Beispiel #1
0
static
rc_t parse_read(SRF_context *ctx, ZTR_Context *ztr_ctx, const uint8_t *data, size_t size)
{
    rc_t rc = 0;
    size_t parsed;
    uint8_t flags;
    pstring readId;
    ztr_raw_t ztr_raw;
    ztr_t ztr;
    enum ztr_chunk_type type;
    fe_context_t* fe = (fe_context_t*)ctx;

    *(void **)&fe->sequence =
    *(void **)&fe->quality1 =
    *(void **)&fe->quality4 =
    *(void **)&fe->signal =
    *(void **)&fe->intensity = 
    *(void **)&fe->noise = NULL;
    
    rc = SRF_ParseReadChunk(data, size, &parsed, &flags, &readId);
    if(rc) {
        rc = RC(rcSRA, rcFormatter, rcParsing, rcData, rc);
        return SRALoaderFile_LOG(ctx->file, klogErr, rc, "corrupt", NULL);
    }
    if(fe->defered != NULL)
        ZTR_AddToBuffer(ztr_ctx, fe->defered, fe->defered_len);
    ZTR_AddToBuffer(ztr_ctx, data + parsed, size - parsed);
    if(fe->defered == NULL) {
        rc = ZTR_ParseBlock(ztr_ctx, &ztr_raw);
        if(rc == 0)
            goto PARSE_BLOCK;
        rc = ZTR_ParseHeader(ztr_ctx);
        if(rc) {
            return SRALoaderFile_LOG(ctx->file, klogErr, rc, "corrupt", NULL);
        }
    }
    
    while (!ZTR_BufferIsEmpty(ztr_ctx)) {
        rc = ZTR_ParseBlock(ztr_ctx, &ztr_raw);
    PARSE_BLOCK:
        if(rc != 0 || (rc = ZTR_ProcessBlock(ztr_ctx, &ztr_raw, &ztr, &type)) != 0 ) {
            return SRALoaderFile_LOG(ctx->file, klogErr, rc, "corrupt", NULL);
        }
        
        switch (type) {
            case READ:
                if(ztr.sequence->datatype != i8) {
                    rc = RC(rcSRA, rcFormatter, rcParsing, rcData, rcUnexpected);
                    return SRALoaderFile_LOG(ctx->file, klogErr, rc, "invalid data type for sequence data", NULL);
                }
                fe->sequence = ztr;
                break;
            case QUALITY1:
                if(ztr.quality1->datatype != i8) {
                    rc = RC(rcSRA, rcFormatter, rcParsing, rcData, rcUnexpected);
                    return SRALoaderFile_LOG(ctx->file, klogErr, rc, "invalid data type for quality1 data", NULL);
                }
                fe->quality1 = ztr;
                break;
            case QUALITY4:
                if(ztr.quality4->datatype != i8) {
                    rc = RC(rcSRA, rcFormatter, rcParsing, rcData, rcUnexpected);
                    return SRALoaderFile_LOG(ctx->file, klogErr, rc, "invalid data type for quality4 data", NULL);
                }
                fe->quality4 = ztr;
                break;
            case SIGNAL4:
                if(ztr.signal4->Type != NULL && strncmp(ztr.signal4->Type, "SLXI", 4) == 0 ) {
                    if( !fe->skip_intensity ) {
                        fe->intensity = ztr;
                    } else if(ztr.signal4){
			if(ztr.signal4->data) free(ztr.signal4->data);
			free(ztr.signal4);
		    }
                } else if(ztr.signal4->Type != NULL && strncmp(ztr.signal4->Type, "SLXN", 4) == 0 ) {
                    if( !fe->skip_noise ) {
                        fe->noise = ztr;
                    } else if(ztr.signal4){
			if(ztr.signal4->data) free(ztr.signal4->data);
			free(ztr.signal4);
                    }
                } else if( !fe->skip_signal ) {
                    fe->signal = ztr;
		} else if(ztr.signal4){
			if(ztr.signal4->data) free(ztr.signal4->data);
			free(ztr.signal4);
                }
                break;

            default:
                free(*(void **)&ztr);

            case none:
            case ignore:
                if(ztr_raw.data) {
                    free(ztr_raw.data);
                }
                break;
        }
	if(ztr_raw.meta){
		free(ztr_raw.meta);
		ztr_raw.meta=NULL;
	}
    }
    
