boolean faSizeNext(struct lineFile *lf, char retLine[512], int *retSize)
/* Get > line and size for next fa record. Return FALSE at end of file. */
{
char *line;
int size = 0;
/* Fetch first record . */
for (;;)
{
if (!lineFileNext(lf, &line, NULL))
return FALSE;
if (line[0] == '>')
break;
}
strncpy(retLine, line, 512);
/* Loop around counting DNA-looking characters until next record. */
for (;;)
{
if (!lineFileNext(lf, &line, NULL))
break;
if (line[0] == '>')
{
lineFileReuse(lf);
break;
}
size += alphaCount(line);
}
*retSize = size;
return TRUE;
}
void lineRange(char *fileName, int start, int count)
/* lineRange - Get a range of lines from file. */
{
struct lineFile *lf = lineFileOpen(fileName, FALSE);
char *line;
int lineSize;
int i;
if (count == 0 || start == 0)
errAbort("Expecting positive number for start, count in command line");
/* Skip over first lines. */
for (i=1; i<start; ++i)
{
if (!lineFileNext(lf, &line, &lineSize))
errAbort("%s doesn't have %d lines", fileName, start);
}
/* Print coutn lines. */
for (i=0; i<count; ++i)
{
if (!lineFileNext(lf, &line, &lineSize))
break;
mustWrite(stdout, line, lineSize);
}
lineFileClose(&lf);
}
boolean read_fastq_auto(struct fastq_auto *fq, struct lineFile *lf, boolean just_seq_qual)
/* fill in fastq struct from open lineFile. return FALSE if at EOF */
/* set just_seq_qual=TRUE to skip loading everything except the sequence */
/* and quality information. */
{
char *line;
int len = 0;
boolean neof = lineFileNext(lf, &line, &len);
if (neof)
{
int i;
int qual_size;
/* should be header */
if ((len <= 0) || (line[0] != '@'))
errAbort("Expecting header. Problem on line %d\n", lf->lineIx);
if (!just_seq_qual)
{
char *words[7];
int numWords;
numWords = chopByChar(line, ':', words, 6);
strcpy(fq->machine, words[0] + 1);
fq->flow_cell = sqlSigned(words[1]);
fq->tile = sqlSigned(words[2]);
fq->tile_x = sqlSigned(words[3]);
words[5] = chopPrefixAt(words[4], '#');
words[6] = chopPrefixAt(words[5], '/');
fq->tile_y = sqlSigned(words[4]);
fq->multiplex_index = sqlSigned(words[5]);
fq->pair_num = sqlSigned(words[6]);
}
/* read the sequence */
fq->seq[0] = '\0';
while ((neof = lineFileNext(lf, &line, &len)) && (len > 0) && (line[0] != '+'))
strcat(fq->seq, line);
if (!neof)
errAbort("incomplete fastq file. early EOF");
fq->seq_size = strlen(fq->seq);
/* at the point of the quality header. who cares, read the quality */
fq->qual[0] = '\0';
while ((neof = lineFileNext(lf, &line, &len)) && (len > 0) && (line[0] != '@'))
strcat(fq->qual, line);
if ((len > 0) && (line[0] == '@'))
lineFileReuse(lf);
qual_size = strlen(fq->qual);
if (qual_size != fq->seq_size)
errAbort("something wrong line %d. sequence size (%d) should match quality size (%d)\n",
lf->lineIx, fq->seq_size, qual_size);
/* convert Illumina 1.3+ quals to Sanger */
for (i = 0; i < qual_size; i++)
fq->qual[i] -= 64;
}
else
return FALSE;
return TRUE;
}
void viewWaba(char *wabName)
/* Show human readable waba alignment. */
{
