static void readTaggedNumLine(struct lineFile *lf, char *tag,
int count, int *intOut, double *floatOut)
/* Read in a line that starts with tag and then has count numbers.
* Complain and die if tag is unexpected or other problem occurs.
* Put output as integers and/or floating point into intOut and
* floatOut. */
{
char *line;
int i = 0;
char *word;
if (!lineFileNextReal(lf, &line))
lineFileUnexpectedEnd(lf);
word = nextWord(&line);
if (!sameWord(tag, word))
errAbort("Expecting %s got %s line %d of %s",
tag, word, lf->lineIx, lf->fileName);
for (i = 0; i < count; ++i)
{
word = nextWord(&line);
if (word == NULL)
errAbort("Not enough numbers line %d of %s", lf->lineIx, lf->fileName);
if (!isdigit(word[0]))
errAbort("Expecting number got %s line %d of %s",
word, lf->lineIx, lf->fileName);
if (intOut)
intOut[i] = atoi(word);
if (floatOut)
floatOut[i] = atof(word);
}
word = nextWord(&line);
if (word != NULL)
errAbort("Too many numbers line %d of %s", lf->lineIx, lf->fileName);
}
void makeMotifs(char *inFile, struct hash *tfHash, char *outFile)
/* Parse input motifs and save them to outFile in dnaMotif format. */
{
struct lineFile *lf = lineFileOpen(inFile, TRUE);
FILE *f = mustOpen(outFile, "w");
struct hashEl *hel;
for (;;)
{
char *line;
char *words[256], *word;
int wordCount;
struct dnaMotif *motif;
if (!lineFileSkipTo(lf, "Probability matrix for"))
break;
lineFileNeedNext(lf, &line, NULL);
wordCount = chopLine(line, words);
if (wordCount >= ArraySize(words))
errAbort("Line %d of %s is too long\n", lf->lineIx, lf->fileName);
if (!sameString(words[0], "#"))
badFormat(lf);
AllocVar(motif);
motif->columnCount = wordCount-1;
readBaseProbs(lf, words, "#A", &motif->aProb, motif->columnCount);
readBaseProbs(lf, words, "#C", &motif->cProb, motif->columnCount);
readBaseProbs(lf, words, "#T", &motif->tProb, motif->columnCount);
readBaseProbs(lf, words, "#G", &motif->gProb, motif->columnCount);
if (!lineFileSkipTo(lf, "Source:"))
lineFileUnexpectedEnd(lf);
lineFileReuse(lf);
lineFileNeedNext(lf, &line, NULL);
word = nextWord(&line);
word = nextWord(&line);
if (word == NULL)
errAbort("Short Source: line %d of %s", lf->lineIx, lf->fileName);
motif->name = cloneString(word);
hel = hashLookup(tfHash, motif->name);
if (hel == NULL)
errAbort("%s in %s but not GFFs", motif->name, lf->fileName);
hel->val = motif;
dnaMotifTabOut(motif, f);
}
carefulClose(&f);
lineFileClose(&lf);
}
static struct joinerSet *parseIdentifierSet(struct lineFile *lf,
char *line, struct hash *symHash, struct dyString *dyBuf)
/* Parse out one joiner record - keep going until blank line or
* end of file. */
{
struct joinerSet *js;
struct joinerField *jf;
char *word, *e;
char *parts[3];
int partCount;
/* Parse through first line - first word is name. */
word = nextWord(&line);
if (word == NULL || strchr(word, '=') != NULL)
errAbort("joiner without name line %d of %s\n", lf->lineIx, lf->fileName);
AllocVar(js);
js->name = cloneString(word);
js->lineIx = lf->lineIx;
while ((word = nextWord(&line)) != NULL)
{
char *e = strchr(word, '=');
if (e != NULL)
*e++ = 0;
if (sameString(word, "typeOf"))
{
js->typeOf = cloneSpecified(lf, word, e);
}
else if (sameString(word, "external"))
{
js->external = cloneSpecified(lf, word, e);
}
else if (sameString(word, "fuzzy"))
{
js->isFuzzy = TRUE;
}
else if (sameString(word, "dependency"))
