int main(int argc, char *argv[])
/* Process command line. */
{
optionInit(&argc, argv, options);
if (argc != 3)
usage();
clSeq = optionVal("seq", clSeq);
clStart = optionInt("start", clStart);
clEnd = optionInt("end", clEnd);
clSeqList = optionVal("seqList", clSeqList);
clBpt = optionVal("bpt", clBpt);
clBed = optionVal("bed", clBed);
clBedPos = optionExists("bedPos");
noMask = optionExists("noMask");
udcSetDefaultDir(optionVal("udcDir", udcDefaultDir()));
if (clBedPos && !clBed)
errAbort("the -bedPos option requires the -bed option");
if (clBed != NULL)
{
if (clSeqList != NULL)
errAbort("Can only have seqList or bed options, not both.");
if (clSeq != NULL)
errAbort("Can only have seq or bed options, not both.");
}
if ((clStart > clEnd) && (clSeq == NULL))
errAbort("must specify -seq with -start and -end");
if ((clSeq != NULL) && (clSeqList != NULL))
errAbort("can't specify both -seq and -seqList");
dnaUtilOpen();
twoBitToFa(argv[1], argv[2]);
return 0;
}
enum metaBigFileType isBigWigOrBed(char* filename)
/* Peak at a file to see if it's bigWig */
{
enum metaBigFileType ret = isNotBig;
bits32 magic;
struct udcFile* udc;
udc = udcFileOpen(filename, udcDefaultDir());
/* Read magic number at head of file and use it to see if we are proper file type, and
* see if we are byte-swapped. */
udcMustRead(udc, &magic, sizeof(magic));
if (magic == bigWigSig)
ret = isaBigWig;
else if (magic == bigBedSig)
ret = isaBigBed;
if (ret != isNotBig) {
udcFileClose(&udc);
return ret;
}
magic = byteSwap32(magic);
if (magic == bigWigSig)
ret = isaBigWig;
else if (magic == bigBedSig)
ret = isaBigBed;
udcFileClose(&udc);
return ret;
}
int main(int argc, char *argv[])
/* Process command line. */
{
optionInit(&argc, argv, options);
if (argc != 2)
usage();
udcSetDefaultDir(optionVal("udcDir", udcDefaultDir()));
twoBitDup(argv[1]);
return 0;
}
int main(int argc, char *argv[])
/* Process command line. */
{
optionInit(&argc, argv, options);
if (argc != 2)
usage();
cacheDir = optionVal("cacheDir", udcDefaultDir());
testOnly = optionExists("test");
doCleanup(argv[1]);
return 0;
}
char *getSparseFileName(char *url)
/* Return the path to sparseData cache file for url. */
{
struct slName *sl, *cacheFiles = udcFileCacheFiles(url, udcDefaultDir());
char *sparseFileName = NULL;
for (sl = cacheFiles; sl != NULL; sl = sl->next)
if (endsWith(sl->name, "sparseData"))
sparseFileName = sl->name;
if (sparseFileName == NULL)
errAbort("can't find sparseData file in udcFileCacheFiles(%s) results", url);
return sparseFileName;
}
int main(int argc, char *argv[])
/* Process command line. */
{
optionInit(&argc, argv, options);
udcSetDefaultDir(optionVal("udcDir", udcDefaultDir()));
if (argc != 2)
usage();
bigWigInfo(argv[1]);
if (verboseLevel() > 1)
printVmPeak();
return 0;
}
int main(int argc, char *argv[])
/* Process command line. */
{
optionInit(&argc, argv, options);
if (argc != 3)
usage();
clChrom = optionVal("chrom", clChrom);
clStart = optionInt("start", clStart);
clEnd = optionInt("end", clEnd);
udcSetDefaultDir(optionVal("udcDir", udcDefaultDir()));
bigWigToWig(argv[1], argv[2]);
return 0;
}
void usage()
/* Explain usage and exit. */
{
errAbort(
"udcCleanup - Clean up old unused files in udcCache.\n"
"usage:\n"
" udcCleanup maxUnusedDays\n"
"example:\n"
" udcCleanup 7.5\n"
"options:\n"
" -cacheDir=dir use the indicated cache dir instead of the default (%s)\n"
