/*
* Extract stored datum (and possible null category) from GIN tuple
*/
Datum
gintuple_get_key(GinState *ginstate, IndexTuple tuple,
GinNullCategory *category)
{
Datum res;
bool isnull;
if (ginstate->oneCol)
{
/*
* Single column index doesn't store attribute numbers in tuples
*/
res = index_getattr(tuple, FirstOffsetNumber, ginstate->origTupdesc,
&isnull);
}
else
{
/*
* Since the datum type depends on which index column it's from, we
* must be careful to use the right tuple descriptor here.
*/
OffsetNumber colN = gintuple_get_attrnum(ginstate, tuple);
res = index_getattr(tuple, OffsetNumberNext(FirstOffsetNumber),
ginstate->tupdesc[colN - 1],
&isnull);
}
if (isnull)
*category = GinGetNullCategory(tuple, ginstate);
else
*category = GIN_CAT_NORM_KEY;
return res;
}
/*
* Extract stored datum from GIN tuple
*/
Datum
gin_index_getattr(GinState *ginstate, IndexTuple tuple)
{
bool isnull;
Datum res;
if (ginstate->oneCol)
{
/*
* Single column index doesn't store attribute numbers in tuples
*/
res = index_getattr(tuple, FirstOffsetNumber, ginstate->origTupdesc,
&isnull);
}
else
{
/*
* Since the datum type depends on which index column it's from, we
* must be careful to use the right tuple descriptor here.
*/
OffsetNumber colN = gintuple_get_attrnum(ginstate, tuple);
res = index_getattr(tuple, OffsetNumberNext(FirstOffsetNumber),
ginstate->tupdesc[colN - 1],
&isnull);
}
Assert(!isnull);
return res;
}
static int
compare_indextuple(const IndexTuple itup1, const IndexTuple itup2,
ScanKey entry, int keysz, TupleDesc tupdes, bool *hasnull)
{
int i;
int32 compare;
*hasnull = false;
for (i = 1; i <= keysz; i++, entry++)
{
Datum attrDatum1,
attrDatum2;
bool isNull1,
isNull2;
attrDatum1 = index_getattr(itup1, i, tupdes, &isNull1);
attrDatum2 = index_getattr(itup2, i, tupdes, &isNull2);
if (isNull1)
{
*hasnull = true;
if (isNull2)
compare = 0; /* NULL "=" NULL */
else if (entry->sk_flags & SK_BT_NULLS_FIRST)
compare = -1; /* NULL "<" NOT_NULL */
else
compare = 1; /* NULL ">" NOT_NULL */
}
else if (isNull2)
{
*hasnull = true;
if (entry->sk_flags & SK_BT_NULLS_FIRST)
compare = 1; /* NOT_NULL ">" NULL */
else
compare = -1; /* NOT_NULL "<" NULL */
}
else
{
compare =
#if PG_VERSION_NUM >= 90100
DatumGetInt32(FunctionCall2Coll(&entry->sk_func,
entry->sk_collation,
attrDatum1,
attrDatum2));
#else
DatumGetInt32(FunctionCall2(&entry->sk_func,
attrDatum1,
attrDatum2));
#endif
if (entry->sk_flags & SK_BT_DESC)
compare = -compare;
}
if (compare != 0)
return compare;
}
return 0;
}
/*
* Extract attribute (column) number of stored entry from GIN tuple
*/
OffsetNumber
gintuple_get_attrnum(GinState *ginstate, IndexTuple tuple)
{
OffsetNumber colN;
if (ginstate->oneCol)
{
/* column number is not stored explicitly */
colN = FirstOffsetNumber;
}
else
{
Datum res;
bool isnull;
/*
* First attribute is always int16, so we can safely use any tuple
* descriptor to obtain first attribute of tuple
*/
res = index_getattr(tuple, FirstOffsetNumber, ginstate->tupdesc[0],
&isnull);
Assert(!isnull);
colN = DatumGetUInt16(res);
Assert(colN >= FirstOffsetNumber && colN <= ginstate->origTupdesc->natts);
}
return colN;
}
/*
* Write itup vector to page, has no control of free space
*/
static OffsetNumber
gistwritebuffer(Relation r, Page page, IndexTuple *itup,
int len, OffsetNumber off)
{
OffsetNumber l = InvalidOffsetNumber;
int i;
#ifdef GIST_PAGEADDITEM
GISTENTRY tmpdentry;
IndexTuple newtup;
bool IsNull;
#endif
for (i = 0; i < len; i++)
{
#ifdef GIST_PAGEADDITEM
l = gistPageAddItem(giststate, r, page,
(Item) itup[i], IndexTupleSize(itup[i]),
off, LP_USED, &tmpdentry, &newtup);
off = OffsetNumberNext(off);
if (DatumGetPointer(tmpdentry.key) != NULL &&
tmpdentry.key != index_getattr(itup[i], 1, r->rd_att, &IsNull))
pfree(DatumGetPointer(tmpdentry.key));
if (itup[i] != newtup)
pfree(newtup);
#else
l = PageAddItem(page, (Item) itup[i], IndexTupleSize(itup[i]),
off, LP_USED);
if (l == InvalidOffsetNumber)
elog(ERROR, "failed to add index item to \"%s\"",
RelationGetRelationName(r));
#endif
}
return l;
}
/*
* Make unions of keys in IndexTuple vector (one union datum per index column).
* Union Datums are returned into the attr/isnull arrays.
* Resulting Datums aren't compressed.
*/
void
gistMakeUnionItVec(GISTSTATE *giststate, IndexTuple *itvec, int len,
Datum *attr, bool *isnull)
{
int i;
GistEntryVector *evec;
int attrsize;
evec = (GistEntryVector *) palloc((len + 2) * sizeof(GISTENTRY) + GEVHDRSZ);
for (i = 0; i < giststate->tupdesc->natts; i++)
{
int j;
/* Collect non-null datums for this column */
evec->n = 0;
for (j = 0; j < len; j++)
{
Datum datum;
bool IsNull;
datum = index_getattr(itvec[j], i + 1, giststate->tupdesc, &IsNull);
if (IsNull)
continue;
gistdentryinit(giststate, i,
evec->vector + evec->n,
datum,
NULL, NULL, (OffsetNumber) 0,
false, IsNull);
evec->n++;
}
/* If this column was all NULLs, the union is NULL */
if (evec->n == 0)
{
attr[i] = (Datum) 0;
isnull[i] = true;
}
else
{
if (evec->n == 1)
{
/* unionFn may expect at least two inputs */
evec->n = 2;
evec->vector[1] = evec->vector[0];
}
/* Make union and store in attr array */
attr[i] = FunctionCall2Coll(&giststate->unionFn[i],
giststate->supportCollation[i],
PointerGetDatum(evec),
PointerGetDatum(&attrsize));
isnull[i] = false;
}
}
}
static void
readindextuple(readindexinfo *info, Relation irel, Relation hrel, Datum *values, bool *nulls)
{
BlockNumber blkno;
Page page;
OffsetNumber offnum;
IndexTuple itup;
HeapTupleData htup;
Buffer hbuf;
AttrNumber attno;
TupleDesc tupdesc = RelationGetDescr(irel);
blkno = info->blkno;
page = info->page;
offnum = info->offnum;
itup = (IndexTuple) PageGetItem(page, PageGetItemId(page, offnum));
htup.t_self = itup->t_tid;
values[1] = ItemPointerGetDatum(&itup->t_tid);
if (hrel == NULL)
values[2] = PointerGetDatum(cstring_to_text(AOTupleIdToString((AOTupleId *)&itup->t_tid)));
else
values[2] = PointerGetDatum(cstring_to_text("N/A"));
values[3] = PointerGetDatum(istatus_text(PageGetItemId(page, offnum)));
if (hrel != NULL)
{
if (heap_fetch(hrel, SnapshotAny, &htup, &hbuf, true, NULL))
values[4] = PointerGetDatum(hstatus_text(htup.t_data, true));
else if (htup.t_data)
values[4] = PointerGetDatum(hstatus_text(htup.t_data, false));
else
values[4] = PointerGetDatum(cstring_to_text("NOT_FOUND"));
ReleaseBuffer(hbuf);
}
else
values[4] = PointerGetDatum(cstring_to_text("N/A"));
for (attno = 1; attno <= tupdesc->natts; attno++)
{
bool isnull;
values[FIXED_COLUMN+attno-1] = index_getattr(itup, attno, tupdesc, &isnull);
nulls[FIXED_COLUMN+attno-1] = isnull;
}
}
/*
* Convert an index tuple into Datum/isnull arrays.
