/*
* Open block directory relation, initialize scan keys and minipages
* for ALTER TABLE ADD COLUMN operation.
*/
void
AppendOnlyBlockDirectory_Init_addCol(
AppendOnlyBlockDirectory *blockDirectory,
AppendOnlyEntry *aoEntry,
Snapshot appendOnlyMetaDataSnapshot,
FileSegInfo *segmentFileInfo,
Relation aoRel,
int segno,
int numColumnGroups,
bool isAOCol)
{
Assert(aoEntry != NULL);
blockDirectory->aoRel = aoRel;
blockDirectory->appendOnlyMetaDataSnapshot = appendOnlyMetaDataSnapshot;
if (!OidIsValid(aoEntry->blkdirrelid))
{
Assert(!OidIsValid(aoEntry->blkdiridxid));
blockDirectory->blkdirRel = NULL;
blockDirectory->blkdirIdx = NULL;
blockDirectory->numColumnGroups = 0;
return;
}
blockDirectory->segmentFileInfo = NULL;
blockDirectory->totalSegfiles = -1;
blockDirectory->currentSegmentFileInfo = segmentFileInfo;
blockDirectory->currentSegmentFileNum = segno;
blockDirectory->numColumnGroups = numColumnGroups;
blockDirectory->isAOCol = isAOCol;
blockDirectory->proj = NULL;
Assert(OidIsValid(aoEntry->blkdirrelid));
/*
* TODO: refactor the *_addCol* interface so that opening of
* blockdirectory relation and index, init_internal and
* corresponding cleanup in *_End_addCol() is called only once
* during the add-column operation. Currently, this is being
* called for every appendonly segment.
*/
blockDirectory->blkdirRel =
heap_open(aoEntry->blkdirrelid, RowExclusiveLock);
Assert(OidIsValid(aoEntry->blkdiridxid));
blockDirectory->blkdirIdx =
index_open(aoEntry->blkdiridxid, RowExclusiveLock);
init_internal(blockDirectory);
}
/* ----------
* toast_delete_datum -
*
* Delete a single external stored value.
* ----------
*/
static void
toast_delete_datum(Relation rel, Datum value)
{
struct varlena *attr = (struct varlena *) DatumGetPointer(value);
struct varatt_external toast_pointer;
Relation toastrel;
Relation toastidx;
ScanKeyData toastkey;
SysScanDesc toastscan;
HeapTuple toasttup;
if (!VARATT_IS_EXTERNAL(attr))
return;
/* Must copy to access aligned fields */
VARATT_EXTERNAL_GET_POINTER(toast_pointer, attr);
/*
* Open the toast relation and its index
*/
toastrel = heap_open(toast_pointer.va_toastrelid, RowExclusiveLock);
toastidx = index_open(toastrel->rd_rel->reltoastidxid, RowExclusiveLock);
/*
* Setup a scan key to find chunks with matching va_valueid
*/
ScanKeyInit(&toastkey,
(AttrNumber) 1,
BTEqualStrategyNumber, F_OIDEQ,
ObjectIdGetDatum(toast_pointer.va_valueid));
/*
* Find all the chunks. (We don't actually care whether we see them in
* sequence or not, but since we've already locked the index we might as
* well use systable_beginscan_ordered.)
*/
toastscan = systable_beginscan_ordered(toastrel, toastidx,
SnapshotToast, 1, &toastkey);
while ((toasttup = systable_getnext_ordered(toastscan, ForwardScanDirection)) != NULL)
{
/*
* Have a chunk, delete it
*/
simple_heap_delete(toastrel, &toasttup->t_self);
}
/*
* End scan and close relations
*/
systable_endscan_ordered(toastscan);
index_close(toastidx, RowExclusiveLock);
heap_close(toastrel, RowExclusiveLock);
}
/*
* AppendOnlyBlockDirectory_Init_forSearch
*
* Initialize the block directory to handle the lookup.
*
* If the block directory relation for this appendonly relation
* does not exist before calling this function, set blkdirRel
* and blkdirIdx to NULL, and return.
*/
void
AppendOnlyBlockDirectory_Init_forSearch(
AppendOnlyBlockDirectory *blockDirectory,
AppendOnlyEntry *aoEntry,
Snapshot appendOnlyMetaDataSnapshot,
FileSegInfo **segmentFileInfo,
int totalSegfiles,
Relation aoRel,
int numColumnGroups,
bool isAOCol,
bool *proj)
{
Assert(aoEntry != NULL);
blockDirectory->aoRel = aoRel;
if (!OidIsValid(aoEntry->blkdirrelid))
{
Assert(!OidIsValid(aoEntry->blkdiridxid));
blockDirectory->blkdirRel = NULL;
blockDirectory->blkdirIdx = NULL;
return;
}
ereportif(Debug_appendonly_print_blockdirectory, LOG,
(errmsg("Append-only block directory init for search: "
"(totalSegfiles, numColumnGroups, isAOCol)="
"(%d, %d, %d)",
totalSegfiles, numColumnGroups, isAOCol)));
blockDirectory->segmentFileInfo = segmentFileInfo;
blockDirectory->totalSegfiles = totalSegfiles;
blockDirectory->aoRel = aoRel;
blockDirectory->appendOnlyMetaDataSnapshot = appendOnlyMetaDataSnapshot;
blockDirectory->numColumnGroups = numColumnGroups;
blockDirectory->isAOCol = isAOCol;
blockDirectory->proj = proj;
blockDirectory->currentSegmentFileNum = -1;
Assert(OidIsValid(aoEntry->blkdirrelid));
blockDirectory->blkdirRel =
heap_open(aoEntry->blkdirrelid, AccessShareLock);
Assert(OidIsValid(aoEntry->blkdiridxid));
blockDirectory->blkdirIdx =
index_open(aoEntry->blkdiridxid, AccessShareLock);
init_internal(blockDirectory);
}
/*
* GetNewOid
* Generate a new OID that is unique within the given relation.
*
* Caller must have a suitable lock on the relation.
*
* Uniqueness is promised only if the relation has a unique index on OID.
* This is true for all system catalogs that have OIDs, but might not be
* true for user tables. Note that we are effectively assuming that the
* table has a relatively small number of entries (much less than 2^32)
* and there aren't very long runs of consecutive existing OIDs. Again,
* this is reasonable for system catalogs but less so for user tables.
*
* Since the OID is not immediately inserted into the table, there is a
* race condition here; but a problem could occur only if someone else
* managed to cycle through 2^32 OIDs and generate the same OID before we
* finish inserting our row. This seems unlikely to be a problem. Note
* that if we had to *commit* the row to end the race condition, the risk
* would be rather higher; therefore we use SnapshotDirty in the test,
* so that we will see uncommitted rows.
*/
Oid
GetNewOid(Relation relation)
{
Oid newOid;
Oid oidIndex;
Relation indexrel;
/* If relation doesn't have OIDs at all, caller is confused */
Assert(relation->rd_rel->relhasoids);
/* In bootstrap mode, we don't have any indexes to use */
if (IsBootstrapProcessingMode())
return GetNewObjectId();
/* The relcache will cache the identity of the OID index for us */
oidIndex = RelationGetOidIndex(relation);
/* If no OID index, just hand back the next OID counter value */
if (!OidIsValid(oidIndex))
{
/*
* System catalogs that have OIDs should *always* have a unique OID
* index; we should only take this path for user tables. Give a
* warning if it looks like somebody forgot an index.
*/
if (IsSystemRelation(relation))
elog(WARNING, "generating possibly-non-unique OID for \"%s\"",
RelationGetRelationName(relation));
return GetNewObjectId();
}
/* Otherwise, use the index to find a nonconflicting OID */
indexrel = index_open(oidIndex, AccessShareLock);
newOid = GetNewOidWithIndex(relation, indexrel);
index_close(indexrel, AccessShareLock);
/*
* Most catalog objects need to have the same OID in the master and all
* segments. When creating a new object, the master should allocate the
* OID and tell the segments to use the same, so segments should have no
* need to ever allocate OIDs on their own. Therefore, give a WARNING if
* GetNewOid() is called in a segment. (There are a few exceptions, see
* RelationNeedsSynchronizedOIDs()).
*/
if (Gp_role == GP_ROLE_EXECUTE && RelationNeedsSynchronizedOIDs(relation))
elog(PANIC, "allocated OID %u for relation \"%s\" in segment",
newOid, RelationGetRelationName(relation));
return newOid;
}
/* ----------
* toast_delete_datum -
*
* Delete a single external stored value.
* ----------
*/
static void
toast_delete_datum(Relation rel __attribute__((unused)), Datum value)
{
varattrib *attr = (varattrib *) DatumGetPointer(value);
Relation toastrel;
Relation toastidx;
ScanKeyData toastkey;
IndexScanDesc toastscan;
HeapTuple toasttup;
if (!VARATT_IS_EXTERNAL(attr))
return;
/*
* Open the toast relation and its index
*/
toastrel = heap_open(attr->va_external.va_toastrelid,
RowExclusiveLock);
toastidx = index_open(toastrel->rd_rel->reltoastidxid, RowExclusiveLock);
/*
* Setup a scan key to fetch from the index by va_valueid (we don't
* particularly care whether we see them in sequence or not)
*/
ScanKeyInit(&toastkey,
(AttrNumber) 1,
BTEqualStrategyNumber, F_OIDEQ,
ObjectIdGetDatum(attr->va_external.va_valueid));
/*
* Find all the chunks. (We don't actually care whether we see them in
* sequence or not, but since we've already locked the index we might as
* well use systable_beginscan_ordered.)
*/
toastscan = index_beginscan(toastrel, toastidx,
SnapshotToast, 1, &toastkey);
while ((toasttup = index_getnext(toastscan, ForwardScanDirection)) != NULL)
{
/*
* Have a chunk, delete it
*/
simple_heap_delete(toastrel, &toasttup->t_self);
}
/*
* End scan and close relations
*/
index_endscan(toastscan);
index_close(toastidx, RowExclusiveLock);
heap_close(toastrel, RowExclusiveLock);
}
/*
* systable_beginscan --- set up for heap-or-index scan
*
* rel: catalog to scan, already opened and suitably locked
* indexId: OID of index to conditionally use
* indexOK: if false, forces a heap scan (see notes below)
* snapshot: time qual to use (usually should be SnapshotNow)
* nkeys, key: scan keys
*
* The attribute numbers in the scan key should be set for the heap case.
* If we choose to index, we reset them to 1..n to reference the index
* columns. Note this means there must be one scankey qualification per
* index column! This is checked by the Asserts in the normal, index-using
* case, but won't be checked if the heapscan path is taken.
*
* The routine checks the normal cases for whether an indexscan is safe,
* but caller can make additional checks and pass indexOK=false if needed.
* In standard case indexOK can simply be constant TRUE.
*/
SysScanDesc
systable_beginscan(Relation heapRelation,
Oid indexId,
bool indexOK,
Snapshot snapshot,
int nkeys, ScanKey key)
{
SysScanDesc sysscan;
Relation irel;
if (indexOK &&
!IgnoreSystemIndexes &&
!ReindexIsProcessingIndex(indexId))
irel = index_open(indexId, AccessShareLock);
else
irel = NULL;
sysscan = (SysScanDesc) palloc(sizeof(SysScanDescData));
sysscan->heap_rel = heapRelation;
sysscan->irel = irel;
if (irel)
{
int i;
/*
* Change attribute numbers to be index column numbers.
*
* This code could be generalized to search for the index key numbers
* to substitute, but for now there's no need.
*/
for (i = 0; i < nkeys; i++)
{
Assert(key[i].sk_attno == irel->rd_index->indkey.values[i]);
key[i].sk_attno = i + 1;
}
sysscan->iscan = index_beginscan(heapRelation, irel,
snapshot, nkeys, key);
sysscan->scan = NULL;
}
else
{
sysscan->scan = heap_beginscan(heapRelation, snapshot, nkeys, key);
sysscan->iscan = NULL;
}
return sysscan;
}
/* ----------
* toast_delete_datum -
*
* Delete a single external stored value.
