Added test for arrays

.github/workflows/build.yml

@@ -28,7 +28,7 @@ jobs:
           dev-files: true
       - run: make
         env:
-          PG_CFLAGS: -DUSE_ASSERT_CHECKING -Wall -Wextra -Werror -Wno-unused-parameter -Wno-sign-compare
+          PG_CFLAGS: -Wall -Wextra -Werror -Wno-unused-parameter -Wno-sign-compare
       - run: |
           export PG_CONFIG=`which pg_config`
           sudo --preserve-env=PG_CONFIG make install
@@ -49,7 +49,7 @@ jobs:
           postgres-version: 14
       - run: make
         env:
-          PG_CFLAGS: -DUSE_ASSERT_CHECKING -Wall -Wextra -Werror -Wno-unused-parameter
+          PG_CFLAGS: -Wall -Wextra -Werror -Wno-unused-parameter
       - run: make install
       - run: make installcheck
       - if: ${{ failure() }}
@@ -60,9 +60,7 @@ jobs:
           wget -q https://github.com/postgres/postgres/archive/refs/tags/REL_14_10.tar.gz
           tar xf REL_14_10.tar.gz
       - run: make prove_installcheck PROVE_FLAGS="-I ./postgres-REL_14_10/src/test/perl" PERL5LIB="/Users/runner/perl5/lib/perl5"
-      - run: make clean && /usr/local/opt/llvm@15/bin/scan-build --status-bugs make
-        env:
-          PG_CFLAGS: -DUSE_ASSERT_CHECKING
+      - run: make clean && /usr/local/opt/llvm@15/bin/scan-build --status-bugs make PG_CFLAGS="-DUSE_ASSERT_CHECKING"
   windows:
     runs-on: windows-latest
     if: ${{ !startsWith(github.ref_name, 'mac') }}
@@ -73,7 +71,6 @@ jobs:
           postgres-version: 14
       - run: |
           call "C:\Program Files\Microsoft Visual Studio\2022\Enterprise\VC\Auxiliary\Build\vcvars64.bat" && ^
-          cd %TEMP% && ^
           nmake /NOLOGO /F Makefile.win && ^
           nmake /NOLOGO /F Makefile.win install && ^
           nmake /NOLOGO /F Makefile.win installcheck && ^
@@ -100,15 +97,4 @@ jobs:
           sudo -u postgres make installcheck
           sudo -u postgres make prove_installcheck
         env:
-          PG_CFLAGS: -DUSE_ASSERT_CHECKING -Wall -Wextra -Werror -Wno-unused-parameter -Wno-sign-compare
-  valgrind:
-    if: ${{ !startsWith(github.ref_name, 'mac') && !startsWith(github.ref_name, 'windows') }}
-    runs-on: ubuntu-latest
-    steps:
-      - uses: actions/checkout@v4
-      - uses: ankane/setup-postgres-valgrind@v1
-        with:
-          postgres-version: 16
-      - run: make
-      - run: sudo --preserve-env=PG_CONFIG make install
-      - run: make installcheck
+          PG_CFLAGS: -Wall -Wextra -Werror -Wno-unused-parameter -Wno-sign-compare

CHANGELOG.md

@@ -1,16 +1,6 @@
-## 0.7.0 (unreleased)
-
-- Added subscript function for vectors
-
-## 0.6.2 (unreleased)
-
-- Reduced lock contention with parallel HNSW index builds
-
-## 0.6.1 (2024-03-04)
+## 0.6.1 (unreleased)
 
 - Fixed error with `ANALYZE` and vectors with different dimensions
-- Fixed error with `shared_preload_libraries`
-- Fixed vector subtraction being marked as commutative
 
 ## 0.6.0 (2024-01-29)
 

LICENSE

@@ -1,4 +1,4 @@
-Portions Copyright (c) 1996-2024, PostgreSQL Global Development Group
+Portions Copyright (c) 1996-2023, PostgreSQL Global Development Group
 
 Portions Copyright (c) 1994, The Regents of the University of California
 

META.json

@@ -2,7 +2,7 @@
 	"name": "vector",
 	"abstract": "Open-source vector similarity search for Postgres",
 	"description": "Supports L2 distance, inner product, and cosine distance",
-	"version": "0.6.1",
+	"version": "0.6.0",
 	"maintainer": [
 		"Andrew Kane <andrew@ankane.org>"
 	],
@@ -20,7 +20,7 @@
 		"vector": {
 			"file": "sql/vector.sql",
 			"docfile": "README.md",
-			"version": "0.6.1",
+			"version": "0.6.0",
 			"abstract": "Open-source vector similarity search for Postgres"
 		}
 	},

Makefile

@@ -1,5 +1,5 @@
 EXTENSION = vector
-EXTVERSION = 0.6.1
+EXTVERSION = 0.6.0
 
 MODULE_big = vector
 DATA = $(wildcard sql/*--*.sql)
@@ -71,9 +71,11 @@ PG_MAJOR ?= 16
 .PHONY: docker
 
 docker:
-	docker build --pull --no-cache --build-arg PG_MAJOR=$(PG_MAJOR) -t pgvector/pgvector:pg$(PG_MAJOR) -t pgvector/pgvector:$(EXTVERSION)-pg$(PG_MAJOR) .
+	docker build --pull --no-cache --build-arg PG_MAJOR=$(PG_MAJOR) -t pgvector/pgvector:pg$(PG_MAJOR) .
+	docker build --build-arg PG_MAJOR=$(PG_MAJOR) -t pgvector/pgvector:$(EXTVERSION)-pg$(PG_MAJOR) .
 
 .PHONY: docker-release
 
 docker-release:
-	docker buildx build --push --pull --no-cache --platform linux/amd64,linux/arm64 --build-arg PG_MAJOR=$(PG_MAJOR) -t pgvector/pgvector:pg$(PG_MAJOR) -t pgvector/pgvector:$(EXTVERSION)-pg$(PG_MAJOR) .
+	docker buildx build --push --pull --no-cache --platform linux/amd64,linux/arm64 --build-arg PG_MAJOR=$(PG_MAJOR) -t pgvector/pgvector:pg$(PG_MAJOR) .
+	docker buildx build --push --platform linux/amd64,linux/arm64 --build-arg PG_MAJOR=$(PG_MAJOR) -t pgvector/pgvector:$(EXTVERSION)-pg$(PG_MAJOR) .