" -test - don't actually clean up, but do still figure cleanup bytes\n"
, udcDefaultDir()
);
}
int main(int argc, char *argv[])
/* Process command line. */
{
optionInit(&argc, argv, options);
if (argc != 2)
usage();
udcSetDefaultDir(optionVal("udcDir", udcDefaultDir()));
char *hubUrl = optionVal("addHub", NULL);
if (hubUrl != NULL)
return hubPublicAdd(argv[1],hubUrl);
else
return hubPublicCheck(argv[1]);
}
int main(int argc, char *argv[])
/* Process command line. */
{
optionInit(&argc, argv, options);
clChrom = optionVal("chrom", clChrom);
clStart = optionInt("start", clStart);
clEnd = optionInt("end", clEnd);
maxItems = optionInt("maxItems", maxItems);
udcSetDefaultDir(optionVal("udcDir", udcDefaultDir()));
if (argc != 3)
usage();
bigBedToBed(argv[1], argv[2]);
if (verboseLevel() > 1)
printVmPeak();
return 0;
}
int main(int argc, char *argv[])
/* Process command line. */
{
optionInit(&argc, argv, options);
if (argc != 2)
usage();
udcSetCacheTimeout(cacheTime);
// UDC cache dir: first check for hg.conf setting, then override with command line option if given.
setUdcCacheDir();
udcSetDefaultDir(optionVal("udcDir", udcDefaultDir()));
if (trackHubCrawl(argv[1]))
{
return 1;
}
return 0;
}
struct crTreeFile *crTreeFileOpen(char *fileName)
/* Open up r-tree index file - reading headers and verifying things. */
{
/* Open file and allocate structure to hold info from header etc. */
struct udcFile *udc = udcFileOpen(fileName, udcDefaultDir());
struct crTreeFile *crt = needMem(sizeof(*crt));
fileName = crt->fileName = cloneString(fileName);
crt->udc = udc;
/* Read magic number at head of file and use it to see if we are proper file type, and
* see if we are byte-swapped. */
bits32 magic;
boolean isSwapped = FALSE;
udcMustReadOne(udc, magic);
if (magic != crTreeSig)
{
magic = byteSwap32(magic);
isSwapped = crt->isSwapped = TRUE;
if (magic != crTreeSig)
errAbort("%s is not a chromosome r-tree index file", fileName);
}
/* Read rest of high level header including notably the offsets to the
* chromosome and range indexes. */
bits32 reserved32;
udcMustReadOne(udc, reserved32);
crt->chromOffset = udcReadBits64(udc, isSwapped);
crt->cirOffset = udcReadBits64(udc, isSwapped);
/* Read in the chromosome index header. */
udcSeek(udc, crt->chromOffset);
crt->chromBpt = bptFileAttach(fileName, udc);
/* Read in range index header. */
udcSeek(udc, crt->cirOffset);
crt->cir = cirTreeFileAttach(fileName, udc);
return crt;
}
int main(int argc, char *argv[])
/* Process command line. */
{
optionInit(&argc, argv, options);
udcSetDefaultDir(optionVal("udcDir", udcDefaultDir()));
if (optionExists("fields"))
{
if (argc < 2)
usage();
bigBedFields(argv[1]);
}
else
{
summaryType = optionVal("type", summaryType);
if (argc != 6)
usage();
bigBedSummary(argv[1], argv[2], sqlUnsigned(argv[3]), sqlUnsigned(argv[4]), sqlUnsigned(argv[5]));
}
if (verboseLevel() > 1)
printVmPeak();
return 0;
}
int main(int argc, char *argv[])
/* Process command line. */
{
optionInit(&argc, argv, options);
blockSize = optionInt("blockSize", blockSize);
itemsPerSlot = optionInt("itemsPerSlot", itemsPerSlot);
asFile = optionVal("as", asFile);
doCompress = !optionExists("unc");
sizesIs2Bit = optionExists("sizesIs2Bit");
extraIndex = optionVal("extraIndex", NULL);
tabSep = optionExists("tab");
udcDir = optionVal("udcDir", udcDefaultDir());
if (argc != 4)
usage();
udcSetDefaultDir(udcDir);