*
* The caller must allocate sufficient storage for the output arrays.
* (INDEX_MAX_KEYS entries should be enough.)
*/
void
index_deform_tuple(IndexTuple tup, TupleDesc tupleDescriptor,
Datum *values, bool *isnull)
{
int i;
/* Assert to protect callers who allocate fixed-size arrays */
Assert(tupleDescriptor->natts <= INDEX_MAX_KEYS);
for (i = 0; i < tupleDescriptor->natts; i++)
{
values[i] = index_getattr(tup, i + 1, tupleDescriptor, &isnull[i]);
}
}
/*
* Decompress all keys in tuple
*/
void
gistDeCompressAtt(GISTSTATE *giststate, Relation r, IndexTuple tuple, Page p,
OffsetNumber o, GISTENTRY *attdata, bool *isnull)
{
int i;
for (i = 0; i < r->rd_att->natts; i++)
{
Datum datum = index_getattr(tuple, i + 1, giststate->tupdesc, &isnull[i]);
gistdentryinit(giststate, i, &attdata[i],
datum, r, p, o,
FALSE, isnull[i]);
}
}
/*
* Convert an index tuple into Datum/isnull arrays.
*
* The caller must allocate sufficient storage for the output arrays.
* (INDEX_MAX_KEYS entries should be enough.)
*/
void
index_deform_tuple(
struct index_tuple* tup,
struct tuple *tuple,
datum_t* values,
bool* isnull)
{
int i;
/* ASSERT to protect callers who allocate fixed-size arrays */
ASSERT(tuple->natts <= INDEX_MAX_KEYS);
for (i = 0; i < tuple->natts; i++) {
values[i] = index_getattr(tup, i + 1, tuple, &isnull[i]);
}
}
/*
* Fetch all keys in tuple.
* Returns a new HeapTuple containing the originally-indexed data.
*/
HeapTuple
gistFetchTuple(GISTSTATE *giststate, Relation r, IndexTuple tuple)
{
MemoryContext oldcxt = MemoryContextSwitchTo(giststate->tempCxt);
Datum fetchatt[INDEX_MAX_KEYS];
bool isnull[INDEX_MAX_KEYS];
int i;
for (i = 0; i < r->rd_att->natts; i++)
{
Datum datum;
datum = index_getattr(tuple, i + 1, giststate->tupdesc, &isnull[i]);
if (giststate->fetchFn[i].fn_oid != InvalidOid)
{
if (!isnull[i])
fetchatt[i] = gistFetchAtt(giststate, i, datum, r);
else
fetchatt[i] = (Datum) 0;
}
else if (giststate->compressFn[i].fn_oid == InvalidOid)
{
/*
* If opclass does not provide compress method that could change
* original value, att is necessarily stored in original form.
*/
if (!isnull[i])
fetchatt[i] = datum;
else
fetchatt[i] = (Datum) 0;
}
else
{
/*
* Index-only scans not supported for this column. Since the
* planner chose an index-only scan anyway, it is not interested
* in this column, and we can replace it with a NULL.
*/
isnull[i] = true;
fetchatt[i] = (Datum) 0;
}
}
MemoryContextSwitchTo(oldcxt);
return heap_form_tuple(giststate->fetchTupdesc, fetchatt, isnull);
}
/*
* _hash_checkqual -- does the index tuple satisfy the scan conditions?
*/
bool
_hash_checkqual(IndexScanDesc scan, IndexTuple itup)
{
/*
* Currently, we can't check any of the scan conditions since we do not
* have the original index entry value to supply to the sk_func. Always
* return true; we expect that hashgettuple already set the recheck flag
* to make the main indexscan code do it.
*/
#ifdef NOT_USED
TupleDesc tupdesc = RelationGetDescr(scan->indexRelation);
ScanKey key = scan->keyData;
int scanKeySize = scan->numberOfKeys;
while (scanKeySize > 0)
{
Datum datum;
bool isNull;
Datum test;
datum = index_getattr(itup,
key->sk_attno,
tupdesc,
&isNull);
/* assume sk_func is strict */
if (isNull)
return false;
if (key->sk_flags & SK_ISNULL)
return false;
test = FunctionCall2Coll(&key->sk_func, key->sk_collation,
datum, key->sk_argument);
if (!DatumGetBool(test))
return false;
key++;
scanKeySize--;
}
#endif
return true;
}
/*
* Take a compressed entry, and install it on a page. Since we now know
* where the entry will live, we decompress it and recompress it using
* that knowledge (some compression routines may want to fish around
* on the page, for example, or do something special for leaf nodes.)
*/
static OffsetNumber
gistPageAddItem(GISTSTATE *giststate,
Relation r,
Page page,
Item item,
Size size,
OffsetNumber offsetNumber,
ItemIdFlags flags,
GISTENTRY *dentry,
IndexTuple *newtup)
{
GISTENTRY tmpcentry;
IndexTuple itup = (IndexTuple) item;
OffsetNumber retval;
Datum datum;
bool IsNull;
/*
* recompress the item given that we now know the exact page and
* offset for insertion
*/
datum = index_getattr(itup, 1, r->rd_att, &IsNull);
gistdentryinit(giststate, 0, dentry, datum,
(Relation) 0, (Page) 0,
(OffsetNumber) InvalidOffsetNumber,
ATTSIZE(datum, r, 1, IsNull),
FALSE, IsNull);
gistcentryinit(giststate, 0, &tmpcentry, dentry->key, r, page,
offsetNumber, dentry->bytes, FALSE);
*newtup = gist_tuple_replacekey(r, tmpcentry, itup);
retval = PageAddItem(page, (Item) *newtup, IndexTupleSize(*newtup),
offsetNumber, flags);
if (retval == InvalidOffsetNumber)
elog(ERROR, "failed to add index item to \"%s\"",
RelationGetRelationName(r));
/* be tidy */
if (DatumGetPointer(tmpcentry.key) != NULL &&
tmpcentry.key != dentry->key &&
tmpcentry.key != datum)
pfree(DatumGetPointer(tmpcentry.key));
return (retval);
}
/*
* _bt_mkscankey
* Build an insertion scan key that contains comparison data from itup
* as well as comparator routines appropriate to the key datatypes.
*
* The result is intended for use with _bt_compare().
