BWT* Load_Indexes(char *BWTINDEX,char *OCCFILE, char *SAFILE, MMPool* & mmPool)
{
int PoolSize = 524288;
MMMasterInitialize(3, 0, FALSE, NULL);
mmPool = MMPoolCreate(PoolSize);
return BWTLoad(mmPool, BWTINDEX, OCCFILE, SAFILE, NULL, NULL, NULL);//Load FM index
}
Idx2BWT * BWTLoad2BWT(const char * indexFilePrefix, const char * saFileNameExtension) {
Idx2BWT * idx2BWT = malloc(sizeof(Idx2BWT));
BWT * bwt;
BWT * rev_bwt;
HSP * hsp;
char bwtFilename[MAX_INDEX_FILENAME_LENGTH];
char bwtOccFilename[MAX_INDEX_FILENAME_LENGTH];
char saFilename[MAX_INDEX_FILENAME_LENGTH];
char rev_bwtFilename[MAX_INDEX_FILENAME_LENGTH];
char rev_bwtOccFilename[MAX_INDEX_FILENAME_LENGTH];
char packedDnaFilename[MAX_INDEX_FILENAME_LENGTH];
char annotationFilename[MAX_INDEX_FILENAME_LENGTH];
char ambiguityFilename[MAX_INDEX_FILENAME_LENGTH];
char translateFilename[MAX_INDEX_FILENAME_LENGTH];
strcpy(bwtFilename,indexFilePrefix);
strcpy(bwtOccFilename,indexFilePrefix);
strcpy(saFilename,indexFilePrefix);
strcpy(rev_bwtFilename,indexFilePrefix);
strcpy(rev_bwtOccFilename,indexFilePrefix);
strcpy(packedDnaFilename,indexFilePrefix);
strcpy(annotationFilename,indexFilePrefix);
strcpy(ambiguityFilename,indexFilePrefix);
strcpy(translateFilename,indexFilePrefix);
strcat(bwtFilename,".bwt");
strcat(bwtOccFilename,".fmv");
strcat(saFilename,saFileNameExtension);
strcat(rev_bwtFilename,".rev.bwt");
strcat(rev_bwtOccFilename,".rev.fmv");
strcat(packedDnaFilename,".pac");
strcat(annotationFilename,".ann");
strcat(ambiguityFilename,".amb");
strcat(translateFilename,".tra");
MMMasterInitialize(3, 0, FALSE, NULL);
MMPool * mmPool = MMPoolCreate(2097152);
bwt = BWTLoad(mmPool, bwtFilename, bwtOccFilename, saFilename, NULL, NULL, NULL);
rev_bwt = BWTLoad(mmPool, rev_bwtFilename, rev_bwtOccFilename, NULL, NULL, NULL, NULL);
hsp = HSPLoad(mmPool, packedDnaFilename, annotationFilename, ambiguityFilename,translateFilename, 1);
HSPFillCharMap(idx2BWT->charMap);
HSPFillComplementMap(idx2BWT->complementMap);
idx2BWT->bwt = bwt;
idx2BWT->rev_bwt = rev_bwt;
idx2BWT->hsp = hsp;
idx2BWT->mmPool = mmPool;
return idx2BWT;
}
int main(int argc, char** argv) {
char c;
unsigned int i, j, k;
MMPool *mmPool;
dictionary *programInput, *ini;
char argumentText[11] = "argument:0";
double startTime;
double elapsedTime = 0, totalElapsedTime = 0;
BWT *bwt;
HSP *hsp;
unsigned char charMap[256];
unsigned char *convertedKey;
unsigned int *packedKey;
unsigned int found;
unsigned int numberOfPattern, numberOfPatternFound;
unsigned int saIndexLeft, saIndexRight;
SaIndexGroupNew *saIndexGroup;
SaIndexList *tempSaIndex1, *tempSaIndex2;
unsigned int numberOfSaIndexGroup;
HitList *hitList;
unsigned int textPositionMatched, textPositionRetrieved;
unsigned long long totalTextPositionMatched, totalTextPositionRetrieved;
BWTSaRetrievalStatistics BWTSaRetrievalStatistics;
FILE *logFile;
init(textPositionRetrieved); // to avoid compiler warning only
