/* Licensed Materials - Property of IBM DB2 Storage Engine Enablement Copyright IBM Corporation 2007,2008 All rights reserved Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: (a) Redistributions of source code must retain this list of conditions, the copyright notice in section {d} below, and the disclaimer following this list of conditions. (b) Redistributions in binary form must reproduce this list of conditions, the copyright notice in section (d) below, and the disclaimer following this list of conditions, in the documentation and/or other materials provided with the distribution. (c) The name of IBM may not be used to endorse or promote products derived from this software without specific prior written permission. (d) The text of the required copyright notice is: Licensed Materials - Property of IBM DB2 Storage Engine Enablement Copyright IBM Corporation 2007,2008 All rights reserved THIS SOFTWARE IS PROVIDED BY IBM CORPORATION "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL IBM CORPORATION BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ #include "ha_ibmdb2i.h" /* Helper function for records_in_range. Input: Bitmap of used key parts. Output: Number of used key parts. */ static inline int getKeyCntFromMap(key_part_map keypart_map) { int cnt = 0; while (keypart_map) { keypart_map = keypart_map >> 1; cnt++; } return (cnt); } /** @brief Given a starting key and an ending key, estimate the number of rows that will exist between the two keys. INPUT inx Index to use min_key Min key. Is NULL if no min range max_key Max key. Is NULL if no max range NOTES min_key.flag can have one of the following values: HA_READ_KEY_EXACT Include the key in the range HA_READ_AFTER_KEY Don't include key in range max_key.flag can have one of the following values: HA_READ_BEFORE_KEY Don't include key in range HA_READ_AFTER_KEY Include all 'end_key' values in the range RETURN HA_POS_ERROR Error or the storage engine cannot estimate the number of rows 1 There are no matching keys in the given range n > 0 There are approximately n rows in the range */ ha_rows ha_ibmdb2i::records_in_range(uint inx, key_range *min_key, key_range *max_key) { DBUG_ENTER("ha_ibmdb2i::records_in_range"); int rc = 0; // Return code ha_rows rows = 0; // Row count returned to caller of this method uint32 spcLen; // Length of space passed to DB2 uint32 keyCnt; // Number of fields in the key composite uint32 literalCnt = 0; // Number of literals uint32 boundsOff; // Offset from beginning of space to range bounds uint32 litDefOff; // Offset from beginning of space to literal definitions uint32 literalsOff; // Offset from beginning of space to literal values uint32 cutoff = 0; // Early exit cutoff (currently not used) uint64 recCnt; // Row count from DB2 uint16 rtnCode; // Return code from DB2 Bounds* boundsPtr; // Pointer to a pair of range bounds Bound* boundPtr; // Pointer to a single (high or low) range bound LitDef* litDefPtr; // Pointer to a literal definition char* literalsPtr; // Pointer to the start of all literal values char* literalPtr; // Pointer to the start of this literal value char* tempPtr; // Temporary pointer char* tempMinPtr; // Temporary pointer into min_key int minKeyCnt = 0; // Number of fields in the min_key composite int maxKeyCnt = 0; // Number of fields in the max_key composite size_t tempLen = 0; // Temporary length uint16 DB2FieldWidth = 0; // DB2 field width uint32 workFieldLen = 0; // Length of workarea needed for CCSID conversions bool overrideInclusion; // Indicator for inclusion/exclusion char* endOfLiteralPtr; // Pointer to the end of this literal char* endOfMinPtr; // Pointer to end of min_key uint16 endByte = 0; // End byte of char or graphic literal (padding not included) bool reuseLiteral; // Indicator that hi and lo bounds use same literal char* minPtr = NULL; // Work pointer for traversing min_key char* maxPtr = NULL; // Work pointer for traversing max_key /* Handle the special case of 'x < null' anywhere in the key range. There are no values less than null, but return 1 so that MySQL does not assume the empty set