* ----------
*/
static void
toast_delete_datum(Relation rel, Datum value)
{
varattrib *attr = (varattrib *) DatumGetPointer(value);
Relation toastrel;
Relation toastidx;
ScanKeyData toastkey;
IndexScanDesc toastscan;
HeapTuple toasttup;
if (!VARATT_IS_EXTERNAL(attr))
return;
/*
* Open the toast relation and it's index
*/
toastrel = heap_open(attr->va_content.va_external.va_toastrelid,
RowExclusiveLock);
toastidx = index_open(toastrel->rd_rel->reltoastidxid);
/*
* Setup a scan key to fetch from the index by va_valueid (we don't
* particularly care whether we see them in sequence or not)
*/
ScanKeyEntryInitialize(&toastkey,
(bits16) 0,
(AttrNumber) 1,
(RegProcedure) F_OIDEQ,
ObjectIdGetDatum(attr->va_content.va_external.va_valueid));
/*
* Find the chunks by index
*/
toastscan = index_beginscan(toastrel, toastidx, SnapshotToast,
1, &toastkey);
while ((toasttup = index_getnext(toastscan, ForwardScanDirection)) != NULL)
{
/*
* Have a chunk, delete it
*/
simple_heap_delete(toastrel, &toasttup->t_self);
}
/*
* End scan and close relations
*/
index_endscan(toastscan);
index_close(toastidx);
heap_close(toastrel, RowExclusiveLock);
}
/*
* InitCatCachePhase2 -- external interface for CatalogCacheInitializeCache
*
* One reason to call this routine is to ensure that the relcache has
* created entries for all the catalogs and indexes referenced by catcaches.
* Therefore, provide an option to open the index as well as fixing the
* cache itself. An exception is the indexes on pg_am, which we don't use
* (cf. IndexScanOK).
*/
void
InitCatCachePhase2(CatCache *cache, bool touch_index)
{
if (cache->cc_tupdesc == NULL)
CatalogCacheInitializeCache(cache);
if (touch_index &&
cache->id != AMOID &&
cache->id != AMNAME)
{
Relation idesc;
idesc = index_open(cache->cc_indexoid, AccessShareLock);
index_close(idesc, AccessShareLock);
}
}
/*
* Inits the visimap store.
* The store is ready for usage after this function call.
*
* Assumes a zero-allocated visimap store data structure.
* Assumes that the visimap memory context is active.
*/
void
AppendOnlyVisimapStore_Init(
AppendOnlyVisimapStore *visiMapStore,
Oid visimapRelid,
Oid visimapIdxid,
LOCKMODE lockmode,
Snapshot snapshot,
MemoryContext memoryContext)
{
TupleDesc heapTupleDesc;
ScanKey scanKey;
Assert(visiMapStore);
Assert(CurrentMemoryContext == memoryContext);
Assert(OidIsValid(visimapRelid));
Assert(OidIsValid(visimapIdxid));
visiMapStore->snapshot = snapshot;
visiMapStore->memoryContext = memoryContext;
visiMapStore->visimapRelation = heap_open(
visimapRelid, lockmode);
visiMapStore->visimapIndex = index_open(
visimapIdxid, lockmode);
heapTupleDesc =
RelationGetDescr(visiMapStore->visimapRelation);
Assert(heapTupleDesc->natts == Natts_pg_aovisimap);
visiMapStore->scanKeys = palloc0(sizeof(ScanKeyData) * APPENDONLY_VISIMAP_INDEX_SCAN_KEY_NUM);
// scan key: segno
scanKey = visiMapStore->scanKeys;
ScanKeyInit(scanKey,
Anum_pg_aovisimap_segno, /* segno */
BTEqualStrategyNumber,
F_INT4EQ,
0);
// scan key: firstRowNum
scanKey++;
ScanKeyInit(scanKey,
Anum_pg_aovisimap_firstrownum, /* attribute number to scan */
BTEqualStrategyNumber, /* strategy */
F_INT8EQ, /* reg proc to use */
0);
}
/*
* SQL-callable function to clean the insert pending list
*/
Datum
gin_clean_pending_list(PG_FUNCTION_ARGS)
{
Oid indexoid = PG_GETARG_OID(0);
Relation indexRel = index_open(indexoid, AccessShareLock);
IndexBulkDeleteResult stats;
GinState ginstate;
if (RecoveryInProgress())
ereport(ERROR,
(errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
errmsg("recovery is in progress"),
errhint("GIN pending list cannot be cleaned up during recovery.")));
/* Must be a GIN index */
if (indexRel->rd_rel->relkind != RELKIND_INDEX ||
indexRel->rd_rel->relam != GIN_AM_OID)
ereport(ERROR,
(errcode(ERRCODE_WRONG_OBJECT_TYPE),
errmsg("\"%s\" is not a GIN index",
RelationGetRelationName(indexRel))));
/*
* Reject attempts to read non-local temporary relations; we would be
* likely to get wrong data since we have no visibility into the owning
* session's local buffers.
*/
if (RELATION_IS_OTHER_TEMP(indexRel))
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("cannot access temporary indexes of other sessions")));
/* User must own the index (comparable to privileges needed for VACUUM) */
if (!pg_class_ownercheck(indexoid, GetUserId()))
aclcheck_error(ACLCHECK_NOT_OWNER, ACL_KIND_CLASS,
RelationGetRelationName(indexRel));
memset(&stats, 0, sizeof(stats));
initGinState(&ginstate, indexRel);
ginInsertCleanup(&ginstate, true, true, &stats);
index_close(indexRel, AccessShareLock);
PG_RETURN_INT64((int64) stats.pages_deleted);
}
/*
* enum_endpoint: common code for enum_first/enum_last
*/
static Oid
enum_endpoint(Oid enumtypoid, ScanDirection direction)
{
Relation enum_rel;
Relation enum_idx;
SysScanDesc enum_scan;
HeapTuple enum_tuple;
ScanKeyData skey;
Oid minmax;
/*
* Find the first/last enum member using pg_enum_typid_sortorder_index.
* Note we must not use the syscache. See comments for RenumberEnumType
* in catalog/pg_enum.c for more info.
*/
ScanKeyInit(&skey,
Anum_pg_enum_enumtypid,
BTEqualStrategyNumber, F_OIDEQ,
ObjectIdGetDatum(enumtypoid));
enum_rel = heap_open(EnumRelationId, AccessShareLock);
enum_idx = index_open(EnumTypIdSortOrderIndexId, AccessShareLock);
enum_scan = systable_beginscan_ordered(enum_rel, enum_idx, NULL,
1, &skey);
enum_tuple = systable_getnext_ordered(enum_scan, direction);
if (HeapTupleIsValid(enum_tuple))
{
/* check it's safe to use in SQL */
check_safe_enum_use(enum_tuple);
minmax = HeapTupleGetOid(enum_tuple);
}
else
{
/* should only happen with an empty enum */
minmax = InvalidOid;
}
systable_endscan_ordered(enum_scan);
index_close(enum_idx, AccessShareLock);
heap_close(enum_rel, AccessShareLock);
return minmax;
}
/*
* SQL-callable function to scan through an index and summarize all ranges
* that are not currently summarized.
*/
Datum
brin_summarize_new_values(PG_FUNCTION_ARGS)
{
Oid indexoid = PG_GETARG_OID(0);
Relation indexRel;
Relation heapRel;
double numSummarized = 0;
heapRel = heap_open(IndexGetRelation(indexoid, false),
ShareUpdateExclusiveLock);
indexRel = index_open(indexoid, ShareUpdateExclusiveLock);
brinsummarize(indexRel, heapRel, &numSummarized, NULL);
relation_close(indexRel, ShareUpdateExclusiveLock);
relation_close(heapRel, ShareUpdateExclusiveLock);
PG_RETURN_INT32((int32) numSummarized);
}
/*
* AppendOnlyBlockDirectory_DeleteSegmentFile
*
* Deletes all block directory entries for given segment file of an
* append-only relation.
*/
void
AppendOnlyBlockDirectory_DeleteSegmentFile(
AppendOnlyEntry *aoEntry,
Snapshot snapshot,
int segno,
int columnGroupNo)
{
Assert(aoEntry);
Assert(OidIsValid(aoEntry->blkdirrelid));
Assert(OidIsValid(aoEntry->blkdiridxid));
Relation blkdirRel = heap_open(aoEntry->blkdirrelid, RowExclusiveLock);
Relation blkdirIdx = index_open(aoEntry->blkdiridxid, RowExclusiveLock);
ScanKeyData scanKey;
ScanKeyInit(&scanKey,
1, /* segno */
BTEqualStrategyNumber,
F_INT4EQ,
Int32GetDatum(segno));
IndexScanDesc indexScan = index_beginscan(
blkdirRel,
blkdirIdx,
snapshot,
1,
&scanKey);
HeapTuple tuple = NULL;
while ((tuple = index_getnext(indexScan, ForwardScanDirection)) != NULL)
{
simple_heap_delete(blkdirRel,
&tuple->t_self);
}
index_endscan(indexScan);
index_close(blkdirIdx, RowExclusiveLock);
heap_close(blkdirRel, RowExclusiveLock);
}
/*
* enum_endpoint: common code for enum_first/enum_last
*/
static Oid
enum_endpoint(Oid enumtypoid, ScanDirection direction)
{
Relation enum_rel;
Relation enum_idx;
SysScanDesc enum_scan;
HeapTuple enum_tuple;
ScanKeyData skey;
Oid minmax;
/*
* Find the first/last enum member using pg_enum_typid_sortorder_index.
* Note we must not use the syscache, and must use an MVCC snapshot here.
* See comments for RenumberEnumType in catalog/pg_enum.c for more info.
*/
ScanKeyInit(&skey,
Anum_pg_enum_enumtypid,
BTEqualStrategyNumber, F_OIDEQ,
ObjectIdGetDatum(enumtypoid));
enum_rel = heap_open(EnumRelationId, AccessShareLock);
enum_idx = index_open(EnumTypIdSortOrderIndexId, AccessShareLock);
enum_scan = systable_beginscan_ordered(enum_rel, enum_idx,
GetTransactionSnapshot(),
1, &skey);
enum_tuple = systable_getnext_ordered(enum_scan, direction);
if (HeapTupleIsValid(enum_tuple))
minmax = HeapTupleGetOid(enum_tuple);
else
minmax = InvalidOid;
systable_endscan_ordered(enum_scan);
index_close(enum_idx, AccessShareLock);
heap_close(enum_rel, AccessShareLock);
return minmax;
}
void
rebuildheap(Oid OIDNewHeap, Oid OIDOldHeap, Oid OIDOldIndex)
{
Relation LocalNewHeap, LocalOldHeap, LocalOldIndex;
IndexScanDesc ScanDesc;
RetrieveIndexResult ScanResult;
ItemPointer HeapTid;
HeapTuple LocalHeapTuple;
Buffer LocalBuffer;
Oid OIDNewHeapInsert;
/*
* Open the relations I need. Scan through the OldHeap on the OldIndex and
* insert each tuple into the NewHeap.
*/
LocalNewHeap=(Relation)heap_open(OIDNewHeap);
LocalOldHeap=(Relation)heap_open(OIDOldHeap);
LocalOldIndex=(Relation)index_open(OIDOldIndex);
ScanDesc=index_beginscan(LocalOldIndex, false, 0, (ScanKey) NULL);
while ((ScanResult =
index_getnext(ScanDesc, ForwardScanDirection)) != NULL) {
HeapTid = &ScanResult->heap_iptr;
LocalHeapTuple = heap_fetch(LocalOldHeap, 0, HeapTid, &LocalBuffer);
OIDNewHeapInsert =
heap_insert(LocalNewHeap, LocalHeapTuple);
pfree(ScanResult);
ReleaseBuffer(LocalBuffer);
}
index_close(LocalOldIndex);
heap_close(LocalOldHeap);
heap_close(LocalNewHeap);
}
/*
* InitCatCachePhase2 -- external interface for CatalogCacheInitializeCache
*
* One reason to call this routine is to ensure that the relcache has
* created entries for all the catalogs and indexes referenced by catcaches.
* Therefore, provide an option to open the index as well as fixing the
* cache itself. An exception is the indexes on pg_am, which we don't use
* (cf. IndexScanOK).
*/
void
InitCatCachePhase2(CatCache *cache, bool touch_index)
{
if (cache->cc_tupdesc == NULL)
CatalogCacheInitializeCache(cache);
if (touch_index &&
cache->id != AMOID &&
cache->id != AMNAME)
{
Relation idesc;
/*
* We must lock the underlying catalog before opening the index to
* avoid deadlock, since index_open could possibly result in reading
* this same catalog, and if anyone else is exclusive-locking this
* catalog and index they'll be doing it in that order.