*/
ScanKey
_bt_mkscankey(Relation rel, IndexTuple itup)
{
ScanKey skey;
TupleDesc itupdesc;
int natts;
int16 *indoption;
int i;
itupdesc = RelationGetDescr(rel);
natts = RelationGetNumberOfAttributes(rel);
indoption = rel->rd_indoption;
skey = (ScanKey) palloc(natts * sizeof(ScanKeyData));
for (i = 0; i < natts; i++)
{
FmgrInfo *procinfo;
Datum arg;
bool null;
int flags;
/*
* We can use the cached (default) support procs since no cross-type
* comparison can be needed.
*/
procinfo = index_getprocinfo(rel, i + 1, BTORDER_PROC);
arg = index_getattr(itup, i + 1, itupdesc, &null);
flags = (null ? SK_ISNULL : 0) | (indoption[i] << SK_BT_INDOPTION_SHIFT);
ScanKeyEntryInitializeWithInfo(&skey[i],
flags,
(AttrNumber) (i + 1),
InvalidStrategy,
InvalidOid,
rel->rd_indcollation[i],
procinfo,
arg);
}
return skey;
}
/*
** Given an IndexTuple to be inserted on a page, this routine replaces
** the key with another key, which may involve generating a new IndexTuple
** if the sizes don't match or if the null status changes.
**
** XXX this only works for a single-column index tuple!
*/
static IndexTuple
gist_tuple_replacekey(Relation r, GISTENTRY entry, IndexTuple t)
{
bool IsNull;
Datum datum = index_getattr(t, 1, r->rd_att, &IsNull);
/*
* If new entry fits in index tuple, copy it in. To avoid worrying
* about null-value bitmask, pass it off to the general
* index_formtuple routine if either the previous or new value is
* NULL.
*/
if (!IsNull && DatumGetPointer(entry.key) != NULL &&
(Size) entry.bytes <= ATTSIZE(datum, r, 1, IsNull))
{
memcpy(DatumGetPointer(datum),
DatumGetPointer(entry.key),
entry.bytes);
/* clear out old size */
t->t_info &= ~INDEX_SIZE_MASK;
/* or in new size */
t->t_info |= MAXALIGN(entry.bytes + sizeof(IndexTupleData));
return t;
}
else
{
/* generate a new index tuple for the compressed entry */
TupleDesc tupDesc = r->rd_att;
IndexTuple newtup;
char isnull;
isnull = DatumGetPointer(entry.key) != NULL ? ' ' : 'n';
newtup = (IndexTuple) index_formtuple(tupDesc,
&(entry.key),
&isnull);
newtup->t_tid = t->t_tid;
return newtup;
}
}
/*
* StoreIndexTuple
* Fill the slot with data from the index tuple.
*
* At some point this might be generally-useful functionality, but
* right now we don't need it elsewhere.
*/
static void
StoreIndexTuple(TupleTableSlot *slot, IndexTuple itup, TupleDesc itupdesc)
{
int nindexatts = itupdesc->natts;
Datum *values = slot->tts_values;
bool *isnull = slot->tts_isnull;
int i;
/*
* Note: we must use the tupdesc supplied by the AM in index_getattr, not
* the slot's tupdesc, in case the latter has different datatypes (this
* happens for btree name_ops in particular). They'd better have the same
* number of columns though, as well as being datatype-compatible which is
* something we can't so easily check.
*/
Assert(slot->tts_tupleDescriptor->natts == nindexatts);
ExecClearTuple(slot);
for (i = 0; i < nindexatts; i++)
values[i] = index_getattr(itup, i + 1, itupdesc, &isnull[i]);
ExecStoreVirtualTuple(slot);
}
/*
* bt_mk_scankey
* Build an insertion scan key that contains comparison data from itup
* as well as comparator routines appropriate to the key datatypes.
*
* The result is intended for use with bt_compare().
*/
struct scankey *bt_mk_scankey(struct relation* rel, struct index_tuple* itup)
{
struct scankey *skey;
struct tuple *itupdesc;
int natts;
int16 *indoption;
int i;
itupdesc = REL_DESC(rel);
natts = REL_GET_NR_ATTR(rel);
indoption = rel->rd_indoption;
skey = (struct scankey *)palloc(natts * sizeof(struct scankey));
for (i = 0; i < natts; i++) {
struct fmgr_info *procinfo;
datum_t arg;
bool null;
int flags;
/*
* We can use the cached (default) support procs since no cross-type
* comparison can be needed.
*/
procinfo = index_getprocinfo(rel, i + 1, BT_ORDER_PROC);
arg = index_getattr(itup, i + 1, itupdesc, &null);
flags = (null ? SK_ISNULL : 0) | (indoption[i] << SK_BT_INDEX_OPT_SHIFT);
scankey_init_info(&skey[i],
flags,
(attr_nr_t)(i + 1),
INVALID_STRAT,
INVALID_OID,
rel->rd_indcollation[i],
procinfo,
arg);
}
return skey;
}
/*
* Adds array of item pointers to tuple's posting list or
* creates posting tree and tuple pointed to tree in a case
* of not enough space. Max size of tuple is defined in
* GinFormTuple().
*/
static IndexTuple
addItemPointersToTuple(Relation index, GinState *ginstate, GinBtreeStack *stack,
IndexTuple old, ItemPointerData *items, uint32 nitem, bool isBuild)
{
bool isnull;
Datum key = index_getattr(old, FirstOffsetNumber, ginstate->tupdesc, &isnull);
IndexTuple res = GinFormTuple(ginstate, key, NULL, nitem + GinGetNPosting(old));
if (res)
{
/* good, small enough */
MergeItemPointers(GinGetPosting(res),
GinGetPosting(old), GinGetNPosting(old),
items, nitem
);
GinSetNPosting(res, nitem + GinGetNPosting(old));
}
else
{
BlockNumber postingRoot;
GinPostingTreeScan *gdi;
/* posting list becomes big, so we need to make posting's tree */
res = GinFormTuple(ginstate, key, NULL, 0);
postingRoot = createPostingTree(index, GinGetPosting(old), GinGetNPosting(old));
GinSetPostingTree(res, postingRoot);
gdi = prepareScanPostingTree(index, postingRoot, FALSE);
gdi->btree.isBuild = isBuild;
insertItemPointer(gdi, items, nitem);
pfree(gdi);
}
return res;
}
/*
* _hash_checkqual -- does the index tuple satisfy the scan conditions?
*/
bool
_hash_checkqual(IndexScanDesc scan, IndexTuple itup)
{
TupleDesc tupdesc = RelationGetDescr(scan->indexRelation);
ScanKey key = scan->keyData;
int scanKeySize = scan->numberOfKeys;
IncrIndexProcessed();
while (scanKeySize > 0)
{
Datum datum;
bool isNull;
Datum test;
datum = index_getattr(itup,
key->sk_attno,
tupdesc,
&isNull);
/* assume sk_func is strict */
if (isNull)
return false;
if (key->sk_flags & SK_ISNULL)
return false;
test = FunctionCall2(&key->sk_func, datum, key->sk_argument);
if (!DatumGetBool(test))
return false;
key++;
scanKeySize--;
}
return true;
}
static void
gistDeCompressAtt(GISTSTATE *giststate, Relation r, IndexTuple tuple, Page p,
OffsetNumber o, GISTENTRY attdata[], bool decompvec[], bool isnull[])
{
int i;
Datum datum;
for (i = 0; i < r->rd_att->natts; i++)
{
datum = index_getattr(tuple, i + 1, giststate->tupdesc, &isnull[i]);
gistdentryinit(giststate, i, &attdata[i],
datum, r, p, o,
ATTSIZE(datum, giststate->tupdesc, i + 1, isnull[i]), FALSE, isnull[i]);
if (isAttByVal(giststate, i))
decompvec[i] = FALSE;
else
{
if (attdata[i].key == datum || isnull[i])
decompvec[i] = FALSE;
else
decompvec[i] = TRUE;
}
}
}
/*
* For a newly inserted heap tid, check if an entry with this tid
* already exists in a unique index. If it does, abort the inserting
* transaction.
