/*
* setNewRelfilenode - assign a new relfilenode value to the relation
*
* Caller must already hold exclusive lock on the relation.
*/
void
setNewRelfilenode(Relation relation)
{
Oid newrelfilenode;
Relation pg_class;
HeapTuple tuple;
Form_pg_class rd_rel;
RelationData workrel;
/* Can't change relfilenode for nailed tables (indexes ok though) */
Assert(!relation->rd_isnailed ||
relation->rd_rel->relkind == RELKIND_INDEX);
/* Can't change for shared tables or indexes */
Assert(!relation->rd_rel->relisshared);
/* Allocate a new relfilenode */
newrelfilenode = newoid();
/*
* Find the pg_class tuple for the given relation. This is not used
* during bootstrap, so okay to use heap_update always.
*/
pg_class = heap_openr(RelationRelationName, RowExclusiveLock);
tuple = SearchSysCacheCopy(RELOID,
ObjectIdGetDatum(RelationGetRelid(relation)),
0, 0, 0);
if (!HeapTupleIsValid(tuple))
elog(ERROR, "could not find tuple for relation %u",
RelationGetRelid(relation));
rd_rel = (Form_pg_class) GETSTRUCT(tuple);
/* create another storage file. Is it a little ugly ? */
/* NOTE: any conflict in relfilenode value will be caught here */
memcpy((char *) &workrel, relation, sizeof(RelationData));
workrel.rd_fd = -1;
workrel.rd_node.relNode = newrelfilenode;
heap_storage_create(&workrel);
smgrclose(DEFAULT_SMGR, &workrel);
/* schedule unlinking old relfilenode */
smgrunlink(DEFAULT_SMGR, relation);
/* update the pg_class row */
rd_rel->relfilenode = newrelfilenode;
simple_heap_update(pg_class, &tuple->t_self, tuple);
CatalogUpdateIndexes(pg_class, tuple);
heap_freetuple(tuple);
heap_close(pg_class, RowExclusiveLock);
/* Make sure the relfilenode change is visible */
CommandCounterIncrement();
}
/*
* inv_create -- create a new large object.
*
* Arguments:
* flags
*
* Returns:
* large object descriptor, appropriately filled in.
*/
LargeObjectDesc *
inv_create(int flags)
{
Oid file_oid;
LargeObjectDesc *retval;
/*
* Allocate an OID to be the LO's identifier.
*/
file_oid = newoid();
/* Check for duplicate (shouldn't happen) */
if (LargeObjectExists(file_oid))
elog(ERROR, "large object %u already exists", file_oid);
/*
* Create the LO by writing an empty first page for it in
* pg_largeobject
*/
LargeObjectCreate(file_oid);
/*
* Advance command counter so that new tuple will be seen by later
* large-object operations in this transaction.
*/
CommandCounterIncrement();
/*
* Prepare LargeObjectDesc data structure for accessing LO
*/
retval = (LargeObjectDesc *) palloc(sizeof(LargeObjectDesc));
retval->id = file_oid;
retval->subid = GetCurrentSubTransactionId();
retval->offset = 0;
if (flags & INV_WRITE)
retval->flags = IFS_WRLOCK | IFS_RDLOCK;
else if (flags & INV_READ)
retval->flags = IFS_RDLOCK;
else
elog(ERROR, "invalid flags: %d", flags);
return retval;
}
/* ----------
* 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);
}