init(textPositionMatched); // to avoid compiler warning only
programInput = ParseInput(argc, argv);
ini = ParseIniFile();
ProcessIni();
ValidateIni();
PrintIni();
if (Confirmation == TRUE) {
printf("BWT Search. Press Y to go or N to cancel. ");
c = (char)getchar();
while (c != 'y' && c != 'Y' && c != 'n' && c!= 'N') {
c = (char)getchar();
}
if (c == 'n' || c == 'N') {
exit(0);
}
}
startTime = setStartTime();
MMMasterInitialize(1, 0, FALSE, NULL);
mmPool = MMPoolCreate(PoolSize);
// Load Database
printf("Loading Database..");
fflush(stdout);
bwt = BWTLoad(mmPool, BWTCodeFileName, BWTOccValueFileName, SaValueFileName, NULL, SaIndexFileName, NULL);
HSPFillCharMap(charMap);
hsp = HSPLoad(mmPool, PackedDNAFileName, AnnotationFileName, AmbiguityFileName, MAX_SEARCH_PATTERN_LENGTH / CHAR_PER_WORD);
if (bwt->textLength != hsp->dnaLength) {
fprintf(stderr, "BWT-BLAST: Database length inconsistent!\n");
exit(1);
}
if (LogFileName[0] != '\0') {
logFile = fopen64(LogFileName, "w");
if (logFile == NULL) {
fprintf(stderr, "Cannot open log file!\n");
exit(1);
}
} else {
logFile = NULL;
}
packedKey = MMPoolDispatch(mmPool, WordPackedLengthFromText(MAX_SEARCH_PATTERN_LENGTH, BIT_PER_CHAR));
convertedKey = MMPoolDispatch(mmPool, MAX_SEARCH_PATTERN_LENGTH);
saIndexGroup = MMUnitAllocate(MaxNumberOfHitGroups * sizeof(SaIndexGroup));
hitList = MMUnitAllocate(MaxNumberOfTextPosition * sizeof(HitList));
tempSaIndex1 = MMUnitAllocate(MaxNumberOfTextPosition * sizeof(SaIndexList));
tempSaIndex2 = MMUnitAllocate(MaxNumberOfTextPosition * sizeof(SaIndexList));
elapsedTime = getElapsedTime(startTime) - totalElapsedTime;
printf("Elapsed time = ");
printElapsedTime(stdout, FALSE, FALSE, TRUE, 4, elapsedTime);
totalElapsedTime += elapsedTime;
printf("\n");
// Process search pattern files
for (i=1; i<=NumberOfSearchPatternFile; i++) {
argumentText[9] = '0' + (char)(i + 2);
iniparser_copystring(programInput, argumentText, PatternFileName, PatternFileName, MAX_FILENAME_LEN);
printf("Loading search pattern : %s\n", PatternFileName);
numberOfPattern = ProcessSearchPattern();
printf("Finished loading search pattern.\n");
elapsedTime = getElapsedTime(startTime) - totalElapsedTime;
printf("Elapsed time = ");
printElapsedTime(stdout, FALSE, FALSE, TRUE, 4, elapsedTime);
totalElapsedTime += elapsedTime;
printf("\n");
if (LogFileName[0] != '\0') {
fprintf(logFile, "Searching for pattern : %s\n", PatternFileName);
}
if (SABinarySearch == TRUE) {
printf("Start forward search with SA index.\n");
if (LogFileName[0] != '\0') {
fprintf(logFile, "Forward search with SA index.\n");
}
numberOfPatternFound = 0;
totalTextPositionMatched = 0;
totalTextPositionRetrieved = 0;
BWTInitializeSaRetrievalStatistics(&BWTSaRetrievalStatistics);
for (j=0; j<numberOfPattern; j++) {
//ConvertTextToWordPacked(searchPattern + searchPatternPosition[j], packedKey, charMap, ALPHABET_SIZE,