for the query. */ if (min_key != NULL && max_key != NULL && min_key->flag == HA_READ_AFTER_KEY && max_key->flag == HA_READ_BEFORE_KEY && min_key->length == max_key->length && (memcmp((uchar*)min_key->key,(uchar*)max_key->key,min_key->length)==0)) { DBUG_PRINT("ha_ibmdb2i::records_in_range",("Estimate 1 row for key %d; special case: < null", inx)); DBUG_RETURN((ha_rows) 1 ); } /* Determine the number of fields in the key composite. */ if (min_key) { minKeyCnt = getKeyCntFromMap(min_key->keypart_map); minPtr = (char*)min_key->key; } if (max_key) { maxKeyCnt = getKeyCntFromMap(max_key->keypart_map); maxPtr = (char*)max_key->key; } keyCnt = maxKeyCnt >= minKeyCnt ? maxKeyCnt : minKeyCnt; /* Handle the special case where MySQL does not pass either a min or max key range. In this case, set the key count to 1 (knowing that there is at least one key field) to flow through and create one bounds structure. When both the min and max key ranges are nil, the bounds structure will specify positive and negative infinity and DB2 will estimate the total number of rows. */ if (keyCnt == 0) keyCnt = 1; /* Allocate the space needed to pass range information to DB2. The space must be large enough to store the following: - one pair of bounds (high and low) per field in the key composite - one literal definition per literal value - the literal values - work area for literal CCSID conversions Since we don't know yet how many of these structures are needed, allocate enough space for the maximum that we will possibly need. The workarea for the literal conversion must be big enough to hold the largest of the DB2 key fields. */ KEY& curKey = table->key_info[inx]; for (int i = 0; i < keyCnt; i++) { DB2FieldWidth = db2Table->db2Field(curKey.key_part[i].field->field_index).getByteLengthInRecord(); if (DB2FieldWidth > workFieldLen) workFieldLen = DB2FieldWidth; // Get length of largest DB2 field tempLen = tempLen + DB2FieldWidth; // Tally the DB2 field lengths } spcLen = (sizeof(Bounds)*keyCnt) + (sizeof(LitDef)*keyCnt*2) + (tempLen*2) + workFieldLen; ValidatedPointer spcPtr(spcLen); // Pointer to space passed to DB2 memset(spcPtr, 0, spcLen); // Clear the allocated space /* Set addressability to the various sections of the DB2 interface space. */ boundsOff = 0; // Range bounds are at the start of the space litDefOff = sizeof(Bounds) * keyCnt; // Literal defs follow all the range bounds literalsOff = litDefOff + (sizeof(LitDef) * keyCnt * 2); // Literal values are last boundsPtr = (Bounds_t*)(void*)spcPtr; // Address first bounds structure tempPtr = (char*)((char*)spcPtr + litDefOff); litDefPtr = (LitDef_t*)tempPtr; // Address first literal definition tempPtr = (char*)((char*)spcPtr + literalsOff); literalsPtr = (char*)tempPtr; // Address start of literal values literalPtr = literalsPtr; // Address first literal value /* For each key part, build the low (min) and high (max) DB2 range bounds. If literals are specified in the MySQL range, build DB2 literal definitions and store the literal values for access by DB2. If no value is specified for a key part, assume infinity. Negative infinity will cause processing to start at the first index entry. Positive infinity will cause processing to end at the last index entry. When infinity is specified in a bound, inclusion/exclusion and position are ignored, and there is no literal definition or literal value for the bound. If the keypart value is null, the null indicator is set in the range bound and the other fields in the bound are ignored. When the bound is null, only index entries with the null value will be included in the estimate. If one bound is null, both bounds must be null. When the bound is not null, the data offset and length must be set, and the literal value stored for access by DB2. */ for (int partsInUse = 0; partsInUse < keyCnt; ++partsInUse) { Field *field= curKey.key_part[partsInUse].field; overrideInclusion = false; reuseLiteral = false; endOfLiteralPtr = NULL; /* Build the low bound for the key range. */ if ((partsInUse + 1) > minKeyCnt) // if no min_key info for this part