*/
LockRelationOid(cache->cc_reloid, AccessShareLock);
idesc = index_open(cache->cc_indexoid, AccessShareLock);
index_close(idesc, AccessShareLock);
UnlockRelationOid(cache->cc_reloid, AccessShareLock);
}
}
/* ----------------------------------------------------------------
* ExecInitBitmapIndexScan
*
* Initializes the index scan's state information.
* ----------------------------------------------------------------
*/
BitmapIndexScanState *
ExecInitBitmapIndexScan(BitmapIndexScan *node, EState *estate, int eflags)
{
BitmapIndexScanState *indexstate;
bool relistarget;
/* check for unsupported flags */
Assert(!(eflags & (EXEC_FLAG_BACKWARD | EXEC_FLAG_MARK)));
/*
* create state structure
*/
indexstate = makeNode(BitmapIndexScanState);
indexstate->ss.ps.plan = (Plan *) node;
indexstate->ss.ps.state = estate;
/* normally we don't make the result bitmap till runtime */
indexstate->biss_result = NULL;
/*
* Miscellaneous initialization
*
* We do not need a standard exprcontext for this node, though we may
* decide below to create a runtime-key exprcontext
*/
/*
* initialize child expressions
*
* We don't need to initialize targetlist or qual since neither are used.
*
* Note: we don't initialize all of the indexqual expression, only the
* sub-parts corresponding to runtime keys (see below).
*/
/*
* We do not open or lock the base relation here. We assume that an
* ancestor BitmapHeapScan node is holding AccessShareLock (or better) on
* the heap relation throughout the execution of the plan tree.
*/
indexstate->ss.ss_currentRelation = NULL;
indexstate->ss.ss_currentScanDesc = NULL;
/*
* If we are just doing EXPLAIN (ie, aren't going to run the plan), stop
* here. This allows an index-advisor plugin to EXPLAIN a plan containing
* references to nonexistent indexes.
*/
if (eflags & EXEC_FLAG_EXPLAIN_ONLY)
return indexstate;
/*
* Open the index relation.
*
* If the parent table is one of the target relations of the query, then
* InitPlan already opened and write-locked the index, so we can avoid
* taking another lock here. Otherwise we need a normal reader's lock.
*/
relistarget = ExecRelationIsTargetRelation(estate, node->scan.scanrelid);
indexstate->biss_RelationDesc = index_open(node->indexid,
relistarget ? NoLock : AccessShareLock);
/*
* Initialize index-specific scan state
*/
indexstate->biss_RuntimeKeysReady = false;
indexstate->biss_RuntimeKeys = NULL;
indexstate->biss_NumRuntimeKeys = 0;
/*
* build the index scan keys from the index qualification
*/
ExecIndexBuildScanKeys((PlanState *) indexstate,
indexstate->biss_RelationDesc,
node->indexqual,
false,
&indexstate->biss_ScanKeys,
&indexstate->biss_NumScanKeys,
&indexstate->biss_RuntimeKeys,
&indexstate->biss_NumRuntimeKeys,
&indexstate->biss_ArrayKeys,
&indexstate->biss_NumArrayKeys);
/*
* If we have runtime keys or array keys, we need an ExprContext to
* evaluate them. We could just create a "standard" plan node exprcontext,
* but to keep the code looking similar to nodeIndexscan.c, it seems
* better to stick with the approach of using a separate ExprContext.
*/
if (indexstate->biss_NumRuntimeKeys != 0 ||
indexstate->biss_NumArrayKeys != 0)
{
ExprContext *stdecontext = indexstate->ss.ps.ps_ExprContext;
ExecAssignExprContext(estate, &indexstate->ss.ps);
indexstate->biss_RuntimeContext = indexstate->ss.ps.ps_ExprContext;
indexstate->ss.ps.ps_ExprContext = stdecontext;
}
else
{
indexstate->biss_RuntimeContext = NULL;
}
/*
* Initialize scan descriptor.
*/
indexstate->biss_ScanDesc =
index_beginscan_bitmap(indexstate->biss_RelationDesc,
estate->es_snapshot,
indexstate->biss_NumScanKeys);
/*
* If no run-time keys to calculate, go ahead and pass the scankeys to the
* index AM.
*/
if (indexstate->biss_NumRuntimeKeys == 0 &&
indexstate->biss_NumArrayKeys == 0)
index_rescan(indexstate->biss_ScanDesc,
indexstate->biss_ScanKeys, indexstate->biss_NumScanKeys,
NULL, 0);
/*
* all done.
*/
return indexstate;
}
/*
* get_relation_info -
* Retrieves catalog information for a given relation.
*
* Given the Oid of the relation, return the following info into fields
* of the RelOptInfo struct:
*
* min_attr lowest valid AttrNumber
* max_attr highest valid AttrNumber
* indexlist list of IndexOptInfos for relation's indexes
* pages number of pages
* tuples number of tuples
*
* Also, initialize the attr_needed[] and attr_widths[] arrays. In most
* cases these are left as zeroes, but sometimes we need to compute attr
* widths here, and we may as well cache the results for costsize.c.
*
* If inhparent is true, all we need to do is set up the attr arrays:
* the RelOptInfo actually represents the appendrel formed by an inheritance
* tree, and so the parent rel's physical size and index information isn't
* important for it.
*/
void
get_relation_info(PlannerInfo *root, Oid relationObjectId, bool inhparent,
RelOptInfo *rel)
{
Index varno = rel->relid;
Relation relation;
bool hasindex;
List *indexinfos = NIL;
/*
* We need not lock the relation since it was already locked, either by
* the rewriter or when expand_inherited_rtentry() added it to the query's
* rangetable.
*/
relation = heap_open(relationObjectId, NoLock);
rel->min_attr = FirstLowInvalidHeapAttributeNumber + 1;
rel->max_attr = RelationGetNumberOfAttributes(relation);
rel->reltablespace = RelationGetForm(relation)->reltablespace;
Assert(rel->max_attr >= rel->min_attr);
rel->attr_needed = (Relids *)
palloc0((rel->max_attr - rel->min_attr + 1) * sizeof(Relids));
rel->attr_widths = (int32 *)
palloc0((rel->max_attr - rel->min_attr + 1) * sizeof(int32));
/*
* Estimate relation size --- unless it's an inheritance parent, in which
* case the size will be computed later in set_append_rel_pathlist, and we
* must leave it zero for now to avoid bollixing the total_table_pages
* calculation.
*/
if (!inhparent)
estimate_rel_size(relation, rel->attr_widths - rel->min_attr,
&rel->pages, &rel->tuples);
/*
* Make list of indexes. Ignore indexes on system catalogs if told to.
* Don't bother with indexes for an inheritance parent, either.
*/
if (inhparent ||
(IgnoreSystemIndexes && IsSystemClass(relation->rd_rel)))
hasindex = false;
else
hasindex = relation->rd_rel->relhasindex;
if (hasindex)
{
List *indexoidlist;
ListCell *l;
LOCKMODE lmode;
indexoidlist = RelationGetIndexList(relation);
/*
* For each index, we get the same type of lock that the executor will
* need, and do not release it. This saves a couple of trips to the
* shared lock manager while not creating any real loss of
* concurrency, because no schema changes could be happening on the
* index while we hold lock on the parent rel, and neither lock type
* blocks any other kind of index operation.
*/
if (rel->relid == root->parse->resultRelation)
lmode = RowExclusiveLock;
else
lmode = AccessShareLock;
foreach(l, indexoidlist)
{
Oid indexoid = lfirst_oid(l);
Relation indexRelation;
Form_pg_index index;
IndexOptInfo *info;
int ncolumns;
int i;
/*
* Extract info from the relation descriptor for the index.
*/
indexRelation = index_open(indexoid, lmode);
index = indexRelation->rd_index;
/*
* Ignore invalid indexes, since they can't safely be used for
* queries. Note that this is OK because the data structure we
* are constructing is only used by the planner --- the executor
* still needs to insert into "invalid" indexes!
*/
if (!index->indisvalid)
{
index_close(indexRelation, NoLock);
continue;
}
/*
* If the index is valid, but cannot yet be used, ignore it; but
* mark the plan we are generating as transient. See
* src/backend/access/heap/README.HOT for discussion.
*/
if (index->indcheckxmin &&
!TransactionIdPrecedes(HeapTupleHeaderGetXmin(indexRelation->rd_indextuple->t_data),
TransactionXmin))
{
root->glob->transientPlan = true;
index_close(indexRelation, NoLock);
continue;
}
info = makeNode(IndexOptInfo);
info->indexoid = index->indexrelid;
info->reltablespace =
RelationGetForm(indexRelation)->reltablespace;
info->rel = rel;
info->ncolumns = ncolumns = index->indnatts;
/*
* Allocate per-column info arrays. To save a few palloc cycles
* we allocate all the Oid-type arrays in one request. Note that
* the opfamily array needs an extra, terminating zero at the end.
* We pre-zero the ordering info in case the index is unordered.
*/
info->indexkeys = (int *) palloc(sizeof(int) * ncolumns);
info->opfamily = (Oid *) palloc0(sizeof(Oid) * (4 * ncolumns + 1));
info->opcintype = info->opfamily + (ncolumns + 1);
info->fwdsortop = info->opcintype + ncolumns;
info->revsortop = info->fwdsortop + ncolumns;
info->nulls_first = (bool *) palloc0(sizeof(bool) * ncolumns);
for (i = 0; i < ncolumns; i++)
{
info->indexkeys[i] = index->indkey.values[i];
info->opfamily[i] = indexRelation->rd_opfamily[i];
info->opcintype[i] = indexRelation->rd_opcintype[i];
}
info->relam = indexRelation->rd_rel->relam;
info->amcostestimate = indexRelation->rd_am->amcostestimate;
info->amoptionalkey = indexRelation->rd_am->amoptionalkey;
info->amsearchnulls = indexRelation->rd_am->amsearchnulls;
info->amhasgettuple = OidIsValid(indexRelation->rd_am->amgettuple);
info->amhasgetbitmap = OidIsValid(indexRelation->rd_am->amgetbitmap);
/*
* Fetch the ordering operators associated with the index, if any.
* We expect that all ordering-capable indexes use btree's
* strategy numbers for the ordering operators.
*/
if (indexRelation->rd_am->amcanorder)
{
int nstrat = indexRelation->rd_am->amstrategies;
for (i = 0; i < ncolumns; i++)
{
int16 opt = indexRelation->rd_indoption[i];
int fwdstrat;
int revstrat;
if (opt & INDOPTION_DESC)
{
fwdstrat = BTGreaterStrategyNumber;
revstrat = BTLessStrategyNumber;
}
else
{
fwdstrat = BTLessStrategyNumber;
revstrat = BTGreaterStrategyNumber;
}
/*
* Index AM must have a fixed set of strategies for it to
* make sense to specify amcanorder, so we need not allow
* the case amstrategies == 0.
*/
if (fwdstrat > 0)
{
Assert(fwdstrat <= nstrat);
info->fwdsortop[i] = indexRelation->rd_operator[i * nstrat + fwdstrat - 1];
}
if (revstrat > 0)
{
Assert(revstrat <= nstrat);
info->revsortop[i] = indexRelation->rd_operator[i * nstrat + revstrat - 1];
}
info->nulls_first[i] = (opt & INDOPTION_NULLS_FIRST) != 0;
}
}
/*
* Fetch the index expressions and predicate, if any. We must
* modify the copies we obtain from the relcache to have the
* correct varno for the parent relation, so that they match up
* correctly against qual clauses.
*/
info->indexprs = RelationGetIndexExpressions(indexRelation);
info->indpred = RelationGetIndexPredicate(indexRelation);
if (info->indexprs && varno != 1)
ChangeVarNodes((Node *) info->indexprs, 1, varno, 0);
if (info->indpred && varno != 1)
ChangeVarNodes((Node *) info->indpred, 1, varno, 0);
info->predOK = false; /* set later in indxpath.c */
info->unique = index->indisunique;
/*
* Estimate the index size. If it's not a partial index, we lock
* the number-of-tuples estimate to equal the parent table; if it
* is partial then we have to use the same methods as we would for
* a table, except we can be sure that the index is not larger
* than the table.