// searchPatternPosition[j+1] - searchPatternPosition[j]);
//found = BWTForwardSearchSaIndex(packedKey, searchPatternPosition[j+1] - searchPatternPosition[j],
// bwt, hsp->packedDNA, &saIndexLeft, &saIndexRight);
ConvertTextToCode(searchPattern + searchPatternPosition[j], convertedKey, charMap,
searchPatternPosition[j+1] - searchPatternPosition[j]);
found = BWTSaBinarySearch(convertedKey, searchPatternPosition[j+1] - searchPatternPosition[j],
bwt, hsp->packedDNA, &saIndexLeft, &saIndexRight, packedKey);
if (found) {
numberOfPatternFound++;
textPositionMatched = saIndexRight - saIndexLeft + 1;
totalTextPositionMatched += textPositionMatched;
if (FindTextPosition) {
saIndexGroup->startSaIndex = saIndexLeft;
if (textPositionMatched <= MaxNumberOfTextPosition) {
saIndexGroup->numOfMatch = textPositionMatched;
} else {
saIndexGroup->numOfMatch = MaxNumberOfTextPosition;
printf("Max no of text positions reached! Only %u out of %u text positions retrieved.\n", MaxNumberOfTextPosition, textPositionMatched);
}
saIndexGroup->posQuery = 0;
saIndexGroup->info = 0;
textPositionRetrieved = BWTTextPosition(bwt, saIndexGroup, 1, hitList, tempSaIndex1, tempSaIndex2, &BWTSaRetrievalStatistics, FALSE);
totalTextPositionRetrieved += textPositionRetrieved;
}
}
if (LogFileName[0] != '\0') {
if (found) {
fprintf(logFile, "Pattern number %u found. SA Index from %u to %u. ",
j + 1, saIndexLeft, saIndexRight);
if (FindTextPosition) {
fprintf(logFile, "%u matches of %u located. ", textPositionRetrieved, textPositionMatched);
for (k=0; k<textPositionRetrieved; k++) {
fprintf(logFile, "%u ", hitList[k].posText);
}
}
} else {
fprintf(logFile, "Pattern number %u not found.", j + 1);
}
fprintf(logFile, "\n");
}
}
printf("Finished forward search with SA index.\n");
printf("%u patterns out of %u found. %llu of %llu matches located.\n", numberOfPatternFound, numberOfPattern, totalTextPositionRetrieved, totalTextPositionMatched);
if (FindTextPosition) {
printf("Hits: BWT/Cached/Diag/Dup : %u/%u/%u/%u\n",
BWTSaRetrievalStatistics.bwtSaRetrieved, BWTSaRetrievalStatistics.cachedSaRetrieved,
BWTSaRetrievalStatistics.saDiagonalLinked, BWTSaRetrievalStatistics.saDuplicated);
}
elapsedTime = getElapsedTime(startTime) - totalElapsedTime;
printf("Elapsed time = ");
printElapsedTime(stdout, FALSE, FALSE, TRUE, 4, elapsedTime);
totalElapsedTime += elapsedTime;
printf("\n");
if (LogFileName[0] != '\0') {
fprintf(logFile, "Finished forward search with SA index.\n");
fprintf(logFile, "%u patterns out of %u found. %llu of %llu matches located.\n", numberOfPatternFound, numberOfPattern, totalTextPositionRetrieved, totalTextPositionMatched);
if (FindTextPosition) {
fprintf(logFile, "Hits: BWT/Cached/Diag/Dup : %u/%u/%u/%u\n",
BWTSaRetrievalStatistics.bwtSaRetrieved, BWTSaRetrievalStatistics.cachedSaRetrieved,
BWTSaRetrievalStatistics.saDiagonalLinked, BWTSaRetrievalStatistics.saDuplicated);
}
fprintf(logFile, "Elapsed time = ");