*/
static void
_bt_validate_tid(Relation irel, ItemPointer h_tid)
{
MIRROREDLOCK_BUFMGR_DECLARE;
BlockNumber blkno;
BlockNumber num_pages;
Buffer buf;
Page page;
BTPageOpaque opaque;
IndexTuple itup;
OffsetNumber maxoff,
minoff,
offnum;
elog(DEBUG1, "validating tid (%d,%d) for index (%s)",
ItemPointerGetBlockNumber(h_tid), ItemPointerGetOffsetNumber(h_tid),
RelationGetRelationName(irel));
blkno = BTREE_METAPAGE + 1;
num_pages = RelationGetNumberOfBlocks(irel);
MIRROREDLOCK_BUFMGR_LOCK;
for (; blkno < num_pages; blkno++)
{
buf = ReadBuffer(irel, blkno);
page = BufferGetPage(buf);
opaque = (BTPageOpaque) PageGetSpecialPointer(page);
if (!PageIsNew(page))
_bt_checkpage(irel, buf);
if (P_ISLEAF(opaque))
{
minoff = P_FIRSTDATAKEY(opaque);
maxoff = PageGetMaxOffsetNumber(page);
for (offnum = minoff;
offnum <= maxoff;
offnum = OffsetNumberNext(offnum))
{
itup = (IndexTuple) PageGetItem(page,
PageGetItemId(page, offnum));
if (ItemPointerEquals(&itup->t_tid, h_tid))
{
Form_pg_attribute key_att = RelationGetDescr(irel)->attrs[0];
Oid key = InvalidOid;
bool isnull;
if (key_att->atttypid == OIDOID)
{
key = DatumGetInt32(
index_getattr(itup, 1, RelationGetDescr(irel), &isnull));
elog(ERROR, "found tid (%d,%d), %s (%d) already in index (%s)",
ItemPointerGetBlockNumber(h_tid), ItemPointerGetOffsetNumber(h_tid),
NameStr(key_att->attname), key, RelationGetRelationName(irel));
}
else
{
elog(ERROR, "found tid (%d,%d) already in index (%s)",
ItemPointerGetBlockNumber(h_tid), ItemPointerGetOffsetNumber(h_tid),
RelationGetRelationName(irel));
}
}
}
}
ReleaseBuffer(buf);
}
MIRROREDLOCK_BUFMGR_UNLOCK;
}
/*
* Post vacuum, iterate over all entries in index, check if the h_tid
* of each entry exists and is not dead. For specific system tables,
* also ensure that the key in index entry matches the corresponding
* attribute in the heap tuple.
*/
void
_bt_validate_vacuum(Relation irel, Relation hrel, TransactionId oldest_xmin)
{
MIRROREDLOCK_BUFMGR_DECLARE;
BlockNumber blkno;
BlockNumber num_pages;
Buffer ibuf = InvalidBuffer;
Buffer hbuf = InvalidBuffer;
Page ipage;
BTPageOpaque opaque;
IndexTuple itup;
HeapTupleData htup;
OffsetNumber maxoff,
minoff,
offnum;
Oid ioid,
hoid;
bool isnull;
blkno = BTREE_METAPAGE + 1;
num_pages = RelationGetNumberOfBlocks(irel);
elog(LOG, "btvalidatevacuum: index %s, heap %s",
RelationGetRelationName(irel), RelationGetRelationName(hrel));
MIRROREDLOCK_BUFMGR_LOCK;
for (; blkno < num_pages; blkno++)
{
ibuf = ReadBuffer(irel, blkno);
ipage = BufferGetPage(ibuf);
opaque = (BTPageOpaque) PageGetSpecialPointer(ipage);
if (!PageIsNew(ipage))
_bt_checkpage(irel, ibuf);
if (P_ISLEAF(opaque))
{
minoff = P_FIRSTDATAKEY(opaque);
maxoff = PageGetMaxOffsetNumber(ipage);
for (offnum = minoff;
offnum <= maxoff;
offnum = OffsetNumberNext(offnum))
{
itup = (IndexTuple) PageGetItem(ipage,
PageGetItemId(ipage, offnum));
ItemPointerCopy(&itup->t_tid, &htup.t_self);
/*
* TODO: construct a tid bitmap based on index tids
* and fetch heap tids in order afterwards. That will
* also allow validating if a heap tid appears twice
* in a unique index.
*/
if (!heap_release_fetch(hrel, SnapshotAny, &htup,
&hbuf, true, NULL))
{
elog(ERROR, "btvalidatevacuum: tid (%d,%d) from index %s "
"not found in heap %s",
ItemPointerGetBlockNumber(&itup->t_tid),
ItemPointerGetOffsetNumber(&itup->t_tid),
RelationGetRelationName(irel),
RelationGetRelationName(hrel));
}
switch (HeapTupleSatisfiesVacuum(hrel, htup.t_data, oldest_xmin, hbuf))
{
case HEAPTUPLE_RECENTLY_DEAD:
case HEAPTUPLE_LIVE:
case HEAPTUPLE_INSERT_IN_PROGRESS:
case HEAPTUPLE_DELETE_IN_PROGRESS:
/* these tuples are considered alive by vacuum */
break;
case HEAPTUPLE_DEAD:
elog(ERROR, "btvalidatevacuum: vacuum did not remove "
"dead tuple (%d,%d) from heap %s and index %s",
ItemPointerGetBlockNumber(&itup->t_tid),
ItemPointerGetOffsetNumber(&itup->t_tid),
RelationGetRelationName(hrel),
RelationGetRelationName(irel));
break;
default:
elog(ERROR, "btvalidatevacuum: invalid visibility");
break;
}
switch(RelationGetRelid(irel))
{
case DatabaseOidIndexId:
case TypeOidIndexId:
case ClassOidIndexId:
case ConstraintOidIndexId:
hoid = HeapTupleGetOid(&htup);
ioid = index_getattr(itup, 1, RelationGetDescr(irel), &isnull);
if (hoid != ioid)
{
elog(ERROR,
"btvalidatevacuum: index oid(%d) != heap oid(%d)"
" tuple (%d,%d) index %s", ioid, hoid,
ItemPointerGetBlockNumber(&itup->t_tid),
ItemPointerGetOffsetNumber(&itup->t_tid),
RelationGetRelationName(irel));
}
break;
case GpRelationNodeOidIndexId:
hoid = heap_getattr(&htup, 1, RelationGetDescr(hrel), &isnull);
ioid = index_getattr(itup, 1, RelationGetDescr(irel), &isnull);
if (hoid != ioid)
{
elog(ERROR,
"btvalidatevacuum: index oid(%d) != heap oid(%d)"
" tuple (%d,%d) index %s", ioid, hoid,
ItemPointerGetBlockNumber(&itup->t_tid),
ItemPointerGetOffsetNumber(&itup->t_tid),
RelationGetRelationName(irel));
}
int4 hsegno = heap_getattr(&htup, 2, RelationGetDescr(hrel), &isnull);
int4 isegno = index_getattr(itup, 2, RelationGetDescr(irel), &isnull);
if (isegno != hsegno)
{
elog(ERROR,
"btvalidatevacuum: index segno(%d) != heap segno(%d)"
" tuple (%d,%d) index %s", isegno, hsegno,
ItemPointerGetBlockNumber(&itup->t_tid),
ItemPointerGetOffsetNumber(&itup->t_tid),
RelationGetRelationName(irel));
}
break;
default:
break;
}
if (RelationGetNamespace(irel) == PG_AOSEGMENT_NAMESPACE)
{
int4 isegno = index_getattr(itup, 1, RelationGetDescr(irel), &isnull);
int4 hsegno = heap_getattr(&htup, 1, RelationGetDescr(hrel), &isnull);
if (isegno != hsegno)
{
elog(ERROR,
"btvalidatevacuum: index segno(%d) != heap segno(%d)"
" tuple (%d,%d) index %s", isegno, hsegno,
ItemPointerGetBlockNumber(&itup->t_tid),
ItemPointerGetOffsetNumber(&itup->t_tid),
RelationGetRelationName(irel));
}
}
}
}
if (BufferIsValid(ibuf))
ReleaseBuffer(ibuf);
}
if (BufferIsValid(hbuf))
ReleaseBuffer(hbuf);
MIRROREDLOCK_BUFMGR_UNLOCK;
}
/*----------
* _bt_compare() -- Compare scankey to a particular tuple on the page.
