static
rc_t read_quality(const char* data, size_t data_sz, IlluminaRead* read)
{
rc_t rc = 0;
if( (rc = pstring_assign(&read->qual, data, data_sz)) == 0 ) {
if( (rc = pstring_quality_convert(&read->qual, eExperimentQualityEncoding_Decimal, 0, -128, 127)) == 0 ) {
read->qual_type = ILLUMINAWRITER_COLMASK_QUALITY_LOGODDS4;
}
}
return rc;
}
/* reads from a file data for a sinlge spot, data may be partial */
static
rc_t read_next_spot(FastqLoaderFmt* self, FastqFileInfo* file)
{
rc_t rc = 0;
if( file->spot->ready ) {
/* data still not used */
return 0;
}
FileReadData_init(file->spot, false);
FileReadData_init(&file->spot[1], false);
if( (rc = file_read_line(file, true)) != 0 ) {
return SRALoaderFile_LOG(file->file, klogErr, rc, "$(msg)", "msg=reading more data");
} else if( file->line == NULL ) {
return 0; /* eof */
}
if( find_seq_qual_by_sep(self, file, ':') || find_seq_qual_by_sep(self, file, ' ') ) {
/* single line forms */
file->line = NULL; /* line consumed */
file->spot->ready = true;
} else if( file->line[0] == '>' || file->line[0] == '@' ) {
/* 4 or 8 line format */
FileReadData sd;
uint8_t word = 0, best_word = 0;
uint8_t score = 0, best_score = 0;
/* find and parse spot name on defline */
do {
score = parse_spot_name(file->file, &sd, &file->line[1], file->line_len - 1, ++word);
if( score > best_score ) {
if( (rc = pstring_copy(&file->spot->name, &sd.name)) != 0 ||
(rc = pstring_copy(&file->spot->barcode, &sd.barcode)) != 0 ) {
return SRALoaderFile_LOG(file->file, klogErr, rc, "$(msg)", "msg=copying read name");
}
file->spot->read.read_id = sd.read.read_id;
best_score = score;
best_word = word; /* used below for quality defline parsing */
}
} while(score != 0);
if( best_score == 0 ) {
rc = RC(rcSRA, rcFormatter, rcReading, rcId, rcNotFound);
return SRALoaderFile_LOG(file->file, klogErr, rc, "$(msg)", "msg=spot name not detected");
}
if( file->line[0] == '@' ) {
if( (rc = read_spot_data_3lines(file, file->spot, best_word, best_score, file->qualType)) != 0 ) {
return rc;
}
/* now we MAY have 5th line in buffer so we can check if it's 2nd read for 8 lines format */
if( file->line_len != 0 && file->line != NULL && file->line[0] == '@' ) {
/* try to find read id on next line */
FileReadData_init(&file->spot[1], false);
if( parse_spot_name(file->file, &file->spot[1], &file->line[1], file->line_len - 1, best_word) == best_score ) {
if( pstring_cmp(&file->spot->name, &file->spot[1].name) == 0 &&
pstring_cmp(&file->spot->barcode, &file->spot[1].barcode) == 0 &&
file->spot->read.read_id != file->spot[1].read.read_id ) {
/* since it is different read id with same name and barcode, fill up second read */
if( (rc = read_spot_data_3lines(file, &file->spot[1], best_word, best_score, file->qualType)) != 0 ) {
return rc;
}
}
}
}
} else {
/* 2 line seq or quality form */
file->line = NULL; /* line consumed */
/* read sequence/quality */
if( (rc = read_multiline_seq_or_qual(file, '>', &file->spot->read.seq)) != 0 ) {
return SRALoaderFile_LOG(file->file, klogErr, rc, "$(msg)", "msg=reading seq/qual data");
}
if( file->spot->read.seq.len == 0 ) {
return SRALoaderFile_LOG(file->file, klogErr, rc, "$(msg)", "msg=empty string reading seq/qual data");
} else if( !pstring_is_fasta(&file->spot->read.seq) ) {
/* swap */
