heaptuple.c

postgresql8.3.4源码,开源数据库

/*-------------------------------------------------------------------------
 *
 * heaptuple.c
 *	  This file contains heap tuple accessor and mutator routines, as well
 *	  as various tuple utilities.
 *
 * NOTE: there is massive duplication of code in this module to
 * support both the convention that a null is marked by a bool TRUE,
 * and the convention that a null is marked by a char 'n'.	The latter
 * convention is deprecated but it'll probably be a long time before
 * we can get rid of it entirely.
 *
 *
 * Some notes about varlenas and this code:
 *
 * Before Postgres 8.3 varlenas always had a 4-byte length header, and
 * therefore always needed 4-byte alignment (at least).  This wasted space
 * for short varlenas, for example CHAR(1) took 5 bytes and could need up to
 * 3 additional padding bytes for alignment.
 *
 * Now, a short varlena (up to 126 data bytes) is reduced to a 1-byte header
 * and we don't align it.  To hide this from datatype-specific functions that
 * don't want to deal with it, such a datum is considered "toasted" and will
 * be expanded back to the normal 4-byte-header format by pg_detoast_datum.
 * (In performance-critical code paths we can use pg_detoast_datum_packed
 * and the appropriate access macros to avoid that overhead.)  Note that this
 * conversion is performed directly in heap_form_tuple (or heap_formtuple),
 * without explicitly invoking the toaster.
 *
 * This change will break any code that assumes it needn't detoast values
 * that have been put into a tuple but never sent to disk.	Hopefully there
 * are few such places.
 *
 * Varlenas still have alignment 'i' (or 'd') in pg_type/pg_attribute, since
 * that's the normal requirement for the untoasted format.  But we ignore that
 * for the 1-byte-header format.  This means that the actual start position
 * of a varlena datum may vary depending on which format it has.  To determine
 * what is stored, we have to require that alignment padding bytes be zero.
 * (Postgres actually has always zeroed them, but now it's required!)  Since
 * the first byte of a 1-byte-header varlena can never be zero, we can examine
 * the first byte after the previous datum to tell if it's a pad byte or the
 * start of a 1-byte-header varlena.
 *
 * Note that while formerly we could rely on the first varlena column of a
 * system catalog to be at the offset suggested by the C struct for the
 * catalog, this is now risky: it's only safe if the preceding field is
 * word-aligned, so that there will never be any padding.
 *
 * We don't pack varlenas whose attstorage is 'p', since the data type
 * isn't expecting to have to detoast values.  This is used in particular
 * by oidvector and int2vector, which are used in the system catalogs
 * and we'd like to still refer to them via C struct offsets.
 *
 *
 * Portions Copyright (c) 1996-2008, PostgreSQL Global Development Group
 * Portions Copyright (c) 1994, Regents of the University of California
 *
 *
 * IDENTIFICATION
 *	  $PostgreSQL: pgsql/src/backend/access/common/heaptuple.c,v 1.120 2008/01/01 19:45:45 momjian Exp $
 *
 *-------------------------------------------------------------------------
 */

#include "postgres.h"

#include "access/heapam.h"
#include "access/tuptoaster.h"
#include "executor/tuptable.h"


/* Does att's datatype allow packing into the 1-byte-header varlena format? */
#define ATT_IS_PACKABLE(att) \
	((att)->attlen == -1 && (att)->attstorage != 'p')
/* Use this if it's already known varlena */
#define VARLENA_ATT_IS_PACKABLE(att) \
	((att)->attstorage != 'p')


/* ----------------------------------------------------------------
 *						misc support routines
 * ----------------------------------------------------------------
 */


/*
 * heap_compute_data_size
 *		Determine size of the data area of a tuple to be constructed
 */
Size
heap_compute_data_size(TupleDesc tupleDesc,
					   Datum *values,
					   bool *isnull)
{
	Size		data_length = 0;
	int			i;
	int			numberOfAttributes = tupleDesc->natts;
	Form_pg_attribute *att = tupleDesc->attrs;

	for (i = 0; i < numberOfAttributes; i++)
	{
		Datum		val;

		if (isnull[i])
			continue;

		val = values[i];

		if (ATT_IS_PACKABLE(att[i]) &&
			VARATT_CAN_MAKE_SHORT(DatumGetPointer(val)))
		{
			/*
			 * we're anticipating converting to a short varlena header, so
			 * adjust length and don't count any alignment
			 */
			data_length += VARATT_CONVERTED_SHORT_SIZE(DatumGetPointer(val));
		}
		else
		{
			data_length = att_align_datum(data_length, att[i]->attalign,
										  att[i]->attlen, val);
			data_length = att_addlength_datum(data_length, att[i]->attlen,
											  val);
		}
	}

	return data_length;
}

/* ----------------
 *		ComputeDataSize
 *
 * Determine size of the data area of a tuple to be constructed
 *
 * OLD API with char 'n'/' ' convention for indicating nulls
 * ----------------
 */
static Size
ComputeDataSize(TupleDesc tupleDesc,
				Datum *values,
				char *nulls)
{
	Size		data_length = 0;
	int			i;
	int			numberOfAttributes = tupleDesc->natts;
	Form_pg_attribute *att = tupleDesc->attrs;