*
* The passed scankey must be an insertion-type scankey (see nbtree/README),
* but it can omit the rightmost column(s) of the index.
*
* keysz: number of key conditions to be checked (might be less than the
* number of index columns!)
* page/offnum: location of btree item to be compared to.
*
* This routine returns:
* <0 if scankey < tuple at offnum;
* 0 if scankey == tuple at offnum;
* >0 if scankey > tuple at offnum.
* NULLs in the keys are treated as sortable values. Therefore
* "equality" does not necessarily mean that the item should be
* returned to the caller as a matching key!
*
* CRUCIAL NOTE: on a non-leaf page, the first data key is assumed to be
* "minus infinity": this routine will always claim it is less than the
* scankey. The actual key value stored (if any, which there probably isn't)
* does not matter. This convention allows us to implement the Lehman and
* Yao convention that the first down-link pointer is before the first key.
* See backend/access/nbtree/README for details.
*----------
*/
int32
_bt_compare(Relation rel,
int keysz,
ScanKey scankey,
Page page,
OffsetNumber offnum)
{
TupleDesc itupdesc = RelationGetDescr(rel);
BTPageOpaque opaque = (BTPageOpaque) PageGetSpecialPointer(page);
IndexTuple itup;
int i;
/*
* Force result ">" if target item is first data item on an internal page
* --- see NOTE above.
*/
if (!P_ISLEAF(opaque) && offnum == P_FIRSTDATAKEY(opaque))
return 1;
itup = (IndexTuple) PageGetItem(page, PageGetItemId(page, offnum));
/*
* The scan key is set up with the attribute number associated with each
* term in the key. It is important that, if the index is multi-key, the
* scan contain the first k key attributes, and that they be in order. If
* you think about how multi-key ordering works, you'll understand why
* this is.
*
* We don't test for violation of this condition here, however. The
* initial setup for the index scan had better have gotten it right (see
* _bt_first).
*/
for (i = 1; i <= keysz; i++)
{
Datum datum;
bool isNull;
int32 result;
datum = index_getattr(itup, scankey->sk_attno, itupdesc, &isNull);
/* see comments about NULLs handling in btbuild */
if (scankey->sk_flags & SK_ISNULL) /* key is NULL */
{
if (isNull)
result = 0; /* NULL "=" NULL */
else if (scankey->sk_flags & SK_BT_NULLS_FIRST)
result = -1; /* NULL "<" NOT_NULL */
else
result = 1; /* NULL ">" NOT_NULL */
}
else if (isNull) /* key is NOT_NULL and item is NULL */
{
if (scankey->sk_flags & SK_BT_NULLS_FIRST)
result = 1; /* NOT_NULL ">" NULL */
else
result = -1; /* NOT_NULL "<" NULL */
}
else
{
/*
* The sk_func needs to be passed the index value as left arg and
* the sk_argument as right arg (they might be of different
* types). Since it is convenient for callers to think of
* _bt_compare as comparing the scankey to the index item, we have
* to flip the sign of the comparison result. (Unless it's a DESC
* column, in which case we *don't* flip the sign.)
*/
result = DatumGetInt32(FunctionCall2Coll(&scankey->sk_func,
scankey->sk_collation,
datum,
scankey->sk_argument));
if (!(scankey->sk_flags & SK_BT_DESC))
result = -result;
}
/* if the keys are unequal, return the difference */
if (result != 0)
return result;
scankey++;
}
/* if we get here, the keys are equal */
return 0;
}
/*
* Read tuples in correct sort order from tuplesort, and load them into
* btree leaves.
*/
static void
_bt_load(BTWriteState *wstate, BTSpool *btspool, BTSpool *btspool2)
{
BTPageState *state = NULL;
bool merge = (btspool2 != NULL);
IndexTuple itup,
itup2 = NULL;
bool load1;
TupleDesc tupdes = RelationGetDescr(wstate->index);
int i,
keysz = RelationGetNumberOfAttributes(wstate->index);
ScanKey indexScanKey = NULL;
SortSupport sortKeys;
if (merge)
{
/*
* Another BTSpool for dead tuples exists. Now we have to merge
* btspool and btspool2.
*/
/* the preparation of merge */
itup = tuplesort_getindextuple(btspool->sortstate, true);
itup2 = tuplesort_getindextuple(btspool2->sortstate, true);
indexScanKey = _bt_mkscankey_nodata(wstate->index);
/* Prepare SortSupport data for each column */
sortKeys = (SortSupport) palloc0(keysz * sizeof(SortSupportData));
for (i = 0; i < keysz; i++)
{
SortSupport sortKey = sortKeys + i;
ScanKey scanKey = indexScanKey + i;
int16 strategy;
sortKey->ssup_cxt = CurrentMemoryContext;
sortKey->ssup_collation = scanKey->sk_collation;
sortKey->ssup_nulls_first =
(scanKey->sk_flags & SK_BT_NULLS_FIRST) != 0;
sortKey->ssup_attno = scanKey->sk_attno;
/* Abbreviation is not supported here */
sortKey->abbreviate = false;
AssertState(sortKey->ssup_attno != 0);
strategy = (scanKey->sk_flags & SK_BT_DESC) != 0 ?
BTGreaterStrategyNumber : BTLessStrategyNumber;
PrepareSortSupportFromIndexRel(wstate->index, strategy, sortKey);
}
_bt_freeskey(indexScanKey);
for (;;)
{
load1 = true; /* load BTSpool next ? */
if (itup2 == NULL)
{
if (itup == NULL)
break;
}
else if (itup != NULL)
{
for (i = 1; i <= keysz; i++)
{
SortSupport entry;
Datum attrDatum1,
attrDatum2;
bool isNull1,
isNull2;
int32 compare;
entry = sortKeys + i - 1;
attrDatum1 = index_getattr(itup, i, tupdes, &isNull1);
attrDatum2 = index_getattr(itup2, i, tupdes, &isNull2);
compare = ApplySortComparator(attrDatum1, isNull1,
attrDatum2, isNull2,
entry);
if (compare > 0)
{
load1 = false;
break;
}
else if (compare < 0)
break;
}
}
else
load1 = false;
/* When we see first tuple, create first index page */
if (state == NULL)
state = _bt_pagestate(wstate, 0);
if (load1)
{
_bt_buildadd(wstate, state, itup);
itup = tuplesort_getindextuple(btspool->sortstate, true);
}
else
{
_bt_buildadd(wstate, state, itup2);
itup2 = tuplesort_getindextuple(btspool2->sortstate, true);
}
}
pfree(sortKeys);
}
else
{
/* merge is unnecessary */
while ((itup = tuplesort_getindextuple(btspool->sortstate,
true)) != NULL)
{
/* When we see first tuple, create first index page */
if (state == NULL)
state = _bt_pagestate(wstate, 0);
_bt_buildadd(wstate, state, itup);
}
}
/* Close down final pages and write the metapage */
_bt_uppershutdown(wstate, state);
/*
* If the index is WAL-logged, we must fsync it down to disk before it's
* safe to commit the transaction. (For a non-WAL-logged index we don't
* care since the index will be uninteresting after a crash anyway.)