*/
if (info->indpred == NIL)
{
info->pages = RelationGetNumberOfBlocks(indexRelation);
info->tuples = rel->tuples;
}
else
{
estimate_rel_size(indexRelation, NULL,
&info->pages, &info->tuples);
if (info->tuples > rel->tuples)
info->tuples = rel->tuples;
}
index_close(indexRelation, NoLock);
indexinfos = lcons(info, indexinfos);
}
list_free(indexoidlist);
}
/*
* get_relation_info -
* Retrieves catalog information for a given relation.
*
* Given the Oid of the relation, return the following info into fields
* of the RelOptInfo struct:
*
* min_attr lowest valid AttrNumber
* max_attr highest valid AttrNumber
* indexlist list of IndexOptInfos for relation's indexes
* pages number of pages
* tuples number of tuples
*
* Also, initialize the attr_needed[] and attr_widths[] arrays. In most
* cases these are left as zeroes, but sometimes we need to compute attr
* widths here, and we may as well cache the results for costsize.c.
*
* If inhparent is true, all we need to do is set up the attr arrays:
* the RelOptInfo actually represents the appendrel formed by an inheritance
* tree, and so the parent rel's physical size and index information isn't
* important for it.
*/
void
get_relation_info(PlannerInfo *root, Oid relationObjectId, bool inhparent,
RelOptInfo *rel)
{
Index varno = rel->relid;
Relation relation;
bool hasindex;
List *indexinfos = NIL;
/*
* We need not lock the relation since it was already locked, either by
* the rewriter or when expand_inherited_rtentry() added it to the query's
* rangetable.
*/
relation = heap_open(relationObjectId, NoLock);
/* Temporary and unlogged relations are inaccessible during recovery. */
if (!RelationNeedsWAL(relation) && RecoveryInProgress())
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("cannot access temporary or unlogged relations during recovery")));
rel->min_attr = FirstLowInvalidHeapAttributeNumber + 1;
rel->max_attr = RelationGetNumberOfAttributes(relation);
rel->reltablespace = RelationGetForm(relation)->reltablespace;
Assert(rel->max_attr >= rel->min_attr);
rel->attr_needed = (Relids *)
palloc0((rel->max_attr - rel->min_attr + 1) * sizeof(Relids));
rel->attr_widths = (int32 *)
palloc0((rel->max_attr - rel->min_attr + 1) * sizeof(int32));
/*
* Estimate relation size --- unless it's an inheritance parent, in which
* case the size will be computed later in set_append_rel_pathlist, and we
* must leave it zero for now to avoid bollixing the total_table_pages
* calculation.
*/
if (!inhparent)
estimate_rel_size(relation, rel->attr_widths - rel->min_attr,
&rel->pages, &rel->tuples, &rel->allvisfrac);
/*
* Make list of indexes. Ignore indexes on system catalogs if told to.
* Don't bother with indexes for an inheritance parent, either.
*/
if (inhparent ||
(IgnoreSystemIndexes && IsSystemClass(relation->rd_rel)))
hasindex = false;
else
hasindex = relation->rd_rel->relhasindex;
if (hasindex)
{
List *indexoidlist;
ListCell *l;
LOCKMODE lmode;
indexoidlist = RelationGetIndexList(relation);
/*
* For each index, we get the same type of lock that the executor will
* need, and do not release it. This saves a couple of trips to the
* shared lock manager while not creating any real loss of
* concurrency, because no schema changes could be happening on the
* index while we hold lock on the parent rel, and neither lock type
* blocks any other kind of index operation.
*/
if (rel->relid == root->parse->resultRelation)
lmode = RowExclusiveLock;
else
lmode = AccessShareLock;
foreach(l, indexoidlist)
{
Oid indexoid = lfirst_oid(l);
Relation indexRelation;
Form_pg_index index;
IndexOptInfo *info;
int ncolumns;
int i;
/*
* Extract info from the relation descriptor for the index.
*/
indexRelation = index_open(indexoid, lmode);
index = indexRelation->rd_index;
/*
* Ignore invalid indexes, since they can't safely be used for
* queries. Note that this is OK because the data structure we
* are constructing is only used by the planner --- the executor
* still needs to insert into "invalid" indexes!
*/
if (!index->indisvalid)
{
index_close(indexRelation, NoLock);
continue;
}
/*
* If the index is valid, but cannot yet be used, ignore it; but
* mark the plan we are generating as transient. See
* src/backend/access/heap/README.HOT for discussion.
*/
if (index->indcheckxmin &&
!TransactionIdPrecedes(HeapTupleHeaderGetXmin(indexRelation->rd_indextuple->t_data),
TransactionXmin))
{
root->glob->transientPlan = true;
index_close(indexRelation, NoLock);
continue;
}
info = makeNode(IndexOptInfo);
info->indexoid = index->indexrelid;
info->reltablespace =
RelationGetForm(indexRelation)->reltablespace;
info->rel = rel;
info->ncolumns = ncolumns = index->indnatts;
info->indexkeys = (int *) palloc(sizeof(int) * ncolumns);
info->indexcollations = (Oid *) palloc(sizeof(Oid) * ncolumns);
info->opfamily = (Oid *) palloc(sizeof(Oid) * ncolumns);
info->opcintype = (Oid *) palloc(sizeof(Oid) * ncolumns);
for (i = 0; i < ncolumns; i++)
{
info->indexkeys[i] = index->indkey.values[i];
info->indexcollations[i] = indexRelation->rd_indcollation[i];
info->opfamily[i] = indexRelation->rd_opfamily[i];
info->opcintype[i] = indexRelation->rd_opcintype[i];
}
info->relam = indexRelation->rd_rel->relam;
info->amcostestimate = indexRelation->rd_am->amcostestimate;
info->canreturn = index_can_return(indexRelation);
info->amcanorderbyop = indexRelation->rd_am->amcanorderbyop;
info->amoptionalkey = indexRelation->rd_am->amoptionalkey;
info->amsearcharray = indexRelation->rd_am->amsearcharray;
info->amsearchnulls = indexRelation->rd_am->amsearchnulls;
info->amhasgettuple = OidIsValid(indexRelation->rd_am->amgettuple);
info->amhasgetbitmap = OidIsValid(indexRelation->rd_am->amgetbitmap);
/*
* Fetch the ordering information for the index, if any.
*/
if (info->relam == BTREE_AM_OID)
{
/*
* If it's a btree index, we can use its opfamily OIDs
* directly as the sort ordering opfamily OIDs.
*/
Assert(indexRelation->rd_am->amcanorder);
info->sortopfamily = info->opfamily;
info->reverse_sort = (bool *) palloc(sizeof(bool) * ncolumns);
info->nulls_first = (bool *) palloc(sizeof(bool) * ncolumns);
for (i = 0; i < ncolumns; i++)
{
int16 opt = indexRelation->rd_indoption[i];
info->reverse_sort[i] = (opt & INDOPTION_DESC) != 0;
info->nulls_first[i] = (opt & INDOPTION_NULLS_FIRST) != 0;
}
}
else if (indexRelation->rd_am->amcanorder)
{
/*
* Otherwise, identify the corresponding btree opfamilies by
* trying to map this index's "<" operators into btree. Since
* "<" uniquely defines the behavior of a sort order, this is
* a sufficient test.
*
* XXX This method is rather slow and also requires the
* undesirable assumption that the other index AM numbers its
* strategies the same as btree. It'd be better to have a way
* to explicitly declare the corresponding btree opfamily for
* each opfamily of the other index type. But given the lack
* of current or foreseeable amcanorder index types, it's not
* worth expending more effort on now.
*/
info->sortopfamily = (Oid *) palloc(sizeof(Oid) * ncolumns);
info->reverse_sort = (bool *) palloc(sizeof(bool) * ncolumns);
info->nulls_first = (bool *) palloc(sizeof(bool) * ncolumns);
for (i = 0; i < ncolumns; i++)
{
int16 opt = indexRelation->rd_indoption[i];
Oid ltopr;
Oid btopfamily;
Oid btopcintype;
int16 btstrategy;
info->reverse_sort[i] = (opt & INDOPTION_DESC) != 0;
info->nulls_first[i] = (opt & INDOPTION_NULLS_FIRST) != 0;
ltopr = get_opfamily_member(info->opfamily[i],
info->opcintype[i],
info->opcintype[i],
BTLessStrategyNumber);
if (OidIsValid(ltopr) &&
get_ordering_op_properties(ltopr,
&btopfamily,
&btopcintype,
&btstrategy) &&
btopcintype == info->opcintype[i] &&
btstrategy == BTLessStrategyNumber)
{
/* Successful mapping */
info->sortopfamily[i] = btopfamily;
}
else
{
/* Fail ... quietly treat index as unordered */
info->sortopfamily = NULL;
info->reverse_sort = NULL;
info->nulls_first = NULL;
break;
}
}
}
else
{
info->sortopfamily = NULL;
info->reverse_sort = NULL;
info->nulls_first = NULL;
}
/*
* Fetch the index expressions and predicate, if any. We must
* modify the copies we obtain from the relcache to have the
* correct varno for the parent relation, so that they match up
* correctly against qual clauses.
*/
info->indexprs = RelationGetIndexExpressions(indexRelation);
info->indpred = RelationGetIndexPredicate(indexRelation);
if (info->indexprs && varno != 1)
ChangeVarNodes((Node *) info->indexprs, 1, varno, 0);
if (info->indpred && varno != 1)
ChangeVarNodes((Node *) info->indpred, 1, varno, 0);
/* Build targetlist using the completed indexprs data */
info->indextlist = build_index_tlist(root, info, relation);
info->predOK = false; /* set later in indxpath.c */
info->unique = index->indisunique;
info->immediate = index->indimmediate;
info->hypothetical = false;
/*
* Estimate the index size. If it's not a partial index, we lock
* the number-of-tuples estimate to equal the parent table; if it
* is partial then we have to use the same methods as we would for
* a table, except we can be sure that the index is not larger
* than the table.
*/
if (info->indpred == NIL)
{
info->pages = RelationGetNumberOfBlocks(indexRelation);
info->tuples = rel->tuples;
}
else
{
double allvisfrac; /* dummy */
estimate_rel_size(indexRelation, NULL,
&info->pages, &info->tuples, &allvisfrac);
if (info->tuples > rel->tuples)
info->tuples = rel->tuples;
}
index_close(indexRelation, NoLock);
indexinfos = lcons(info, indexinfos);
}
list_free(indexoidlist);
}
Datum
spgstat(PG_FUNCTION_ARGS)
{
text *name=PG_GETARG_TEXT_P(0);
char *relname=text_to_cstring(name);
RangeVar *relvar;
Relation index;
List *relname_list;
Oid relOid;
BlockNumber blkno = SPGIST_HEAD_BLKNO;
BlockNumber totalPages = 0,
innerPages = 0,
emptyPages = 0;
double usedSpace = 0.0;
char res[1024];
int bufferSize = -1;
int64 innerTuples = 0,
leafTuples = 0;
relname_list = stringToQualifiedNameList(relname);
relvar = makeRangeVarFromNameList(relname_list);
relOid = RangeVarGetRelid(relvar, false);
index = index_open(relOid, AccessExclusiveLock);
if ( index->rd_am == NULL )
elog(ERROR, "Relation %s.%s is not an index",
get_namespace_name(RelationGetNamespace(index)),
RelationGetRelationName(index) );
totalPages = RelationGetNumberOfBlocks(index);
for(blkno=SPGIST_HEAD_BLKNO; blkno<totalPages; blkno++)
{
Buffer buffer;
Page page;
buffer = ReadBuffer(index, blkno);
LockBuffer(buffer, BUFFER_LOCK_SHARE);
page = BufferGetPage(buffer);
if (SpGistPageIsLeaf(page))
{
leafTuples += SpGistPageGetMaxOffset(page);
}
else
{
innerPages++;
innerTuples += SpGistPageGetMaxOffset(page);
}
if (bufferSize < 0)
bufferSize = BufferGetPageSize(buffer) - MAXALIGN(sizeof(SpGistPageOpaqueData)) -
SizeOfPageHeaderData;
usedSpace += bufferSize - (PageGetFreeSpace(page) + sizeof(ItemIdData));
if (PageGetFreeSpace(page) + sizeof(ItemIdData) == bufferSize)
emptyPages++;
UnlockReleaseBuffer(buffer);
}
index_close(index, AccessExclusiveLock);
totalPages--; /* metapage */
snprintf(res, sizeof(res),
"totalPages: %u\n"
"innerPages: %u\n"
"leafPages: %u\n"
"emptyPages: %u\n"
"usedSpace: %.2f kbytes\n"
"freeSpace: %.2f kbytes\n"
"fillRatio: %.2f%c\n"
"leafTuples: %lld\n"
"innerTuples: %lld",
totalPages, innerPages, totalPages - innerPages, emptyPages,
usedSpace / 1024.0,
(( (double) bufferSize ) * ( (double) totalPages ) - usedSpace) / 1024,
100.0 * ( usedSpace / (( (double) bufferSize ) * ( (double) totalPages )) ),
'%',
leafTuples, innerTuples
);
PG_RETURN_TEXT_P(CStringGetTextDatum(res));
}
/* ------------------------------------------------------
* pgstathashindex()
*
* Usage: SELECT * FROM pgstathashindex('hashindex');
* ------------------------------------------------------
*/
Datum
pgstathashindex(PG_FUNCTION_ARGS)
{
Oid relid = PG_GETARG_OID(0);
BlockNumber nblocks;
BlockNumber blkno;
Relation rel;
HashIndexStat stats;
BufferAccessStrategy bstrategy;
HeapTuple tuple;
TupleDesc tupleDesc;
Datum values[8];
bool nulls[8];
Buffer metabuf;
HashMetaPage metap;
float8 free_percent;
uint64 total_space;
rel = index_open(relid, AccessShareLock);
/* index_open() checks that it's an index */
if (!IS_HASH(rel))
ereport(ERROR,
(errcode(ERRCODE_WRONG_OBJECT_TYPE),
errmsg("relation \"%s\" is not a HASH index",
RelationGetRelationName(rel))));
/*
* Reject attempts to read non-local temporary relations; we would be
* likely to get wrong data since we have no visibility into the owning
* session's local buffers.
