Expand description
Static argument type specifications for all built-in functions.
Each constant defines the accepted argument types per position.
An empty accepted slice means any type is accepted (wildcard).
variadic: true on the last entry means the argument repeats.
Statics§
- APPROX_
COUNT_ ARGS - APPROX_
COUNT_ DISTINCT_ ARGS - APPROX_
PERCENTILE_ ARGS - APPROX_
TOPK_ ARGS - ARRAY_
1_ ARGS array_length(arr)/cardinality(arr)/array_distinct(arr)/array_reverse(arr)— a single array argument.- ARRAY_
AGG_ ARGS - ARRAY_
CAT_ ARGS array_concat(arr1, arr2)/array_cat(arr1, arr2).- ARRAY_
ELEMENTWISE_ ARGS - ARRAY_
ELEM_ ARGS array_append(arr, value)/array_remove(arr, value)/array_contains(arr, value)/array_position(arr, value)— an array plus an element of any type.- ARRAY_
MAINT_ ARGS - ARRAY_
PREPEND_ ARGS array_prepend(value, arr)— value-first argument order, matchingnodedb_query::functions::array::try_eval’s"array_prepend"arm.- ARRAY_
PROJECT_ ARGS - ARRAY_
SLICE_ ARGS - AVG_
ARGS - BM25_
SCORE_ ARGS - COALESCE_
ARGS - CONCAT_
ARGS - COUNT_
ARGS - count(*) / count(expr)
- DATETIME_
1_ ARGS datetime(value)/to_datetime(value)/unix_secs(value)/epoch_secs(value)/unix_millis(value)/epoch_millis(value)/duration(value)/to_duration(value)/decimal(value)/to_decimal(value)— a single value whose accepted shape (string, integer micros, or an existing timestamp/duration) varies by function, so the position is left untyped (any) rather than over-constrained.- DATE_
ARITH_ ARGS date_add(ts, duration)/datetime_add(ts, duration)/date_sub(ts, duration)/datetime_sub(ts, duration).- DATE_
DIFF_ ARGS date_diff(ts1, ts2)/datediff(ts1, ts2).- DATE_
PART_ ARGS extract(part, ts)/date_part(part, ts)/date_trunc(part, ts)/datetrunc(part, ts)— a text field-name selector plus the timestamp-like value to read it from.- DISPERSION_
ARGS stddev/varianceand their_samp/_popforms.- DOC_
ARRAY_ CONTAINS_ ARGS - DOC_
EXISTS_ ARGS - DOC_
GET_ ARGS - FIRST_
LAST_ VALUE_ ARGS - GRAPH_
SCORE_ ARGS graph_score(node_id_col, seed_id, depth => N, label => 'edge_label').- GREATEST_
LEAST_ ARGS greatest(expr, ...)/least(expr, ...)— one or more arguments of any (comparable) type.- GROUPING_
ARGS GROUPING(col [, col2, ...])— accepts 1 or more column references.- ID_
CHECK_ ARGS is_uuid(value)/is_ulid(value)/is_cuid2(value)/is_nanoid(value)/id_type(value)/uuid_version(value)/ulid_timestamp(value)— a single text value to classify. Shared shape across all detector functions.- JSON_
1_ ARGS json_keys(json)/json_values(json)/json_length(json)/json_len(json)/json_type(json)— a single JSON-shaped value.- JSON_
ARRAY_ ARGS json_array(value, ...)— zero or more values of any type.- JSON_
CONTAINS_ ARGS json_contains(container, needle).- JSON_
EXISTS_ ARGS - JSON_
EXTRACT_ ARGS json_extract(json, path)/json_get(json, path).- JSON_
MERGE_ ARGS json_merge(base, overlay)/json_patch(base, overlay).- JSON_
OBJECT_ ARGS json_object(key, value, ...)— a flat, variadic key/value list.- JSON_
QUERY_ ARGS - JSON_
REMOVE_ ARGS json_remove(json, key).- JSON_
SET_ ARGS json_set(json, key, value).- JSON_
VALUE_ ARGS - LAG_
LEAD_ ARGS - LENGTH_
ARGS - MAKE_
ARRAY_ ARGS - MATH_
1_ ARGS - MATH_
2_ ARGS mod(a, b)/power(base, exp)/pow(base, exp)— two required numeric arguments.- MAX_
ARGS - MIN_
ARGS - MULTI_
VECTOR_ SCORE_ ARGS - MULTI_
VECTOR_ SEARCH_ ARGS - NANOID_
ARGS nanoid(length?)— optional custom length, defaults to the standard 21-character id when omitted.- NAV_
ARGS - NDB_
CHUNK_ TEXT_ ARGS - NO_ARGS
- NTH_
VALUE_ ARGS - NTILE_
ARGS - NULLIF_
ARGS - PG_
CATALOG_ 2_ ARGS - PG_
FTS_ MATCH_ ARGS - PG_
JSON_ 2_ ARGS - PG_
JSON_ BOOL_ 2_ ARGS - PG_
TO_ TSQUERY_ ARGS - PG_
TO_ TSVECTOR_ ARGS - PG_
TS_ HEADLINE_ ARGS - PG_
TS_ RANK_ ARGS - REPLACE_
ARGS - ROUND_
ARGS - RRF_
SCORE_ ARGS - Two-source
rrf_score(rank1, rank2, k1?, k2?)— vector + text. Accepts 2–4 arguments; the planner validates arity at plan time. - RRF_
SCORE_ TRIPLE_ ARGS - Three-source
rrf_score(rank1, rank2, rank3, k1?, k2?, k3?)— vector + text + graph. Shares the same function name; arity dispatch happens in the planner. - SEARCH_
SCORE_ ARGS - SEQUENCE_
NAME_ ARGS nextval(name)/currval(name)— one sequence name.- SETVAL_
ARGS setval(name, value)— a sequence name and the value it must take.- SPARSE_
SCORE_ ARGS - STRING_
1_ ARGS - SUBSTRING_
ARGS - SUM_
ARGS - TEXT_
MATCH_ ARGS - TIME_
BUCKET_ ARGS - TO_
JSONB_ ARGS to_jsonb(value)— any value, or*for the whole row.- TS_
CORRELATE_ ARGS - TS_
EMA_ ARGS - TS_
LAG_ LEAD_ ARGS - TS_
MOVING_ AVG_ ARGS - TS_
PERCENTILE_ ARGS - TS_
STDDEV_ ARGS - TS_
WINDOW_ 1_ ARGS - TYPEOF_
ARGS typeof(expr)/type_of(expr)— accepts any value, including NULL.- VECTOR_
DISTANCE_ ARGS