void chainIndex(char *inChain, char *outIndex)
/* chainIndex - Create simple two column file index for chain. */
{
struct lineFile *lf = lineFileOpen(inChain, TRUE);
FILE *f = mustOpen(outIndex, "w");
struct chain *chain, *lastChain = NULL;
long pos = 0;
struct hash *uniqHash = hashNew(16);
while ((chain = chainRead(lf)) != NULL)
{
if (lastChain == NULL || !sameString(chain->tName, lastChain->tName))
{
if (hashLookup(uniqHash, chain->tName))
{
errAbort("%s is not sorted, %s repeated with intervening %s",
inChain, chain->tName, lastChain->tName);
}
hashAddInt(uniqHash, chain->tName, pos);
fprintf(f, "%lx\t%s\n", pos, chain->tName);
}
chainFree(&lastChain);
lastChain = chain;
pos = lineFileTell(lf);
}
}
void getOffset(char *directoryName, char *chromName, char *outputFileName)
{
FILE *outputFileHandle = mustOpen(outputFileName, "w");
struct lineFile *lf = NULL;
char *line;
off_t offset;
char *row[9], *rsId[2];
char inputFileName[64];
safef(inputFileName, sizeof(inputFileName), "%s/%s.fa", directoryName, chromName);
lf = lineFileOpen(inputFileName, TRUE);
while (lineFileNext(lf, &line, NULL))
{
if (line[0] == '>')
{
chopString(line, "|", row, ArraySize(row));
chopString(row[2], " ", rsId, ArraySize(rsId));
offset = lineFileTell(lf);
fprintf(outputFileHandle, "%s\t%s\t%ld\n", rsId[0], chromName, offset);
}
}
carefulClose(&outputFileHandle);
lineFileClose(&lf);
}
void wigSort(char *input, char *output)
/* wigSort - Sort a wig file.. */
{
struct lineFile *lf = lineFileOpen(input, TRUE);
struct pos *pos, *posList = NULL;
char *line;
while (lineFileNextReal(lf, &line))
{
verbose(2, "processing %s\n", line);
AllocVar(pos);
pos->fileOffset = lineFileTell(lf);
if (posList != NULL)
posList->fileSize = pos->fileOffset - posList->fileOffset;
slAddHead(&posList, pos);
if (stringIn("chrom=", line))
{
parseSteppedSection(lf, line, pos);
}
else
{
/* Check for bed... */
char *words[5];
int wordCount = chopLine(line, words);
if (wordCount != 4)
errAbort("Unrecognized format line %d of %s:\n", lf->lineIx, lf->fileName);
pos->chrom = cloneString(words[0]);
pos->start = lineFileNeedNum(lf, words, 1);
}
}
if (posList != NULL)
{
posList->fileSize = lineFileTell(lf) - posList->fileOffset;
slReverse(&posList);
slSort(&posList, posCmp);
}
lineFileClose(&lf);
FILE *in = mustOpen(input, "r");
FILE *out = mustOpen(output, "w");
for (pos = posList; pos != NULL; pos = pos->next)
{
fseek(in, pos->fileOffset, SEEK_SET);
copyFileBytes(in, out, pos->fileSize);
}
carefulClose(&in);
carefulClose(&out);
}
struct chromIxRange *chromIxRangeLoadAll(char *fileName, bits64 *retFileSize)
/* Load all chromIxRange from a whitespace-separated file.