*/
if (RELATION_IS_OTHER_TEMP(rel))
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("cannot access temporary indexes of other sessions")));
/* Get the information we need from the metapage. */
memset(&stats, 0, sizeof(stats));
metabuf = _hash_getbuf(rel, HASH_METAPAGE, HASH_READ, LH_META_PAGE);
metap = HashPageGetMeta(BufferGetPage(metabuf));
stats.version = metap->hashm_version;
stats.space_per_page = metap->hashm_bsize;
_hash_relbuf(rel, metabuf);
/* Get the current relation length */
nblocks = RelationGetNumberOfBlocks(rel);
/* prepare access strategy for this index */
bstrategy = GetAccessStrategy(BAS_BULKREAD);
/* Start from blkno 1 as 0th block is metapage */
for (blkno = 1; blkno < nblocks; blkno++)
{
Buffer buf;
Page page;
CHECK_FOR_INTERRUPTS();
buf = ReadBufferExtended(rel, MAIN_FORKNUM, blkno, RBM_NORMAL,
bstrategy);
LockBuffer(buf, BUFFER_LOCK_SHARE);
page = (Page) BufferGetPage(buf);
if (PageIsNew(page))
stats.unused_pages++;
else if (PageGetSpecialSize(page) !=
MAXALIGN(sizeof(HashPageOpaqueData)))
ereport(ERROR,
(errcode(ERRCODE_INDEX_CORRUPTED),
errmsg("index \"%s\" contains corrupted page at block %u",
RelationGetRelationName(rel),
BufferGetBlockNumber(buf))));
else
{
HashPageOpaque opaque;
int pagetype;
opaque = (HashPageOpaque) PageGetSpecialPointer(page);
pagetype = opaque->hasho_flag & LH_PAGE_TYPE;
if (pagetype == LH_BUCKET_PAGE)
{
stats.bucket_pages++;
GetHashPageStats(page, &stats);
}
else if (pagetype == LH_OVERFLOW_PAGE)
{
stats.overflow_pages++;
GetHashPageStats(page, &stats);
}
else if (pagetype == LH_BITMAP_PAGE)
stats.bitmap_pages++;
else if (pagetype == LH_UNUSED_PAGE)
stats.unused_pages++;
else
ereport(ERROR,
(errcode(ERRCODE_INDEX_CORRUPTED),
errmsg("unexpected page type 0x%04X in HASH index \"%s\" block %u",
opaque->hasho_flag, RelationGetRelationName(rel),
BufferGetBlockNumber(buf))));
}
UnlockReleaseBuffer(buf);
}
/* Done accessing the index */
index_close(rel, AccessShareLock);
/* Count unused pages as free space. */
stats.free_space += stats.unused_pages * stats.space_per_page;
/*
* Total space available for tuples excludes the metapage and the bitmap
* pages.
*/
total_space = (nblocks - (stats.bitmap_pages + 1)) * stats.space_per_page;
if (total_space == 0)
free_percent = 0.0;
else
free_percent = 100.0 * stats.free_space / total_space;
/*
* Build a tuple descriptor for our result type
*/
if (get_call_result_type(fcinfo, NULL, &tupleDesc) != TYPEFUNC_COMPOSITE)
elog(ERROR, "return type must be a row type");
tupleDesc = BlessTupleDesc(tupleDesc);
/*
* Build and return the tuple
*/
MemSet(nulls, 0, sizeof(nulls));
values[0] = Int32GetDatum(stats.version);
values[1] = Int64GetDatum((int64) stats.bucket_pages);
values[2] = Int64GetDatum((int64) stats.overflow_pages);
values[3] = Int64GetDatum((int64) stats.bitmap_pages);
values[4] = Int64GetDatum((int64) stats.unused_pages);
values[5] = Int64GetDatum(stats.live_items);
values[6] = Int64GetDatum(stats.dead_items);
values[7] = Float8GetDatum(free_percent);
tuple = heap_form_tuple(tupleDesc, values, nulls);
PG_RETURN_DATUM(HeapTupleGetDatum(tuple));
}
/* ----------
* toast_fetch_datum_slice -
*
* Reconstruct a segment of a Datum from the chunks saved
* in the toast relation
* ----------
*/
static struct varlena *
toast_fetch_datum_slice(struct varlena * attr, int32 sliceoffset, int32 length)
{
Relation toastrel;
Relation toastidx;
ScanKeyData toastkey[3];
int nscankeys;
SysScanDesc toastscan;
HeapTuple ttup;
TupleDesc toasttupDesc;
struct varlena *result;
struct varatt_external toast_pointer;
int32 attrsize;
int32 residx;
int32 nextidx;
int numchunks;
int startchunk;
int endchunk;
int32 startoffset;
int32 endoffset;
int totalchunks;
Pointer chunk;
bool isnull;
char *chunkdata;
int32 chunksize;
int32 chcpystrt;
int32 chcpyend;
Assert(VARATT_IS_EXTERNAL(attr));
/* Must copy to access aligned fields */
VARATT_EXTERNAL_GET_POINTER(toast_pointer, attr);
/*
* It's nonsense to fetch slices of a compressed datum -- this isn't lo_*
* we can't return a compressed datum which is meaningful to toast later
*/
Assert(!VARATT_EXTERNAL_IS_COMPRESSED(toast_pointer));
attrsize = toast_pointer.va_extsize;
totalchunks = ((attrsize - 1) / TOAST_MAX_CHUNK_SIZE) + 1;
if (sliceoffset >= attrsize)
{
sliceoffset = 0;
length = 0;
}
if (((sliceoffset + length) > attrsize) || length < 0)
length = attrsize - sliceoffset;
result = (struct varlena *) palloc(length + VARHDRSZ);
if (VARATT_EXTERNAL_IS_COMPRESSED(toast_pointer))
SET_VARSIZE_COMPRESSED(result, length + VARHDRSZ);
else
SET_VARSIZE(result, length + VARHDRSZ);
if (length == 0)
return result; /* Can save a lot of work at this point! */
startchunk = sliceoffset / TOAST_MAX_CHUNK_SIZE;
endchunk = (sliceoffset + length - 1) / TOAST_MAX_CHUNK_SIZE;
numchunks = (endchunk - startchunk) + 1;
startoffset = sliceoffset % TOAST_MAX_CHUNK_SIZE;
endoffset = (sliceoffset + length - 1) % TOAST_MAX_CHUNK_SIZE;
/*
* Open the toast relation and its index
*/
toastrel = heap_open(toast_pointer.va_toastrelid, AccessShareLock);
toasttupDesc = toastrel->rd_att;
toastidx = index_open(toastrel->rd_rel->reltoastidxid, AccessShareLock);
/*
* Setup a scan key to fetch from the index. This is either two keys or
* three depending on the number of chunks.
*/
ScanKeyInit(&toastkey[0],
(AttrNumber) 1,
BTEqualStrategyNumber, F_OIDEQ,
ObjectIdGetDatum(toast_pointer.va_valueid));
/*
* Use equality condition for one chunk, a range condition otherwise:
*/
if (numchunks == 1)
{
ScanKeyInit(&toastkey[1],
(AttrNumber) 2,
BTEqualStrategyNumber, F_INT4EQ,
Int32GetDatum(startchunk));
nscankeys = 2;
}
else
{
ScanKeyInit(&toastkey[1],
(AttrNumber) 2,
BTGreaterEqualStrategyNumber, F_INT4GE,
Int32GetDatum(startchunk));
ScanKeyInit(&toastkey[2],
(AttrNumber) 2,
BTLessEqualStrategyNumber, F_INT4LE,
Int32GetDatum(endchunk));
nscankeys = 3;
}
/*
* Read the chunks by index
*
* The index is on (valueid, chunkidx) so they will come in order
*/
nextidx = startchunk;
toastscan = systable_beginscan_ordered(toastrel, toastidx,
SnapshotToast, nscankeys, toastkey);
while ((ttup = systable_getnext_ordered(toastscan, ForwardScanDirection)) != NULL)
{
/*
* Have a chunk, extract the sequence number and the data
*/
residx = DatumGetInt32(fastgetattr(ttup, 2, toasttupDesc, &isnull));
Assert(!isnull);
chunk = DatumGetPointer(fastgetattr(ttup, 3, toasttupDesc, &isnull));
Assert(!isnull);
if (!VARATT_IS_EXTENDED(chunk))
{
chunksize = VARSIZE(chunk) - VARHDRSZ;
chunkdata = VARDATA(chunk);
}
else if (VARATT_IS_SHORT(chunk))
{
/* could happen due to heap_form_tuple doing its thing */
chunksize = VARSIZE_SHORT(chunk) - VARHDRSZ_SHORT;
chunkdata = VARDATA_SHORT(chunk);
}
else
{
/* should never happen */
elog(ERROR, "found toasted toast chunk for toast value %u in %s",
toast_pointer.va_valueid,
RelationGetRelationName(toastrel));
chunksize = 0; /* keep compiler quiet */
chunkdata = NULL;
}
/*
* Some checks on the data we've found
*/
if ((residx != nextidx) || (residx > endchunk) || (residx < startchunk))
elog(ERROR, "unexpected chunk number %d (expected %d) for toast value %u in %s",
residx, nextidx,
toast_pointer.va_valueid,
RelationGetRelationName(toastrel));
if (residx < totalchunks - 1)
{
if (chunksize != TOAST_MAX_CHUNK_SIZE)
elog(ERROR, "unexpected chunk size %d (expected %d) in chunk %d of %d for toast value %u in %s when fetching slice",
chunksize, (int) TOAST_MAX_CHUNK_SIZE,
residx, totalchunks,
toast_pointer.va_valueid,
RelationGetRelationName(toastrel));
}
else if (residx == totalchunks - 1)
{
if ((residx * TOAST_MAX_CHUNK_SIZE + chunksize) != attrsize)
elog(ERROR, "unexpected chunk size %d (expected %d) in final chunk %d for toast value %u in %s when fetching slice",
chunksize,
(int) (attrsize - residx * TOAST_MAX_CHUNK_SIZE),
residx,
toast_pointer.va_valueid,
RelationGetRelationName(toastrel));
}
else
elog(ERROR, "unexpected chunk number %d (out of range %d..%d) for toast value %u in %s",
residx,
0, totalchunks - 1,
toast_pointer.va_valueid,
RelationGetRelationName(toastrel));
/*
* Copy the data into proper place in our result
*/
chcpystrt = 0;
chcpyend = chunksize - 1;
if (residx == startchunk)
chcpystrt = startoffset;
if (residx == endchunk)
chcpyend = endoffset;
memcpy(VARDATA(result) +
(residx * TOAST_MAX_CHUNK_SIZE - sliceoffset) + chcpystrt,
chunkdata + chcpystrt,
(chcpyend - chcpystrt) + 1);
nextidx++;
}
/*
* Final checks that we successfully fetched the datum
*/
if (nextidx != (endchunk + 1))
elog(ERROR, "missing chunk number %d for toast value %u in %s",
nextidx,
toast_pointer.va_valueid,
RelationGetRelationName(toastrel));
/*
* End scan and close relations
*/
systable_endscan_ordered(toastscan);
index_close(toastidx, AccessShareLock);
heap_close(toastrel, AccessShareLock);
return result;
}
/* ----------
* toast_save_datum -
*
* Save one single datum into the secondary relation and return
* a Datum reference for it.