*
* It's obvious that we must do this when not WAL-logging the build. It's
* less obvious that we have to do it even if we did WAL-log the index
* pages. The reason is that since we're building outside shared buffers,
* a CHECKPOINT occurring during the build has no way to flush the
* previously written data to disk (indeed it won't know the index even
* exists). A crash later on would replay WAL from the checkpoint,
* therefore it wouldn't replay our earlier WAL entries. If we do not
* fsync those pages here, they might still not be on disk when the crash
* occurs.
*/
if (RelationNeedsWAL(wstate->index))
{
RelationOpenSmgr(wstate->index);
smgrimmedsync(wstate->index->rd_smgr, MAIN_FORKNUM);
}
}
/*----------
* _bt_compare() -- Compare scankey to a particular tuple on the page.
*
* keysz: number of key conditions to be checked (might be less than the
* total length of the scan key!)
* page/offnum: location of btree item to be compared to.
*
* This routine returns:
* <0 if scankey < tuple at offnum;
* 0 if scankey == tuple at offnum;
* >0 if scankey > tuple at offnum.
* NULLs in the keys are treated as sortable values. Therefore
* "equality" does not necessarily mean that the item should be
* returned to the caller as a matching key!
*
* CRUCIAL NOTE: on a non-leaf page, the first data key is assumed to be
* "minus infinity": this routine will always claim it is less than the
* scankey. The actual key value stored (if any, which there probably isn't)
* does not matter. This convention allows us to implement the Lehman and
* Yao convention that the first down-link pointer is before the first key.
* See backend/access/nbtree/README for details.
*----------
*/
int32
_bt_compare(Relation rel,
int keysz,
ScanKey scankey,
Page page,
OffsetNumber offnum)
{
TupleDesc itupdesc = RelationGetDescr(rel);
BTPageOpaque opaque = (BTPageOpaque) PageGetSpecialPointer(page);
BTItem btitem;
IndexTuple itup;
int i;
/*
* Force result ">" if target item is first data item on an internal
* page --- see NOTE above.
*/
if (!P_ISLEAF(opaque) && offnum == P_FIRSTDATAKEY(opaque))
return 1;
btitem = (BTItem) PageGetItem(page, PageGetItemId(page, offnum));
itup = &(btitem->bti_itup);
/*
* The scan key is set up with the attribute number associated with
* each term in the key. It is important that, if the index is
* multi-key, the scan contain the first k key attributes, and that
* they be in order. If you think about how multi-key ordering works,
* you'll understand why this is.
*
* We don't test for violation of this condition here, however. The
* initial setup for the index scan had better have gotten it right
* (see _bt_first).
*/
for (i = 0; i < keysz; i++)
{
ScanKey entry = &scankey[i];
Datum datum;
bool isNull;
int32 result;
datum = index_getattr(itup, entry->sk_attno, itupdesc, &isNull);
/* see comments about NULLs handling in btbuild */
if (entry->sk_flags & SK_ISNULL) /* key is NULL */
{
if (isNull)
result = 0; /* NULL "=" NULL */
else
result = 1; /* NULL ">" NOT_NULL */
}
else if (isNull) /* key is NOT_NULL and item is NULL */
{
result = -1; /* NOT_NULL "<" NULL */
}
else
{
result = DatumGetInt32(FunctionCall2(&entry->sk_func,
entry->sk_argument,
datum));
}
/* if the keys are unequal, return the difference */
if (result != 0)
return result;
}
/* if we get here, the keys are equal */
return 0;
}
/*
* Test whether an indextuple satisfies a row-comparison scan condition.
*
* Return true if so, false if not. If not, also clear *continuescan if
* it's not possible for any future tuples in the current scan direction
* to pass the qual.
*
* This is a subroutine for _bt_checkkeys, which see for more info.
*/
static bool
_bt_check_rowcompare(ScanKey skey, IndexTuple tuple, TupleDesc tupdesc,
ScanDirection dir, bool *continuescan)
{
ScanKey subkey = (ScanKey) DatumGetPointer(skey->sk_argument);
int32 cmpresult = 0;
bool result;
/* First subkey should be same as the header says */
Assert(subkey->sk_attno == skey->sk_attno);
/* Loop over columns of the row condition */
for (;;)
{
Datum datum;
bool isNull;
Assert(subkey->sk_flags & SK_ROW_MEMBER);
datum = index_getattr(tuple,
subkey->sk_attno,
tupdesc,
&isNull);
if (isNull)
{
if (subkey->sk_flags & SK_BT_NULLS_FIRST)
{
/*
* Since NULLs are sorted before non-NULLs, we know we have
* reached the lower limit of the range of values for this
* index attr. On a backward scan, we can stop if this qual is
* one of the "must match" subset. On a forward scan,
* however, we should keep going.
*/
if ((subkey->sk_flags & SK_BT_REQBKWD) &&
ScanDirectionIsBackward(dir))
*continuescan = false;
}
else
{
/*
* Since NULLs are sorted after non-NULLs, we know we have
* reached the upper limit of the range of values for this
* index attr. On a forward scan, we can stop if this qual is
* one of the "must match" subset. On a backward scan,
* however, we should keep going.
*/
if ((subkey->sk_flags & SK_BT_REQFWD) &&
ScanDirectionIsForward(dir))
*continuescan = false;
}
/*
* In any case, this indextuple doesn't match the qual.
*/
return false;
}
if (subkey->sk_flags & SK_ISNULL)
{
/*
* Unlike the simple-scankey case, this isn't a disallowed case.
* But it can never match. If all the earlier row comparison
* columns are required for the scan direction, we can stop the
* scan, because there can't be another tuple that will succeed.
*/
if (subkey != (ScanKey) DatumGetPointer(skey->sk_argument))
subkey--;
if ((subkey->sk_flags & SK_BT_REQFWD) &&
ScanDirectionIsForward(dir))
*continuescan = false;
else if ((subkey->sk_flags & SK_BT_REQBKWD) &&
ScanDirectionIsBackward(dir))
*continuescan = false;
return false;
}
/* Perform the test --- three-way comparison not bool operator */
cmpresult = DatumGetInt32(FunctionCall2(&subkey->sk_func,
datum,
subkey->sk_argument));
if (subkey->sk_flags & SK_BT_DESC)
cmpresult = -cmpresult;
/* Done comparing if unequal, else advance to next column */
if (cmpresult != 0)
break;
if (subkey->sk_flags & SK_ROW_END)
break;
subkey++;
}
/*
* At this point cmpresult indicates the overall result of the row
* comparison, and subkey points to the deciding column (or the last
* column if the result is "=").
