/*!
* Event-handler for button "Add".
*/
void
AP_UnixDialog_Spell::onAddClicked ()
{
UT_DEBUGMSG (("ROB: AP_UnixDialog_Spell::onAddClicked()\n"));
addToDict();
ignoreWord();
}
int
main(int argc, char **argv) {
uint32 pNum = 0;
uint32 pAlloc = 8388608;
uint32 estID = ~uint32ZERO;
bool *found = 0L;
// From fixPolishesIID.c
IIDdict = 0L;
SEQdict = 0L;
GENdict = 0L;
// Incorporated from sortPolishes
mergeFilesLen = 0;
mergeFilesMax = sysconf(_SC_OPEN_MAX);
mergeFiles = new FILE * [mergeFilesMax];
mergeNames = new char * [mergeFilesMax];
mergePolishes = new sim4polish * [mergeFilesMax];
// Default to printing stats on stdout.
sFile = stdout;
int arg = 1;
while (arg < argc) {
if (strcmp(argv[arg], "-n") == 0) {
pAlloc = strtouint32(argv[++arg], 0L);
} else if (strcmp(argv[arg], "-fpart") == 0) {
arg++;
fprintf(stderr, "reading query deflines from '%s'\n", argv[arg]);
IIDdict = dict_create(DICTCOUNT_T_MAX, headerCompare);
addToDict(IIDdict, argv[arg]);
} else if (strcmp(argv[arg], "-g") == 0) {
++arg;
fprintf(stderr, "reading genomic deflines from '%s'\n", argv[arg]);
GENdict = dict_create(DICTCOUNT_T_MAX, headerCompare);
addToDict(GENdict, argv[arg]);
} else if (strcmp(argv[arg], "-F") == 0) {
++arg;
fprintf(stderr, "reading query deflines from '%s'\n", argv[arg]);
SEQdict = dict_create(DICTCOUNT_T_MAX, headerCompare);
addToDict(SEQdict, argv[arg]);
} else if (strcmp(argv[arg], "-f") == 0) {
++arg;
SEQ = new seqCache(argv[arg]);
} else if (strcmp(argv[arg], "-q") == 0) {
++arg;
QLT = new seqCache(argv[arg]);
} else if (strcmp(argv[arg], "-filter") == 0) {
filter = atof(argv[++arg]);
doFiltering = true;
} else if (strcmp(argv[arg], "-output") == 0) {
char cmd[1024] = {0};
errno = 0;
++arg;
if (strcmp(argv[arg] + strlen(argv[arg]) - 4, ".bz2") == 0) {
sprintf(cmd, "bzip2 -1c > %s", argv[arg]);
oFile = popen(cmd, "w");
oFileIsPipe = 1;
} else if (strcmp(argv[arg] + strlen(argv[arg]) - 3, ".gz") == 0) {
sprintf(cmd, "gzip -1c > %s", argv[arg]);
oFile = popen(cmd, "w");
oFileIsPipe = 1;
} else {
fprintf(stderr, "Got %s, not .bz2 not .gz!\n", argv[arg]);
exit(1);
}
if (errno)
fprintf(stderr, "Failed to open '%s': %s\n", cmd, strerror(errno));
doFiltering = true;
} else if (strcmp(argv[arg], "-scores") == 0) {
errno = 0;
sFile = fopen(argv[++arg], "w");
if (errno)
fprintf(stderr, "Failed to open '%s': %s\n", argv[arg-1], strerror(errno));
doFiltering = true;
} else if (strcmp(argv[arg], "-unique") == 0) {
char cmd[1024] = {0};
errno = 0;
arg++;
if (strcmp(argv[arg] + strlen(argv[arg]) - 4, ".bz2") == 0)
sprintf(cmd, "bzip2 -1c > %s", argv[arg]);
else if (strcmp(argv[arg] + strlen(argv[arg]) - 3, ".gz") == 0)
sprintf(cmd, "gzip -1c > %s", argv[arg]);
else
sprintf(cmd, "cat > %s", argv[arg]);
uFile = popen(cmd, "w");
if (errno)
fprintf(stderr, "Failed to open '%s': %s\n", cmd, strerror(errno));
doFiltering = true;
} else if (strncmp(argv[arg], "-M", 2) == 0) {
arg++;