Makefile.win

@@ -1,5 +1,5 @@
 EXTENSION = vector
-EXTVERSION = 0.6.1
+EXTVERSION = 0.6.0
 
 OBJS = src\hnsw.obj src\hnswbuild.obj src\hnswinsert.obj src\hnswscan.obj src\hnswutils.obj src\hnswvacuum.obj src\ivfbuild.obj src\ivfflat.obj src\ivfinsert.obj src\ivfkmeans.obj src\ivfscan.obj src\ivfutils.obj src\ivfvacuum.obj src\vector.obj
 HEADERS = src\vector.h

README.md

@@ -20,15 +20,15 @@ Compile and install the extension (supports Postgres 12+)
 
 ```sh
 cd /tmp
-git clone --branch v0.6.1 https://github.com/pgvector/pgvector.git
+git clone --branch v0.6.0 https://github.com/pgvector/pgvector.git
 cd pgvector
 make
 make install # may need sudo
 ```
 
-See the [installation notes](#installation-notes---linux-and-mac) if you run into issues
+See the [installation notes](#installation-notes) if you run into issues
 
-You can also install it with [Docker](#docker), [Homebrew](#homebrew), [PGXN](#pgxn), [APT](#apt), [Yum](#yum), [pkg](#pkg), or [conda-forge](#conda-forge), and it comes preinstalled with [Postgres.app](#postgresapp) and many [hosted providers](#hosted-postgres). There are also instructions for [GitHub Actions](https://github.com/pgvector/setup-pgvector).
+You can also install it with [Docker](#docker), [Homebrew](#homebrew), [PGXN](#pgxn), [APT](#apt), [Yum](#yum), or [conda-forge](#conda-forge), and it comes preinstalled with [Postgres.app](#postgresapp) and many [hosted providers](#hosted-postgres). There are also instructions for [GitHub Actions](https://github.com/pgvector/setup-pgvector).
 
 ### Windows
 
@@ -44,15 +44,12 @@ Then use `nmake` to build:
 
 ```cmd
 set "PGROOT=C:\Program Files\PostgreSQL\16"
-cd %TEMP%
-git clone --branch v0.6.1 https://github.com/pgvector/pgvector.git
+git clone --branch v0.6.0 https://github.com/pgvector/pgvector.git
 cd pgvector
 nmake /F Makefile.win
 nmake /F Makefile.win install
 ```
 
-See the [installation notes](#installation-notes---windows) if you run into issues
-
 You can also install it with [Docker](#docker) or [conda-forge](#conda-forge).
 
 ## Getting Started
@@ -105,12 +102,6 @@ Insert vectors
 INSERT INTO items (embedding) VALUES ('[1,2,3]'), ('[4,5,6]');
 ```
 
-Or load vectors in bulk using `COPY` ([example](https://github.com/pgvector/pgvector-python/blob/master/examples/bulk_loading.py))
-
-```sql
-COPY items (embedding) FROM STDIN WITH (FORMAT BINARY);
-```
-
 Upsert vectors
 
 ```sql
@@ -273,8 +264,6 @@ HINT:  Increase maintenance_work_mem to speed up builds.
 
 Note: Do not set `maintenance_work_mem` so high that it exhausts the memory on the server
 
-Like other index types, it’s faster to create an index after loading your initial data
-
 Starting with 0.6.0, you can also speed up index creation by increasing the number of parallel workers (2 by default)
 
 ```sql
@@ -413,39 +402,13 @@ You can use [Reciprocal Rank Fusion](https://github.com/pgvector/pgvector-python
 
 ## Performance
 
-### Tuning
-
-Use a tool like [PgTune](https://pgtune.leopard.in.ua/) to set initial values for Postgres server parameters.
-
-### Loading
-
-Use `COPY` for bulk loading data ([example](https://github.com/pgvector/pgvector-python/blob/master/examples/bulk_loading.py)).
-
-```sql
-COPY items (embedding) FROM STDIN WITH (FORMAT BINARY);
-```
-
-Add any indexes *after* loading the initial data for best performance.
-
-### Indexing
-
-See index build time for [HNSW](#index-build-time) and [IVFFlat](#index-build-time-1).
-
-In production environments, create indexes concurrently to avoid blocking writes.
-
-```sql
-CREATE INDEX CONCURRENTLY ...
-```
-
-### Querying
-
 Use `EXPLAIN ANALYZE` to debug performance.
 
 ```sql
 EXPLAIN ANALYZE SELECT * FROM items ORDER BY embedding <-> '[3,1,2]' LIMIT 5;
 ```
 
-#### Exact Search
+### Exact Search
 
 To speed up queries without an index, increase `max_parallel_workers_per_gather`.
 
@@ -459,7 +422,7 @@ If vectors are normalized to length 1 (like [OpenAI embeddings](https://platform
 SELECT * FROM items ORDER BY embedding <#> '[3,1,2]' LIMIT 5;
 ```
 
-#### Approximate Search
+### Approximate Search
 
 To speed up queries with an IVFFlat index, increase the number of inverted lists (at the expense of recall).
 