    while(rc == 0) {
        if(*(void **)&fe->sequence == NULL) {
            rc = RC(rcSRA, rcFormatter, rcParsing, rcConstraint, rcViolated);
            SRALoaderFile_LOG(ctx->file, klogErr, rc, "missing sequence data", NULL);
            break;
        }
        if(*(void **)&fe->quality4 == NULL && *(void **)&fe->quality1 == NULL) {
            rc = RC(rcSRA, rcFormatter, rcParsing, rcConstraint, rcViolated);
            SRALoaderFile_LOG(ctx->file, klogErr, rc, "missing quality data", NULL);
            break;
        }

        if( (rc = ILL_ZTR_Decompress(ztr_ctx, BASE, fe->sequence, fe->sequence)) != 0 ||
            (rc = pstring_assign(&fe->read.seq, fe->sequence.sequence->data, fe->sequence.sequence->datasize)) != 0 ) {
            SRALoaderFile_LOG(ctx->file, klogErr, rc, "failed to decompress sequence data", NULL);
            break;
        }
        
        if( *(void **)&fe->quality4 != NULL ) {
            if( (rc = ILL_ZTR_Decompress(ztr_ctx, CNF4, fe->quality4, fe->sequence)) != 0 ||
                (rc = pstring_assign(&fe->read.qual, fe->quality4.quality4->data, fe->quality4.quality4->datasize)) != 0 ) {
                SRALoaderFile_LOG(ctx->file, klogErr, rc, "failed to decompress quality4 data", NULL);
                break;
            }
            fe->read.qual_type = ILLUMINAWRITER_COLMASK_QUALITY_LOGODDS4;
        } else if( *(void **)&fe->quality1 != NULL ) {
            if( (rc = ILL_ZTR_Decompress(ztr_ctx, CNF1, fe->quality1, fe->sequence)) != 0 ||
                (rc = pstring_assign(&fe->read.qual, fe->quality1.quality1->data, fe->quality1.quality4->datasize)) != 0 ) {
                SRALoaderFile_LOG(ctx->file, klogErr, rc, "failed to decompress quality1 data", NULL);
                break;
            }
            fe->read.qual_type = ILLUMINAWRITER_COLMASK_QUALITY_PHRED;
        }
        if( *(void **)&fe->signal != NULL ) {
            if( (rc = ILL_ZTR_Decompress(ztr_ctx, SMP4, fe->signal, fe->sequence)) != 0 ||
                (rc = pstring_assign(&fe->read.signal, fe->signal.signal4->data, fe->signal.signal4->datasize)) != 0 ) {
                SRALoaderFile_LOG(ctx->file, klogErr, rc, "failed to decompress signal data", NULL);
                break;
            }
        }
        if( *(void **)&fe->intensity != NULL ) {
            if( (rc = ILL_ZTR_Decompress(ztr_ctx, SMP4, fe->intensity, fe->sequence)) != 0 ||
                (rc = pstring_assign(&fe->read.intensity, fe->intensity.signal4->data, fe->intensity.signal4->datasize)) != 0 ) {
                SRALoaderFile_LOG(ctx->file, klogErr, rc, "failed to decompress intensity data", NULL);
                break;
            }
        }
        if( *(void **)&fe->noise != NULL ) {
            if( (rc = ILL_ZTR_Decompress(ztr_ctx, SMP4, fe->noise, fe->sequence)) != 0 ||
                (rc = pstring_assign(&fe->read.noise, fe->noise.signal4->data, fe->noise.signal4->datasize)) != 0 ) {
                SRALoaderFile_LOG(ctx->file, klogErr, rc, "failed to decompress noise data", NULL);
                break;
            }
        }
        rc = fe_new_read(fe, flags, &readId);
        break;
    }
    if(fe->sequence.sequence) {
        if(fe->sequence.sequence->data)
            free(fe->sequence.sequence->data);
        free(fe->sequence.sequence);
    }
    if(fe->quality1.quality1) {
        if(fe->quality1.quality1->data)
            free(fe->quality1.quality1->data);
        free(fe->quality1.quality1);
    }
    if(fe->quality4.quality4) {
        if(fe->quality4.quality4->data)
            free(fe->quality4.quality4->data);
        free(fe->quality4.quality4);
    }
    if(fe->signal.signal4) {
        if(fe->signal.signal4->data)
            free(fe->signal.signal4->data);
        free(fe->signal.signal4);
    }
    if(fe->intensity.signal4) {
        if(fe->intensity.signal4->data)
            free(fe->intensity.signal4->data);
        free(fe->intensity.signal4);
    }
    if(fe->noise.signal4) {
        if(fe->noise.signal4->data)
            free(fe->noise.signal4->data);
        free(fe->noise.signal4);
    }
    return rc;
}
Beispiel #2
0
static
rc_t parse_v1_read(SRF_context *ctx, ZTR_Context *ztr_ctx, const uint8_t *data, size_t size)
{
    rc_t rc = 0;
    size_t i, parsed;
    ztr_raw_t ztr_raw;
    ztr_t ztr;
    enum ztr_chunk_type type;
    fe_context_t* fe = (fe_context_t*)ctx;