* Dispose of this with chromIxRangeFreeList(). */
{
struct chromIxRange *list = NULL, *el;
struct lineFile *lf = lineFileOpen(fileName, TRUE);
char *row[3];
while (lineFileRow(lf, row))
{
AllocVar(el);
el->chromIx = sqlUnsigned(row[0]);
el->start = sqlUnsigned(row[1]);
el->end = sqlUnsigned(row[2]);
el->fileOffset = lineFileTell(lf);
slAddHead(&list, el);
}
*retFileSize = lineFileTell(lf);
lineFileClose(&lf);
slReverse(&list);
return list;
}
void axtIndex(char *in, char *out)
/* axtIndex - Create summary file for axt. */
{
struct lineFile *lf = lineFileOpen(in, TRUE);
FILE *f = mustOpen(out, "w");
struct axt *axt;
for (;;)
{
off_t pos = lineFileTell(lf);
axt = axtRead(lf);
if (axt == NULL)
break;
fprintf(f, "%d %d %lld\n", axt->tStart, axt->tEnd - axt->tStart, (unsigned long long) pos);
axtFree(&axt);
}
carefulClose(&f);
}
void addFa(char *file, struct hash *fileHash, struct hash *seqHash)
/* Add a fa file to hashes. */
{
struct lineFile *lf = lineFileOpen(file, TRUE);
char *line, *name;
char *rFile = hashStoreName(fileHash, file);
while (lineFileNext(lf, &line, NULL))
{
if (line[0] == '>')
{
struct seqFilePos *sfp;
line += 1;
name = nextWord(&line);
if (name == NULL)
errAbort("bad line %d of %s", lf->lineIx, lf->fileName);
AllocVar(sfp);
hashAddSaveName(seqHash, name, sfp, &sfp->name);
sfp->file = rFile;
sfp->pos = lineFileTell(lf);
}
}
lineFileClose(&lf);
}
struct segBlock *segNextWithPos(struct segFile *sf, off_t *retOffset)
/* Return next segment in segment file or NULL if at end. If retOffset
* is not NULL, return start offset of record in file. */
{
struct lineFile *lf = sf->lf;
struct segBlock *block;
struct segComp *comp, *tail = NULL;
char *line, *name, *row[6], *val, *word;
int wordCount;
/* Loop until we get a segment paragraph or reach end of file. */
for (;;)
{
/* Get segment header line. If it's not there assume end of file. */
if (!nextLine(lf, &line))
{
lineFileClose(&sf->lf);
return NULL;
}
/* Parse segment header line. */
word = nextWord(&line);
if (word == NULL)
continue; /* Ignore blank lines. */
if (sameString(word, "b"))
{
if (retOffset != NULL)
*retOffset = lineFileTell(sf->lf);
AllocVar(block);
/* Parse name=val. */
while ((word = nextWord(&line)) != NULL)
{
name = word;
val = strchr(word, '=');
if (val == NULL)
errAbort("Missing = after %s line %d of %s",
name, lf->lineIx, lf->fileName);
*val++ = 0;
if (sameString(name, "name"))
block->name = cloneString(val);
else if (sameString(name, "val"))
block->val = atoi(val);
}
/* Parse segment components until blank line. */
for (;;)
{
if (!nextLine(lf, &line))
errAbort("Unexpected end of file %s", lf->fileName);
word = nextWord(&line);
if (word == NULL)
break;
if (sameString(word, "s"))
{
/* Chop line up by white space. This involves a few +=1's
* because we have already chopped out the first word. */
row[0] = word;
wordCount = chopByWhite(line, row+1, ArraySize(row)-1) +1;
lineFileExpectWords(lf, ArraySize(row), wordCount);
AllocVar(comp);
/* Convert ASCII text representation to segComp structure. */
comp->src = cloneString(row[1]);
comp->start = lineFileNeedNum(lf, row, 2);
comp->size = lineFileNeedNum(lf, row, 3);
comp->strand = row[4][0];
comp->srcSize = lineFileNeedNum(lf, row, 5);
/* Do some sanity checking. */
if (comp->size <= 0 || comp->srcSize <= 0)
errAbort("Got a negative or zero size line %d of %s",
lf->lineIx, lf->fileName);
if (comp->start < 0 || comp->start + comp->size > comp->srcSize)
errAbort("Coordinates out of range line %d of %s",
lf->lineIx, lf->fileName);
if (comp->strand != '+' && comp->strand != '-')
errAbort("Invalid strand line %d of %s",
lf->lineIx, lf->fileName);
/* Add the new component to the current list. */
if (block->components == NULL)
block->components = comp;
else
tail->next = comp;
tail = comp;
}
}
return block;
}
else /* Skip over paragraph we don't understand. */
{
for (;;)
{
if (!nextLine(lf, &line))
return NULL;
if (nextWord(&line) == NULL)
break;
}
}
}
}