@@ -467,7 +430,7 @@ To speed up queries with an IVFFlat index, increase the number of inverted lists
 CREATE INDEX ON items USING ivfflat (embedding vector_l2_ops) WITH (lists = 1000);
 ```
 
-### Vacuuming
+## Vacuuming
 
 Vacuuming can take a while for HNSW indexes. Speed it up by reindexing first.
 
@@ -476,41 +439,6 @@ REINDEX INDEX CONCURRENTLY index_name;
 VACUUM table_name;
 ```
 
-## Monitoring
-
-Monitor performance with [pg_stat_statements](https://www.postgresql.org/docs/current/pgstatstatements.html) (be sure to add it to `shared_preload_libraries`).
-
-```sql
-CREATE EXTENSION pg_stat_statements;
-```
-
-Get the most time-consuming queries with:
-
-```sql
-SELECT query, calls, ROUND((total_plan_time + total_exec_time) / calls) AS avg_time_ms,
-    ROUND((total_plan_time + total_exec_time) / 60000) AS total_time_min
-    FROM pg_stat_statements ORDER BY total_plan_time + total_exec_time DESC LIMIT 20;
-```
-
-Note: Replace `total_plan_time + total_exec_time` with `total_time` for Postgres < 13
-
-Monitor recall by comparing results from approximate search with exact search.
-
-```sql
-BEGIN;
-SET LOCAL enable_indexscan = off; -- use exact search
-SELECT ...
-COMMIT;
-```
-
-## Scaling
-
-Scale pgvector the same way you scale Postgres.
-
-Scale vertically by increasing memory, CPU, and storage on a single instance. Use existing tools to [tune parameters](#tuning) and [monitor performance](#monitoring).
-
-Scale horizontally with [replicas](https://www.postgresql.org/docs/current/hot-standby.html), or use [Citus](https://github.com/citusdata/citus) or another approach for sharding ([example](https://github.com/pgvector/pgvector-python/blob/master/examples/citus.py)).
-
 ## Languages
 
 Use pgvector from any language with a Postgres client. You can even generate and store vectors in one language and query them in another.
@@ -616,17 +544,7 @@ SELECT pg_size_pretty(pg_relation_size('index_name'));
 
 #### Why isn’t a query using an index?
 
-The query needs to have an `ORDER BY` and `LIMIT`, and the `ORDER BY` must be the result of a distance operator, not an expression.
-
-```sql
--- index
-ORDER BY embedding <=> '[3,1,2]' LIMIT 5;
-
--- no index
-ORDER BY 1 - (embedding <=> '[3,1,2]') DESC LIMIT 5;
-```
-
-You can encourage the planner to use an index for a query with:
+The cost estimation in pgvector < 0.4.3 does not always work well with the planner. You can encourage the planner to use an index for a query with:
 
 ```sql
 BEGIN;
@@ -655,10 +573,6 @@ or choose to store vectors inline:
 ALTER TABLE items ALTER COLUMN embedding SET STORAGE PLAIN;
 ```
 
-#### Why are there less results for a query after adding an HNSW index?
-
-Results are limited by the size of the dynamic candidate list (`hnsw.ef_search`). There may be even less results due to dead tuples or filtering conditions in the query. We recommend setting `hnsw.ef_search` to at least twice the `LIMIT` of the query. If you need more than 500 results, use an IVFFlat index instead.
-
 #### Why are there less results for a query after adding an IVFFlat index?
 
 The index was likely created with too little data for the number of lists. Drop the index until the table has more data.
@@ -667,8 +581,6 @@ The index was likely created with too little data for the number of lists. Drop
 DROP INDEX index_name;
 ```
 
-Results can also be limited by the number of probes (`ivfflat.probes`).
-
 ## Reference
 
 ### Vector Type
@@ -704,7 +616,7 @@ Function | Description | Added
 avg(vector) → vector | average |
 sum(vector) → vector | sum | 0.5.0
 
-## Installation Notes - Linux and Mac
+## Installation Notes
 
 ### Postgres Location
 
@@ -744,26 +656,6 @@ Note: Replace `16` with your Postgres server version
 
 If compilation fails and the output includes `warning: no such sysroot directory` on Mac, reinstall Xcode Command Line Tools.
 
-### Portability
-
-By default, pgvector compiles with `-march=native` on some platforms for best performance. However, this can lead to `Illegal instruction` errors if trying to run the compiled extension on a different machine.
-
-To compile for portability, use:
-
-```sh
-make OPTFLAGS=""
-```
-
-## Installation Notes - Windows
-
-### Missing Header
-
-If compilation fails with `Cannot open include file: 'postgres.h': No such file or directory`, make sure `PGROOT` is correct.
-
-### Permissions
-
-If installation fails with `Access is denied`, re-run the installation instructions as an administrator.
-
 ## Additional Installation Methods
 
 ### Docker
@@ -779,7 +671,7 @@ This adds pgvector to the [Postgres image](https://hub.docker.com/_/postgres) (r
 You can also build the image manually:
 
 ```sh
-git clone --branch v0.6.1 https://github.com/pgvector/pgvector.git
+git clone --branch v0.6.0 https://github.com/pgvector/pgvector.git
 cd pgvector
 docker build --build-arg PG_MAJOR=16 -t myuser/pgvector .
 ```
@@ -824,21 +716,6 @@ sudo dnf install pgvector_16
 
 Note: Replace `16` with your Postgres server version
 
-### pkg
-
-Install the FreeBSD package with:
-
-```sh
-pkg install postgresql15-pg_vector
-```
-
-or the port with:
-
-```sh
-cd /usr/ports/databases/pgvector
-make install
-```
-
 ### conda-forge
 
 With Conda Postgres, install from [conda-forge](https://anaconda.org/conda-forge/pgvector) with:
@@ -948,12 +825,6 @@ make installcheck REGRESS=functions                            # regression test
 make prove_installcheck PROVE_TESTS=test/t/001_ivfflat_wal.pl  # TAP test
 ```
 
-To enable assertions:
-
-```sh
-make clean && PG_CFLAGS="-DUSE_ASSERT_CHECKING" make && make install
-```
-
 To enable benchmarking:
 
 ```sh
@@ -966,6 +837,12 @@ To show memory usage:
 make clean && PG_CFLAGS="-DHNSW_MEMORY -DIVFFLAT_MEMORY" make && make install
 ```
 