struct lineFile *lf = lineFileOpen(wabName, TRUE);
int lineSize;
char *line;
char *qSym;
char *tSym;
char *hSym;
int symCount;
int wordCount, partCount;
char *words[16], *parts[4];
int qStart, qEnd, tStart, tEnd;
char strand;
while (lineFileNext(lf, &line, &lineSize))
{
printf("%s\n", line);
wordCount = chopLine(line, words);
if (wordCount != 10)
errAbort("Funny info line %d of %s\n", lf->lineIx, lf->fileName);
partCount = chopString(words[6], ":-", parts, ArraySize(parts));
if (partCount != 3)
errAbort("Bad query range line %d of %s\n", lf->lineIx, lf->fileName);
qStart = atoi(parts[1]);
qEnd = atoi(parts[2]);
strand = words[7][0];
partCount = chopString(words[8], ":-", parts, ArraySize(parts));
if (partCount != 3)
errAbort("Bad target range line %d of %s\n", lf->lineIx, lf->fileName);
tStart = atoi(parts[1]);
tEnd = atoi(parts[2]);
if (!lineFileNext(lf, &line, &lineSize))
errAbort("Unexpected EOF.");
symCount = strlen(line);
qSym = cloneString(line);
if (!lineFileNext(lf, &line, &lineSize))
errAbort("Unexpected EOF.");
tSym = cloneString(line);
if (!lineFileNext(lf, &line, &lineSize))
errAbort("Unexpected EOF.");
hSym = cloneString(line);
if (strand == '+')
xenShowAli(qSym, tSym, hSym, symCount, stdout, qStart, tStart, '+', '+', 60);
else
xenShowAli(qSym, tSym, hSym, symCount, stdout, qEnd, tStart, '-', '+', 60);
freeMem(hSym);
freeMem(tSym);
freeMem(qSym);
}
lineFileClose(&lf);
}
void oneGenieFile(char *fileName, struct hash *uniq, FILE *f)
/* Process one genie peptide prediction file into known and alt tab files. */
{
struct lineFile *lf = lineFileOpen(fileName, TRUE);
char *line;
int lineSize;
boolean firstTime = TRUE;
char *trans;
boolean skip = FALSE;
/* Do cursory sanity check. */
if (!lineFileNext(lf, &line, &lineSize))
errAbort("%s is empty", fileName);
if (line[0] != '>')
errAbort("%s is badly formatted, doesn't begin with '>'", fileName);
lineFileReuse(lf);
while (lineFileNext(lf, &line, &lineSize))
{
if (line[0] == '>')
{
/* End last line. */
if (firstTime)
firstTime = FALSE;
else
fputc('\n', f);
trans = firstWordInLine(line+1);
if (abbr != NULL && startsWith(abbr, trans))
trans += strlen(abbr);
if (hashLookupUpperCase(uniq, trans) != NULL)
{
warn("Duplicate (case insensitive) '%s' line %d of %s. Ignoring all but first.", trans, lf->lineIx, lf->fileName);
skip = TRUE;
}
else
{
char *upperCase;
upperCase = cloneString(trans);
touppers(upperCase);
hashAdd(uniq, upperCase, NULL);
freeMem(upperCase);
fprintf(f, "%s\t", trans);
skip = FALSE;
}
}
else if (!skip)
{
mustWrite(f, line, lineSize-1);
}
}
fputc('\n', f);
lineFileClose(&lf);
}
void genericOne(char *fileName, struct hash *uniq, FILE *f)
/* Process one ensemble peptide prediction file into tab delimited
* output f, using uniq hash to make sure no dupes. */
{
struct lineFile *lf = lineFileOpen(fileName, TRUE);
char *line;
int lineSize;
boolean firstTime = TRUE;
char *trans, transBuf[128];
/* Do cursory sanity check. */
if (!lineFileNext(lf, &line, &lineSize))