printElapsedTime(logFile, FALSE, FALSE, TRUE, 4, elapsedTime);
fprintf(logFile, "\n");
}
}
if (BackwardSearch == TRUE) {
printf("Start backward search.\n");
if (LogFileName[0] != '\0') {
fprintf(logFile, "Backward search.\n");
}
numberOfPatternFound = 0;
totalTextPositionMatched = 0;
totalTextPositionRetrieved = 0;
BWTInitializeSaRetrievalStatistics(&BWTSaRetrievalStatistics);
for (j=0; j<numberOfPattern; j++) {
ConvertTextToCode(searchPattern + searchPatternPosition[j], convertedKey, charMap,
searchPatternPosition[j+1] - searchPatternPosition[j]);
found = BWTBackwardSearch(convertedKey, searchPatternPosition[j+1] - searchPatternPosition[j],
bwt, &saIndexLeft, &saIndexRight);
if (found) {
numberOfPatternFound++;
textPositionMatched = saIndexRight - saIndexLeft + 1;
totalTextPositionMatched += textPositionMatched;
if (FindTextPosition) {
saIndexGroup->startSaIndex = saIndexLeft;
if (textPositionMatched <= MaxNumberOfTextPosition) {
saIndexGroup->numOfMatch = textPositionMatched;
} else {
saIndexGroup->numOfMatch = MaxNumberOfTextPosition;
printf("Max no of text positions reached! Only %u out of %u text positions retrieved.\n", MaxNumberOfTextPosition, textPositionMatched);
}
saIndexGroup->posQuery = 0;
saIndexGroup->info = 0;
textPositionRetrieved = BWTTextPosition(bwt, saIndexGroup, 1, hitList, tempSaIndex1, tempSaIndex2, &BWTSaRetrievalStatistics, FALSE);
totalTextPositionRetrieved += textPositionRetrieved;
}
}
if (LogFileName[0] != '\0') {
if (found) {
fprintf(logFile, "Pattern number %u found. SA Index from %u to %u. Total %u occurrences ", j + 1, saIndexLeft, saIndexRight, saIndexRight - saIndexLeft + 1);
if (FindTextPosition) {
fprintf(logFile, "%u matches of %u located. ", textPositionRetrieved, textPositionMatched);
for (k=0; k<textPositionRetrieved; k++) {
fprintf(logFile, "%u ", hitList[k].posText);
}
}
} else {
fprintf(logFile, "Pattern number %u not found.", j + 1);
}
fprintf(logFile, "\n");
}
}
printf("Finished backward search.\n");
printf("%u patterns out of %u found. %llu of %llu matches located.\n", numberOfPatternFound, numberOfPattern, totalTextPositionRetrieved, totalTextPositionMatched);
if (FindTextPosition) {
printf("Hits: BWT/Cached/Diag/Dup : %u/%u/%u/%u\n",
BWTSaRetrievalStatistics.bwtSaRetrieved, BWTSaRetrievalStatistics.cachedSaRetrieved,
BWTSaRetrievalStatistics.saDiagonalLinked, BWTSaRetrievalStatistics.saDuplicated);
}
elapsedTime = getElapsedTime(startTime) - totalElapsedTime;
printf("Elapsed time = ");
printElapsedTime(stdout, FALSE, FALSE, TRUE, 4, elapsedTime);
totalElapsedTime += elapsedTime;
printf("\n");
if (LogFileName[0] != '\0') {
fprintf(logFile, "Finished backward search.\n");
fprintf(logFile, "%u patterns out of %u found. %llu of %llu matches located.\n", numberOfPatternFound, numberOfPattern, totalTextPositionRetrieved, totalTextPositionMatched);
if (FindTextPosition) {
fprintf(logFile, "Hits: BWT/Cached/Diag/Dup : %u/%u/%u/%u\n",