if( (rc = pstring_copy(&file->spot->read.qual, &file->spot->read.seq)) == 0 ) {
file->spot->read.qual_type = file->qualType;
pstring_clear(&file->spot->read.seq);
}
}
file->spot->ready = true;
}
} else {
rc = RC(rcSRA, rcFormatter, rcReading, rcFile, rcInvalid);
return SRALoaderFile_LOG(file->file, klogErr, rc, "$(msg)", "msg=file corrupt or format unknown");
}
if( rc == 0 ) {
int k;
for(k = 0; k < 2; k++) {
FileReadData* rd = &file->spot[k];
if( rd->ready && rd->read.qual_type != ILLUMINAWRITER_COLMASK_NOTSET ) {
if( file->qualOffset == 0 ) {
/* detect and remember */
file->qualOffset = 33;
file->qualMax = 94;
rc = pstring_quality_convert(&rd->read.qual, file->qualEnc, file->qualOffset, file->qualMin, file->qualMax);
if( GetRCState(rc) == rcOutofrange ) {
file->qualOffset = 64;
file->qualMax = 61;
rc = pstring_quality_convert(&rd->read.qual, file->qualEnc, file->qualOffset, file->qualMin, file->qualMax);
}
} else {
if(file->qualOffset == 33) file->qualMax = 94;
rc = pstring_quality_convert(&rd->read.qual, file->qualEnc, file->qualOffset, file->qualMin, file->qualMax);
}
if( rc != 0 ) {
return SRALoaderFile_LOG(file->file, klogErr, rc, "$(msg)", "msg=converting quality");
}
}
}
}
return 0;
}
/*
* assumes tab separated file:
* first 2 postiions concatinated with "_" into spot prefix
* nextg 4 postiions concatinated with ":" into spot id: lane:tile:x:y
* 7th (index) ignored
* 8th is read id
* 9th fasta
* 10th quality
* 11th (optional) read filter
*/
static
rc_t parse_qseq(IlluminaFileInfo* file, const char* data, size_t data_sz)
{
rc_t rc = 0;
const char* t, *str = data, *end = data + data_sz;
int tabs = 0;
do {
if( (t = memchr(str, '\t', end - str)) != NULL ) {
switch(++tabs) {
case 1:
rc = pstring_assign(&file->name, str, t - str);
break;
case 2:
if( (rc = pstring_append(&file->name, "_", 1)) == 0 ) {
rc = pstring_append(&file->name, str, t - str);
}
break;
case 3:
case 4:
case 5:
case 6:
errno = 0;
file->coord[tabs - 3] = strtol(str, NULL, 10);
if( errno != 0 ) {
file->coord[tabs - 3] = 0;
}
if( (rc = pstring_append(&file->name, ":", 1)) == 0 ) {
rc = pstring_append(&file->name, str, t - str);
}
break;
case 7:
if( t - str != 1 || (*str != '0' && *str != '1') ) {
rc = pstring_assign(&file->barcode, str, t - str);
}
break;
case 8:
if( t - str != 1 || !isdigit(*str) ) {
rc = RC(rcSRA, rcFormatter, rcReading, rcData, rcInvalid);
} else {
file->read.read_id = *str - '0';
if( file->read.read_id == 0 ) {
file->read.read_id = ILLUMINAWRITER_READID_NONE;
}
}
break;
case 9:
rc = pstring_assign(&file->read.seq, str, t - str);
break;
case 10:
file->read.qual_type = ILLUMINAWRITER_COLMASK_QUALITY_PHRED;
rc = pstring_assign(&file->read.qual, str, t - str);
break;
}
str = ++t;
}
} while( rc == 0 && t != NULL && str < end );
if( rc == 0 ) {
if( tabs == 9 ) {
file->read.qual_type = ILLUMINAWRITER_COLMASK_QUALITY_PHRED;
rc = pstring_assign(&file->read.qual, str, end - str);
} else if( tabs == 10 ) {
if( end - str != 1 ) {
rc = RC(rcSRA, rcFormatter, rcReading, rcData, rcInvalid);
} else if( *str == '1' ) {
file->read.filter = SRA_READ_FILTER_PASS;
} else if( *str == '0' ) {
file->read.filter = SRA_READ_FILTER_REJECT;
} else {
rc = RC(rcSRA, rcFormatter, rcReading, rcData, rcInvalid);
}
} else {
