Updated changelog [skip ci]

.github/workflows/build.yml

@@ -73,7 +73,6 @@ 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 && ^

CHANGELOG.md

@@ -1,7 +1,3 @@
-## 0.7.0 (unreleased)
-
-- Added subscript function for vectors
-
 ## 0.6.2 (unreleased)
 
 - Reduced lock contention with parallel HNSW index builds

README.md

@@ -26,7 +26,7 @@ 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).
 
@@ -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
 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
@@ -413,39 +410,13 @@ 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 +430,7 @@ 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 +438,7 @@ 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 +447,6 @@ 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 +552,7 @@ 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.
+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
--- 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:
 
 ```sql
 BEGIN;
@@ -704,7 +630,7 @@ Function | Description | Added
 avg(vector) → vector | average |
 sum(vector) → vector | sum | 0.5.0
 
-## Installation Notes - Linux and Mac
+## Installation Notes
 
 ### Postgres Location
 
@@ -746,24 +672,12 @@ If compilation fails and the output includes `warning: no such sysroot directory
 
 ### 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
@@ -948,12 +862,6 @@ 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 +874,12 @@ 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--0.6.2--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
 );
 

src/hnsw.h

@@ -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
 {

src/hnswbuild.c

@@ -436,7 +436,7 @@ InsertTupleInMemory(HnswBuildState * buildstate, HnswElement element)
 	int			m = buildstate->m;
 	char	   *base = buildstate->hnswarea;
 
-	/* Wait if another process needs exclusive lock on entry lock */
+	/* Wait if another process needs exclusive lock */
 	LWLockAcquire(entryWaitLock, LW_EXCLUSIVE);
 	LWLockRelease(entryWaitLock);
 
@@ -450,7 +450,7 @@ InsertTupleInMemory(HnswBuildState * buildstate, HnswElement element)
 		/* Release shared lock */
 		LWLockRelease(entryLock);
 
-		/* Tell other processes to wait and get exclusive lock */
+		/* Get exclusive lock */
 		LWLockAcquire(entryWaitLock, LW_EXCLUSIVE);
 		LWLockAcquire(entryLock, LW_EXCLUSIVE);
 		LWLockRelease(entryWaitLock);

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]';