	for (i = 0; i < numberOfAttributes; i++)
	{
		Datum		val;

		if (nulls[i] != ' ')
			continue;

		val = values[i];

		if (ATT_IS_PACKABLE(att[i]) &&
			VARATT_CAN_MAKE_SHORT(DatumGetPointer(val)))
		{
			/*
			 * we're anticipating converting to a short varlena header, so
			 * adjust length and don't count any alignment
			 */
			data_length += VARATT_CONVERTED_SHORT_SIZE(DatumGetPointer(val));
		}
		else
		{
			data_length = att_align_datum(data_length, att[i]->attalign,
										  att[i]->attlen, val);
			data_length = att_addlength_datum(data_length, att[i]->attlen,
											  val);
		}
	}

	return data_length;
}

/*
 * heap_fill_tuple
 *		Load data portion of a tuple from values/isnull arrays
 *
 * We also fill the null bitmap (if any) and set the infomask bits
 * that reflect the tuple's data contents.
 *
 * NOTE: it is now REQUIRED that the caller have pre-zeroed the data area.
 */
void
heap_fill_tuple(TupleDesc tupleDesc,
				Datum *values, bool *isnull,
				char *data, Size data_size,
				uint16 *infomask, bits8 *bit)
{
	bits8	   *bitP;
	int			bitmask;
	int			i;
	int			numberOfAttributes = tupleDesc->natts;
	Form_pg_attribute *att = tupleDesc->attrs;

#ifdef USE_ASSERT_CHECKING
	char	   *start = data;
#endif

	if (bit != NULL)
	{
		bitP = &bit[-1];
		bitmask = HIGHBIT;
	}
	else
	{
		/* just to keep compiler quiet */
		bitP = NULL;
		bitmask = 0;
	}

	*infomask &= ~(HEAP_HASNULL | HEAP_HASVARWIDTH | HEAP_HASEXTERNAL);

	for (i = 0; i < numberOfAttributes; i++)
	{
		Size		data_length;

		if (bit != NULL)
		{
			if (bitmask != HIGHBIT)
				bitmask <<= 1;
			else
			{
				bitP += 1;
				*bitP = 0x0;
				bitmask = 1;
			}

			if (isnull[i])
			{
				*infomask |= HEAP_HASNULL;
				continue;
			}

			*bitP |= bitmask;
		}

		/*
		 * XXX we use the att_align macros on the pointer value itself, not on
		 * an offset.  This is a bit of a hack.
		 */

		if (att[i]->attbyval)
		{
			/* pass-by-value */
			data = (char *) att_align_nominal((long) data, att[i]->attalign);
			store_att_byval(data, values[i], att[i]->attlen);
			data_length = att[i]->attlen;
		}
		else if (att[i]->attlen == -1)
		{
			/* varlena */
			Pointer		val = DatumGetPointer(values[i]);

			*infomask |= HEAP_HASVARWIDTH;
			if (VARATT_IS_EXTERNAL(val))
			{
				*infomask |= HEAP_HASEXTERNAL;
				/* no alignment, since it's short by definition */
				data_length = VARSIZE_EXTERNAL(val);
				memcpy(data, val, data_length);
			}
			else if (VARATT_IS_SHORT(val))
			{
				/* no alignment for short varlenas */
				data_length = VARSIZE_SHORT(val);
				memcpy(data, val, data_length);
			}
			else if (VARLENA_ATT_IS_PACKABLE(att[i]) &&
					 VARATT_CAN_MAKE_SHORT(val))
			{
				/* convert to short varlena -- no alignment */
				data_length = VARATT_CONVERTED_SHORT_SIZE(val);
				SET_VARSIZE_SHORT(data, data_length);
				memcpy(data + 1, VARDATA(val), data_length - 1);
			}
			else
			{
				/* full 4-byte header varlena */
				data = (char *) att_align_nominal((long) data,
												  att[i]->attalign);
				data_length = VARSIZE(val);
				memcpy(data, val, data_length);
			}
		}
		else if (att[i]->attlen == -2)
		{
			/* cstring ... never needs alignment */
			*infomask |= HEAP_HASVARWIDTH;
			Assert(att[i]->attalign == 'c');
			data_length = strlen(DatumGetCString(values[i])) + 1;
			memcpy(data, DatumGetPointer(values[i]), data_length);
		}
		else
		{
			/* fixed-length pass-by-reference */
			data = (char *) att_align_nominal((long) data, att[i]->attalign);
			Assert(att[i]->attlen > 0);
			data_length = att[i]->attlen;
			memcpy(data, DatumGetPointer(values[i]), data_length);
		}

		data += data_length;
	}

	Assert((data - start) == data_size);
}

/* ----------------
 *		DataFill
 *
 * Load data portion of a tuple from values/nulls arrays
 *
 * OLD API with char 'n'/' ' convention for indicating nulls
 * ----------------
 */
static void
DataFill(TupleDesc tupleDesc,
		 Datum *values, char *nulls,
		 char *data, Size data_size,
		 uint16 *infomask, bits8 *bit)
{
	bits8	   *bitP;
	int			bitmask;
	int			i;
	int			numberOfAttributes = tupleDesc->natts;
	Form_pg_attribute *att = tupleDesc->attrs;

