Updated readme [skip ci]

CHANGELOG.md

@@ -1,3 +1,7 @@
+## 0.7.0 (unreleased)
+
+- Added `subvector` function
+
 ## 0.6.2 (2024-03-18)
 
 - Reduced lock contention with parallel HNSW index builds

README.md

@@ -604,6 +604,18 @@ and query with:
 SELECT * FROM items ORDER BY embedding::vector(3) <-> '[3,1,2]' LIMIT 5;
 ```
 
+#### Are binary vectors supported?
+
+You can store binary vectors and perform exact nearest neighbor search by Hamming distance in Postgres without an extension ([example](https://github.com/pgvector/pgvector-python/blob/master/examples/hash_image_search.py)).
+
+```tsql
+CREATE TABLE items (id bigserial PRIMARY KEY, embedding bit(3));
+INSERT INTO items (embedding) VALUES (B'000'), (B'111');
+SELECT * FROM items ORDER BY bit_count(embedding # B'101') LIMIT 5;
+```
+
+Indexing is not currently supported.
+
 #### Do indexes need to fit into memory?
 
 No, but like other index types, you’ll likely see better performance if they do. You can get the size of an index with:
@@ -694,6 +706,7 @@ cosine_distance(vector, vector) → double precision | cosine distance |
 inner_product(vector, vector) → double precision | inner product |
 l2_distance(vector, vector) → double precision | Euclidean distance |
 l1_distance(vector, vector) → double precision | taxicab distance | 0.5.0
+subvector(vector, integer, integer) → vector | subvector | 0.7.0 [unreleased]
 vector_dims(vector) → integer | number of dimensions |
 vector_norm(vector) → double precision | Euclidean norm |
 

sql/vector--0.6.2--0.7.0.sql

@@ -0,0 +1,5 @@
+-- complain if script is sourced in psql, rather than via CREATE EXTENSION
+\echo Use "ALTER EXTENSION vector UPDATE TO '0.7.0'" to load this file. \quit
+
+CREATE FUNCTION subvector(vector, int, int) RETURNS vector
+	AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;

sql/vector.sql

@@ -58,6 +58,9 @@ CREATE FUNCTION vector_sub(vector, vector) RETURNS vector
 CREATE FUNCTION vector_mul(vector, vector) RETURNS vector
 	AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
 
+CREATE FUNCTION subvector(vector, int, int) RETURNS vector
+	AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
+
 -- private functions
 
 CREATE FUNCTION vector_lt(vector, vector) RETURNS bool

src/vector.c

@@ -860,6 +860,37 @@ vector_mul(PG_FUNCTION_ARGS)
 	PG_RETURN_POINTER(result);
 }
 
+/*
+ * Get a subset of a vector
+ */
+PGDLLEXPORT PG_FUNCTION_INFO_V1(subvector);
+Datum
+subvector(PG_FUNCTION_ARGS)
+{
+	Vector	   *a = PG_GETARG_VECTOR_P(0);
+	int32		start = PG_GETARG_INT32(1);
+	int32		count = PG_GETARG_INT32(2);
+	int32		end = start + count;
+	float	   *ax = a->x;
+	Vector	   *result;
+	int			dim;
+
+	if (start < 1)
+		start = 1;
+
+	if (end > a->dim)
+		end = a->dim + 1;
+
+	dim = end - start;
+	CheckDim(dim);
+	result = InitVector(dim);
+
+	for (int i = 0; i < dim; i++)