*/
switch (subkey->sk_strategy)
{
/* EQ and NE cases aren't allowed here */
case BTLessStrategyNumber:
result = (cmpresult < 0);
break;
case BTLessEqualStrategyNumber:
result = (cmpresult <= 0);
break;
case BTGreaterEqualStrategyNumber:
result = (cmpresult >= 0);
break;
case BTGreaterStrategyNumber:
result = (cmpresult > 0);
break;
default:
elog(ERROR, "unrecognized RowCompareType: %d",
(int) subkey->sk_strategy);
result = 0; /* keep compiler quiet */
break;
}
if (!result)
{
/*
* Tuple fails this qual. If it's a required qual for the current
* scan direction, then we can conclude no further tuples will pass,
* either. Note we have to look at the deciding column, not
* necessarily the first or last column of the row condition.
*/
if ((subkey->sk_flags & SK_BT_REQFWD) &&
ScanDirectionIsForward(dir))
*continuescan = false;
else if ((subkey->sk_flags & SK_BT_REQBKWD) &&
ScanDirectionIsBackward(dir))
*continuescan = false;
}
return result;
}
/*
* Test whether an indextuple satisfies all the scankey conditions.
*
* If so, copy its TID into scan->xs_ctup.t_self, and return TRUE.
* If not, return FALSE (xs_ctup is not changed).
*
* If the tuple fails to pass the qual, we also determine whether there's
* any need to continue the scan beyond this tuple, and set *continuescan
* accordingly. See comments for _bt_preprocess_keys(), above, about how
* this is done.
*
* scan: index scan descriptor (containing a search-type scankey)
* page: buffer page containing index tuple
* offnum: offset number of index tuple (must be a valid item!)
* dir: direction we are scanning in
* continuescan: output parameter (will be set correctly in all cases)
*/
bool
_bt_checkkeys(IndexScanDesc scan,
Page page, OffsetNumber offnum,
ScanDirection dir, bool *continuescan)
{
ItemId iid = PageGetItemId(page, offnum);
bool tuple_valid;
IndexTuple tuple;
TupleDesc tupdesc;
BTScanOpaque so;
int keysz;
int ikey;
ScanKey key;
*continuescan = true; /* default assumption */
/*
* If the scan specifies not to return killed tuples, then we treat a
* killed tuple as not passing the qual. Most of the time, it's a win to
* not bother examining the tuple's index keys, but just return
* immediately with continuescan = true to proceed to the next tuple.
* However, if this is the last tuple on the page, we should check the
* index keys to prevent uselessly advancing to the next page.
*/
if (scan->ignore_killed_tuples && ItemIdIsDead(iid))
{
/* return immediately if there are more tuples on the page */
if (ScanDirectionIsForward(dir))
{
if (offnum < PageGetMaxOffsetNumber(page))
return false;
}
else
{
BTPageOpaque opaque = (BTPageOpaque) PageGetSpecialPointer(page);
if (offnum > P_FIRSTDATAKEY(opaque))
return false;
}
/*
* OK, we want to check the keys, but we'll return FALSE even if the
* tuple passes the key tests.
*/
tuple_valid = false;
}
else
tuple_valid = true;
tuple = (IndexTuple) PageGetItem(page, iid);
IncrIndexProcessed();
tupdesc = RelationGetDescr(scan->indexRelation);
so = (BTScanOpaque) scan->opaque;
keysz = so->numberOfKeys;
for (key = so->keyData, ikey = 0; ikey < keysz; key++, ikey++)
{
Datum datum;
bool isNull;
Datum test;
/* row-comparison keys need special processing */
if (key->sk_flags & SK_ROW_HEADER)
{
if (_bt_check_rowcompare(key, tuple, tupdesc, dir, continuescan))
continue;
return false;
}
datum = index_getattr(tuple,
key->sk_attno,
tupdesc,
&isNull);
if (key->sk_flags & SK_ISNULL)
{
/* Handle IS NULL tests */
Assert(key->sk_flags & SK_SEARCHNULL);
if (isNull)
continue; /* tuple satisfies this qual */
/*
* Tuple fails this qual. If it's a required qual for the current
* scan direction, then we can conclude no further tuples will
* pass, either.
*/
if ((key->sk_flags & SK_BT_REQFWD) &&
ScanDirectionIsForward(dir))
*continuescan = false;
else if ((key->sk_flags & SK_BT_REQBKWD) &&
ScanDirectionIsBackward(dir))
*continuescan = false;
/*
* In any case, this indextuple doesn't match the qual.
*/
return false;
}
if (isNull)
{
if (key->sk_flags & SK_BT_NULLS_FIRST)
{
/*
* Since NULLs are sorted before non-NULLs, we know we have
* reached the lower limit of the range of values for this
* index attr. On a backward scan, we can stop if this qual
* is one of the "must match" subset. On a forward scan,
* however, we should keep going.
*/
if ((key->sk_flags & SK_BT_REQBKWD) &&
ScanDirectionIsBackward(dir))
*continuescan = false;
}
else
{
/*
* Since NULLs are sorted after non-NULLs, we know we have
* reached the upper limit of the range of values for this
* index attr. On a forward scan, we can stop if this qual is
* one of the "must match" subset. On a backward scan,
* however, we should keep going.
*/
if ((key->sk_flags & SK_BT_REQFWD) &&
ScanDirectionIsForward(dir))
*continuescan = false;
}
/*
* In any case, this indextuple doesn't match the qual.
*/
return false;
}
test = FunctionCall2(&key->sk_func, datum, key->sk_argument);
if (!DatumGetBool(test))
{
/*
* Tuple fails this qual. If it's a required qual for the current
* scan direction, then we can conclude no further tuples will
* pass, either.
*
* Note: because we stop the scan as soon as any required equality
* qual fails, it is critical that equality quals be used for the
* initial positioning in _bt_first() when they are available. See
* comments in _bt_first().
*/
if ((key->sk_flags & SK_BT_REQFWD) &&
ScanDirectionIsForward(dir))
*continuescan = false;
else if ((key->sk_flags & SK_BT_REQBKWD) &&
ScanDirectionIsBackward(dir))
*continuescan = false;
/*
* In any case, this indextuple doesn't match the qual.
*/
return false;
}
}
/* If we get here, the tuple passes all index quals. */
if (tuple_valid)
scan->xs_ctup.t_self = tuple->t_tid;
return tuple_valid;
}
/*
* gistindex_keytest() -- does this index tuple satisfy the scan key(s)?
*
* The index tuple might represent either a heap tuple or a lower index page,
* depending on whether the containing page is a leaf page or not.
*
* On success return for a heap tuple, *recheck_p is set to indicate
* whether recheck is needed. We recheck if any of the consistent() functions
* request it. recheck is not interesting when examining a non-leaf entry,
* since we must visit the lower index page if there's any doubt.
*
* If we are doing an ordered scan, so->distances[] is filled with distance
* data from the distance() functions before returning success.
*
* We must decompress the key in the IndexTuple before passing it to the
* sk_funcs (which actually are the opclass Consistent or Distance methods).
*
* Note that this function is always invoked in a short-lived memory context,
* so we don't need to worry about cleaning up allocated memory, either here
* or in the implementation of any Consistent or Distance methods.
*/
static bool
gistindex_keytest(IndexScanDesc scan,
IndexTuple tuple,
Page page,
OffsetNumber offset,
bool *recheck_p)
{
GISTScanOpaque so = (GISTScanOpaque) scan->opaque;
GISTSTATE *giststate = so->giststate;
ScanKey key = scan->keyData;
int keySize = scan->numberOfKeys;
double *distance_p;
Relation r = scan->indexRelation;
*recheck_p = false;
/*
* If it's a leftover invalid tuple from pre-9.1, treat it as a match with
* minimum possible distances. This means we'll always follow it to the
* referenced page.