    uint8_t flags;
    pstring readId;
    EAbisolidReadType read_type;
    pstring label;

    AbsolidRead read[ABSOLID_FMT_MAX_NUM_READS];
        
    if( fe->region.nreads == 0 ) {
        rc = RC(rcSRA, rcFormatter, rcParsing, rcData, rcNotFound);
        return SRALoaderFile_LOG(ctx->file, klogErr, rc, "missing region chunk before 1st read chunk", NULL);
    }
    if( (rc = SRF_ParseReadChunk(data, size, &parsed, &flags, &readId)) != 0 ) {
        rc = RC(rcSRA, rcFormatter, rcParsing, rcData, rc);
        return SRALoaderFile_LOG(ctx->file, klogErr, rc, "SRF parsing failure", NULL);
    }
    ABI_ZTR_AddToBuffer(ztr_ctx, data + parsed, size - parsed);

    /* readId will have spotname */
    if( (rc = fe_new_read(fe, &readId, &read_type, &label)) != 0 ) {
        return SRALoaderFile_LOG(ctx->file, klogErr, rc, "parsing spot name suffix", NULL);
    }
    for(i = 0; i < sizeof(read) / sizeof(read[0]); i++) {
        AbsolidRead_Init(&read[i]);
    }
    while(!ABI_ZTR_BufferIsEmpty(ztr_ctx)) {
        if( (rc = ABI_ZTR_ParseBlock(ztr_ctx, &ztr_raw)) != 0 ||
            (rc = ABI_ZTR_ProcessBlock(ztr_ctx, &ztr_raw, &ztr, &type)) != 0 ) {
            SRALoaderFile_LOG(ctx->file, klogErr, rc, "ZTR parsing failure", NULL);
            break;
        }
        switch (type) {
        case BASE:
            if(ztr.sequence->datatype != i8) {
                rc = RC(rcSRA, rcFormatter, rcParsing, rcData, rcUnexpected);
                SRALoaderFile_LOG(ctx->file, klogErr, rc, "read: expected 8-bit datatype", NULL);
            } else if( read_type > eAbisolidReadType_SPOT ) {
                int read_number = AbisolidReadType2ReadNumber[read_type];
                if( (rc = pstring_assign(&read[read_number].seq, ztr.sequence->data, ztr.sequence->datasize)) == 0 ) {
                    /* grab 1st, may be the only cs_key */
                    read[read_number].cs_key = fe->region.cs_key[0];
                    for(i = 1; i < fe->region.nreads; i++) {
                        if( read_type == fe->region.type[i] ) {
                            read[read_number].cs_key = fe->region.cs_key[i];
                            break;
                        }
                    }
                    SRF_set_read_filter(&read[read_number].filter, flags);
                    rc = pstring_copy(&read[read_number].label, &label);
                    DEBUG_MSG(3, ("SRF READ: '%s'\n", read[read_number].seq.data));
                }
                if( rc != 0 ) {
                    SRALoaderFile_LOG(ctx->file, klogErr, rc, "copying read", NULL);
                }
            } else {
                for(i = 0; rc == 0 && i < fe->region.nreads; i++) {
                    int read_number = AbisolidReadType2ReadNumber[fe->region.type[i]];
                    size_t len = (i + 1 >= fe->region.nreads ? ztr.sequence->datasize : fe->region.start[i + 1]) - fe->region.start[i];
                    rc = pstring_assign(&read[read_number].seq, &ztr.sequence->data[fe->region.start[i]], len);
                    read[read_number].cs_key = fe->region.cs_key[i];
                    SRF_set_read_filter(&read[read_number].filter, flags);
                    if( fe->region.label[i].len != 0 ) {
                        rc = pstring_copy(&read[read_number].label, &fe->region.label[i]);
                    }
                    DEBUG_MSG(3, ("SRF READ[%u]: '%s'\n", i, read[read_number].seq.data));
                }
                if( rc != 0 ) {
                    SRALoaderFile_LOG(ctx->file, klogErr, rc, "copying reads", NULL);
                }
            }
            break;
        case CNF1:
            if(ztr.quality1->datatype != i8) {
                rc = RC(rcSRA, rcFormatter, rcParsing, rcData, rcUnexpected);
                SRALoaderFile_LOG(ctx->file, klogErr, rc, "quality: expected 8-bit datatype", NULL);
            } else if( read_type > eAbisolidReadType_SPOT ) {
                int read_number = AbisolidReadType2ReadNumber[read_type];
                if( (rc = pstring_assign(&read[read_number].qual, ztr.quality1->data, ztr.quality1->datasize)) == 0 ) {
                    DEBUG_MSG(3, ("SRF QUAL: %u bytes\n", read[read_number].qual.len));
                }
                if( rc != 0 ) {
                    SRALoaderFile_LOG(ctx->file, klogErr, rc, "copying quality", NULL);
                }
            } else {