* ----------
*/
static Datum
toast_save_datum(Relation rel, Datum value, int options)
{
Relation toastrel;
Relation toastidx;
HeapTuple toasttup;
TupleDesc toasttupDesc;
Datum t_values[3];
bool t_isnull[3];
CommandId mycid = GetCurrentCommandId(true);
struct varlena *result;
struct varatt_external toast_pointer;
struct
{
struct varlena hdr;
char data[TOAST_MAX_CHUNK_SIZE]; /* make struct big enough */
int32 align_it; /* ensure struct is aligned well enough */
} chunk_data;
int32 chunk_size;
int32 chunk_seq = 0;
char *data_p;
int32 data_todo;
Pointer dval = DatumGetPointer(value);
/*
* Open the toast relation and its index. We can use the index to check
* uniqueness of the OID we assign to the toasted item, even though it has
* additional columns besides OID.
*/
toastrel = heap_open(rel->rd_rel->reltoastrelid, RowExclusiveLock);
toasttupDesc = toastrel->rd_att;
toastidx = index_open(toastrel->rd_rel->reltoastidxid, RowExclusiveLock);
/*
* Get the data pointer and length, and compute va_rawsize and va_extsize.
*
* va_rawsize is the size of the equivalent fully uncompressed datum, so
* we have to adjust for short headers.
*
* va_extsize is the actual size of the data payload in the toast records.
*/
if (VARATT_IS_SHORT(dval))
{
data_p = VARDATA_SHORT(dval);
data_todo = VARSIZE_SHORT(dval) - VARHDRSZ_SHORT;
toast_pointer.va_rawsize = data_todo + VARHDRSZ; /* as if not short */
toast_pointer.va_extsize = data_todo;
}
else if (VARATT_IS_COMPRESSED(dval))
{
data_p = VARDATA(dval);
data_todo = VARSIZE(dval) - VARHDRSZ;
/* rawsize in a compressed datum is just the size of the payload */
toast_pointer.va_rawsize = VARRAWSIZE_4B_C(dval) + VARHDRSZ;
toast_pointer.va_extsize = data_todo;
/* Assert that the numbers look like it's compressed */
Assert(VARATT_EXTERNAL_IS_COMPRESSED(toast_pointer));
}
else
{
data_p = VARDATA(dval);
data_todo = VARSIZE(dval) - VARHDRSZ;
toast_pointer.va_rawsize = VARSIZE(dval);
toast_pointer.va_extsize = data_todo;
}
/*
* Insert the correct table OID into the result TOAST pointer.
*
* Normally this is the actual OID of the target toast table, but during
* table-rewriting operations such as CLUSTER, we have to insert the OID
* of the table's real permanent toast table instead. rd_toastoid is set
* if we have to substitute such an OID.
*/
if (OidIsValid(rel->rd_toastoid))
toast_pointer.va_toastrelid = rel->rd_toastoid;
else
toast_pointer.va_toastrelid = RelationGetRelid(toastrel);
/*
* Choose an unused OID within the toast table for this toast value.
*/
toast_pointer.va_valueid = GetNewOidWithIndex(toastrel,
RelationGetRelid(toastidx),
(AttrNumber) 1);
/*
* Initialize constant parts of the tuple data
*/
t_values[0] = ObjectIdGetDatum(toast_pointer.va_valueid);
t_values[2] = PointerGetDatum(&chunk_data);
t_isnull[0] = false;
t_isnull[1] = false;
t_isnull[2] = false;
/*
* Split up the item into chunks
*/
while (data_todo > 0)
{
/*
* Calculate the size of this chunk
*/
chunk_size = Min(TOAST_MAX_CHUNK_SIZE, data_todo);
/*
* Build a tuple and store it
*/
t_values[1] = Int32GetDatum(chunk_seq++);
SET_VARSIZE(&chunk_data, chunk_size + VARHDRSZ);
memcpy(VARDATA(&chunk_data), data_p, chunk_size);
toasttup = heap_form_tuple(toasttupDesc, t_values, t_isnull);
heap_insert(toastrel, toasttup, mycid, options, NULL);
/*
* Create the index entry. We cheat a little here by not using
* FormIndexDatum: this relies on the knowledge that the index columns
* are the same as the initial columns of the table.
*
* Note also that there had better not be any user-created index on
* the TOAST table, since we don't bother to update anything else.
*/
index_insert(toastidx, t_values, t_isnull,
&(toasttup->t_self),
toastrel,
toastidx->rd_index->indisunique ?
UNIQUE_CHECK_YES : UNIQUE_CHECK_NO);
/*
* Free memory
*/
heap_freetuple(toasttup);
/*
* Move on to next chunk
*/
data_todo -= chunk_size;
data_p += chunk_size;
}
/*
* Done - close toast relation
*/
index_close(toastidx, RowExclusiveLock);
heap_close(toastrel, RowExclusiveLock);
/*
* Create the TOAST pointer value that we'll return
*/
result = (struct varlena *) palloc(TOAST_POINTER_SIZE);
SET_VARSIZE_EXTERNAL(result, TOAST_POINTER_SIZE);
memcpy(VARDATA_EXTERNAL(result), &toast_pointer, sizeof(toast_pointer));
return PointerGetDatum(result);
}
/* ----------
* toast_save_datum -
*
* Save one single datum into the secondary relation and return
* a varattrib reference for it.
* ----------
*/
static Datum
toast_save_datum(Relation rel, Datum value)
{
Relation toastrel;
Relation toastidx;
HeapTuple toasttup;
InsertIndexResult idxres;
TupleDesc toasttupDesc;
Datum t_values[3];
char t_nulls[3];
varattrib *result;
struct
{
struct varlena hdr;
char data[TOAST_MAX_CHUNK_SIZE];
} chunk_data;
int32 chunk_size;
int32 chunk_seq = 0;
char *data_p;
int32 data_todo;
/*
* Create the varattrib reference
*/
result = (varattrib *) palloc(sizeof(varattrib));
result->va_header = sizeof(varattrib) | VARATT_FLAG_EXTERNAL;
if (VARATT_IS_COMPRESSED(value))
{
result->va_header |= VARATT_FLAG_COMPRESSED;
result->va_content.va_external.va_rawsize =
((varattrib *) value)->va_content.va_compressed.va_rawsize;
}
else
result->va_content.va_external.va_rawsize = VARATT_SIZE(value);
result->va_content.va_external.va_extsize =
VARATT_SIZE(value) - VARHDRSZ;
result->va_content.va_external.va_valueid = newoid();
result->va_content.va_external.va_toastrelid =
rel->rd_rel->reltoastrelid;
/*
* Initialize constant parts of the tuple data
*/
t_values[0] = ObjectIdGetDatum(result->va_content.va_external.va_valueid);
t_values[2] = PointerGetDatum(&chunk_data);
t_nulls[0] = ' ';
t_nulls[1] = ' ';
t_nulls[2] = ' ';
/*
* Get the data to process
*/
data_p = VARATT_DATA(value);
data_todo = VARATT_SIZE(value) - VARHDRSZ;
/*
* Open the toast relation
*/
toastrel = heap_open(rel->rd_rel->reltoastrelid, RowExclusiveLock);
toasttupDesc = toastrel->rd_att;
toastidx = index_open(toastrel->rd_rel->reltoastidxid);
/*
* Split up the item into chunks
*/
while (data_todo > 0)
{
/*
* Calculate the size of this chunk
*/
chunk_size = Min(TOAST_MAX_CHUNK_SIZE, data_todo);
/*
* Build a tuple and store it
*/
t_values[1] = Int32GetDatum(chunk_seq++);
VARATT_SIZEP(&chunk_data) = chunk_size + VARHDRSZ;
memcpy(VARATT_DATA(&chunk_data), data_p, chunk_size);
toasttup = heap_formtuple(toasttupDesc, t_values, t_nulls);
if (!HeapTupleIsValid(toasttup))
elog(ERROR, "failed to build TOAST tuple");
simple_heap_insert(toastrel, toasttup);
/*
* Create the index entry. We cheat a little here by not using
* FormIndexDatum: this relies on the knowledge that the index
* columns are the same as the initial columns of the table.
*
* Note also that there had better not be any user-created index on
* the TOAST table, since we don't bother to update anything else.
*/
idxres = index_insert(toastidx, t_values, t_nulls,
&(toasttup->t_self),
toastrel, toastidx->rd_index->indisunique);
if (idxres == NULL)
elog(ERROR, "failed to insert index entry for TOAST tuple");
/*
* Free memory
*/
pfree(idxres);
heap_freetuple(toasttup);
/*
* Move on to next chunk
*/
data_todo -= chunk_size;
data_p += chunk_size;
}
/*
* Done - close toast relation and return the reference
*/
index_close(toastidx);
heap_close(toastrel, RowExclusiveLock);
return PointerGetDatum(result);
}
/* ----------
* toast_fetch_datum_slice -
*
* Reconstruct a segment of a varattrib from the chunks saved
* in the toast relation
* ----------
*/
static varattrib *
toast_fetch_datum_slice(varattrib *attr, int32 sliceoffset, int32 length)
{
Relation toastrel;
Relation toastidx;
ScanKeyData toastkey[3];
int nscankeys;
IndexScanDesc toastscan;
HeapTuple ttup;
TupleDesc toasttupDesc;
varattrib *result;
int32 attrsize;
int32 residx;
int32 nextidx;
int numchunks;
int startchunk;
int endchunk;
int32 startoffset;
int32 endoffset;
int totalchunks;
Pointer chunk;
bool isnull;
int32 chunksize;
int32 chcpystrt;
int32 chcpyend;
attrsize = attr->va_content.va_external.va_extsize;
totalchunks = ((attrsize - 1) / TOAST_MAX_CHUNK_SIZE) + 1;
if (sliceoffset >= attrsize)
{
sliceoffset = 0;
length = 0;
}
if (((sliceoffset + length) > attrsize) || length < 0)
length = attrsize - sliceoffset;
result = (varattrib *) palloc(length + VARHDRSZ);
VARATT_SIZEP(result) = length + VARHDRSZ;
if (VARATT_IS_COMPRESSED(attr))
VARATT_SIZEP(result) |= VARATT_FLAG_COMPRESSED;
if (length == 0)
return (result); /* Can save a lot of work at this point! */
startchunk = sliceoffset / TOAST_MAX_CHUNK_SIZE;
endchunk = (sliceoffset + length - 1) / TOAST_MAX_CHUNK_SIZE;
numchunks = (endchunk - startchunk) + 1;
startoffset = sliceoffset % TOAST_MAX_CHUNK_SIZE;
endoffset = (sliceoffset + length - 1) % TOAST_MAX_CHUNK_SIZE;
/*
* Open the toast relation and it's index
*/
toastrel = heap_open(attr->va_content.va_external.va_toastrelid,
AccessShareLock);
toasttupDesc = toastrel->rd_att;
toastidx = index_open(toastrel->rd_rel->reltoastidxid);
/*
* Setup a scan key to fetch from the index. This is either two keys
* or three depending on the number of chunks.