+To enable assertions:
+
+```sh
+make clean && PG_CFLAGS="-DUSE_ASSERT_CHECKING" make && make install
+```
+
 To get k-means metrics:
 
 ```sh

sql/vector--0.6.0--0.6.1.sql

@@ -1,16 +0,0 @@
--- complain if script is sourced in psql, rather than via CREATE EXTENSION
-\echo Use "ALTER EXTENSION vector UPDATE TO '0.6.1'" to load this file. \quit
-
-DROP OPERATOR - (vector, vector);
-
-CREATE OPERATOR - (
-	LEFTARG = vector, RIGHTARG = vector, PROCEDURE = vector_sub
-);
-
-ALTER OPERATOR <= (vector, vector) SET (
-	RESTRICT = scalarlesel, JOIN = scalarlejoinsel
-);
-
-ALTER OPERATOR >= (vector, vector) SET (
-	RESTRICT = scalargesel, JOIN = scalargejoinsel
-);

sql/vector--0.6.2--0.7.0.sql

@@ -1,7 +0,0 @@
--- 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 vector_subscript(internal) RETURNS internal
-	AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
-
-ALTER TYPE vector SET (SUBSCRIPT = vector_subscript);

sql/vector.sql

@@ -20,16 +20,12 @@ CREATE FUNCTION vector_recv(internal, oid, integer) RETURNS vector
 CREATE FUNCTION vector_send(vector) RETURNS bytea
 	AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
 
-CREATE FUNCTION vector_subscript(internal) RETURNS internal
-	AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
-
 CREATE TYPE vector (
 	INPUT     = vector_in,
 	OUTPUT    = vector_out,
 	TYPMOD_IN = vector_typmod_in,
 	RECEIVE   = vector_recv,
 	SEND      = vector_send,
-	SUBSCRIPT = vector_subscript,
 	STORAGE   = external
 );
 
@@ -184,7 +180,8 @@ CREATE OPERATOR + (
 );
 
 CREATE OPERATOR - (
-	LEFTARG = vector, RIGHTARG = vector, PROCEDURE = vector_sub
+	LEFTARG = vector, RIGHTARG = vector, PROCEDURE = vector_sub,
+	COMMUTATOR = -
 );
 
 CREATE OPERATOR * (
@@ -198,10 +195,11 @@ CREATE OPERATOR < (
 	RESTRICT = scalarltsel, JOIN = scalarltjoinsel
 );
 
+-- should use scalarlesel and scalarlejoinsel, but not supported in Postgres < 11
 CREATE OPERATOR <= (
 	LEFTARG = vector, RIGHTARG = vector, PROCEDURE = vector_le,
 	COMMUTATOR = >= , NEGATOR = > ,
-	RESTRICT = scalarlesel, JOIN = scalarlejoinsel
+	RESTRICT = scalarltsel, JOIN = scalarltjoinsel
 );
 
 CREATE OPERATOR = (
@@ -216,10 +214,11 @@ CREATE OPERATOR <> (
 	RESTRICT = eqsel, JOIN = eqjoinsel
 );
 
+-- should use scalargesel and scalargejoinsel, but not supported in Postgres < 11
 CREATE OPERATOR >= (
 	LEFTARG = vector, RIGHTARG = vector, PROCEDURE = vector_ge,
 	COMMUTATOR = <= , NEGATOR = < ,
-	RESTRICT = scalargesel, JOIN = scalargejoinsel
+	RESTRICT = scalargtsel, JOIN = scalargtjoinsel
 );
 
 CREATE OPERATOR > (

src/hnsw.c

@@ -8,7 +8,6 @@
 #include "commands/progress.h"
 #include "commands/vacuum.h"
 #include "hnsw.h"
-#include "miscadmin.h"
 #include "utils/guc.h"
 #include "utils/selfuncs.h"
 
@@ -29,7 +28,7 @@ static relopt_kind hnsw_relopt_kind;
  * this grows bigger, we should use a shmem_request_hook and
  * RequestAddinShmemSpace() to pre-reserve space for this.
  */
-void
+static void
 HnswInitLockTranche(void)
 {
 	int		   *tranche_ids;
@@ -54,8 +53,7 @@ HnswInitLockTranche(void)
 void
 HnswInit(void)
 {
-	if (!process_shared_preload_libraries_in_progress)
-		HnswInitLockTranche();
+	HnswInitLockTranche();
 
 	hnsw_relopt_kind = add_reloption_kind();
 	add_int_reloption(hnsw_relopt_kind, "m", "Max number of connections",

src/hnsw.h

@@ -80,7 +80,7 @@
 
 #if PG_VERSION_NUM < 130000
 #define list_delete_last(list) list_truncate(list, list_length(list) - 1)
-#define list_sort(list, cmp) ((list) = list_qsort(list, cmp))
+#define list_sort(list, cmp) list_qsort(list, cmp)
 #endif
 
 #define HnswIsElementTuple(tup) ((tup)->type == HNSW_ELEMENT_TUPLE_TYPE)
@@ -129,7 +129,7 @@ HnswPtrDeclare(HnswNeighborArray, HnswNeighborArrayRelptr, HnswNeighborArrayPtr)
 HnswPtrDeclare(HnswNeighborArrayPtr, HnswNeighborsRelptr, HnswNeighborsPtr);
 HnswPtrDeclare(char, DatumRelptr, DatumPtr);
 
-struct HnswElementData
+typedef struct HnswElementData
 {
 	HnswElementPtr next;
 	ItemPointerData heaptids[HNSW_HEAPTIDS];
@@ -144,7 +144,7 @@ struct HnswElementData
 	BlockNumber neighborPage;
 	DatumPtr	value;
 	LWLock		lock;
-};
+}			HnswElementData;
 
 typedef HnswElementData * HnswElement;
 
@@ -155,12 +155,12 @@ typedef struct HnswCandidate
 	bool		closer;
 }			HnswCandidate;
 
-struct HnswNeighborArray
+typedef struct HnswNeighborArray
 {
 	int			length;
 	bool		closerSet;
 	HnswCandidate items[FLEXIBLE_ARRAY_MEMBER];
-};
+}			HnswNeighborArray;
 
 typedef struct HnswPairingHeapNode
 {
@@ -185,7 +185,6 @@ typedef struct HnswGraph
 