*
* GPDB: the virtual TIDs created for AO tables use the full range of
* offset numbers from 0 to 65535. So a tuple on leaf page that looks like
* an invalid tuple, is actually ok.
*/
if (!GistPageIsLeaf(page) && GistTupleIsInvalid(tuple))
{
int i;
for (i = 0; i < scan->numberOfOrderBys; i++)
so->distances[i] = -get_float8_infinity();
return true;
}
/* Check whether it matches according to the Consistent functions */
while (keySize > 0)
{
Datum datum;
bool isNull;
datum = index_getattr(tuple,
key->sk_attno,
giststate->tupdesc,
&isNull);
if (key->sk_flags & SK_ISNULL)
{
/*
* On non-leaf page we can't conclude that child hasn't NULL
* values because of assumption in GiST: union (VAL, NULL) is VAL.
* But if on non-leaf page key IS NULL, then all children are
* NULL.
*/
if (key->sk_flags & SK_SEARCHNULL)
{
if (GistPageIsLeaf(page) && !isNull)
return false;
}
else
{
Assert(key->sk_flags & SK_SEARCHNOTNULL);
if (isNull)
return false;
}
}
else if (isNull)
{
return false;
}
else
{
Datum test;
bool recheck;
GISTENTRY de;
gistdentryinit(giststate, key->sk_attno - 1, &de,
datum, r, page, offset,
FALSE, isNull);
/*
* Call the Consistent function to evaluate the test. The
* arguments are the index datum (as a GISTENTRY*), the comparison
* datum, the comparison operator's strategy number and subtype
* from pg_amop, and the recheck flag.
*
* (Presently there's no need to pass the subtype since it'll
* always be zero, but might as well pass it for possible future
* use.)
*
* We initialize the recheck flag to true (the safest assumption)
* in case the Consistent function forgets to set it.
*/
recheck = true;
test = FunctionCall5Coll(&key->sk_func,
key->sk_collation,
PointerGetDatum(&de),
key->sk_argument,
Int32GetDatum(key->sk_strategy),
ObjectIdGetDatum(key->sk_subtype),
PointerGetDatum(&recheck));
if (!DatumGetBool(test))
return false;
*recheck_p |= recheck;
}
key++;
keySize--;
}
/* OK, it passes --- now let's compute the distances */
key = scan->orderByData;
distance_p = so->distances;
keySize = scan->numberOfOrderBys;
while (keySize > 0)
{
Datum datum;
bool isNull;
datum = index_getattr(tuple,
key->sk_attno,
giststate->tupdesc,
&isNull);
if ((key->sk_flags & SK_ISNULL) || isNull)
{
/* Assume distance computes as null and sorts to the end */
*distance_p = get_float8_infinity();
}
else
{
Datum dist;
GISTENTRY de;
gistdentryinit(giststate, key->sk_attno - 1, &de,
datum, r, page, offset,
FALSE, isNull);
/*
* Call the Distance function to evaluate the distance. The
* arguments are the index datum (as a GISTENTRY*), the comparison
* datum, and the ordering operator's strategy number and subtype
* from pg_amop.
*
* (Presently there's no need to pass the subtype since it'll
* always be zero, but might as well pass it for possible future
* use.)
*
* Note that Distance functions don't get a recheck argument. We
* can't tolerate lossy distance calculations on leaf tuples;
* there is no opportunity to re-sort the tuples afterwards.
*/
dist = FunctionCall4Coll(&key->sk_func,
key->sk_collation,
PointerGetDatum(&de),
key->sk_argument,
Int32GetDatum(key->sk_strategy),
ObjectIdGetDatum(key->sk_subtype));
*distance_p = DatumGetFloat8(dist);
}
key++;
distance_p++;
keySize--;
}
return true;
}
/*
* find entry with lowest penalty
*/
OffsetNumber
gistchoose(Relation r, Page p, IndexTuple it, /* it has compressed entry */
GISTSTATE *giststate)
{
OffsetNumber maxoff;
OffsetNumber i;
OffsetNumber which;
float sum_grow,
which_grow[INDEX_MAX_KEYS];
GISTENTRY entry,
identry[INDEX_MAX_KEYS];
bool isnull[INDEX_MAX_KEYS];
maxoff = PageGetMaxOffsetNumber(p);
*which_grow = -1.0;
which = InvalidOffsetNumber;
sum_grow = 1;
gistDeCompressAtt(giststate, r,
it, NULL, (OffsetNumber) 0,
identry, isnull);
Assert(maxoff >= FirstOffsetNumber);
Assert(!GistPageIsLeaf(p));
for (i = FirstOffsetNumber; i <= maxoff && sum_grow; i = OffsetNumberNext(i))
{
int j;
IndexTuple itup = (IndexTuple) PageGetItem(p, PageGetItemId(p, i));
if (!GistPageIsLeaf(p) && GistTupleIsInvalid(itup))
{
ereport(LOG,
(errmsg("index \"%s\" needs VACUUM or REINDEX to finish crash recovery",
RelationGetRelationName(r))));
continue;
}
sum_grow = 0;
for (j = 0; j < r->rd_att->natts; j++)
{
Datum datum;
float usize;
bool IsNull;
datum = index_getattr(itup, j + 1, giststate->tupdesc, &IsNull);
gistdentryinit(giststate, j, &entry, datum, r, p, i,
FALSE, IsNull);
usize = gistpenalty(giststate, j, &entry, IsNull,
&identry[j], isnull[j]);
if (which_grow[j] < 0 || usize < which_grow[j])
{
which = i;
which_grow[j] = usize;
if (j < r->rd_att->natts - 1 && i == FirstOffsetNumber)
which_grow[j + 1] = -1;
sum_grow += which_grow[j];
}
else if (which_grow[j] == usize)
sum_grow += usize;
else
{
sum_grow = 1;
break;
}
}
}
if (which == InvalidOffsetNumber)
which = FirstOffsetNumber;
return which;
}
bool
gistMakeUnionItVec(GISTSTATE *giststate, IndexTuple *itvec, int len, int startkey,
Datum *attr, bool *isnull)
{
int i;
GistEntryVector *evec;
int attrsize;
evec = (GistEntryVector *) palloc((len + 2) * sizeof(GISTENTRY) + GEVHDRSZ);
for (i = startkey; i < giststate->tupdesc->natts; i++)
{
int j;
evec->n = 0;
if (!isnull[i])
{
gistentryinit(evec->vector[evec->n], attr[i],
NULL, NULL, (OffsetNumber) 0,
FALSE);
evec->n++;
}
for (j = 0; j < len; j++)
{
Datum datum;
bool IsNull;
if (GistTupleIsInvalid(itvec[j]))
return FALSE; /* signals that union with invalid tuple =>
* result is invalid */
datum = index_getattr(itvec[j], i + 1, giststate->tupdesc, &IsNull);
if (IsNull)
continue;
gistdentryinit(giststate, i,
evec->vector + evec->n,
datum,
NULL, NULL, (OffsetNumber) 0,
FALSE, IsNull);
evec->n++;
}
/* If this tuple vector was all NULLs, the union is NULL */
if (evec->n == 0)
{
attr[i] = (Datum) 0;
isnull[i] = TRUE;
}
else
{
if (evec->n == 1)
{
evec->n = 2;
evec->vector[1] = evec->vector[0];
}
/* Make union and store in attr array */
attr[i] = FunctionCall2(&giststate->unionFn[i],
PointerGetDatum(evec),
PointerGetDatum(&attrsize));
isnull[i] = FALSE;
}
}
return TRUE;
}