*/
ScanKeyEntryInitialize(&toastkey[0],
(bits16) 0,
(AttrNumber) 1,
(RegProcedure) F_OIDEQ,
ObjectIdGetDatum(attr->va_content.va_external.va_valueid));
/*
* Now dependent on number of chunks:
*/
if (numchunks == 1)
{
ScanKeyEntryInitialize(&toastkey[1],
(bits16) 0,
(AttrNumber) 2,
(RegProcedure) F_INT4EQ,
Int32GetDatum(startchunk));
nscankeys = 2;
}
else
{
ScanKeyEntryInitialize(&toastkey[1],
(bits16) 0,
(AttrNumber) 2,
(RegProcedure) F_INT4GE,
Int32GetDatum(startchunk));
ScanKeyEntryInitialize(&toastkey[2],
(bits16) 0,
(AttrNumber) 2,
(RegProcedure) F_INT4LE,
Int32GetDatum(endchunk));
nscankeys = 3;
}
/*
* Read the chunks by index
*
* The index is on (valueid, chunkidx) so they will come in order
*/
nextidx = startchunk;
toastscan = index_beginscan(toastrel, toastidx, SnapshotToast,
nscankeys, toastkey);
while ((ttup = index_getnext(toastscan, ForwardScanDirection)) != NULL)
{
/*
* Have a chunk, extract the sequence number and the data
*/
residx = DatumGetInt32(heap_getattr(ttup, 2, toasttupDesc, &isnull));
Assert(!isnull);
chunk = DatumGetPointer(heap_getattr(ttup, 3, toasttupDesc, &isnull));
Assert(!isnull);
chunksize = VARATT_SIZE(chunk) - VARHDRSZ;
/*
* Some checks on the data we've found
*/
if ((residx != nextidx) || (residx > endchunk) || (residx < startchunk))
elog(ERROR, "unexpected chunk number %d (expected %d) for toast value %u",
residx, nextidx,
attr->va_content.va_external.va_valueid);
if (residx < totalchunks - 1)
{
if (chunksize != TOAST_MAX_CHUNK_SIZE)
elog(ERROR, "unexpected chunk size %d in chunk %d for toast value %u",
chunksize, residx,
attr->va_content.va_external.va_valueid);
}
else
{
if ((residx * TOAST_MAX_CHUNK_SIZE + chunksize) != attrsize)
elog(ERROR, "unexpected chunk size %d in chunk %d for toast value %u",
chunksize, residx,
attr->va_content.va_external.va_valueid);
}
/*
* Copy the data into proper place in our result
*/
chcpystrt = 0;
chcpyend = chunksize - 1;
if (residx == startchunk)
chcpystrt = startoffset;
if (residx == endchunk)
chcpyend = endoffset;
memcpy(((char *) VARATT_DATA(result)) +
(residx * TOAST_MAX_CHUNK_SIZE - sliceoffset) + chcpystrt,
VARATT_DATA(chunk) + chcpystrt,
(chcpyend - chcpystrt) + 1);
nextidx++;
}
/*
* Final checks that we successfully fetched the datum
*/
if (nextidx != (endchunk + 1))
elog(ERROR, "missing chunk number %d for toast value %u",
nextidx,
attr->va_content.va_external.va_valueid);
/*
* End scan and close relations
*/
index_endscan(toastscan);
index_close(toastidx);
heap_close(toastrel, AccessShareLock);
return result;
}
/* ----------
* toast_fetch_datum -
*
* Reconstruct an in memory varattrib from the chunks saved
* in the toast relation
* ----------
*/
static varattrib *
toast_fetch_datum(varattrib *attr)
{
Relation toastrel;
Relation toastidx;
ScanKeyData toastkey;
IndexScanDesc toastscan;
HeapTuple ttup;
TupleDesc toasttupDesc;
varattrib *result;
int32 ressize;
int32 residx,
nextidx;
int32 numchunks;
Pointer chunk;
bool isnull;
int32 chunksize;
ressize = attr->va_content.va_external.va_extsize;
numchunks = ((ressize - 1) / TOAST_MAX_CHUNK_SIZE) + 1;
result = (varattrib *) palloc(ressize + VARHDRSZ);
VARATT_SIZEP(result) = ressize + VARHDRSZ;
if (VARATT_IS_COMPRESSED(attr))
VARATT_SIZEP(result) |= VARATT_FLAG_COMPRESSED;
/*
* Open the toast relation and its index
*/
toastrel = heap_open(attr->va_content.va_external.va_toastrelid,
AccessShareLock);
toasttupDesc = toastrel->rd_att;
toastidx = index_open(toastrel->rd_rel->reltoastidxid);
/*
* Setup a scan key to fetch from the index by va_valueid
*/
ScanKeyEntryInitialize(&toastkey,
(bits16) 0,
(AttrNumber) 1,
(RegProcedure) F_OIDEQ,
ObjectIdGetDatum(attr->va_content.va_external.va_valueid));
/*
* Read the chunks by index
*
* Note that because the index is actually on (valueid, chunkidx) we will
* see the chunks in chunkidx order, even though we didn't explicitly
* ask for it.
*/
nextidx = 0;
toastscan = index_beginscan(toastrel, toastidx, SnapshotToast,
1, &toastkey);
while ((ttup = index_getnext(toastscan, ForwardScanDirection)) != NULL)
{
/*
* Have a chunk, extract the sequence number and the data
*/
residx = DatumGetInt32(heap_getattr(ttup, 2, toasttupDesc, &isnull));
Assert(!isnull);
chunk = DatumGetPointer(heap_getattr(ttup, 3, toasttupDesc, &isnull));
Assert(!isnull);
chunksize = VARATT_SIZE(chunk) - VARHDRSZ;
/*
* Some checks on the data we've found
*/
if (residx != nextidx)
elog(ERROR, "unexpected chunk number %d (expected %d) for toast value %u",
residx, nextidx,
attr->va_content.va_external.va_valueid);
if (residx < numchunks - 1)
{
if (chunksize != TOAST_MAX_CHUNK_SIZE)
elog(ERROR, "unexpected chunk size %d in chunk %d for toast value %u",
chunksize, residx,
attr->va_content.va_external.va_valueid);
}
else if (residx < numchunks)
{
if ((residx * TOAST_MAX_CHUNK_SIZE + chunksize) != ressize)
elog(ERROR, "unexpected chunk size %d in chunk %d for toast value %u",
chunksize, residx,
attr->va_content.va_external.va_valueid);
}
else
elog(ERROR, "unexpected chunk number %d for toast value %u",
residx,
attr->va_content.va_external.va_valueid);
/*
* Copy the data into proper place in our result
*/
memcpy(((char *) VARATT_DATA(result)) + residx * TOAST_MAX_CHUNK_SIZE,
VARATT_DATA(chunk),
chunksize);
nextidx++;
}
/*
* Final checks that we successfully fetched the datum
*/
if (nextidx != numchunks)
elog(ERROR, "missing chunk number %d for toast value %u",
nextidx,
attr->va_content.va_external.va_valueid);
/*
* End scan and close relations
*/
index_endscan(toastscan);
index_close(toastidx);
heap_close(toastrel, AccessShareLock);
return result;
}
/* ----------------------------------------------------------------
* ExecInitIndexScan
*
* Initializes the index scan's state information, creates
* scan keys, and opens the base and index relations.
*
* Note: index scans have 2 sets of state information because
* we have to keep track of the base relation and the
* index relation.
* ----------------------------------------------------------------
*/
IndexScanState *
ExecInitIndexScan(IndexScan *node, EState *estate, int eflags)
{
IndexScanState *indexstate;
Relation currentRelation;
bool relistarget;
/*
* create state structure
*/
indexstate = makeNode(IndexScanState);
indexstate->ss.ps.plan = (Plan *) node;
indexstate->ss.ps.state = estate;
/*
* Miscellaneous initialization
*
* create expression context for node
*/
ExecAssignExprContext(estate, &indexstate->ss.ps);
indexstate->ss.ps.ps_TupFromTlist = false;
/*
* initialize child expressions
*
* Note: we don't initialize all of the indexqual expression, only the
* sub-parts corresponding to runtime keys (see below). Likewise for
* indexorderby, if any. But the indexqualorig expression is always
* initialized even though it will only be used in some uncommon cases ---
* would be nice to improve that. (Problem is that any SubPlans present
* in the expression must be found now...)
*/
indexstate->ss.ps.targetlist = (List *)
ExecInitExpr((Expr *) node->scan.plan.targetlist,
(PlanState *) indexstate);
indexstate->ss.ps.qual = (List *)
ExecInitExpr((Expr *) node->scan.plan.qual,
(PlanState *) indexstate);
indexstate->indexqualorig = (List *)
ExecInitExpr((Expr *) node->indexqualorig,
(PlanState *) indexstate);
/*
* tuple table initialization
*/
ExecInitResultTupleSlot(estate, &indexstate->ss.ps);
ExecInitScanTupleSlot(estate, &indexstate->ss);
/*
* open the base relation and acquire appropriate lock on it.
*/
currentRelation = ExecOpenScanRelation(estate, node->scan.scanrelid);
indexstate->ss.ss_currentRelation = currentRelation;
indexstate->ss.ss_currentScanDesc = NULL; /* no heap scan here */
/*
* get the scan type from the relation descriptor.
*/
ExecAssignScanType(&indexstate->ss, RelationGetDescr(currentRelation));
/*
* Initialize result tuple type and projection info.
*/
ExecAssignResultTypeFromTL(&indexstate->ss.ps);
ExecAssignScanProjectionInfo(&indexstate->ss);
/*
* If we are just doing EXPLAIN (ie, aren't going to run the plan), stop
* here. This allows an index-advisor plugin to EXPLAIN a plan containing
* references to nonexistent indexes.
*/
if (eflags & EXEC_FLAG_EXPLAIN_ONLY)
return indexstate;
/*
* Open the index relation.
*
* If the parent table is one of the target relations of the query, then
* InitPlan already opened and write-locked the index, so we can avoid
* taking another lock here. Otherwise we need a normal reader's lock.
*/
relistarget = ExecRelationIsTargetRelation(estate, node->scan.scanrelid);
indexstate->iss_RelationDesc = index_open(node->indexid,
relistarget ? NoLock : AccessShareLock);
/*
* Initialize index-specific scan state
*/
indexstate->iss_RuntimeKeysReady = false;
indexstate->iss_RuntimeKeys = NULL;
indexstate->iss_NumRuntimeKeys = 0;
/*
* build the index scan keys from the index qualification
*/
ExecIndexBuildScanKeys((PlanState *) indexstate,
indexstate->iss_RelationDesc,
node->scan.scanrelid,
node->indexqual,
false,
&indexstate->iss_ScanKeys,
&indexstate->iss_NumScanKeys,
&indexstate->iss_RuntimeKeys,
&indexstate->iss_NumRuntimeKeys,
NULL, /* no ArrayKeys */
NULL);
/*
* any ORDER BY exprs have to be turned into scankeys in the same way
*/
ExecIndexBuildScanKeys((PlanState *) indexstate,
indexstate->iss_RelationDesc,
node->scan.scanrelid,
node->indexorderby,
true,
&indexstate->iss_OrderByKeys,
&indexstate->iss_NumOrderByKeys,
&indexstate->iss_RuntimeKeys,
&indexstate->iss_NumRuntimeKeys,
NULL, /* no ArrayKeys */
NULL);
/*
* If we have runtime keys, we need an ExprContext to evaluate them. The
* node's standard context won't do because we want to reset that context
* for every tuple. So, build another context just like the other one...
* -tgl 7/11/00
*/
if (indexstate->iss_NumRuntimeKeys != 0)
{
ExprContext *stdecontext = indexstate->ss.ps.ps_ExprContext;
ExecAssignExprContext(estate, &indexstate->ss.ps);
indexstate->iss_RuntimeContext = indexstate->ss.ps.ps_ExprContext;
indexstate->ss.ps.ps_ExprContext = stdecontext;
}
else
{
indexstate->iss_RuntimeContext = NULL;
}
/*
* Initialize scan descriptor.
*/
indexstate->iss_ScanDesc = index_beginscan(currentRelation,
indexstate->iss_RelationDesc,
estate->es_snapshot,
indexstate->iss_NumScanKeys,
indexstate->iss_NumOrderByKeys);
/*
* If no run-time keys to calculate, go ahead and pass the scankeys to the
* index AM.
*/
if (indexstate->iss_NumRuntimeKeys == 0)
index_rescan(indexstate->iss_ScanDesc,
indexstate->iss_ScanKeys, indexstate->iss_NumScanKeys,
indexstate->iss_OrderByKeys, indexstate->iss_NumOrderByKeys);
/*
* all done.