errAbort("%s is empty", fileName);
if (line[0] != '>')
errAbort("%s is badly formatted, doesn't begin with '>'", fileName);
lineFileReuse(lf);
while (lineFileNext(lf, &line, &lineSize))
{
if (line[0] == '>')
{
char *upperCase;
/* End last line. */
if (firstTime)
firstTime = FALSE;
else
fputc('\n', f);
trans = firstWordInLine(line+1);
if (abbr != NULL && startsWith(abbr, trans))
trans += strlen(abbr);
if (suffix != NULL)
{
safef(transBuf, sizeof(transBuf), "%s%s", trans, suffix);
trans = transBuf;
}
if (hashLookupUpperCase(uniq, trans) != NULL)
errAbort("Duplicate (case insensitive) '%s' line %d of %s", trans, lf->lineIx, lf->fileName);
upperCase = cloneString(trans);
touppers(upperCase);
hashAdd(uniq, upperCase, NULL);
freeMem(upperCase);
fprintf(f, "%s\t", trans);
}
else
{
mustWrite(f, line, lineSize-1);
}
}
fputc('\n', f);
lineFileClose(&lf);
}
void fastq_stats_read(char *filename, struct fastq_stats *fqs)
/* read from a file */
{
struct lineFile *lf = lineFileOpen(filename, TRUE);
char *line;
char *words[FASTQ_MAX_LEN];
int numWords = 0;
int i, j;
if (!lineFileNext(lf, &line, NULL))
errAbort("bad file, total count section");
fqs->total_count = sqlUnsignedLong(line);
if (!lineFileNext(lf, &line, NULL))
errAbort("bad file, skipped count section");
fqs->all_b = sqlUnsignedLong(line);
if (!lineFileNext(lf, &line, NULL))
errAbort("bad file, skipped count section");
fqs->skipped_count = sqlUnsignedLong(line);
if (!lineFileNext(lf, &line, NULL))
errAbort("bad file, skipped count section");
fqs->kept_count = sqlUnsignedLong(line);
if (!lineFileNext(lf, &line, NULL))
errAbort("bad file, longest read section");
fqs->longest_read = sqlUnsigned(line);
numWords = lineFileChopTab(lf, words);
if (numWords != fqs->longest_read)
errAbort("bad file: original lengths section shoud have %u cols, has %d", fqs->longest_read, numWords);
for (i = 0; i < numWords; i++)
fqs->original_lengths[i] = sqlUnsignedLong(words[i]);
numWords = lineFileChopTab(lf, words);
if (numWords != fqs->longest_read)
errAbort("bad file: trimmed lengths section shoud have %u cols, has %d", fqs->longest_read, numWords);
for (i = 0; i < numWords; i++)
fqs->trimmed_lengths[i] = sqlUnsignedLong(words[i]);
for (i = 0; i <= FASTQ_SANGER_MAX_QUAL; i++)
{
numWords = lineFileChopTab(lf, words);
if (numWords != fqs->longest_read)
errAbort("bad file: before_quals section line %d shoud have %u cols, has %d", i+1, fqs->longest_read, numWords);
for (j = 0; j < numWords; j++)
fqs->before_quals[j][i] = sqlUnsignedLong(words[j]);
}
for (i = 0; i <= FASTQ_SANGER_MAX_QUAL; i++)
{
numWords = lineFileChopTab(lf, words);
if (numWords != fqs->longest_read)
errAbort("bad file: after_quals section line %d shoud have %u cols, has %d", i+1, fqs->longest_read, numWords);
for (j = 0; j < numWords; j++)
fqs->after_quals[j][i] = sqlUnsignedLong(words[j]);
}
lineFileClose(&lf);
}
struct hash *raNextRecord(struct lineFile *lf)
/* Return a hash containing next record.
