postgresql/contrib/bloom/bloom.h

215 lines
7.0 KiB
C
Raw Normal View History

/*-------------------------------------------------------------------------
*
* bloom.h
* Header for bloom index.
*
* Copyright (c) 2016-2019, PostgreSQL Global Development Group
*
* IDENTIFICATION
* contrib/bloom/bloom.h
*
*-------------------------------------------------------------------------
*/
#ifndef _BLOOM_H_
#define _BLOOM_H_
#include "access/generic_xlog.h"
#include "access/indexam.h"
#include "access/itup.h"
#include "access/xlog.h"
#include "fmgr.h"
#include "nodes/pathnodes.h"
/* Support procedures numbers */
#define BLOOM_HASH_PROC 1
#define BLOOM_NPROC 1
/* Scan strategies */
#define BLOOM_EQUAL_STRATEGY 1
#define BLOOM_NSTRATEGIES 1
/* Opaque for bloom pages */
typedef struct BloomPageOpaqueData
{
2016-06-10 06:02:36 +08:00
OffsetNumber maxoff; /* number of index tuples on page */
uint16 flags; /* see bit definitions below */
uint16 unused; /* placeholder to force maxaligning of size of
* BloomPageOpaqueData and to place
* bloom_page_id exactly at the end of page */
uint16 bloom_page_id; /* for identification of BLOOM indexes */
} BloomPageOpaqueData;
typedef BloomPageOpaqueData *BloomPageOpaque;
/* Bloom page flags */
#define BLOOM_META (1<<0)
#define BLOOM_DELETED (2<<0)
/*
* The page ID is for the convenience of pg_filedump and similar utilities,
* which otherwise would have a hard time telling pages of different index
* types apart. It should be the last 2 bytes on the page. This is more or
* less "free" due to alignment considerations.
*
* See comments above GinPageOpaqueData.
*/
#define BLOOM_PAGE_ID 0xFF83
/* Macros for accessing bloom page structures */
#define BloomPageGetOpaque(page) ((BloomPageOpaque) PageGetSpecialPointer(page))
#define BloomPageGetMaxOffset(page) (BloomPageGetOpaque(page)->maxoff)
#define BloomPageIsMeta(page) \
((BloomPageGetOpaque(page)->flags & BLOOM_META) != 0)
#define BloomPageIsDeleted(page) \
((BloomPageGetOpaque(page)->flags & BLOOM_DELETED) != 0)
#define BloomPageSetDeleted(page) \
(BloomPageGetOpaque(page)->flags |= BLOOM_DELETED)
#define BloomPageSetNonDeleted(page) \
(BloomPageGetOpaque(page)->flags &= ~BLOOM_DELETED)
#define BloomPageGetData(page) ((BloomTuple *)PageGetContents(page))
#define BloomPageGetTuple(state, page, offset) \
((BloomTuple *)(PageGetContents(page) \
+ (state)->sizeOfBloomTuple * ((offset) - 1)))
#define BloomPageGetNextTuple(state, tuple) \
((BloomTuple *)((Pointer)(tuple) + (state)->sizeOfBloomTuple))
/* Preserved page numbers */
#define BLOOM_METAPAGE_BLKNO (0)
Phase 2 of pgindent updates. Change pg_bsd_indent to follow upstream rules for placement of comments to the right of code, and remove pgindent hack that caused comments following #endif to not obey the general rule. Commit e3860ffa4dd0dad0dd9eea4be9cc1412373a8c89 wasn't actually using the published version of pg_bsd_indent, but a hacked-up version that tried to minimize the amount of movement of comments to the right of code. The situation of interest is where such a comment has to be moved to the right of its default placement at column 33 because there's code there. BSD indent has always moved right in units of tab stops in such cases --- but in the previous incarnation, indent was working in 8-space tab stops, while now it knows we use 4-space tabs. So the net result is that in about half the cases, such comments are placed one tab stop left of before. This is better all around: it leaves more room on the line for comment text, and it means that in such cases the comment uniformly starts at the next 4-space tab stop after the code, rather than sometimes one and sometimes two tabs after. Also, ensure that comments following #endif are indented the same as comments following other preprocessor commands such as #else. That inconsistency turns out to have been self-inflicted damage from a poorly-thought-through post-indent "fixup" in pgindent. This patch is much less interesting than the first round of indent changes, but also bulkier, so I thought it best to separate the effects. Discussion: https://postgr.es/m/E1dAmxK-0006EE-1r@gemulon.postgresql.org Discussion: https://postgr.es/m/30527.1495162840@sss.pgh.pa.us
2017-06-22 03:18:54 +08:00
#define BLOOM_HEAD_BLKNO (1) /* first data page */
/*
* We store Bloom signatures as arrays of uint16 words.