struct mafAli *mafNextWithPos(struct mafFile *mf, off_t *retOffset)
/* Return next alignment in FILE or NULL if at end. If retOffset is
* nonNULL, return start offset of record in file. */
{
struct lineFile *lf = mf->lf;
struct mafAli *ali;
char *line, *word;
/* Loop until get an alignment paragraph or reach end of file. */
for (;;)
{
/* Get alignment header line. If it's not there assume end of file. */
if (!nextLine(lf, &line))
{
lineFileClose(&mf->lf);
return NULL;
}
/* Parse alignment header line. */
word = nextWord(&line);
if (word == NULL)
continue; /* Ignore blank lines. */
if (sameString(word, "a"))
{
if (retOffset != NULL)
*retOffset = lineFileTell(mf->lf);
AllocVar(ali);
while ((word = nextWord(&line)) != NULL)
{
/* Parse name=val. */
char *name = word;
char *val = strchr(word, '=');
if (val == NULL)
errAbort("Missing = after %s line 1 of %s", name, lf->fileName);
*val++ = 0;
if (sameString(name, "score"))
ali->score = atof(val);
}
/* Parse alignment components until blank line. */
for (;;)
{
if (!nextLine(lf, &line))
errAbort("Unexpected end of file %s", lf->fileName);
word = nextWord(&line);
if (word == NULL)
break;
if (sameString(word, "s") || sameString(word, "e"))
{
struct mafComp *comp;
int wordCount;
char *row[7];
int textSize;
/* Chop line up by white space. This involves a few +-1's because
* have already chopped out first word. */
row[0] = word;
wordCount = chopByWhite(line, row+1, ArraySize(row)-1) + 1; /* +-1 because of "s" */
lineFileExpectWords(lf, ArraySize(row), wordCount);
AllocVar(comp);
/* Convert ascii text representation to mafComp structure. */
comp->src = cloneString(row[1]);
comp->srcSize = lineFileNeedNum(lf, row, 5);
comp->strand = row[4][0];
comp->start = lineFileNeedNum(lf, row, 2);
if (sameString(word, "e"))
{
comp->size = 0;
comp->rightLen = comp->leftLen = lineFileNeedNum(lf, row, 3);
comp->rightStatus = comp->leftStatus = *row[6];
}
else
{
comp->size = lineFileNeedNum(lf, row, 3);
comp->text = cloneString(row[6]);
textSize = strlen(comp->text);
/* Fill in ali->text size. */
if (ali->textSize == 0)
ali->textSize = textSize;
else if (ali->textSize != textSize)
errAbort("Text size inconsistent (%d vs %d) line %d of %s",
textSize, ali->textSize, lf->lineIx, lf->fileName);
}
/* Do some sanity checking. */
if (comp->srcSize < 0 || comp->size < 0)
errAbort("Got a negative size line %d of %s", lf->lineIx, lf->fileName);
if (comp->start < 0 || comp->start + comp->size > comp->srcSize)
errAbort("Coordinates out of range line %d of %s", lf->lineIx, lf->fileName);
/* Add component to head of list. */
slAddHead(&ali->components, comp);
}
if (sameString(word, "i"))
{
struct mafComp *comp;
int wordCount;
char *row[6];
/* Chop line up by white space. This involves a few +-1's because
* have already chopped out first word. */
row[0] = word;
wordCount = chopByWhite(line, row+1, ArraySize(row)-1) + 1; /* +-1 because of "s" */
lineFileExpectWords(lf, ArraySize(row), wordCount);
if (!sameString(row[1],ali->components->src))
errAbort("i line src mismatch: i is %s :: s is %s\n", row[1], ali->components->src);
comp = ali->components;
comp->leftStatus = *row[2];
comp->leftLen = atoi(row[3]);
comp->rightStatus = *row[4];
comp->rightLen = atoi(row[5]);
}
if (sameString(word, "q"))
{
struct mafComp *comp;
int wordCount;
char *row[3];
/* Chop line up by white space. This involves a few +-1's because
* have already chopped out first word. */
row[0] = word;
wordCount = chopByWhite(line, row+1, ArraySize(row)-1) + 1; /* +-1 because of "s" */
lineFileExpectWords(lf, ArraySize(row), wordCount);
if (!sameString(row[1],ali->components->src))
errAbort("q line src mismatch: q is %s :: s is %s\n", row[1], ali->components->src);
comp = ali->components;
comp->quality = cloneString(row[2]);
}
}
slReverse(&ali->components);
return ali;
}
else /* Skip over paragraph we don't understand. */
{
for (;;)
{
if (!nextLine(lf, &line))
return NULL;
if (nextWord(&line) == NULL)
break;
}
}
}
}