* Returns NULL at end of file. freeHash this
* when done. Note this will free the hash
* keys and values as well, so you'll have to
* cloneMem them if you want them for later. */
{
struct hash *hash = NULL;
char *line, *key, *val;
/* Skip leading empty lines and comments. */
for (;;)
{
if (!lineFileNext(lf, &line, NULL))
return NULL;
line = skipLeadingSpaces(line);
if (line[0] != 0 )
{
if (line[0] == '#')
continue;
else
break;
}
}
lineFileReuse(lf);
for (;;)
{
if (!lineFileNext(lf, &line, NULL))
break;
line = skipLeadingSpaces(line);
if (line[0] == 0)
break;
if (line[0] == '#')
{
if (startsWith("#EOF", line))
return NULL;
else
continue;
}
if (hash == NULL)
hash = newHash(7);
key = nextWord(&line);
val = trimSpaces(line);
if (line == NULL)
line = "";
val = lmCloneString(hash->lm, val);
hashAdd(hash, key, val);
}
return hash;
}
void headRest(char *countString, char *inName)
/* headRest - Return all *but* the first N lines of a file.. */
{
int count;
int i;
char *line;
struct lineFile *lf = lineFileOpen(inName, TRUE);
if (!isdigit(countString[0]))
usage();
count = atoi(countString);
for (i=0; i<count; ++i)
lineFileNext(lf, &line, NULL);
while (lineFileNext(lf, &line, NULL))
puts(line);
}
void oneEnsFile(char *ensFile, struct hash *uniq, struct hash *pToT, FILE *f)
/* Process one ensemble peptide prediction file into tab delimited
* output f, using uniq hash to make sure no dupes. */
{
struct lineFile *lf = lineFileOpen(ensFile, TRUE);
char *line;
int lineSize;
boolean firstTime = TRUE;
char *translation;
/* Do cursory sanity check. */
if (!lineFileNext(lf, &line, &lineSize))
errAbort("%s is empty", ensFile);
if (line[0] != '>')
errAbort("%s is badly formatted, doesn't begin with '>'", ensFile);
lineFileReuse(lf);
while (lineFileNext(lf, &line, &lineSize))
{
if (line[0] == '>')
{
char *upperCase;
char *transcript;
/* End last line. */
if (firstTime)
firstTime = FALSE;
else
fputc('\n', f);
translation = findEnsTrans(lf, line);
if (hashLookupUpperCase(uniq, translation) != NULL)
errAbort("Duplicate (case insensitive) '%s' line %d of %s", translation, lf->lineIx, lf->fileName);
upperCase = cloneString(translation);
touppers(upperCase);
hashAdd(uniq, upperCase, NULL);
freeMem(upperCase);
transcript = hashFindVal(pToT, translation);
if (transcript == NULL)
errAbort("Can't find transcript for %s", translation);
fprintf(f, "%s\t", transcript);
}
else
{
mustWrite(f, line, lineSize-1);
}
}
fputc('\n', f);
lineFileClose(&lf);
}
struct raft *readRaftFile(char *fileName)
/* Read in a raft file. */
{
struct lineFile *lf = lineFileOpen(fileName, TRUE);
struct raft *raftList = NULL, *raft;
struct raftFrag *rf;
int lineSize, wordCount;
char *line, *words[16];
if (!lineFileNext(lf, &line, &lineSize))
errAbort("%s is empty", fileName);
if (!startsWith("ooGreedy version", line))
errAbort("%s isn't a raft file", fileName);
while (lineFileNext(lf, &line, &lineSize))
{
wordCount = chopLine(line, words);
if (wordCount == 0)
{
raft = NULL;
continue;
}
if (wordCount >= 9 && sameString(words[1], "raft"))
{
AllocVar(raft);
raft->name = cloneString(words[0]);
raft->baseCount = atoi(words[2]);
raft->fragCount = atoi(words[4]);
raft->defaultPos = atoi(words[6]);
slAddHead(&raftList, raft);
}