{
js->isDependency = TRUE;
}
else
{
errAbort("Unknown attribute %s line %d of %s", word,
lf->lineIx, lf->fileName);
}
}
/* Parse second line, make sure it is quoted, and save as description. */
line = nextSubbedLine(lf, symHash, dyBuf);
if (line == NULL)
lineFileUnexpectedEnd(lf);
line = trimSpaces(line);
if (line[0] != '"' || lastChar(line) != '"')
errAbort("Expecting quoted line, line %d of %s\n",
lf->lineIx, lf->fileName);
line[strlen(line)-1] = 0;
js->description = cloneString(line+1);
/* Go through subsequent lines. */
while ((line = nextSubbedLine(lf, symHash, dyBuf)) != NULL)
{
/* Keep grabbing until we get a blank line. */
line = skipLeadingSpaces(line);
if (line[0] == 0)
break;
/* First word in line should be database.tabe.field. */
word = nextWord(&line);
partCount = chopString(word, ".", parts, ArraySize(parts));
if (partCount != 3)
errAbort("Expecting database.table.field line %d of %s",
lf->lineIx, lf->fileName);
/* Allocate struct and save table and field. */
AllocVar(jf);
jf->lineIx = lf->lineIx;
jf->table = cloneString(parts[1]);
jf->field = cloneString(parts[2]);
if (js->fieldList == NULL && !js->isFuzzy)
{
jf->isPrimary = TRUE;
jf->unique = TRUE;
jf->full = TRUE;
}
jf->minCheck = 1.0;
slAddHead(&js->fieldList, jf);
/* Database may be a comma-separated list. Parse it here. */
jf->dbList = parseDatabaseList(lf, parts[0]);
/* Look for other fields in subsequent space-separated words. */
while ((word = nextWord(&line)) != NULL)
{
if ((e = strchr(word, '=')) != NULL)
*e++ = 0;
if (sameString("comma", word))
{
jf->separator = cloneString(",");
}
else if (sameString("separator", word))
{
jf->separator = cloneSpecified(lf, word, e);
}
else if (sameString("chopBefore", word))
{
if (e == NULL)
unspecifiedVar(lf, word);
slNameStore(&jf->chopBefore, e);
}
else if (sameString("chopAfter", word))
{
if (e == NULL)
unspecifiedVar(lf, word);
slNameStore(&jf->chopAfter, e);
}
else if (sameString("indexOf", word))
{
jf->indexOf = TRUE;
}
else if (sameString("dupeOk", word))
{
if (!jf->isPrimary)
warn("dupeOk outsite primary key line %d of %s",
lf->lineIx, lf->fileName);
jf->unique = FALSE;
}
else if (sameString("minCheck", word))
{
if (e == NULL)
unspecifiedVar(lf, word);
jf->minCheck = atof(e);
}
else if (sameString("unique", word))
{
jf->unique = TRUE;
}
else if (sameString("full", word))
{
jf->full = TRUE;
}
else if (sameString("splitPrefix", word))
{
jf->splitPrefix = cloneSpecified(lf, word, e);
}
else if (sameString("splitSuffix", word))
{
jf->splitSuffix = cloneSpecified(lf, word, e);
}
else if (sameString("exclude", word))
{
if (e == NULL)
unspecifiedVar(lf, word);
slNameStore(&jf->exclude, e);
}
else
{
errAbort("Unrecognized attribute %s line %d of %s",
word, lf->lineIx, lf->fileName);
}
}
if (jf->indexOf && jf->separator == NULL)
errAbort("indexOf without comma or separator line %d of %s",
lf->lineIx, lf->fileName);
if (jf->isPrimary && jf->separator != NULL)
errAbort("Error line %d of %s\n"
"Primary key can't be a list (comma or separator)."
, lf->lineIx, lf->fileName);
}
slReverse(&js->fieldList);
return js;
}
void lineFileNeedNext(struct lineFile *lf, char **retStart, int *retSize)
/* Fetch next line from file. Squawk and die if it's not there. */
{
if (!lineFileNext(lf, retStart, retSize))
lineFileUnexpectedEnd(lf);
}