while ((arg < argc) && (fileExists(argv[arg]))) {
if (mergeFilesLen >= mergeFilesMax) {
fprintf(stderr, "%s: ERROR! Too many input files! Should be less than %d\n", argv[0], mergeFilesMax);
exit(1);
}
mergeNames[mergeFilesLen] = argv[arg];
mergeFiles[mergeFilesLen++] = openFile(argv[arg], "r");
arg++;
}
arg--;
} else {
fprintf(stderr, "unknown option: %s\n", argv[arg]);
}
arg++;
}
if (doFiltering) {
if (uFile == 0L)
fprintf(stderr, "ERROR: -unique is required\n"), exit(1);
if (sFile == 0L)
fprintf(stderr, "ERROR: -scores is required\n"), exit(1);
if ((filter < 0.0) || (filter > 1.0))
fprintf(stderr, "ERROR: -filter value of %f invalid. 0 <= F <= 100.\n", filter), exit(1);
}
if ((IIDdict == 0L) || (SEQdict == 0L) || (GENdict == 0L)) {
fprintf(stderr, "WARNING! No sequence dictionaries, NOT FIXING IIDs! (supply -fpart, -f and -g)\n");
}
if ((SEQ == 0L) || (QLT == 0L)) {
fprintf(stderr, "I need -f and -q\n");
exit(1);
}
// We no longer require that input polishes be sorted increasingly;
// now they only must be grouped. This remembers if we've seen a
// match or not. At the end, we'll analyze() those we haven't done
// already.
//
found = new bool [ SEQ->getNumberOfSequences() ];
for (uint32 i=0; i<SEQ->getNumberOfSequences(); i++)
found[i] = false;
// Initialize the merge -- if no merge files, nothing done!
//
for (int i=0; i<mergeFilesLen; i++) {
mergePolishes[i] = new sim4polish(mergeFiles[i]);
fixIID(mergePolishes[i], IIDdict);
}
// Read polishes, picking the best when we see a change in the
// estID.
sim4polish **p = new sim4polish * [pAlloc];
sim4polish *q;
while ((q = nextPolish()) != 0L) {
if ((q->_estID != estID) && (pNum > 0)) {
//fprintf(stderr, "PickBest for estID "uint32FMT"\n", estID);
found[estID] = true;
pickBest(p, pNum);
pNum = 0;
}
if (pNum >= pAlloc) {
sim4polish **P = new sim4polish * [pAlloc * 2];
memcpy(p, P, sizeof(sim4polish *) * pAlloc);
delete [] p;
p = P;
pAlloc *= 2;
}
p[pNum++] = q;
estID = q->_estID;
}
if (pNum > 0) {
found[estID] = true;
pickBest(p, pNum);
}
// Attempt cleanup
//
for (int i=0; i<mergeFilesLen; i++)
closeFile(mergeFiles[i], mergeNames[i]);
for (estID=0; estID < SEQ->getNumberOfSequences(); estID++)
if (found[estID] == false)
analyze(estID, 0, SEQ->getSequenceLength(estID), SEQ->getSequenceLength(estID), true, 'M');
delete [] mergeFiles;
delete [] mergeNames;
delete [] mergePolishes;
if (oFile) pclose(oFile);
if (uFile) pclose(uFile);
if (sFile) fclose(sFile);
fprintf(stderr, "Uni:"uint32FMTW(8)" Con:"uint32FMTW(8)" (T:"uint32FMTW(8)" M:"uint32FMTW(8)" I:"uint32FMTW(8)" S:"uint32FMTW(8)" N:"uint32FMTW(8)") Inc:"uint32FMTW(8)" -- Save:"uint32FMTW(8)" Lost:"uint32FMTW(8)"\n",
statOneMatch,
statConsistent, consistentTie, consistentMatches, consistentIdentity, consistentTooShort, consistentNot,
statInconsistent,
statUnique, statLost);
fprintf(stderr, "total: LQ:"uint32FMT" MQ:"uint32FMT" RQ:"uint32FMT"\n",
totLQ, totMQ, totRQ);
return(0);
}