+		result->x[i] = ax[start - 1 + i];
+
+	PG_RETURN_POINTER(result);
+}
+
 /*
  * Internal helper to compare vectors
  */

test/expected/functions.out

@@ -208,6 +208,34 @@ SELECT l1_distance('[3e38]', '[-3e38]');
     Infinity
 (1 row)
 
+SELECT subvector('[1,2,3,4,5]', 1, 3);
+ subvector 
+-----------
+ [1,2,3]
+(1 row)
+
+SELECT subvector('[1,2,3,4,5]', 3, 2);
+ subvector 
+-----------
+ [3,4]
+(1 row)
+
+SELECT subvector('[1,2,3,4,5]', -1, 3);
+ subvector 
+-----------
+ [1]
+(1 row)
+
+SELECT subvector('[1,2,3,4,5]', 3, 9);
+ subvector 
+-----------
+ [3,4,5]
+(1 row)
+
+SELECT subvector('[1,2,3,4,5]', 1, 0);
+ERROR:  vector must have at least 1 dimension
+SELECT subvector('[1,2,3,4,5]', -1, 2);
+ERROR:  vector must have at least 1 dimension
 SELECT avg(v) FROM unnest(ARRAY['[1,2,3]'::vector, '[3,5,7]']) v;
     avg    
 -----------

test/expected/input.out

@@ -116,8 +116,30 @@ SELECT '[1, ,3]'::vector;
 ERROR:  invalid input syntax for type vector: "[1, ,3]"
 LINE 1: SELECT '[1, ,3]'::vector;
                ^
+SELECT '[1,2,3]'::vector(3);
+ vector  
+---------
+ [1,2,3]
+(1 row)
+
 SELECT '[1,2,3]'::vector(2);
 ERROR:  expected 2 dimensions, not 3
+SELECT '[1,2,3]'::vector(3, 2);
+ERROR:  invalid type modifier
+LINE 1: SELECT '[1,2,3]'::vector(3, 2);
+                          ^
+SELECT '[1,2,3]'::vector('a');
+ERROR:  invalid input syntax for type integer: "a"
+LINE 1: SELECT '[1,2,3]'::vector('a');
+                          ^
+SELECT '[1,2,3]'::vector(0);
+ERROR:  dimensions for type vector must be at least 1
+LINE 1: SELECT '[1,2,3]'::vector(0);
+                          ^
+SELECT '[1,2,3]'::vector(16001);
+ERROR:  dimensions for type vector cannot exceed 16000
+LINE 1: SELECT '[1,2,3]'::vector(16001);
+                          ^
 SELECT unnest('{"[1,2,3]", "[4,5,6]"}'::vector[]);
  unnest  
 ---------

test/sql/functions.sql

@@ -48,6 +48,13 @@ SELECT l1_distance('[0,0]', '[0,1]');
 SELECT l1_distance('[1,2]', '[3]');
 SELECT l1_distance('[3e38]', '[-3e38]');
 
+SELECT subvector('[1,2,3,4,5]', 1, 3);
+SELECT subvector('[1,2,3,4,5]', 3, 2);
+SELECT subvector('[1,2,3,4,5]', -1, 3);
+SELECT subvector('[1,2,3,4,5]', 3, 9);
+SELECT subvector('[1,2,3,4,5]', 1, 0);
+SELECT subvector('[1,2,3,4,5]', -1, 2);
+
 SELECT avg(v) FROM unnest(ARRAY['[1,2,3]'::vector, '[3,5,7]']) v;
 SELECT avg(v) FROM unnest(ARRAY['[1,2,3]'::vector, '[3,5,7]', NULL]) v;
 SELECT avg(v) FROM unnest(ARRAY[]::vector[]) v;

test/sql/input.sql

@@ -22,7 +22,13 @@ SELECT '[1,]'::vector;
 SELECT '[1a]'::vector;
 SELECT '[1,,3]'::vector;
 SELECT '[1, ,3]'::vector;
+
+SELECT '[1,2,3]'::vector(3);
 SELECT '[1,2,3]'::vector(2);
+SELECT '[1,2,3]'::vector(3, 2);
+SELECT '[1,2,3]'::vector('a');
+SELECT '[1,2,3]'::vector(0);
+SELECT '[1,2,3]'::vector(16001);
 
 SELECT unnest('{"[1,2,3]", "[4,5,6]"}'::vector[]);
 SELECT '{"[1,2,3]"}'::vector(2)[];