else if (wordCount == 4 || wordCount == 6)
{
AllocVar(rf);
rf->name = cloneString(words[2]);
rf->pos = atoi(words[0]);
rf->strand = words[1][0];
rf->size = atoi(words[3]);
rf->raft = raft;
slAddTail(&raft->rfList, rf);
}
else
errAbort("Bad line %d of %s", lf->lineIx, lf->fileName);
}
lineFileClose(&lf);
slReverse(&raftList);
return raftList;
}
int addFa(FILE *f, char *ctgFaName)
/* Append contents of FA file. Return the number of bases written. */
{
struct lineFile *lf = lineFileOpen(ctgFaName, TRUE);
int lineSize;
char *line, c;
int recordCount = 0;
int baseCount = 0;
int i;
while (lineFileNext(lf, &line, &lineSize))
{
if (line[0] == '>')
{
++recordCount;
if (recordCount > 1)
warn("More than one record in %s\n", ctgFaName);
}
else
{
for (i=0; i<lineSize; ++i)
{
c = line[i];
if (isalpha(c))
{
charOut(f, c);
baseCount++;
}
}
}
}
lineFileClose(&lf);
return(baseCount);
}
void faSomeRecords(char *faIn, char *listName, char *faOut)
/* faSomeRecords - Extract multiple fa records. */
{
struct hash *hash = hashLines(listName);
char *line, *word;
struct lineFile *lf = lineFileOpen(faIn, TRUE);
FILE *f = mustOpen(faOut, "w");
boolean passMe = FALSE;
while (lineFileNext(lf, &line, NULL))
{
if (line[0] == '>')
{
char *d = cloneString(line);
passMe = gExclude;
line += 1;
word = nextWord(&line);
if (word != NULL)
{
if (hashLookup(hash, word))
passMe = !gExclude;
if (passMe)
fprintf(f, "%s\n", d);
}
freeMem(d);
}
else if (passMe)
{
fprintf(f, "%s\n", line);
}
}
}
boolean isValidKey(char *key)
/*
Verify if a key matches our database key
*/
{
struct lineFile *file = NULL;
char *line = NULL;
int len = 0;
if (NULL == key || strlen(key) == 0)
{
return FALSE;
}
file = lineFileOpen(keyFile, TRUE);
while (lineFileNext(file, &line, &len))
{
if (0 == strcmp(key, line))
{
return TRUE;
}
}
return FALSE;
}
struct mapPos *readInfoFile(char *mapName)
/* Read maps from file. */
{
struct lineFile *lf = lineFileOpen(mapName, TRUE);
int lineSize, wordCount;
char *line, *words[16];
struct mapPos *list = NULL, *el;
lineFileNeedNext(lf, &line, &lineSize);
while (lineFileNext(lf, &line, &lineSize))
{
if (line[0] == '#')
continue;
wordCount = chopLine(line, words);
lineFileExpectWords(lf, 3, wordCount);
AllocVar(el);
el->cloneName = cloneString(words[0]);
el->pos = atoi(words[1]);
el->phase = atoi(words[2]);
slAddHead(&list, el);
}
lineFileClose(&lf);
slReverse(&list);
return list;
}
void jkUniq(char *fileName)
/* Remove dupe lines from file. */
{
struct slName *lineList = NULL, *lineEl;
struct lineFile *lf = lineFileOpen(fileName, TRUE);
char *line;
int lineSize;
struct hash *hash = newHash(0);
FILE *f;
while (lineFileNext(lf, &line, &lineSize))
{
if (!hashLookup(hash, line))
{
hashAdd(hash, line, NULL);
lineEl = newSlName(line);
slAddHead(&lineList, lineEl);
}
}
slReverse(&lineList);
lineFileClose(&lf);
f = mustOpen(fileName, "w");
for (lineEl = lineList; lineEl != NULL; lineEl = lineEl->next)
{
fputs(lineEl->name, f);
fputc('\n', f);
}
fclose(f);
slFreeList(&lineList);
freeHash(&hash);
}
void rt1dFind(char *tabFile, char *treeFile, char *chrom, bits32 start, bits32 end)
/* rt1dCreate - find items in 1-D range tree. */
{
struct lineFile *lf = lineFileOpen(tabFile, TRUE);
struct crTreeFile *crf = crTreeFileOpen(treeFile);