BWTSaRetrievalStatistics.bwtSaRetrieved, BWTSaRetrievalStatistics.cachedSaRetrieved,
BWTSaRetrievalStatistics.saDiagonalLinked, BWTSaRetrievalStatistics.saDuplicated);
}
fprintf(logFile, "Elapsed time = ");
printElapsedTime(logFile, FALSE, FALSE, TRUE, 4, elapsedTime);
fprintf(logFile, "\n");
}
}
if (BackwardSearchCheck == TRUE) {
printf("Start backward search with text.\n");
if (LogFileName[0] != '\0') {
fprintf(logFile, "Backward search with text.\n");
}
numberOfPatternFound = 0;
totalTextPositionMatched = 0;
totalTextPositionRetrieved = 0;
BWTInitializeSaRetrievalStatistics(&BWTSaRetrievalStatistics);
for (j=0; j<numberOfPattern; j++) {
ConvertTextToCode(searchPattern + searchPatternPosition[j], convertedKey, charMap,
searchPatternPosition[j+1] - searchPatternPosition[j]);
ConvertTextToWordPacked(searchPattern + searchPatternPosition[j], packedKey, charMap, ALPHABET_SIZE,
searchPatternPosition[j+1] - searchPatternPosition[j]);
found = BWTBackwardSearchCheckWithText(convertedKey, packedKey, searchPatternPosition[j+1] - searchPatternPosition[j],
bwt, hsp->packedDNA, TextCheckCostFactor, MaxNumberOfTextPosition,
hitList, &saIndexLeft, &saIndexRight);
if (found) {
numberOfPatternFound++;
textPositionMatched = saIndexRight - saIndexLeft + 1;
totalTextPositionMatched += textPositionMatched;
// Find text position if text check not used
if (FindTextPosition && saIndexLeft != 0) {
saIndexGroup->startSaIndex = saIndexLeft;
if (textPositionMatched <= MaxNumberOfTextPosition) {
saIndexGroup->numOfMatch = textPositionMatched;
} else {
saIndexGroup->numOfMatch = MaxNumberOfTextPosition;
printf("Max no of text positions reached! Only %u out of %u text positions retrieved.\n", MaxNumberOfTextPosition, textPositionMatched);
}
saIndexGroup->posQuery = 0;
saIndexGroup->info = 0;
textPositionRetrieved = BWTTextPosition(bwt, saIndexGroup, 1, hitList, tempSaIndex1, tempSaIndex2, &BWTSaRetrievalStatistics, FALSE);
totalTextPositionRetrieved += textPositionRetrieved;
} else {
textPositionRetrieved = textPositionMatched;
totalTextPositionRetrieved += textPositionRetrieved;
}
}
if (LogFileName[0] != '\0') {
if (found) {
fprintf(logFile, "Pattern number %u found. ", j + 1);
if (FindTextPosition) {
fprintf(logFile, "%u matches of %u located. ", textPositionRetrieved, textPositionMatched);
for (k=0; k<textPositionRetrieved; k++) {
fprintf(logFile, "%u ", hitList[k].posText);
}
}
} else {
fprintf(logFile, "Pattern number %u not found.", j + 1);
}
fprintf(logFile, "\n");
}
}
printf("Finished backward search with text.\n");
printf("%u patterns out of %u found. %llu of %llu matches located.\n", numberOfPatternFound, numberOfPattern, totalTextPositionRetrieved, totalTextPositionMatched);
if (FindTextPosition) {
printf("Hits: BWT/Cached/Diag/Dup : %u/%u/%u/%u\n",
BWTSaRetrievalStatistics.bwtSaRetrieved, BWTSaRetrievalStatistics.cachedSaRetrieved,