 	/* Entry state */
 	LWLock		entryLock;
-	LWLock		entryWaitLock;
 	HnswElementPtr entryPoint;
 
 	/* Allocations state */
@@ -390,7 +389,6 @@ void		HnswLoadElement(HnswElement element, float *distance, Datum *q, Relation i
 void		HnswSetElementTuple(char *base, HnswElementTuple etup, HnswElement element);
 void		HnswUpdateConnection(char *base, HnswElement element, HnswCandidate * hc, int lm, int lc, int *updateIdx, Relation index, FmgrInfo *procinfo, Oid collation);
 void		HnswLoadNeighbors(HnswElement element, Relation index, int m);
-void		HnswInitLockTranche(void);
 PGDLLEXPORT void HnswParallelBuildMain(dsm_segment *seg, shm_toc *toc);
 
 /* Index access methods */

src/hnswbuild.c

@@ -176,7 +176,7 @@ CreateGraphPages(HnswBuildState * buildstate)
 		Size		etupSize;
 		Size		ntupSize;
 		Size		combinedSize;
-		Pointer		valuePtr = HnswPtrAccess(base, element->value);
+		void	   *valuePtr = HnswPtrAccess(base, element->value);
 
 		/* Update iterator */
 		iter = element->next;
@@ -431,15 +431,10 @@ InsertTupleInMemory(HnswBuildState * buildstate, HnswElement element)
 	HnswGraph  *graph = buildstate->graph;
 	HnswElement entryPoint;
 	LWLock	   *entryLock = &graph->entryLock;
-	LWLock	   *entryWaitLock = &graph->entryWaitLock;
 	int			efConstruction = buildstate->efConstruction;
 	int			m = buildstate->m;
 	char	   *base = buildstate->hnswarea;
 
-	/* Wait if another process needs exclusive lock on entry lock */
-	LWLockAcquire(entryWaitLock, LW_EXCLUSIVE);
-	LWLockRelease(entryWaitLock);
-
 	/* Get entry point */
 	LWLockAcquire(entryLock, LW_SHARED);
 	entryPoint = HnswPtrAccess(base, graph->entryPoint);
@@ -450,10 +445,8 @@ InsertTupleInMemory(HnswBuildState * buildstate, HnswElement element)
 		/* Release shared lock */
 		LWLockRelease(entryLock);
 
-		/* Tell other processes to wait and get exclusive lock */
-		LWLockAcquire(entryWaitLock, LW_EXCLUSIVE);
+		/* Get exclusive lock */
 		LWLockAcquire(entryLock, LW_EXCLUSIVE);
-		LWLockRelease(entryWaitLock);
 
 		/* Get latest entry point after lock is acquired */
 		entryPoint = HnswPtrAccess(base, graph->entryPoint);
@@ -608,9 +601,6 @@ BuildCallback(Relation index, CALLBACK_ITEM_POINTER, Datum *values,
 static void
 InitGraph(HnswGraph * graph, char *base, long memoryTotal)
 {
-	/* Initialize the lock tranche if needed */
-	HnswInitLockTranche();
-
 	HnswPtrStore(base, graph->head, (HnswElement) NULL);
 	HnswPtrStore(base, graph->entryPoint, (HnswElement) NULL);
 	graph->memoryUsed = 0;
@@ -619,7 +609,6 @@ InitGraph(HnswGraph * graph, char *base, long memoryTotal)
 	graph->indtuples = 0;
 	SpinLockInit(&graph->lock);
 	LWLockInitialize(&graph->entryLock, hnsw_lock_tranche_id);
-	LWLockInitialize(&graph->entryWaitLock, hnsw_lock_tranche_id);
 	LWLockInitialize(&graph->allocatorLock, hnsw_lock_tranche_id);
 	LWLockInitialize(&graph->flushLock, hnsw_lock_tranche_id);
 }
@@ -680,10 +669,6 @@ InitBuildState(HnswBuildState * buildstate, Relation heap, Relation index, Index
 	buildstate->efConstruction = HnswGetEfConstruction(index);
 	buildstate->dimensions = TupleDescAttr(index->rd_att, 0)->atttypmod;
 
-	/* Require column to have dimensions to be indexed */
-	if (buildstate->dimensions < 0)
-		elog(ERROR, "column does not have dimensions");
-
 	if (buildstate->dimensions > HNSW_MAX_DIM)
 		elog(ERROR, "column cannot have more than %d dimensions for hnsw index", HNSW_MAX_DIM);
 
@@ -985,14 +970,6 @@ HnswBeginParallel(HnswBuildState * buildstate, bool isconcurrent, int request)
 	/* Report less than allocated so never fails */
 	InitGraph(&hnswshared->graphData, hnswarea, esthnswarea - 1024 * 1024);
 
-	/*
-	 * Avoid base address for relptr for Postgres < 14.5
-	 * https://github.com/postgres/postgres/commit/7201cd18627afc64850537806da7f22150d1a83b
-	 */
-#if PG_VERSION_NUM < 140005
-	hnswshared->graphData.memoryUsed += MAXALIGN(1);
-#endif
-
 	shm_toc_insert(pcxt->toc, PARALLEL_KEY_HNSW_SHARED, hnswshared);
 	shm_toc_insert(pcxt->toc, PARALLEL_KEY_HNSW_AREA, hnswarea);
 

src/hnswutils.c

@@ -860,15 +860,12 @@ HnswSearchLayer(char *base, Datum q, List *ep, int ef, int lc, Relation index, F
 static int
 #if PG_VERSION_NUM >= 130000
 CompareCandidateDistances(const ListCell *a, const ListCell *b)
-{
-	HnswCandidate *hca = lfirst(a);
-	HnswCandidate *hcb = lfirst(b);
 #else
 CompareCandidateDistances(const void *a, const void *b)
-{
-	HnswCandidate *hca = lfirst(*(ListCell **) a);
-	HnswCandidate *hcb = lfirst(*(ListCell **) b);
 #endif
+{