#ifdef USE_ASSERT_CHECKING
	char	   *start = data;
#endif

	if (bit != NULL)
	{
		bitP = &bit[-1];
		bitmask = HIGHBIT;
	}
	else
	{
		/* just to keep compiler quiet */
		bitP = NULL;
		bitmask = 0;
	}

	*infomask &= ~(HEAP_HASNULL | HEAP_HASVARWIDTH | HEAP_HASEXTERNAL);

	for (i = 0; i < numberOfAttributes; i++)
	{
		Size		data_length;

		if (bit != NULL)
		{
			if (bitmask != HIGHBIT)
				bitmask <<= 1;
			else
			{
				bitP += 1;
				*bitP = 0x0;
				bitmask = 1;
			}

			if (nulls[i] == 'n')
			{
				*infomask |= HEAP_HASNULL;
				continue;
			}

			*bitP |= bitmask;
		}

		/*
		 * XXX we use the att_align macros on the pointer value itself, not on
		 * an offset.  This is a bit of a hack.
		 */

		if (att[i]->attbyval)
		{
			/* pass-by-value */
			data = (char *) att_align_nominal((long) data, att[i]->attalign);
			store_att_byval(data, values[i], att[i]->attlen);
			data_length = att[i]->attlen;
		}
		else if (att[i]->attlen == -1)
		{
			/* varlena */
			Pointer		val = DatumGetPointer(values[i]);

			*infomask |= HEAP_HASVARWIDTH;
			if (VARATT_IS_EXTERNAL(val))
			{
				*infomask |= HEAP_HASEXTERNAL;
				/* no alignment, since it's short by definition */
				data_length = VARSIZE_EXTERNAL(val);
				memcpy(data, val, data_length);
			}
			else if (VARATT_IS_SHORT(val))
			{
				/* no alignment for short varlenas */
				data_length = VARSIZE_SHORT(val);
				memcpy(data, val, data_length);
			}
			else if (VARLENA_ATT_IS_PACKABLE(att[i]) &&
					 VARATT_CAN_MAKE_SHORT(val))
			{
				/* convert to short varlena -- no alignment */
				data_length = VARATT_CONVERTED_SHORT_SIZE(val);
				SET_VARSIZE_SHORT(data, data_length);
				memcpy(data + 1, VARDATA(val), data_length - 1);
			}
			else
			{
				/* full 4-byte header varlena */
				data = (char *) att_align_nominal((long) data,
												  att[i]->attalign);
				data_length = VARSIZE(val);
				memcpy(data, val, data_length);
			}
		}
		else if (att[i]->attlen == -2)
		{
			/* cstring ... never needs alignment */
			*infomask |= HEAP_HASVARWIDTH;
			Assert(att[i]->attalign == 'c');
			data_length = strlen(DatumGetCString(values[i])) + 1;
			memcpy(data, DatumGetPointer(values[i]), data_length);
		}
		else
		{
			/* fixed-length pass-by-reference */
			data = (char *) att_align_nominal((long) data, att[i]->attalign);
			Assert(att[i]->attlen > 0);
			data_length = att[i]->attlen;
			memcpy(data, DatumGetPointer(values[i]), data_length);
		}

		data += data_length;
	}

	Assert((data - start) == data_size);
}

/* ----------------------------------------------------------------
 *						heap tuple interface
 * ----------------------------------------------------------------
 */

/* ----------------
 *		heap_attisnull	- returns TRUE iff tuple attribute is not present
 * ----------------
 */
bool
heap_attisnull(HeapTuple tup, int attnum)
{
	if (attnum > (int) HeapTupleHeaderGetNatts(tup->t_data))
		return true;

	if (attnum > 0)
	{
		if (HeapTupleNoNulls(tup))
			return false;
		return att_isnull(attnum - 1, tup->t_data->t_bits);
	}

	switch (attnum)
	{
		case TableOidAttributeNumber:
		case SelfItemPointerAttributeNumber:
		case ObjectIdAttributeNumber:
		case MinTransactionIdAttributeNumber:
		case MinCommandIdAttributeNumber:
		case MaxTransactionIdAttributeNumber:
		case MaxCommandIdAttributeNumber:
			/* these are never null */
			break;

		default:
			elog(ERROR, "invalid attnum: %d", attnum);
	}

	return false;
}

/* ----------------
 *		nocachegetattr
 *
 *		This only gets called from fastgetattr() macro, in cases where
 *		we can't use a cacheoffset and the value is not null.
 *
 *		This caches attribute offsets in the attribute descriptor.
 *
 *		An alternative way to speed things up would be to cache offsets
 *		with the tuple, but that seems more difficult unless you take
 *		the storage hit of actually putting those offsets into the
 *		tuple you send to disk.  Yuck.
 *
 *		This scheme will be slightly slower than that, but should
 *		perform well for queries which hit large #'s of tuples.  After
 *		you cache the offsets once, examining all the other tuples using