BWTSaRetrievalStatistics.saDiagonalLinked, BWTSaRetrievalStatistics.saDuplicated);
}
elapsedTime = getElapsedTime(startTime) - totalElapsedTime;
printf("Elapsed time = ");
printElapsedTime(stdout, FALSE, FALSE, TRUE, 4, elapsedTime);
totalElapsedTime += elapsedTime;
printf("\n");
if (LogFileName[0] != '\0') {
fprintf(logFile, "Finished backward search with text.\n");
fprintf(logFile, "%u patterns out of %u found. %llu of %llu matches located.\n", numberOfPatternFound, numberOfPattern, totalTextPositionRetrieved, totalTextPositionMatched);
if (FindTextPosition) {
fprintf(logFile, "Hits: BWT/Cached/Diag/Dup : %u/%u/%u/%u\n",
BWTSaRetrievalStatistics.bwtSaRetrieved, BWTSaRetrievalStatistics.cachedSaRetrieved,
BWTSaRetrievalStatistics.saDiagonalLinked, BWTSaRetrievalStatistics.saDuplicated);
}
fprintf(logFile, "Elapsed time = ");
printElapsedTime(logFile, FALSE, FALSE, TRUE, 4, elapsedTime);
fprintf(logFile, "\n");
}
}
if (HammingDistSearch == TRUE) {
printf("Start hamming distance %u approximate match.\n", MaxErrorAllowed);
if (LogFileName[0] != '\0') {
fprintf(logFile, "Hamming distance %u approximate match.\n", MaxErrorAllowed);
}
numberOfPatternFound = 0;
totalTextPositionMatched = 0;
totalTextPositionRetrieved = 0;
BWTInitializeSaRetrievalStatistics(&BWTSaRetrievalStatistics);
for (j=0; j<numberOfPattern; j++) {
ConvertTextToCode(searchPattern + searchPatternPosition[j], convertedKey, charMap,
searchPatternPosition[j+1] - searchPatternPosition[j]);
numberOfSaIndexGroup = BWTHammingDistMatchOld(convertedKey, searchPatternPosition[j+1] - searchPatternPosition[j],
bwt, MaxErrorAllowed, saIndexGroup, MaxNumberOfHitGroups, 0, 0);
if (numberOfSaIndexGroup) {
numberOfPatternFound++;
textPositionMatched = 0;
for (k=0; k<numberOfSaIndexGroup; k++) {
textPositionMatched += saIndexGroup[k].numOfMatch;
}
if (textPositionMatched > MaxNumberOfTextPosition) {
textPositionMatched = 0;
for (k=0; k<numberOfSaIndexGroup && textPositionMatched < MaxNumberOfTextPosition; k++) {
textPositionMatched += saIndexGroup[k].numOfMatch;
}
numberOfSaIndexGroup = k - 1;
saIndexGroup[numberOfSaIndexGroup].numOfMatch -= textPositionMatched - MaxNumberOfTextPosition;
for (; k<numberOfSaIndexGroup; k++) {
textPositionMatched += saIndexGroup[k].numOfMatch;
}
printf("Max no of text positions reached! Only %u out of %u text positions retrieved.\n", MaxNumberOfTextPosition, textPositionMatched);
}
totalTextPositionMatched += textPositionMatched;
if (FindTextPosition) {
textPositionRetrieved = BWTTextPosition(bwt, saIndexGroup, numberOfSaIndexGroup, hitList, tempSaIndex1, tempSaIndex2, &BWTSaRetrievalStatistics, FALSE);
totalTextPositionRetrieved += textPositionRetrieved;
}
}
if (LogFileName[0] != '\0') {
if (numberOfSaIndexGroup) {
fprintf(logFile, "Pattern number %u found. %u matches", j + 1, textPositionMatched);
if (FindTextPosition) {
fprintf(logFile, "%u matches of %u located. ", textPositionRetrieved, textPositionMatched);