boundsPtr->LoBound.Infinity[0] = QMY_NEG_INFINITY; // select...where 3 between x and y else { if ((curKey.key_part[partsInUse].null_bit) && (char*)minPtr[0]) { // min_key is null if (max_key == NULL || ((partsInUse + 1) > maxKeyCnt)) // select...where x='ab' and y=null and z != 'c' boundsPtr->LoBound.Infinity[0] = QMY_NEG_INFINITY; // select...where x not null or // select...where x > null else // max_key is not null { if (min_key->flag == HA_READ_KEY_EXACT) boundsPtr->LoBound.IsNull[0] = QMY_YES; // select...where x is null else { if ((char*)maxPtr[0]) boundsPtr->LoBound.IsNull[0] = QMY_YES; // select...where a = null and b < 5 (max-before) // select...where a='a' and b is null and c !='a' (max-after) else boundsPtr->LoBound.Infinity[0] = QMY_NEG_INFINITY; // select...where x < y } } // end min_key is null } else // min_key is not null { if (literalCnt) litDefPtr = litDefPtr + 1; literalCnt = literalCnt + 1; boundsPtr->LoBound.Position = literalCnt; /* Determine inclusion or exclusion. */ if (min_key->flag == HA_READ_KEY_EXACT || //select...where a like 'this%' /* An example for the following conditions is 'select...where a = 5 and b > null'. */ (max_key && (memcmp((uchar*)minPtr,(uchar*)maxPtr, curKey.key_part[partsInUse].store_length)==0))) { if ((min_key->flag != HA_READ_KEY_EXACT) || (max_key && (memcmp((uchar*)minPtr,(uchar*)maxPtr, curKey.key_part[partsInUse].store_length)==0))) overrideInclusion = true; // Need inclusion for both min and max } else boundsPtr->LoBound.Embodiment[0] = QMY_EXCLUSION; litDefPtr->FieldNbr = field->field_index + 1; DB2Field& db2Field = db2Table->db2Field(field->field_index); litDefPtr->DataType = db2Field.getType(); /* Convert the literal to DB2 format */ if ((field->type() != MYSQL_TYPE_BIT) && // Don't do conversion on BIT data (field->charset() != &my_charset_bin) && // Don't do conversion on BINARY data (litDefPtr->DataType == QMY_CHAR || litDefPtr->DataType == QMY_VARCHAR || litDefPtr->DataType == QMY_GRAPHIC || litDefPtr->DataType == QMY_VARGRAPHIC)) { // Most of the code is required by the considerable wrangling needed // to prepare partial keys for use by DB2 // 1. UTF8 (CCSID 1208) data can be copied across unmodified if it is // utf8_bin. Otherwise, we need to convert the min and max // characters into the min and max characters employed // by the DB2 sort sequence. This is complicated by the fact that // the character widths are not always equal. // 2. Likewise, UCS2 (CCSID 13488) data can be copied across unmodified // if it is ucs2_bin or ucs2_general_ci. Otherwise, we need to // convert the min and max characters into the min and max characters // employed by the DB2 sort sequence. // 3. All other data will use standard iconv conversions. If an // unconvertible character is encountered, we assume it is the min // char and fill the remainder of the DB2 key with 0s. This may not // always be accurate, but it is probably sufficient for range // estimations. const char* keyData = minPtr+((curKey.key_part[partsInUse].null_bit)? 1 : 0); char* db2Data = literalPtr; uint16 outLen = db2Field.getByteLengthInRecord(); uint16 inLen; if (litDefPtr->DataType == QMY_VARCHAR || litDefPtr->DataType == QMY_VARGRAPHIC) { inLen = *(uint8*)keyData + ((*(uint8*)(keyData+1)) << 8); keyData += 2; outLen -= sizeof(uint16); db2Data += sizeof(uint16); } else { inLen = field->max_display_length(); } size_t convertedBytes = 0; if (db2Field.getCCSID() == 1208) { DBUG_ASSERT(inLen <= outLen); if (strcmp(field->charset()->name, "utf8_bin")) { const char* end = keyData+inLen; const char* curKey = keyData; char* curDB2 = db2Data; uint32 min = field->charset()->min_sort_char; while ((curKey < end) && (curDB2 < db2Data+outLen-3)) { my_wc_t temp; int len = field->charset()->cset->mb_wc(field->charset(), &temp, (const uchar*)curKey, (const uchar*)end); if (temp != min) { DBUG_ASSERT(len <= 3); switch (len) { case 3: *(curDB2+2) = *(curKey+2); case 2: *(curDB2+1) = *(curKey+1); case 1: *(curDB2) = *(curKey); } curDB2 += len; } else { *(curDB2++) = 0xEF; *(curDB2++) = 0xBF; *(curDB2++) = 0xBF; } curKey += len; } convertedBytes = curDB2 - db2Data; } else { memcpy(db2Data, keyData, inLen); convertedBytes = inLen; } rc = 0; } else if (db2Field.getCCSID() == 13488) { DBUG_ASSERT(inLen <= outLen); if (strcmp(field->charset()->name, "ucs2_bin") && strcmp(field->charset()->name, "ucs2_general_ci")) { const char* end = keyData+inLen; const uint16* curKey = (uint16*)keyData; uint16* curDB2 = (uint16*)db2Data; uint16 min = field->charset()->min_sort_char; while (curKey < (uint16*)end) { if (*curKey != min) *curDB2 = *curKey; else *curDB2 = 0xFFFF; ++curKey; ++curDB2; } } else { memcpy(db2Data, keyData, inLen); } convertedBytes = inLen; rc = 0; } else { rc = convertFieldChars(toDB2, field->field_index, keyData, db2Data, inLen, outLen, &convertedBytes, true); if (rc == DB2I_ERR_ILL_CHAR) { // If an illegal character is encountered, we fill the remainder // of the key with 0x00. This was implemented as a corollary to // Bug#45012, though it should probably remain even after that // bug is fixed. memset(db2Data+convertedBytes, 0x00, outLen-convertedBytes); convertedBytes = outLen; rc = 0; } } if (!rc && (litDefPtr->DataType == QMY_VARGRAPHIC || litDefPtr->DataType == QMY_VARCHAR)) { *(uint16*)(db2Data-sizeof(uint16)) = convertedBytes / (litDefPtr->DataType == QMY_VARGRAPHIC ? 2 : 1); } } else // Non-character fields { rc = convertMySQLtoDB2(field, db2Field, literalPtr, (uchar*)minPtr+((curKey.key_part[partsInUse].null_bit)? 1 : 0)); } if (rc != 0) break; litDefPtr->Offset = (uint32_t)(literalPtr - literalsPtr); litDefPtr->Length = db2Field.getByteLengthInRecord(); literalPtr = literalPtr + litDefPtr->Length; // Bump pointer for next literal } /* If there is a max_key value for this field, and if the max_key value is the same as the min_key value, then the low bound literal can be reused for the high bound literal. This eliminates the overhead of copying and converting the same value twice. */ if (max_key && ((partsInUse + 1) <= maxKeyCnt) && (memcmp((uchar*)minPtr,(uchar*)maxPtr, curKey.key_part[partsInUse].store_length)==0 || endOfLiteralPtr)) reuseLiteral = true; minPtr += curKey.key_part[partsInUse].store_length; } /* Build the high bound for the key range. */ if (max_key == NULL || ((partsInUse + 1) > maxKeyCnt)) boundsPtr->HiBound.Infinity[0] = QMY_POS_INFINITY; else { if ((curKey.key_part[partsInUse].null_bit) && (char*)maxPtr[0]) { if (min_key == NULL) boundsPtr->HiBound.Infinity[0] = QMY_POS_INFINITY; else boundsPtr->HiBound.IsNull[0] = QMY_YES; // select...where x is null } else // max_key field is not null { if (boundsPtr->LoBound.IsNull[0] == QMY_YES) // select where x < 10 or x is null { rc = HA_POS_ERROR; break; } if (!reuseLiteral) { if (literalCnt) litDefPtr = litDefPtr + 1; literalCnt = literalCnt + 1; litDefPtr->FieldNbr = field->field_index + 1; DB2Field& db2Field = db2Table->db2Field(field->field_index); litDefPtr->DataType = db2Field.getType(); /* Convert the literal to DB2 format */ if ((field->type() != MYSQL_TYPE_BIT) && // Don't do conversion on BIT data (field->charset() != &my_charset_bin) && // Don't do conversion on BINARY data (litDefPtr->DataType == QMY_CHAR || litDefPtr->DataType == QMY_VARCHAR || litDefPtr->DataType == QMY_GRAPHIC || litDefPtr->DataType == QMY_VARGRAPHIC)) { // We need to handle char fields in a special way in order to account // for partial keys. Refer to the note above for a description of the // basic design. char* keyData = maxPtr+((curKey.key_part[partsInUse].null_bit)? 1 : 0); char* db2Data = literalPtr; uint16 outLen = db2Field.getByteLengthInRecord(); uint16 inLen; if (litDefPtr->DataType == QMY_VARCHAR || litDefPtr->DataType == QMY_VARGRAPHIC) { inLen = *(uint8*)keyData + ((*(uint8*)(keyData+1)) << 8); keyData += 2; outLen -= sizeof(uint16); db2Data += sizeof(uint16); } else { inLen = field->max_display_length(); } size_t convertedBytes; if (db2Field.getCCSID() == 1208) { if (strcmp(field->charset()->name, "utf8_bin")) { const char* end = keyData+inLen; const char* curKey = keyData; char* curDB2 = db2Data; uint32 max = field->charset()->max_sort_char; while (curKey < end && (curDB2 < db2Data+outLen-3)) { my_wc_t