+	HnswCandidate *hca = lfirst((ListCell *) a);
+	HnswCandidate *hcb = lfirst((ListCell *) b);
 
 	if (hca->distance < hcb->distance)
 		return 1;
@@ -891,15 +888,12 @@ CompareCandidateDistances(const void *a, const void *b)
 static int
 #if PG_VERSION_NUM >= 130000
 CompareCandidateDistancesOffset(const ListCell *a, const ListCell *b)
-{
-	HnswCandidate *hca = lfirst(a);
-	HnswCandidate *hcb = lfirst(b);
 #else
 CompareCandidateDistancesOffset(const void *a, const void *b)
-{
-	HnswCandidate *hca = lfirst(*(ListCell **) a);
-	HnswCandidate *hcb = lfirst(*(ListCell **) b);
 #endif
+{
+	HnswCandidate *hca = lfirst((ListCell *) a);
+	HnswCandidate *hcb = lfirst((ListCell *) b);
 
 	if (hca->distance < hcb->distance)
 		return 1;
@@ -958,9 +952,7 @@ SelectNeighbors(char *base, List *c, int lm, int lc, FmgrInfo *procinfo, Oid col
 {
 	List	   *r = NIL;
 	List	   *w = list_copy(c);
-	HnswCandidate **wd;
-	int			wdlen = 0;
-	int			wdoff = 0;
+	pairingheap *wd;
 	HnswNeighborArray *neighbors = HnswGetNeighbors(base, e2, lc);
 	bool		mustCalculate = !neighbors->closerSet;
 	List	   *added = NIL;
@@ -969,7 +961,7 @@ SelectNeighbors(char *base, List *c, int lm, int lc, FmgrInfo *procinfo, Oid col
 	if (list_length(w) <= lm)
 		return w;
 
-	wd = palloc(sizeof(HnswCandidate *) * list_length(w));
+	wd = pairingheap_allocate(CompareNearestCandidates, NULL);
 
 	/* Ensure order of candidates is deterministic for closer caching */
 	if (sortCandidates)
@@ -1035,21 +1027,21 @@ SelectNeighbors(char *base, List *c, int lm, int lc, FmgrInfo *procinfo, Oid col
 		if (e->closer)
 			r = lappend(r, e);
 		else
-			wd[wdlen++] = e;
+			pairingheap_add(wd, &(CreatePairingHeapNode(e)->ph_node));
 	}
 
 	/* Cached value can only be used in future if sorted deterministically */
 	neighbors->closerSet = sortCandidates;
 
 	/* Keep pruned connections */
-	while (wdoff < wdlen && list_length(r) < lm)
-		r = lappend(r, wd[wdoff++]);
+	while (!pairingheap_is_empty(wd) && list_length(r) < lm)
+		r = lappend(r, ((HnswPairingHeapNode *) pairingheap_remove_first(wd))->inner);
 
 	/* Return pruned for update connections */
 	if (pruned != NULL)
 	{
-		if (wdoff < wdlen)
-			*pruned = wd[wdoff];
+		if (!pairingheap_is_empty(wd))
+			*pruned = ((HnswPairingHeapNode *) pairingheap_first(wd))->inner;
 		else
 			*pruned = linitial(w);
 	}

src/vector.c

@@ -4,18 +4,11 @@
 
 #include "catalog/pg_type.h"
 #include "common/shortest_dec.h"
-#include "executor/execExpr.h"
 #include "fmgr.h"
 #include "hnsw.h"
 #include "ivfflat.h"
 #include "lib/stringinfo.h"
 #include "libpq/pqformat.h"
-#include "nodes/makefuncs.h"
-#include "nodes/nodeFuncs.h"
-#include "nodes/subscripting.h"
-#include "parser/parse_coerce.h"
-#include "parser/parse_expr.h"
-#include "parser/parse_node.h"
 #include "port.h"				/* for strtof() */
 #include "utils/array.h"
 #include "utils/builtins.h"
@@ -425,180 +418,6 @@ vector_send(PG_FUNCTION_ARGS)
 	PG_RETURN_BYTEA_P(pq_endtypsend(&buf));
 }
 
-/*
- * Transform the subscript expressions
- */
-static void
-vector_subscript_transform(SubscriptingRef *sbsref, List *indirection, ParseState *pstate, bool isSlice, bool isAssignment)
-{
-	A_Indices  *ai;
-	Node	   *subexpr;
-
-	if (list_length(indirection) != 1)
-		ereport(ERROR,
-				(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
-				 errmsg("vector allows only one subscript"),
-				 parser_errposition(pstate,
-									exprLocation((Node *) indirection))));
-
-	ai = linitial_node(A_Indices, indirection);
-
-	if (isSlice)
-	{
-		if (ai->lidx)
-		{
-			subexpr = transformExpr(pstate, ai->lidx, pstate->p_expr_kind);
-			/* If it's not int4 already, try to coerce */
-			subexpr = coerce_to_target_type(pstate,
-											subexpr, exprType(subexpr),
-											INT4OID, -1,
-											COERCION_ASSIGNMENT,
-											COERCE_IMPLICIT_CAST,
-											-1);
-			if (subexpr == NULL)
-				ereport(ERROR,
-						(errcode(ERRCODE_DATATYPE_MISMATCH),
-						 errmsg("vector subscript must have type integer"),
-						 parser_errposition(pstate, exprLocation(ai->lidx))));
-		}
-		else if (!ai->is_slice)
-		{
-			/* Make a constant 1 */
-			subexpr = (Node *) makeConst(INT4OID,
-										 -1,
-										 InvalidOid,
-										 sizeof(int32),
-										 Int32GetDatum(1),
-										 false,
-										 true); /* pass by value */
-		}
-		else
-		{
-			/* Slice with omitted lower bound, put NULL into the list */
-			subexpr = NULL;
-		}
-		sbsref->reflowerindexpr = list_make1(subexpr);
-	}
-	else