                for(i = 0; rc == 0 && i < fe->region.nreads; i++) {
                    int read_number = AbisolidReadType2ReadNumber[fe->region.type[i]];
                    size_t len = (i + 1 >= fe->region.nreads ? ztr.quality1->datasize : fe->region.start[i + 1]) - fe->region.start[i];
                    rc = pstring_assign(&read[read_number].qual, &ztr.quality1->data[fe->region.start[i]], len);
                    DEBUG_MSG(3, ("SRF QUAL[%u]: %u bytes\n", i, read[read_number].qual.len));
                }
                if( rc != 0 ) {
                    SRALoaderFile_LOG(ctx->file, klogErr, rc, "copying qualities", NULL);
                }
            }
            break;
        case SAMP:
            if( !fe->skip_signal ) {
                size_t i;
                int stype = ABSOLID_FMT_COLMASK_NOTSET;
                if(ztr.signal->datatype != f32) {
                    rc = RC(rcSRA, rcFormatter, rcParsing, rcData, rcUnexpected);
                    SRALoaderFile_LOG(ctx->file, klogErr, rc, "signal: expected 32-bit float datatype", NULL);
                } else if( (ztr.signal->datasize % sizeof(float)) != 0 ) {
                    rc = RC(rcSRA, rcFormatter, rcParsing, rcData, rcInvalid);
                    SRALoaderFile_LOG(ctx->file, klogErr, rc, "signal: size not 32-bit float aligned", NULL);
                } else if (ztr.signal->channel == NULL) {
                    rc = RC(rcSRA, rcFormatter, rcParsing, rcData, rcIncomplete);
                    SRALoaderFile_LOG(ctx->file, klogErr, rc, "SIGNAL column: missing channel type", NULL);
                } else if(strcmp(ztr.signal->channel, "0FAM") == 0) {
                    stype = ABSOLID_FMT_COLMASK_FAM;
                } else if(strcmp(ztr.signal->channel, "1CY3") == 0) {
                    stype = ABSOLID_FMT_COLMASK_CY3;
                } else if(strcmp(ztr.signal->channel, "2TXR") == 0) {
                    stype = ABSOLID_FMT_COLMASK_TXR;
                } else if(strcmp(ztr.signal->channel, "3CY5") == 0) {
                    stype = ABSOLID_FMT_COLMASK_CY5;
                } else {
                    rc = RC(rcSRA, rcFormatter, rcParsing, rcData, rcUnexpected);
                    SRALoaderFile_LOG(ctx->file, klogErr, rc, "SIGNAL column: unexpected channel type", NULL);
                }
#if __BYTE_ORDER == __LITTLE_ENDIAN
                for(i = 0; rc == 0 && i < ztr.signal->datasize; i += 4) {
                    uint32_t* r = (uint32_t*)&ztr.signal->data[i];
                    *r = bswap_32(*r);
                }
#endif
                if( rc == 0 ) {
                    if( read_type > eAbisolidReadType_SPOT ) {
                        int read_number = AbisolidReadType2ReadNumber[read_type];
                        pstring* d = NULL;
                        switch(stype) {
                            case ABSOLID_FMT_COLMASK_FAM:
                                read[read_number].fs_type = eAbisolidFSignalType_FAM;
                                d = &read[read_number].fxx;
                                break;
                            case ABSOLID_FMT_COLMASK_CY3:
                                d = &read[read_number].cy3;
                               break;
                            case ABSOLID_FMT_COLMASK_TXR:
                                d = &read[read_number].txr;
                                break;
                            case ABSOLID_FMT_COLMASK_CY5:
                                d = &read[read_number].cy5;
                                break;
                        }
                        if( d ) {
                            rc = pstring_assign(d, ztr.signal->data, ztr.signal->datasize);
                            DEBUG_MSG(3, ("SRF SIGNAL[%s]: %u bytes\n", ztr.signal->channel, d->len));
                        } else {
                            rc = RC(rcSRA, rcFormatter, rcParsing, rcData, rcUnrecognized);
                        }
                        if( rc != 0 ) {
                            SRALoaderFile_LOG(ctx->file, klogErr, rc, "copying signal", NULL);
                        }
                    } else {
			 if( fe->region.nreads <= 0 || fe->region.nreads > ABSOLID_FMT_MAX_NUM_READS ) {
				rc = RC(rcSRA, rcFormatter, rcParsing, rcData, rcUnsupported);
				SRALoaderFile_LOG(fe->ctx.file, klogErr, rc, "read count $(c)", PLOG_U8(c), fe->region.nreads);
			 }