if (optionExists("type"))
{
// parse type
char *btype = cloneString(optionVal("type", ""));
char *plus = strchr(btype, '+');
if (plus)
{
*plus++ = 0;
if (isdigit(*plus))
bedP = sqlUnsigned(plus);
}
if (!startsWith("bed", btype))
errAbort("type must begin with \"bed\"");
btype +=3;
bedN = sqlUnsigned(btype);
if (bedN < 3)
errAbort("Bed must be 3 or higher, found %d\n", bedN);
if (bedN > 15)
errAbort("Bed must be 15 or lower, found %d\n", bedN);
}
else
{
if (asText)
errAbort("If you specify the .as file, you must specify the -type as well so that\n"
"the number of standard BED columns is known.");
}
/* If the haven't set bedN and bedP from the type var in the command line, then we sniff it
* out from file. */
char *bedFileName = argv[1];
if (bedN == 0)
{
/* Just read in single line and count fields. */
struct lineFile *lf = lineFileOpen(bedFileName, TRUE);
char *line;
if (!lineFileNextReal(lf, &line))
errAbort("%s is empty", lf->fileName);
int fieldCount;
if (tabSep)
fieldCount = chopByChar(line, '\t', NULL, 0); // Do not use chopString, see GOTCHA
else
fieldCount = chopByWhite(line, NULL, 0);
if (fieldCount > 256)
errAbort("Too many columns in %s, you sure it's a bed file?", lf->fileName);
lineFileClose(&lf);
/* Set up so that it looks like we are straight up bed for that many fields,
* or if more than or maximum defined fields, then for bed15+ */
bedN = fieldCount;
if (bedN > bedKnownFields)
{
bedP = bedN - bedKnownFields;
bedN = bedKnownFields;
}
}
/* Make sure that fields are defined, from bed spec if nowhere else. */
if (asFile)
readInGulp(asFile, &asText, NULL);
else
asText = bedAsDef(bedN, bedN + bedP);
bedToBigBed(bedFileName, argv[2], argv[3]);
optionFree();
if (verboseLevel() > 1)
printVmPeak();
return 0;
}
struct cirTreeFile *cirTreeFileOpen(char *fileName)
/* Open up r-tree index file - reading header and verifying things. */
{
return cirTreeFileAttach(cloneString(fileName), udcFileOpen(fileName, udcDefaultDir()));
}
struct bptFile *bptFileOpen(char *fileName)
/* Open up index file - reading header and verifying things. */
{
return bptFileAttach(cloneString(fileName), udcFileOpen(fileName, udcDefaultDir()));
}
int main(int argc, char *argv[])
/* Process command line. */
{
optionInit(&argc, argv, options);
if (argc != 2 && !optionExists("settings"))
usage();
struct trackHubCheckOptions *checkOptions = NULL;
AllocVar(checkOptions);
checkOptions->specHost = (optionExists("test") ? "genome-test.soe.ucsc.edu" : "genome.ucsc.edu");
checkOptions->specHost = optionVal("specHost", checkOptions->specHost);
checkOptions->printMeta = optionExists("printMeta");
checkOptions->checkFiles = !optionExists("noTracks");
checkOptions->checkSettings = optionExists("checkSettings");
struct trackHubSettingSpec *setting = NULL;
AllocVar(setting);
setting->level = optionVal("level", "all");
if (trackHubSettingLevel(setting) < 0)
{
fprintf(stderr, "ERROR: Unrecognized support level %s\n\n", setting->level);
usage();
}
checkOptions->level = setting->level;
char *version = NULL;
if (optionExists("version"))
version = optionVal("version", NULL);
checkOptions->version = version;
char *extraFile = optionVal("extra", NULL);
if (extraFile != NULL)
addExtras(extraFile, checkOptions);
cacheTime = optionInt("cacheTime", cacheTime);
udcSetCacheTimeout(cacheTime);
// UDC cache dir: first check for hg.conf setting, then override with command line option if given.