temp; int len = field->charset()->cset->mb_wc(field->charset(), &temp, (const uchar*)curKey, (const uchar*)end); if (temp != max) { DBUG_ASSERT(len <= 3); switch (len) { case 3: *(curDB2+2) = *(curKey+2); case 2: *(curDB2+1) = *(curKey+1); case 1: *(curDB2) = *(curKey); } curDB2 += len; } else { *(curDB2++) = 0xE4; *(curDB2++) = 0xB6; *(curDB2++) = 0xBF; } curKey += len; } convertedBytes = curDB2 - db2Data; } else { DBUG_ASSERT(inLen <= outLen); memcpy(db2Data, keyData, inLen); convertedBytes = inLen; } rc = 0; } else if (db2Field.getCCSID() == 13488) { if (strcmp(field->charset()->name, "ucs2_bin") && strcmp(field->charset()->name, "ucs2_general_ci")) { char* end = keyData+inLen; uint16* curKey = (uint16*)keyData; uint16* curDB2 = (uint16*)db2Data; uint16 max = field->charset()->max_sort_char; while (curKey < (uint16*)end) { if (*curKey != max) *curDB2 = *curKey; else *curDB2 = 0x4DBF; ++curKey; ++curDB2; } } else { memcpy(db2Data, keyData, outLen); } rc = 0; } else { size_t substituteChars = 0; rc = convertFieldChars(toDB2, field->field_index, keyData, db2Data, inLen, outLen, &convertedBytes, true, &substituteChars); if (rc == DB2I_ERR_ILL_CHAR) { // If an illegal character is encountered, we fill the remainder // of the key with 0xFF. This was implemented to work around // Bug#45012, though it should probably remain even after that // bug is fixed. memset(db2Data+convertedBytes, 0xFF, outLen-convertedBytes); rc = 0; } else if ((substituteChars && (litDefPtr->DataType == QMY_VARCHAR || litDefPtr->DataType == QMY_CHAR)) || strcmp(field->charset()->name, "cp1251_bulgarian_ci") == 0) { // When iconv translates the max_sort_char with a substitute // character, we have no way to know whether this affects // the sort order of the key. Therefore, to be safe, when // we know that substitute characters have been used in a // single-byte string, we traverse the translated key // in reverse, replacing substitue characters with 0xFF, which // always sorts with the greatest weight in DB2 sort sequences. // cp1251_bulgarian_ci is also handled this way because the // max_sort_char is a control character which does not sort // equivalently in DB2. DBUG_ASSERT(inLen == outLen); char* tmpKey = keyData + inLen - 1; char* tmpDB2 = db2Data + outLen - 1; while (*tmpKey == field->charset()->max_sort_char && *tmpDB2 != 0xFF) { *tmpDB2 = 0xFF; --tmpKey; --tmpDB2; } } } if (!rc && (litDefPtr->DataType == QMY_VARGRAPHIC || litDefPtr->DataType == QMY_VARCHAR)) { *(uint16*)(db2Data-sizeof(uint16)) = outLen / (litDefPtr->DataType == QMY_VARGRAPHIC ? 2 : 1); } } else { rc = convertMySQLtoDB2(field, db2Field, literalPtr, (uchar*)maxPtr+((curKey.key_part[partsInUse].null_bit)? 1 : 0)); } if (rc != 0) break; litDefPtr->Offset = (uint32_t)(literalPtr - literalsPtr); litDefPtr->Length = db2Field.getByteLengthInRecord(); literalPtr = literalPtr + litDefPtr->Length; // Bump pointer for next literal } boundsPtr->HiBound.Position = literalCnt; if (max_key->flag == HA_READ_BEFORE_KEY && !overrideInclusion) boundsPtr->HiBound.Embodiment[0] = QMY_EXCLUSION; } maxPtr += curKey.key_part[partsInUse].store_length; } /* Bump to the next field in the key composite. */ if ((partsInUse+1) < keyCnt) boundsPtr = boundsPtr + 1; } /* Call DB2 to estimate the number of rows in the key range. */ if (rc == 0) { rc = db2i_ileBridge::getBridgeForThread()->recordsInRange((indexHandles[inx] ? indexHandles[inx] : db2Table->indexFile(inx)->getMasterDefnHandle()), spcPtr, keyCnt, literalCnt, boundsOff, litDefOff, literalsOff, cutoff, (uint32_t)(literalPtr - (char*)spcPtr), endByte, &recCnt, &rtnCode); } /* Set the row count and return. Beware that if this method returns a zero row count, MySQL assumes the result set for the query is zero; never return a zero row count. */ if ((rc == 0) && (rtnCode == QMY_SUCCESS || rtnCode == QMY_EARLY_EXIT)) { rows = recCnt ? (ha_rows)recCnt : 1; } rows = (rows > 0 ? rows : HA_POS_ERROR); setIndexReadEstimate(inx, rows); DBUG_PRINT("ha_ibmdb2i::recordsInRange",("Estimate %d rows for key %d", uint32(rows), inx)); DBUG_RETURN(rows); }