*/
typedef uint16 BloomSignatureWord;
#define SIGNWORDBITS ((int) (BITS_PER_BYTE * sizeof(BloomSignatureWord)))
/*
* Default and maximum Bloom signature length in bits.
*/
#define DEFAULT_BLOOM_LENGTH (5 * SIGNWORDBITS)
#define MAX_BLOOM_LENGTH (256 * SIGNWORDBITS)
/*
* Default and maximum signature bits generated per index key.
*/
#define DEFAULT_BLOOM_BITS 2
#define MAX_BLOOM_BITS (MAX_BLOOM_LENGTH - 1)
/* Bloom index options */
typedef struct BloomOptions
{
int32 vl_len_; /* varlena header (do not touch directly!) */
int bloomLength; /* length of signature in words (not bits!) */
Phase 2 of pgindent updates. Change pg_bsd_indent to follow upstream rules for placement of comments to the right of code, and remove pgindent hack that caused comments following #endif to not obey the general rule. Commit e3860ffa4dd0dad0dd9eea4be9cc1412373a8c89 wasn't actually using the published version of pg_bsd_indent, but a hacked-up version that tried to minimize the amount of movement of comments to the right of code. The situation of interest is where such a comment has to be moved to the right of its default placement at column 33 because there's code there. BSD indent has always moved right in units of tab stops in such cases --- but in the previous incarnation, indent was working in 8-space tab stops, while now it knows we use 4-space tabs. So the net result is that in about half the cases, such comments are placed one tab stop left of before. This is better all around: it leaves more room on the line for comment text, and it means that in such cases the comment uniformly starts at the next 4-space tab stop after the code, rather than sometimes one and sometimes two tabs after. Also, ensure that comments following #endif are indented the same as comments following other preprocessor commands such as #else. That inconsistency turns out to have been self-inflicted damage from a poorly-thought-through post-indent "fixup" in pgindent. This patch is much less interesting than the first round of indent changes, but also bulkier, so I thought it best to separate the effects. Discussion: https://postgr.es/m/E1dAmxK-0006EE-1r@gemulon.postgresql.org Discussion: https://postgr.es/m/30527.1495162840@sss.pgh.pa.us
2017-06-22 03:18:54 +08:00
int bitSize[INDEX_MAX_KEYS]; /* # of bits generated for each
* index key */
2016-06-10 06:02:36 +08:00
} BloomOptions;
/*
* FreeBlockNumberArray - array of block numbers sized so that metadata fill
* all space in metapage.
*/
typedef BlockNumber FreeBlockNumberArray[
MAXALIGN_DOWN(
BLCKSZ - SizeOfPageHeaderData - MAXALIGN(sizeof(BloomPageOpaqueData))
- MAXALIGN(sizeof(uint16) * 2 + sizeof(uint32) + sizeof(BloomOptions))
) / sizeof(BlockNumber)
];
/* Metadata of bloom index */
typedef struct BloomMetaPageData
{
uint32 magickNumber;
uint16 nStart;
uint16 nEnd;
BloomOptions opts;
FreeBlockNumberArray notFullPage;
2016-06-10 06:02:36 +08:00
} BloomMetaPageData;
/* Magic number to distinguish bloom pages among anothers */
#define BLOOM_MAGICK_NUMBER (0xDBAC0DED)
/* Number of blocks numbers fit in BloomMetaPageData */
#define BloomMetaBlockN (sizeof(FreeBlockNumberArray) / sizeof(BlockNumber))
#define BloomPageGetMeta(page) ((BloomMetaPageData *) PageGetContents(page))
typedef struct BloomState
{
FmgrInfo hashFn[INDEX_MAX_KEYS];
Collations with nondeterministic comparison This adds a flag "deterministic" to collations. If that is false, such a collation disables various optimizations that assume that strings are equal only if they are byte-wise equal. That then allows use cases such as case-insensitive or accent-insensitive comparisons or handling of strings with different Unicode normal forms. This functionality is only supported with the ICU provider. At least glibc doesn't appear to have any locales that work in a nondeterministic way, so it's not worth supporting this for the libc provider. The term "deterministic comparison" in this context is