for (k=0; k<textPositionRetrieved; k++) {
fprintf(logFile, "%u ", hitList[k].posText);
}
}
} else {
fprintf(logFile, "Pattern number %u not found.", j + 1);
}
fprintf(logFile, "\n");
}
}
printf("Finished hamming distance search.\n");
printf("%u patterns out of %u found. %llu of %llu matches located.\n", numberOfPatternFound, numberOfPattern, totalTextPositionRetrieved, totalTextPositionMatched);
if (FindTextPosition) {
printf("Hits: BWT/Cached/Diag/Dup : %u/%u/%u/%u\n",
BWTSaRetrievalStatistics.bwtSaRetrieved, BWTSaRetrievalStatistics.cachedSaRetrieved,
BWTSaRetrievalStatistics.saDiagonalLinked, BWTSaRetrievalStatistics.saDuplicated);
}
elapsedTime = getElapsedTime(startTime) - totalElapsedTime;
printf("Elapsed time = ");
printElapsedTime(stdout, FALSE, FALSE, TRUE, 4, elapsedTime);
totalElapsedTime += elapsedTime;
printf("\n");
if (LogFileName[0] != '\0') {
fprintf(logFile, "Finished hamming distance search.\n");
fprintf(logFile, "%u patterns out of %u found. %llu of %llu matches located.\n", numberOfPatternFound, numberOfPattern, totalTextPositionRetrieved, totalTextPositionMatched);
if (FindTextPosition) {
fprintf(logFile, "Hits: BWT/Cached/Diag/Dup : %u/%u/%u/%u\n",
BWTSaRetrievalStatistics.bwtSaRetrieved, BWTSaRetrievalStatistics.cachedSaRetrieved,
BWTSaRetrievalStatistics.saDiagonalLinked, BWTSaRetrievalStatistics.saDuplicated);
}
fprintf(logFile, "Elapsed time = ");
printElapsedTime(logFile, FALSE, FALSE, TRUE, 4, elapsedTime);
fprintf(logFile, "\n");
}
}
if (EditDistSearch == TRUE) {
printf("Start edit distance %u approximate match.\n", MaxErrorAllowed);
if (LogFileName[0] != '\0') {
fprintf(logFile, "Edit distance %u approximate match.\n", MaxErrorAllowed);
}
numberOfPatternFound = 0;
totalTextPositionMatched = 0;
totalTextPositionRetrieved = 0;
BWTInitializeSaRetrievalStatistics(&BWTSaRetrievalStatistics);
for (j=0; j<numberOfPattern; j++) {
ConvertTextToCode(searchPattern + searchPatternPosition[j], convertedKey, charMap,
searchPatternPosition[j+1] - searchPatternPosition[j]);
numberOfSaIndexGroup = BWTEditDistMatchOld(convertedKey, searchPatternPosition[j+1] - searchPatternPosition[j],
bwt, MaxErrorAllowed, (SaIndexGroupWithLengthError*)saIndexGroup, MaxNumberOfHitGroups);
if (numberOfSaIndexGroup > MaxNumberOfHitGroups) {
fprintf(stderr, "numberOfSaIndexGroup > MaxNumberOfHitGroups!\n");
}
if (numberOfSaIndexGroup) {
numberOfPatternFound++;
textPositionMatched = 0;
for (k=0; k<numberOfSaIndexGroup; k++) {
textPositionMatched += saIndexGroup[k].numOfMatch;
}
if (textPositionMatched > MaxNumberOfTextPosition) {
textPositionMatched = 0;
for (k=0; k<numberOfSaIndexGroup && textPositionMatched < MaxNumberOfTextPosition; k++) {
textPositionMatched += saIndexGroup[k].numOfMatch;
}
numberOfSaIndexGroup = k - 1;
saIndexGroup[numberOfSaIndexGroup].numOfMatch -= textPositionMatched - MaxNumberOfTextPosition;
for (; k<numberOfSaIndexGroup; k++) {