-		Assert(ai->lidx == NULL && !ai->is_slice);
-
-	if (ai->uidx)
-	{
-		subexpr = transformExpr(pstate, ai->uidx, pstate->p_expr_kind);
-		/* If it's not int4 already, try to coerce */
-		subexpr = coerce_to_target_type(pstate,
-										subexpr, exprType(subexpr),
-										INT4OID, -1,
-										COERCION_ASSIGNMENT,
-										COERCE_IMPLICIT_CAST,
-										-1);
-		if (subexpr == NULL)
-			ereport(ERROR,
-					(errcode(ERRCODE_DATATYPE_MISMATCH),
-					 errmsg("array subscript must have type integer"),
-					 parser_errposition(pstate, exprLocation(ai->uidx))));
-	}
-	else
-	{
-		/* Slice with omitted upper bound, put NULL into the list */
-		Assert(isSlice && ai->is_slice);
-		subexpr = NULL;
-	}
-	sbsref->refupperindexpr = list_make1(subexpr);
-
-	if (isSlice)
-		sbsref->refrestype = sbsref->refcontainertype;
-	else
-		sbsref->refrestype = FLOAT4OID;
-}
-
-/*
- * Fetch a vector element
- */
-static void
-vector_subscript_fetch(ExprState *state, ExprEvalStep *op, ExprContext *econtext)
-{
-	SubscriptingRefState *sbsrefstate = op->d.sbsref.state;
-	Vector	   *vec = DatumGetVector(*op->resvalue);
-	int			index = DatumGetInt32(sbsrefstate->upperindex[0]);
-
-	if (index < 1 || index > vec->dim)
-		*op->resnull = true;
-	else
-		*op->resvalue = Float4GetDatum(vec->x[index - 1]);
-}
-
-/*
- * Fetch a vector slice
- */
-static void
-vector_subscript_fetch_slice(ExprState *state, ExprEvalStep *op, ExprContext *econtext)
-{
-	SubscriptingRefState *sbsrefstate = op->d.sbsref.state;
-
-	if (sbsrefstate->upperprovided[0] && sbsrefstate->upperindexnull[0])
-		*op->resnull = true;
-	else if (sbsrefstate->lowerprovided[0] && sbsrefstate->lowerindexnull[0])
-		*op->resnull = true;
-	else
-	{
-		Vector	   *vec = DatumGetVector(*op->resvalue);
-		int			upperindex = sbsrefstate->upperprovided[0] ? DatumGetInt32(sbsrefstate->upperindex[0]) : vec->dim;
-		int			lowerindex = sbsrefstate->lowerprovided[0] ? DatumGetInt32(sbsrefstate->lowerindex[0]) : 1;
-		int			dim;
-		Vector	   *result;
-
-		if (upperindex > vec->dim)
-			upperindex = vec->dim;
-
-		if (lowerindex < 1)
-			lowerindex = 1;
-
-		dim = upperindex - lowerindex + 1;
-
-		CheckDim(dim);
-
-		result = InitVector(dim);
-		for (int i = 0; i < dim; i++)
-			result->x[i] = vec->x[lowerindex + i - 1];
-
-		*op->resvalue = PointerGetDatum(result);
-	}
-}
-
-/*
- * Set up execution state for a vector subscript operation
- */
-static void
-vector_exec_setup(const SubscriptingRef *sbsref, SubscriptingRefState *sbsrefstate, SubscriptExecSteps *methods)
-{
-	methods->sbs_check_subscripts = NULL;
-	if (sbsrefstate->numlower != 0)
-		methods->sbs_fetch = vector_subscript_fetch_slice;
-	else
-		methods->sbs_fetch = vector_subscript_fetch;
-	methods->sbs_assign = NULL;
-	methods->sbs_fetch_old = NULL;
-}
-
-/*
- * Subscript handler
- */
-PGDLLEXPORT PG_FUNCTION_INFO_V1(vector_subscript);
-Datum
-vector_subscript(PG_FUNCTION_ARGS)
-{
-	static const SubscriptRoutines sbsroutines = {
-		.transform = vector_subscript_transform,
-		.exec_setup = vector_exec_setup,
-		.fetch_strict = true,	/* fetch returns NULL for NULL inputs */
-		.fetch_leakproof = true,	/* fetch returns NULL for bad subscript */
-		.store_leakproof = false	/* ... but assignment throws error */
-	};
-
-	PG_RETURN_POINTER(&sbsroutines);
-}
-
 /*
  * Convert vector to vector
  * This is needed to check the type modifier

test/expected/functions.out

@@ -24,112 +24,6 @@ SELECT '[1e37]'::vector * '[1e37]';
 ERROR:  value out of range: overflow
 SELECT '[1e-37]'::vector * '[1e-37]';
 ERROR:  value out of range: underflow
-SELECT ('[1,2,3]'::vector)[0];
- vector 
---------
-       
-(1 row)
-
-SELECT ('[1,2,3]'::vector)[1];
- vector 
---------
-      1
-(1 row)
-
-SELECT ('[1,2,3]'::vector)[2];
- vector 
---------
-      2
-(1 row)
-
-SELECT ('[1,2,3]'::vector)[3];
- vector 
---------
-      3
-(1 row)
-
-SELECT ('[1,2,3]'::vector)[4];
- vector 
---------
-       
-(1 row)
-
-SELECT ('[1,2,3]'::vector)[1:1];
- vector 
---------
- [1]
-(1 row)
-
-SELECT ('[1,2,3]'::vector)[1:2];
- vector 
---------
- [1,2]
-(1 row)
-
-SELECT ('[1,2,3]'::vector)[2:4];
- vector 
---------
- [2,3]
-(1 row)
-
-SELECT ('[1,2,3]'::vector)[-2:2];
- vector 
---------
- [1,2]
-(1 row)
-
-SELECT ('[1,2,3]'::vector)[2:1];
-ERROR:  vector must have at least 1 dimension
-SELECT ('[1,2,3]'::vector)[:];
- vector  
----------
- [1,2,3]
-(1 row)
-
-SELECT ('[1,2,3]'::vector)[:2];
- vector 
---------
- [1,2]
-(1 row)
-
-SELECT ('[1,2,3]'::vector)[2:];