*/
return indexstate;
}
/* ----------
* toast_fetch_datum -
*
* Reconstruct an in memory Datum from the chunks saved
* in the toast relation
* ----------
*/
static struct varlena *
toast_fetch_datum(struct varlena * attr)
{
Relation toastrel;
Relation toastidx;
ScanKeyData toastkey;
SysScanDesc toastscan;
HeapTuple ttup;
TupleDesc toasttupDesc;
struct varlena *result;
struct varatt_external toast_pointer;
int32 ressize;
int32 residx,
nextidx;
int32 numchunks;
Pointer chunk;
bool isnull;
char *chunkdata;
int32 chunksize;
/* Must copy to access aligned fields */
VARATT_EXTERNAL_GET_POINTER(toast_pointer, attr);
ressize = toast_pointer.va_extsize;
numchunks = ((ressize - 1) / TOAST_MAX_CHUNK_SIZE) + 1;
result = (struct varlena *) palloc(ressize + VARHDRSZ);
if (VARATT_EXTERNAL_IS_COMPRESSED(toast_pointer))
SET_VARSIZE_COMPRESSED(result, ressize + VARHDRSZ);
else
SET_VARSIZE(result, ressize + VARHDRSZ);
/*
* Open the toast relation and its index
*/
toastrel = heap_open(toast_pointer.va_toastrelid, AccessShareLock);
toasttupDesc = toastrel->rd_att;
toastidx = index_open(toastrel->rd_rel->reltoastidxid, AccessShareLock);
/*
* Setup a scan key to fetch from the index by va_valueid
*/
ScanKeyInit(&toastkey,
(AttrNumber) 1,
BTEqualStrategyNumber, F_OIDEQ,
ObjectIdGetDatum(toast_pointer.va_valueid));
/*
* Read the chunks by index
*
* Note that because the index is actually on (valueid, chunkidx) we will
* see the chunks in chunkidx order, even though we didn't explicitly ask
* for it.
*/
nextidx = 0;
toastscan = systable_beginscan_ordered(toastrel, toastidx,
SnapshotToast, 1, &toastkey);
while ((ttup = systable_getnext_ordered(toastscan, ForwardScanDirection)) != NULL)
{
/*
* Have a chunk, extract the sequence number and the data
*/
residx = DatumGetInt32(fastgetattr(ttup, 2, toasttupDesc, &isnull));
Assert(!isnull);
chunk = DatumGetPointer(fastgetattr(ttup, 3, toasttupDesc, &isnull));
Assert(!isnull);
if (!VARATT_IS_EXTENDED(chunk))
{
chunksize = VARSIZE(chunk) - VARHDRSZ;
chunkdata = VARDATA(chunk);
}
else if (VARATT_IS_SHORT(chunk))
{
/* could happen due to heap_form_tuple doing its thing */
chunksize = VARSIZE_SHORT(chunk) - VARHDRSZ_SHORT;
chunkdata = VARDATA_SHORT(chunk);
}
else
{
/* should never happen */
elog(ERROR, "found toasted toast chunk for toast value %u in %s",
toast_pointer.va_valueid,
RelationGetRelationName(toastrel));
chunksize = 0; /* keep compiler quiet */
chunkdata = NULL;
}
/*
* Some checks on the data we've found
*/
if (residx != nextidx)
elog(ERROR, "unexpected chunk number %d (expected %d) for toast value %u in %s",
residx, nextidx,
toast_pointer.va_valueid,
RelationGetRelationName(toastrel));
if (residx < numchunks - 1)
{
if (chunksize != TOAST_MAX_CHUNK_SIZE)
elog(ERROR, "unexpected chunk size %d (expected %d) in chunk %d of %d for toast value %u in %s",
chunksize, (int) TOAST_MAX_CHUNK_SIZE,
residx, numchunks,
toast_pointer.va_valueid,
RelationGetRelationName(toastrel));
}
else if (residx == numchunks - 1)
{
if ((residx * TOAST_MAX_CHUNK_SIZE + chunksize) != ressize)
elog(ERROR, "unexpected chunk size %d (expected %d) in final chunk %d for toast value %u in %s",
chunksize,
(int) (ressize - residx * TOAST_MAX_CHUNK_SIZE),
residx,
toast_pointer.va_valueid,
RelationGetRelationName(toastrel));
}
else
elog(ERROR, "unexpected chunk number %d (out of range %d..%d) for toast value %u in %s",
residx,
0, numchunks - 1,
toast_pointer.va_valueid,
RelationGetRelationName(toastrel));
/*
* Copy the data into proper place in our result
*/
memcpy(VARDATA(result) + residx * TOAST_MAX_CHUNK_SIZE,
chunkdata,
chunksize);
nextidx++;
}
/*
* Final checks that we successfully fetched the datum
*/
if (nextidx != numchunks)
elog(ERROR, "missing chunk number %d for toast value %u in %s",
nextidx,
toast_pointer.va_valueid,
RelationGetRelationName(toastrel));
/*
* End scan and close relations
*/
systable_endscan_ordered(toastscan);
index_close(toastidx, AccessShareLock);
heap_close(toastrel, AccessShareLock);
return result;
}
/*
* Extract all item values from a BRIN index page
*
* Usage: SELECT * FROM brin_page_items(get_raw_page('idx', 1), 'idx'::regclass);
*/
Datum
brin_page_items(PG_FUNCTION_ARGS)
{
bytea *raw_page = PG_GETARG_BYTEA_P(0);
Oid indexRelid = PG_GETARG_OID(1);
ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
TupleDesc tupdesc;
MemoryContext oldcontext;
Tuplestorestate *tupstore;
Relation indexRel;
brin_column_state **columns;
BrinDesc *bdesc;
BrinMemTuple *dtup;
Page page;
OffsetNumber offset;
AttrNumber attno;
bool unusedItem;
if (!superuser())
ereport(ERROR,
(errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
(errmsg("must be superuser to use raw page functions"))));
/* check to see if caller supports us returning a tuplestore */
if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("set-valued function called in context that cannot accept a set")));
if (!(rsinfo->allowedModes & SFRM_Materialize) ||
rsinfo->expectedDesc == NULL)
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("materialize mode required, but it is not allowed in this context")));
/* Build a tuple descriptor for our result type */
if (get_call_result_type(fcinfo, NULL, &tupdesc) != TYPEFUNC_COMPOSITE)
elog(ERROR, "return type must be a row type");
/* Build tuplestore to hold the result rows */
oldcontext = MemoryContextSwitchTo(rsinfo->econtext->ecxt_per_query_memory);
tupstore = tuplestore_begin_heap(true, false, work_mem);
rsinfo->returnMode = SFRM_Materialize;
rsinfo->setResult = tupstore;
rsinfo->setDesc = tupdesc;
MemoryContextSwitchTo(oldcontext);
indexRel = index_open(indexRelid, AccessShareLock);
bdesc = brin_build_desc(indexRel);
/* minimally verify the page we got */
page = verify_brin_page(raw_page, BRIN_PAGETYPE_REGULAR, "regular");
/*
* Initialize output functions for all indexed datatypes; simplifies
* calling them later.
*/
columns = palloc(sizeof(brin_column_state *) * RelationGetDescr(indexRel)->natts);
for (attno = 1; attno <= bdesc->bd_tupdesc->natts; attno++)
{
Oid output;
bool isVarlena;
BrinOpcInfo *opcinfo;
int i;
brin_column_state *column;
opcinfo = bdesc->bd_info[attno - 1];
column = palloc(offsetof(brin_column_state, outputFn) +
sizeof(FmgrInfo) * opcinfo->oi_nstored);
column->nstored = opcinfo->oi_nstored;
for (i = 0; i < opcinfo->oi_nstored; i++)
{
getTypeOutputInfo(opcinfo->oi_typcache[i]->type_id, &output, &isVarlena);
fmgr_info(output, &column->outputFn[i]);
}
columns[attno - 1] = column;
}
offset = FirstOffsetNumber;
unusedItem = false;
dtup = NULL;
for (;;)
{
Datum values[7];
bool nulls[7];
/*
* This loop is called once for every attribute of every tuple in the
* page. At the start of a tuple, we get a NULL dtup; that's our
* signal for obtaining and decoding the next one. If that's not the
* case, we output the next attribute.
*/
if (dtup == NULL)
{
ItemId itemId;
/* verify item status: if there's no data, we can't decode */
itemId = PageGetItemId(page, offset);
if (ItemIdIsUsed(itemId))
{
dtup = brin_deform_tuple(bdesc,
(BrinTuple *) PageGetItem(page, itemId));
attno = 1;
unusedItem = false;
}
else
unusedItem = true;
}
else
attno++;
MemSet(nulls, 0, sizeof(nulls));
if (unusedItem)
{
values[0] = UInt16GetDatum(offset);
nulls[1] = true;
nulls[2] = true;
nulls[3] = true;
nulls[4] = true;
nulls[5] = true;
nulls[6] = true;
}
else
{
int att = attno - 1;
values[0] = UInt16GetDatum(offset);
values[1] = UInt32GetDatum(dtup->bt_blkno);
values[2] = UInt16GetDatum(attno);
values[3] = BoolGetDatum(dtup->bt_columns[att].bv_allnulls);
values[4] = BoolGetDatum(dtup->bt_columns[att].bv_hasnulls);
values[5] = BoolGetDatum(dtup->bt_placeholder);
if (!dtup->bt_columns[att].bv_allnulls)
{
BrinValues *bvalues = &dtup->bt_columns[att];
StringInfoData s;
bool first;
int i;
initStringInfo(&s);
appendStringInfoChar(&s, '{');
first = true;
for (i = 0; i < columns[att]->nstored; i++)
{
char *val;
if (!first)
appendStringInfoString(&s, " .. ");
first = false;
val = OutputFunctionCall(&columns[att]->outputFn[i],
bvalues->bv_values[i]);
appendStringInfoString(&s, val);
pfree(val);
}
appendStringInfoChar(&s, '}');
values[6] = CStringGetTextDatum(s.data);
pfree(s.data);
}
else
{
nulls[6] = true;
}
}
tuplestore_putvalues(tupstore, tupdesc, values, nulls);
/*
* If the item was unused, jump straight to the next one; otherwise,
* the only cleanup needed here is to set our signal to go to the next
* tuple in the following iteration, by freeing the current one.
*/
if (unusedItem)
offset = OffsetNumberNext(offset);
else if (attno >= bdesc->bd_tupdesc->natts)
{
pfree(dtup);
dtup = NULL;
offset = OffsetNumberNext(offset);
}
/*
* If we're beyond the end of the page, we're done.
*/
if (offset > PageGetMaxOffsetNumber(page))
break;
}
/* clean up and return the tuplestore */
brin_free_desc(bdesc);
tuplestore_donestoring(tupstore);
index_close(indexRel, AccessShareLock);
return (Datum) 0;
}
/* ----------------------------------------------------------------
* ExecOpenIndices
*
* Find the indices associated with a result relation, open them,
* and save information about them in the result ResultRelInfo.
*
* At entry, caller has already opened and locked
* resultRelInfo->ri_RelationDesc.
* ----------------------------------------------------------------
*/
void
ExecOpenIndices(ResultRelInfo *resultRelInfo, bool speculative)
{
Relation resultRelation = resultRelInfo->ri_RelationDesc;
List *indexoidlist;
ListCell *l;
int len,
i;
RelationPtr relationDescs;
IndexInfo **indexInfoArray;
resultRelInfo->ri_NumIndices = 0;
/* fast path if no indexes */
if (!RelationGetForm(resultRelation)->relhasindex)
return;
/*
* Get cached list of index OIDs
*/
indexoidlist = RelationGetIndexList(resultRelation);
len = list_length(indexoidlist);
if (len == 0)
return;
/*
* allocate space for result arrays
*/
relationDescs = (RelationPtr) palloc(len * sizeof(Relation));
indexInfoArray = (IndexInfo **) palloc(len * sizeof(IndexInfo *));
resultRelInfo->ri_NumIndices = len;
resultRelInfo->ri_IndexRelationDescs = relationDescs;
resultRelInfo->ri_IndexRelationInfo = indexInfoArray;
/*
* For each index, open the index relation and save pg_index info. We
* acquire RowExclusiveLock, signifying we will update the index.
*
* Note: we do this even if the index is not indisready; it's not worth
* the trouble to optimize for the case where it isn't.
*/
i = 0;
foreach(l, indexoidlist)
{
Oid indexOid = lfirst_oid(l);
Relation indexDesc;
IndexInfo *ii;
indexDesc = index_open(indexOid, RowExclusiveLock);
/* extract index key information from the index's pg_index info */
ii = BuildIndexInfo(indexDesc);
/*
* If the indexes are to be used for speculative insertion, add extra
* information required by unique index entries.
*/
if (speculative && ii->ii_Unique)
BuildSpeculativeIndexInfo(indexDesc, ii);
relationDescs[i] = indexDesc;
indexInfoArray[i] = ii;
i++;
}