struct fileOffsetSize *block, *blockList = crTreeFindOverlappingBlocks(crf, chrom, start, end);
verbose(2, "Got %d overlapping blocks\n", slCount(blockList));
for (block = blockList; block != NULL; block = block->next)
{
verbose(2, "block->offset %llu, block->size %llu\n", block->offset, block->size);
lineFileSeek(lf, block->offset, SEEK_SET);
bits64 sizeUsed = 0;
while (sizeUsed < block->size)
{
char *line;
int size;
if (!lineFileNext(lf, &line, &size))
errAbort("Couldn't read %s\n", lf->fileName);
char *parsedLine = cloneString(line);
char *row[3];
if (chopLine(parsedLine, row) != ArraySize(row))
errAbort("Badly formatted line of %s\n%s", lf->fileName, line);
char *bedChrom = row[0];
bits32 bedStart = sqlUnsigned(row[1]);
bits32 bedEnd = sqlUnsigned(row[2]);
if (sameString(bedChrom, chrom) && rangeIntersection(bedStart, bedEnd, start, end) > 0)
fprintf(stdout, "%s\n", line);
freeMem(parsedLine);
sizeUsed += size;
}
}
crTreeFileClose(&crf);
}
void readHugoMultiTable(char *fileName, struct hugoMulti **retList,
struct hash **retIdHash, struct hash **retSymbolHash)
/* Read in file into list and hashes. Make hash keyed on omim ID
* and on OMIM symbol. */
{
struct hash *idHash = newHash(0);
struct hash *symbolHash = newHash(0);
struct hugoMulti *list = NULL, *el;
struct lineFile *lf = lineFileOpen(fileName, TRUE);
char *words[16];
char *line;
int lineSize, wordCount;
char *name;
while (lineFileNext(lf, &line, &lineSize))
{
if (line[0] == 0 || line[0] == '#')
continue;
wordCount = chopTabs(line, words);
lineFileExpectWords(lf, 12, wordCount);
el = hugoMultiLoad(words);
slAddHead(&list, el);
name = el->omimId;
if (name[0] != 0)
hashAdd(idHash, name, el);
name = el->symbol;
if (name[0] != 0)
hashAdd(symbolHash, name, el);
}
lineFileClose(&lf);
slReverse(&list);
*retList = list;
*retIdHash = idHash;
*retSymbolHash = symbolHash;
}
char *quotedPrintableDecode(char *input)
/* Use Quoted-Printable standard to decode a string. Return decoded
* string which will be freeMem'd. */
{
size_t inplen = strlen(input);
char *result = (char *)needMem(inplen+1);
size_t j=0;
char *line = NULL;
int size = 0;
int i = 0;
boolean newLine = FALSE;
struct lineFile *lf = lineFileOnString("", TRUE, cloneString(input));
while (lineFileNext(lf, &line, &size))
{
newLine = quotedPCollapse(line);
size = strlen(line);
for (i = 0; i < size; )
result[j++] = line[i++];
if (newLine)
result[j++] = '\n';
}
lineFileClose(&lf); /* frees cloned string */
result[j] = 0; /* terminate text string */
return result;
}
static void h1n1DownloadPdb(char *item, char *pdbUrl, struct tempName *tmpPdb)
/* uncompress PDB to trash */
{
int inFd = netOpenHttpExt(pdbUrl, "GET", NULL);
int inFdRedir = 0;
char *pdbUrlRedir = NULL;
if (!netSkipHttpHeaderLinesHandlingRedirect(inFd, pdbUrl, &inFdRedir, &pdbUrlRedir))
errAbort("Unable to access predicted 3D structure file: %s", pdbUrl);
if (pdbUrlRedir != NULL)
{
close(inFd);
inFd = inFdRedir;
freez(&pdbUrlRedir);
}
trashDirFile(tmpPdb, "hgct", item, ".pdb");
FILE *outFh = mustOpen(tmpPdb->forCgi, "w");
struct lineFile *inLf = lineFileDecompressFd(pdbUrl, TRUE, inFd);
char *line;
while (lineFileNext(inLf, &line, NULL))
{
fputs(line, outFh);
fputc('\n', outFh);
}
lineFileClose(&inLf);
carefulClose(&outFh);
}