                        for(i = 0; rc == 0 && i < fe->region.nreads; i++) {
                            pstring* d = NULL;
                            int read_number = AbisolidReadType2ReadNumber[fe->region.type[i]];
                            size_t len = (i + 1 >= fe->region.nreads) ? ztr.signal->datasize : (fe->region.start[i + 1] * sizeof(float));
                            len -= fe->region.start[i] * sizeof(float);
                            switch(stype) {
                                case ABSOLID_FMT_COLMASK_FAM:
                                    read[read_number].fs_type = eAbisolidFSignalType_FAM;
                                    d = &read[read_number].fxx;
                                    break;
                                case ABSOLID_FMT_COLMASK_CY3:
                                    d = &read[read_number].cy3;
                                   break;
                                case ABSOLID_FMT_COLMASK_TXR:
                                    d = &read[read_number].txr;
                                    break;
                                case ABSOLID_FMT_COLMASK_CY5:
                                    d = &read[read_number].cy5;
                                    break;
                            }
                            if( d ) {
                                rc = pstring_assign(d, &ztr.signal->data[fe->region.start[i] * sizeof(float)], len);
                                DEBUG_MSG(3, ("SRF SIGNAL[%s]: %u bytes\n", ztr.signal->channel, d->len));
                            } else {
                                rc = RC(rcSRA, rcFormatter, rcParsing, rcData, rcUnrecognized);
                            }
                        }
                        if( rc != 0 ) {
                            SRALoaderFile_LOG(ctx->file, klogErr, rc, "copying signals", NULL);
                        }
                    }
                }
            }
            break;
        default:
            break;
        }
        if(type != none && type != ignore) {
            free(*(void **)&ztr);
        }
    }
    if(rc == 0) {
        if( read_type <= eAbisolidReadType_SPOT ) {
            rc = SRAWriteAbsolid_Write(fe->writer, ctx->file, &readId, NULL, &read[0], &read[1]);
        } else {
            switch( AbisolidReadType2ReadNumber[read_type] ) {
                case 0:
                    rc = SRAWriteAbsolid_Write(fe->writer, ctx->file, &readId, NULL, &read[0], NULL);
                    break;
                case 1:
                    rc = SRAWriteAbsolid_Write(fe->writer, ctx->file, &readId, NULL, NULL, &read[1]);
                    break;
                default:
                    rc = RC(rcSRA, rcFormatter, rcParsing, rcData, rcUnsupported);
                    SRALoaderFile_LOG(ctx->file, klogErr, rc, "more than 2 reads", NULL);
                    break;
            }
        }
    }
    return rc;
}