, and Window functions.
Aggregation FunctionsCollapse multiple rows into a single value. SQL-equivalents of SUM, COUNT, AVG, plus statistical aggregates like percentile_cont and stdev.
Logical FunctionsBranch on conditions: case, and, or, not, in.
Text FunctionsString manipulation: concat, find, replace, regex helpers, padding, case conversion.
Time Intelligence FunctionsDate/time helpers: truncation (day, month, year), formatting, unix conversion.
AI FunctionsLLM-powered helpers for classification, summarization, similarity. Databricks and Snowflake only.
Null/Zero Handling Functionscoalesce, nullif, safe_divide.
SQL Passthrough FunctionsEscape hatch to call native database functions when AQL doesn't cover what you need.
Miscellaneous Functionscast, is_at_level, and other one-offs.

See also


Open Markdown
Let us know what you think about this document :)