textPositionMatched += saIndexGroup[k].numOfMatch;
}
printf("Max no of text positions reached! Only %u out of %u text positions retrieved.\n", MaxNumberOfTextPosition, textPositionMatched);
}
totalTextPositionMatched += textPositionMatched;
if (FindTextPosition) {
textPositionRetrieved = BWTTextPosition(bwt, saIndexGroup, numberOfSaIndexGroup, hitList, tempSaIndex1, tempSaIndex2, &BWTSaRetrievalStatistics, FALSE);
totalTextPositionRetrieved += textPositionRetrieved;
}
}
if (LogFileName[0] != '\0') {
if (numberOfSaIndexGroup) {
fprintf(logFile, "Pattern number %u found. %u matches. ", j + 1, textPositionMatched);
if (FindTextPosition) {
fprintf(logFile, "%u matches of %u located. ", textPositionRetrieved, textPositionMatched);
for (k=0; k<textPositionRetrieved; k++) {
fprintf(logFile, "%u ", hitList[k].posText);
}
}
} else {
fprintf(logFile, "Pattern number %u not found.", j + 1);
}
fprintf(logFile, "\n");
}
}
printf("Finished edit distance search.\n");
printf("%u patterns out of %u found. %llu of %llu matches located.\n", numberOfPatternFound, numberOfPattern, totalTextPositionRetrieved, totalTextPositionMatched);
if (FindTextPosition) {
printf("Hits: BWT/Cached/Diag/Dup : %u/%u/%u/%u\n",
BWTSaRetrievalStatistics.bwtSaRetrieved, BWTSaRetrievalStatistics.cachedSaRetrieved,
BWTSaRetrievalStatistics.saDiagonalLinked, BWTSaRetrievalStatistics.saDuplicated);
}
elapsedTime = getElapsedTime(startTime) - totalElapsedTime;
printf("Elapsed time = ");
printElapsedTime(stdout, FALSE, FALSE, TRUE, 4, elapsedTime);
totalElapsedTime += elapsedTime;
printf("\n");
if (LogFileName[0] != '\0') {
fprintf(logFile, "Finished edit distance search.\n");
fprintf(logFile, "%u patterns out of %u found. %llu of %llu matches located.\n", numberOfPatternFound, numberOfPattern, totalTextPositionRetrieved, totalTextPositionMatched);
if (FindTextPosition) {
fprintf(logFile, "Hits: BWT/Cached/Diag/Dup : %u/%u/%u/%u\n",
BWTSaRetrievalStatistics.bwtSaRetrieved, BWTSaRetrievalStatistics.cachedSaRetrieved,
BWTSaRetrievalStatistics.saDiagonalLinked, BWTSaRetrievalStatistics.saDuplicated);
}
fprintf(logFile, "Elapsed time = ");
printElapsedTime(logFile, FALSE, FALSE, TRUE, 4, elapsedTime);
fprintf(logFile, "\n");
}
}
}
MMPoolReturn(mmPool, packedKey, WordPackedLengthFromText(MAX_SEARCH_PATTERN_LENGTH, BIT_PER_CHAR));
MMPoolReturn(mmPool, convertedKey, MAX_SEARCH_PATTERN_LENGTH);
HSPFree(mmPool, hsp, MAX_SEARCH_PATTERN_LENGTH / CHAR_PER_WORD);
BWTFree(mmPool, bwt);
if (searchPattern != NULL) {
MMUnitFree(searchPattern, searchPatternAllocated);
}
if (searchPatternPosition != NULL) {
MMUnitFree(searchPatternPosition, searchPatternPositionAllocated);
}
MMUnitFree(saIndexGroup, MaxNumberOfHitGroups * sizeof(unsigned int));
MMUnitFree(hitList, MaxNumberOfTextPosition * sizeof(unsigned int));
MMPoolFree(mmPool);
iniparser_freedict(programInput);
iniparser_freedict(ini);
return 0;
}