rc = RC(rcSRA, rcFormatter, rcReading, rcData, rcInvalid);
}
if( rc == 0 ) {
if( file->read.seq.len != file->read.qual.len ) {
rc = RC(rcSRA, rcFormatter, rcReading, rcData, rcInconsistent);
} else {
rc = pstring_quality_convert(&file->read.qual, eExperimentQualityEncoding_Ascii, 64, 0, 0x7F);
}
}
}
return rc;
}
/* reads from a file data for a sinlge spot, data may be partial */
static
rc_t read_next_spot(HelicosLoaderFmt* self, HelicosFileInfo* file)
{
rc_t rc = 0;
if( file->ready ) {
/* data still not used */
return 0;
}
HelicosFileInfo_init(file);
if( (rc = file_read_line(file, true)) != 0 ) {
return SRALoaderFile_LOG(file->file, klogErr, rc, "$(msg)", "msg=reading more data");
} else if( file->line == NULL ) {
return 0; /* eof */
}
if( file->line[0] == '@' ) { /*** fastq format **/
if( (rc = pstring_assign(&file->name, &file->line[1], file->line_len - 1)) != 0 ) {
return SRALoaderFile_LOG(file->file, klogErr, rc, "$(msg)", "msg=reading name");
}
file->line = NULL;
if( (rc = file_read_line(file, false)) != 0 || file->line_len > sizeof(file->sequence.data)-1 ||
(rc = pstring_assign(&file->sequence, file->line, file->line_len)) != 0 ||
!pstring_is_fasta(&file->sequence) ) {
rc = rc ? rc : RC(rcSRA, rcFormatter, rcReading, rcData, rcUnrecognized);
return SRALoaderFile_LOG(file->file, klogErr, rc, "$(msg)", "msg=reading sequence");
}
file->line = NULL;
if( (rc = file_read_line(file, false)) != 0 ||
file->line[0] != '+' || file->line_len != 1 ) {
rc = RC(rcSRA, rcFormatter, rcReading, rcData, rcCorrupt);
return SRALoaderFile_LOG(file->file, klogErr, rc, "$(msg)", "msg=reading quality defline");
}
file->line = NULL;
if( (rc = file_read_line(file, false)) != 0 || file->line_len > sizeof(file->quality.data)-1 ||
(rc = pstring_assign(&file->quality, file->line, file->line_len)) != 0 ||
(rc = pstring_quality_convert(&file->quality, eExperimentQualityEncoding_Ascii, 33, 0, 0x7F)) != 0 ) {
return SRALoaderFile_LOG(file->file, klogErr, rc, "$(msg)", "msg=reading quality");
}
file->line = NULL;
file->ready = true;
} else if( file->line[0] == '>' ) { /** fasta format **/
if( (rc = pstring_assign(&file->name, &file->line[1], file->line_len - 1)) != 0 ) {
return SRALoaderFile_LOG(file->file, klogErr, rc, "$(msg)", "msg=reading name");
}
file->line = NULL;
if( (rc = file_read_line(file, false)) != 0 || file->line_len > sizeof(file->sequence.data)-1 ||
(rc = pstring_assign(&file->sequence, file->line, file->line_len)) != 0 ||
!pstring_is_fasta(&file->sequence) ) {
rc = rc ? rc : RC(rcSRA, rcFormatter, rcReading, rcData, rcUnrecognized);
return SRALoaderFile_LOG(file->file, klogErr, rc, "$(msg)", "msg=reading sequence");
}
file->line = NULL;
file->quality.len = file->sequence.len;
memset(file->quality.data,14,file->quality.len);
file->ready = true;
} else {
rc = RC(rcSRA, rcFormatter, rcReading, rcFile, rcInvalid);
return SRALoaderFile_LOG(file->file, klogErr, rc, "$(msg)", "msg=expected '@'");
}
#if _DEBUGGING
DEBUG_MSG(3, ("READ: name:'%s', seq[%u]:'%s', qual[%u]\n", file->name.data,
file->sequence.len, file->sequence.data, file->quality.len)); /*
DEBUG_MSG(3, ("READ: name:'%s', seq[%u]:'%s', qual[%u]:'%s'\n", file->name.data,
file->sequence.len, file->sequence.data, file->quality.len, file->quality.data));*/
#endif
return 0;
}