setUdcCacheDir();
udcSetDefaultDir(optionVal("udcDir", udcDefaultDir()));
knetUdcInstall(); // make the htslib library use udc
if (optionExists("settings"))
{
showSettings(checkOptions);
return 0;
}
struct dyString *errors = newDyString(1024);
if (trackHubCheck(argv[1], checkOptions, errors))
{
// uniquify and count errors
struct slName *errs = slNameListFromString(errors->string, '\n');
slUniqify(&errs, slNameCmp, slNameFree);
int errCount = slCount(errs);
printf("Found %d problem%s:\n", errCount, errCount == 1 ? "" : "s");
printf("%s\n", slNameListToString(errs, '\n'));
return 1;
}
return 0;
}
struct bbiFile *bbiFileOpen(char *fileName, bits32 sig, char *typeName)
/* Open up big wig or big bed file. */
{
/* This code needs to agree with code in two other places currently - bigBedFileCreate,
* and bigWigFileCreate. I'm thinking of refactoring to share at least between
* bigBedFileCreate and bigWigFileCreate. It'd be great so it could be structured
* so that it could send the input in one chromosome at a time, and send in the zoom
* stuff only after all the chromosomes are done. This'd potentially reduce the memory
* footprint by a factor of 2 or 4. Still, for now it works. -JK */
struct bbiFile *bbi;
AllocVar(bbi);
bbi->fileName = cloneString(fileName);
struct udcFile *udc = bbi->udc = udcFileOpen(fileName, udcDefaultDir());
/* Read magic number at head of file and use it to see if we are proper file type, and
* see if we are byte-swapped. */
bits32 magic;
boolean isSwapped = FALSE;
udcMustRead(udc, &magic, sizeof(magic));
if (magic != sig)
{
magic = byteSwap32(magic);
isSwapped = TRUE;
if (magic != sig)
errAbort("%s is not a %s file", fileName, typeName);
}
bbi->typeSig = sig;
bbi->isSwapped = isSwapped;
/* Read rest of defined bits of header, byte swapping as needed. */
bbi->version = udcReadBits16(udc, isSwapped);
bbi->zoomLevels = udcReadBits16(udc, isSwapped);
bbi->chromTreeOffset = udcReadBits64(udc, isSwapped);
bbi->unzoomedDataOffset = udcReadBits64(udc, isSwapped);
bbi->unzoomedIndexOffset = udcReadBits64(udc, isSwapped);
bbi->fieldCount = udcReadBits16(udc, isSwapped);
bbi->definedFieldCount = udcReadBits16(udc, isSwapped);
bbi->asOffset = udcReadBits64(udc, isSwapped);
bbi->totalSummaryOffset = udcReadBits64(udc, isSwapped);
bbi->uncompressBufSize = udcReadBits32(udc, isSwapped);
/* Skip over reserved area. */
udcSeek(udc, 64);
/* Read zoom headers. */
int i;
struct bbiZoomLevel *level, *levelList = NULL;
for (i=0; i<bbi->zoomLevels; ++i)
{
AllocVar(level);
level->reductionLevel = udcReadBits32(udc, isSwapped);
level->reserved = udcReadBits32(udc, isSwapped);
level->dataOffset = udcReadBits64(udc, isSwapped);
level->indexOffset = udcReadBits64(udc, isSwapped);
slAddHead(&levelList, level);
}
slReverse(&levelList);
bbi->levelList = levelList;
/* Attach B+ tree of chromosome names and ids. */
udcSeek(udc, bbi->chromTreeOffset);
bbi->chromBpt = bptFileAttach(fileName, udc);
return bbi;
}