void wordUse(char *fileName)
/* wordUse - Make dictionary of all words and count usages, report top ten. */
{
struct lineFile *lf = lineFileOpen(fileName, TRUE);
char *line, *word;
struct hash *hash = newHash(18);
struct wordCount *wcList = NULL, *wc;
int i;
while (lineFileNext(lf, &line, NULL))
{
while ((word = nextWord(&line)) != NULL)
{
wc = hashFindVal(hash, word);
if (wc == NULL)
{
AllocVar(wc);
hashAddSaveName(hash, word, wc, &wc->word);
slAddHead(&wcList, wc);
}
wc->count += 1;
}
}
slSort(&wcList, wordCountCmpCount);
for (i=0, wc=wcList; i<10 && wc != NULL; wc = wc->next, ++i)
printf("%s\t%d\n", wc->word, wc->count);
}
struct bed *loadBedFileWithHeader(char *fileName)
/* Read in a bed file into a bed list, dealing with header for custom track if necessary. */
{
struct bed *bedList = NULL, *bed = NULL;
struct lineFile *lf = lineFileOpen(fileName, TRUE);
char *row[12];
int lineSize;
char *line;
/* Skip the headers. */
while(lineFileNext(lf, &line, &lineSize))
{
if(countChars(line, '\t') > 10)
{
lineFileReuse(lf);
break;
}
}
/* Load in the records. */
while(lineFileRow(lf, row))
{
bed = bedLoad12(row);
slAddHead(&bedList, bed);
}
lineFileClose(&lf);
slReverse(&bedList);
return bedList;
}
struct raRecord *raNext(struct lineFile *lf, struct hash *stringHash)
/* Return next ra record. Returns NULL at end of file. */
{
char *line, *key, *val;
struct raRecord *ra = NULL;
struct nameVal *nv;
for (;;)
{
if (!lineFileNext(lf, &line, NULL))
break;
line = skipLeadingSpaces(line);
if (line[0] == '#')
continue;
if (line[0] == 0)
break;
if (ra == NULL)
AllocVar(ra);
key = nextWord(&line);
AllocVar(nv);
nv->name = hashStoreName(stringHash, key);
val = skipLeadingSpaces(line);
if (val == NULL)
errAbort("Expecting line/value pair line %d of %s"
, lf->lineIx, lf->fileName);
nv->val = hashStoreName(stringHash, val);
slAddHead(&ra->fieldList, nv);
}
if (ra != NULL)
slReverse(&ra->fieldList);
return ra;
}
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 fixGdup(char *inName, char *outName)
/* fixGdup - Reformat genomic dups table a little.. */
{
struct lineFile *lf = lineFileOpen(inName, TRUE);
FILE *f = mustOpen(outName, "w");
int wordCount, lineSize;
char *words[32], *line;
int i;
while (lineFileNext(lf, &line, &lineSize))
{
if (line[0] == '#')
continue;
wordCount = chopTabs(line, words);
if (wordCount == 0)
continue;
lineFileExpectWords(lf, 15, wordCount);
for (i=0; i<3; ++i)
fprintf(f, "%s\t", words[i]);
fprintf(f, "%s:%s\t", words[6], words[7]);
for (i=4; i<9; ++i)
fprintf(f, "%s\t", words[i]);
for (i=10; i<wordCount; ++i)
{
fprintf(f, "%s", words[i]);
if (i == wordCount-1)
fprintf(f, "\n");
else
fprintf(f, "\t");
}
}
}
void addCtgFile(char *liftFileName, struct ctgPos **pCtgList)
/* Create ctgPos's out of liftSpecs in liftFile. */
{
struct lineFile *lf = lineFileOpen(liftFileName, TRUE);
int lineSize, wordCount;
char *line, *words[16];
struct liftSpec lift;
struct ctgPos *ctg;
printf("Processing %s\n", liftFileName);
while (lineFileNext(lf, &line, &lineSize))
{
wordCount = chopLine(line, words);
if (wordCount == 0)
continue;
if (wordCount != 5)
errAbort("Expecting 5 words line %d of %s", lf->lineIx, lf->fileName);
liftSpecStaticLoad(words, &lift);
AllocVar(ctg);