- vector 
---------
- [2,3]
-(1 row)
-
-SELECT ('[1,2,3]'::vector)[:4];
- vector  
----------
- [1,2,3]
-(1 row)
-
-SELECT ('[1,2,3]'::vector)[-2:];
- vector  
----------
- [1,2,3]
-(1 row)
-
-SELECT ('[1,2,3]'::vector)[NULL];
- vector 
---------
-       
-(1 row)
-
-SELECT ('[1,2,3]'::vector)[NULL:2];
- vector 
---------
- 
-(1 row)
-
-SELECT ('[1,2,3]'::vector)[2:NULL];
- vector 
---------
- 
-(1 row)
-
-SELECT ('[1,2,3]'::vector)[1][1];
-ERROR:  vector allows only one subscript
 SELECT '[1,2,3]'::vector = '[1,2,3]';
  ?column? 
 ----------

test/sql/functions.sql

@@ -6,26 +6,6 @@ SELECT '[1,2,3]'::vector * '[4,5,6]';
 SELECT '[1e37]'::vector * '[1e37]';
 SELECT '[1e-37]'::vector * '[1e-37]';
 
-SELECT ('[1,2,3]'::vector)[0];
-SELECT ('[1,2,3]'::vector)[1];
-SELECT ('[1,2,3]'::vector)[2];
-SELECT ('[1,2,3]'::vector)[3];
-SELECT ('[1,2,3]'::vector)[4];
-SELECT ('[1,2,3]'::vector)[1:1];
-SELECT ('[1,2,3]'::vector)[1:2];
-SELECT ('[1,2,3]'::vector)[2:4];
-SELECT ('[1,2,3]'::vector)[-2:2];
-SELECT ('[1,2,3]'::vector)[2:1];
-SELECT ('[1,2,3]'::vector)[:];
-SELECT ('[1,2,3]'::vector)[:2];
-SELECT ('[1,2,3]'::vector)[2:];
-SELECT ('[1,2,3]'::vector)[:4];
-SELECT ('[1,2,3]'::vector)[-2:];
-SELECT ('[1,2,3]'::vector)[NULL];
-SELECT ('[1,2,3]'::vector)[NULL:2];
-SELECT ('[1,2,3]'::vector)[2:NULL];
-SELECT ('[1,2,3]'::vector)[1][1];
-
 SELECT '[1,2,3]'::vector = '[1,2,3]';
 SELECT '[1,2,3]'::vector = '[1,2]';
 

test/t/020_hnsw_array.pl

@@ -0,0 +1,113 @@
+use strict;
+use warnings;
+use PostgresNode;
+use TestLib;
+use Test::More;
+
+my $node;
+my @queries = ();
+my @expected;
+my $limit = 20;
+
+sub test_recall
+{
+	my ($min, $operator) = @_;
+	my $correct = 0;
+	my $total = 0;
+
+	my $explain = $node->safe_psql("postgres", qq(
+		SET enable_seqscan = off;
+		EXPLAIN ANALYZE SELECT i FROM tst ORDER BY v $operator '$queries[0]' LIMIT $limit;
+	));
+	like($explain, qr/Index Scan/);
+
+	for my $i (0 .. $#queries)
+	{
+		my $actual = $node->safe_psql("postgres", qq(
+			SET enable_seqscan = off;
+			SELECT i FROM tst ORDER BY v $operator '$queries[$i]' LIMIT $limit;
+		));
+		my @actual_ids = split("\n", $actual);
+		my %actual_set = map { $_ => 1 } @actual_ids;
+
+		my @expected_ids = split("\n", $expected[$i]);
+
+		foreach (@expected_ids)
+		{
+			if (exists($actual_set{$_}))
+			{
+				$correct++;
+			}
+			$total++;
+		}
+	}
+
+	cmp_ok($correct / $total, ">=", $min, $operator);
+}
+
+# Initialize node
+$node = get_new_node('node');
+$node->init;
+$node->start;
+
+# Create table
+$node->safe_psql("postgres", "CREATE EXTENSION vector;");
+$node->safe_psql("postgres", "CREATE TABLE tst (i int4, v float4[3]);");
+$node->safe_psql("postgres",
+	"INSERT INTO tst SELECT i, ARRAY[random(), random(), random()] FROM generate_series(1, 10000) i;"
+);
+
+$node->safe_psql("postgres", qq(
+	CREATE FUNCTION float4_l2_distance(float4[], float4[]) RETURNS float8
+		AS 'BEGIN RETURN l2_distance(\$1::vector, \$2::vector); END;'
+		LANGUAGE plpgsql IMMUTABLE STRICT PARALLEL SAFE;
+
+	CREATE FUNCTION float4_l2_squared_distance(float4[], float4[]) RETURNS float8
+		AS 'BEGIN RETURN vector_l2_squared_distance(\$1::vector, \$2::vector); END;'
+		LANGUAGE plpgsql IMMUTABLE STRICT PARALLEL SAFE;
+
+	CREATE OPERATOR <-> (
+		LEFTARG = float4[], RIGHTARG = float4[], PROCEDURE = float4_l2_distance,
+		COMMUTATOR = '<->'
+	);
+
+	CREATE OPERATOR CLASS float4_l2_ops
+		FOR TYPE float4[] USING hnsw AS
+		OPERATOR 1 <-> (float4[], float4[]) FOR ORDER BY float_ops,
+		FUNCTION 1 float4_l2_squared_distance(float4[], float4[]);
+));
+
+# Generate queries
+for (1 .. 20)
+{
+	my $r1 = rand();
+	my $r2 = rand();
+	my $r3 = rand();
+	push(@queries, "{$r1,$r2,$r3}");
+}
+
+# Check each index type
+my @operators = ("<->");
+my @opclasses = ("float4_l2_ops");
+
+for my $i (0 .. $#operators)
+{
+	my $operator = $operators[$i];
+	my $opclass = $opclasses[$i];
+
+	# Get exact results
+	@expected = ();
+	foreach (@queries)
+	{
+		my $res = $node->safe_psql("postgres", "SELECT i FROM tst ORDER BY v $operator '$_' LIMIT $limit;");
+		push(@expected, $res);
+	}
+
+	# Add index
+	$node->safe_psql("postgres", "CREATE INDEX ON tst USING hnsw (v $opclass);");
+
+	my $min = $operator eq "<#>" ? 0.80 : 0.99;
+	test_recall($min, $operator);
+}
+
+done_testing();

vector.control

@@ -1,4 +1,4 @@
 comment = 'vector data type and ivfflat and hnsw access methods'
-default_version = '0.6.1'
+default_version = '0.6.0'
 module_pathname = '$libdir/vector'
 relocatable = true