from Unicode Technical Standard #10 (https://unicode.org/reports/tr10/#Deterministic_Comparison). This patch makes changes in three areas: - CREATE COLLATION DDL changes and system catalog changes to support this new flag. - Many executor nodes and auxiliary code are extended to track collations. Previously, this code would just throw away collation information, because the eventually-called user-defined functions didn't use it since they only cared about equality, which didn't need collation information. - String data type functions that do equality comparisons and hashing are changed to take the (non-)deterministic flag into account. For comparison, this just means skipping various shortcuts and tie breakers that use byte-wise comparison. For hashing, we first need to convert the input string to a canonical "sort key" using the ICU analogue of strxfrm(). Reviewed-by: Daniel Verite <daniel@manitou-mail.org> Reviewed-by: Peter Geoghegan <pg@bowt.ie> Discussion: https://www.postgresql.org/message-id/flat/1ccc668f-4cbc-0bef-af67-450b47cdfee7@2ndquadrant.com
2019-03-22 19:09:32 +08:00
Oid collations[INDEX_MAX_KEYS];
BloomOptions opts; /* copy of options on index's metapage */
int32 nColumns;
/*
* sizeOfBloomTuple is index-specific, and it depends on reloptions, so
* precompute it
*/
Size sizeOfBloomTuple;
2016-06-10 06:02:36 +08:00
} BloomState;
#define BloomPageGetFreeSpace(state, page) \
(BLCKSZ - MAXALIGN(SizeOfPageHeaderData) \
- BloomPageGetMaxOffset(page) * (state)->sizeOfBloomTuple \
- MAXALIGN(sizeof(BloomPageOpaqueData)))
/*
* Tuples are very different from all other relations
*/
typedef struct BloomTuple
{
ItemPointerData heapPtr;
BloomSignatureWord sign[FLEXIBLE_ARRAY_MEMBER];
2016-06-10 06:02:36 +08:00
} BloomTuple;
#define BLOOMTUPLEHDRSZ offsetof(BloomTuple, sign)
/* Opaque data structure for bloom index scan */
typedef struct BloomScanOpaqueData
{
2016-06-10 06:02:36 +08:00
BloomSignatureWord *sign; /* Scan signature */
BloomState state;
2016-06-10 06:02:36 +08:00
} BloomScanOpaqueData;
typedef BloomScanOpaqueData *BloomScanOpaque;
/* blutils.c */
extern void _PG_init(void);
2016-06-10 06:02:36 +08:00
extern void initBloomState(BloomState *state, Relation index);
extern void BloomFillMetapage(Relation index, Page metaPage);
extern void BloomInitMetapage(Relation index);
extern void BloomInitPage(Page page, uint16 flags);
extern Buffer BloomNewBuffer(Relation index);
2016-06-10 06:02:36 +08:00
extern void signValue(BloomState *state, BloomSignatureWord *sign, Datum value, int attno);
extern BloomTuple *BloomFormTuple(BloomState *state, ItemPointer iptr, Datum *values, bool *isnull);
extern bool BloomPageAddItem(BloomState *state, Page page, BloomTuple *tuple);
/* blvalidate.c */
extern bool blvalidate(Oid opclassoid);
/* index access method interface functions */
extern bool blinsert(Relation index, Datum *values, bool *isnull,
ItemPointer ht_ctid, Relation heapRel,
IndexUniqueCheck checkUnique,
struct IndexInfo *indexInfo);
extern IndexScanDesc blbeginscan(Relation r, int nkeys, int norderbys);
extern int64 blgetbitmap(IndexScanDesc scan, TIDBitmap *tbm);
extern void blrescan(IndexScanDesc scan, ScanKey scankey, int nscankeys,
ScanKey orderbys, int norderbys);
extern void blendscan(IndexScanDesc scan);
extern IndexBuildResult *blbuild(Relation heap, Relation index,
struct IndexInfo *indexInfo);
extern void blbuildempty(Relation index);
extern IndexBulkDeleteResult *blbulkdelete(IndexVacuumInfo *info,
IndexBulkDeleteResult *stats, IndexBulkDeleteCallback callback,
void *callback_state);
extern IndexBulkDeleteResult *blvacuumcleanup(IndexVacuumInfo *info,
IndexBulkDeleteResult *stats);
extern bytea *bloptions(Datum reloptions, bool validate);
extern void blcostestimate(PlannerInfo *root, IndexPath *path,
double loop_count, Cost *indexStartupCost,
Cost *indexTotalCost, Selectivity *indexSelectivity,
double *indexCorrelation, double *indexPages);
#endif