ctg->contig = cloneString(skipPastSlash(lift.oldName));
ctg->size = lift.oldSize;
ctg->chrom = cloneString(lift.newName);
ctg->chromStart = lift.offset;
ctg->chromEnd = lift.offset + lift.oldSize;
slAddHead(pCtgList, ctg);
}
lineFileClose(&lf);
}
void doCleanSeq(char *inputFileName, char *outputFileName)
{
FILE *outputFileHandle = NULL;
struct lineFile *lf;
char *line;
char *row[9], *rsId[2];
struct hashEl *hel = NULL;
boolean skipping = FALSE;
outputFileHandle = mustOpen(outputFileName, "w");
lf = lineFileOpen(inputFileName, TRUE);
while (lineFileNext(lf, &line, NULL))
{
if (line[0] == '>')
{
skipping = FALSE;
chopString(line, "|", row, ArraySize(row));
chopString(row[2], " ", rsId, ArraySize(rsId));
hel = hashLookup(snpHash, rsId[0]);
if (hel)
skipping = TRUE;
else
{
hashAdd(snpHash, cloneString(rsId[0]), NULL);
fprintf(outputFileHandle, ">%s\n", rsId[0]);
}
}
else if (!skipping)
fprintf(outputFileHandle, "%s\n", line);
}
carefulClose(&outputFileHandle);
lineFileClose(&lf);
}
void netLineFilter(struct lineFile *lf, FILE *f)
/* Do filter one line at a time. */
{
struct hash *nameHash = newHash(0);
char *line, *l;
int d;
while (lineFileNext(lf, &line, NULL))
{
d = countLeadingChars(line, ' ');
l = line + d;
if (startsWith("fill", l) || startsWith("gap", l))
{
struct cnFill *fill = cnFillFromLine(nameHash, lf, l);
if (filterOne(fill))
cnFillWrite(fill, f, d);
cnFillFree(&fill);
}
else
{
fprintf(f, "%s\n", line);
}
}
hashFree(&nameHash);
}
struct dyString *readAndReplaceTableName(char *fileName, char *table)
/* Read file into string. While doing so strip any leading comments
* and insist that the first non-comment line contain the words
* "create table" followed by a table name. Replace the table name,
* and copy the rest of the file verbatem. */
{
struct lineFile *lf = lineFileOpen(fileName, TRUE);
struct dyString *dy = dyStringNew(0);
char *line, *word;
if (!lineFileNextReal(lf, &line))
errAbort("No real lines in %s\n", fileName);
word = nextWord(&line);
if (!sameWord(word, "create"))
errAbort("Expecting first word in file to be CREATE. Got %s", word);
word = nextWord(&line);
if (word == NULL || !sameWord(word, "table"))
errAbort("Expecting second word in file to be table. Got %s", emptyForNull(word));
word = nextWord(&line);
if (word == NULL)
errAbort("Expecting table name on same line as CREATE TABLE");
sqlDyStringPrintf(dy, "CREATE TABLE %s ", table);
if (line != NULL)
dyStringAppend(dy, line);
dyStringAppendC(dy, '\n');
while (lineFileNext(lf, &line, NULL))
{
dyStringAppend(dy, line);
dyStringAppendC(dy, '\n');
}
lineFileClose(&lf);
return dy;
}
void wigSetSpanOrStep(char *input, char *output)
/* wigSetSpanOrStep - Set span or step variables in an ascii format wiggle file. */
{
struct lineFile *lf = lineFileOpen(input, TRUE);
FILE *f = mustOpen(output, "w");
char *line;
int lineSize;
char *fixedPat = "fixedStep";
int fixedPatSize = strlen(fixedPat);
char *varPat = "variableStep";
int varPatSize = strlen(varPat);
while (lineFileNext(lf, &line, &lineSize))
{
if (startsWithWord(fixedPat, line))
handleStepLine(lf, line, fixedPatSize, FALSE, f);
else if (startsWithWord(varPat, line))
handleStepLine(lf, line, varPatSize, TRUE, f);
else
{
mustWrite(f, line, lineSize-1);
fputc('\n', f);
}
}
carefulClose(&f);
}
