Ejemplo n.º 1
0
int main_samview(int argc, char *argv[])
{
 int index;
    for(index = 0; index < argc; index++) {
        printf("The %d is %s\n",index,argv[index]);
    }
    getchar();return 0;
    int c, is_header = 0, is_header_only = 0, ret = 0, compress_level = -1, is_count = 0;
    int is_long_help = 0, n_threads = 0;
    int64_t count = 0;
    samFile *in = 0, *out = 0, *un_out=0;
    bam_hdr_t *header = NULL;
    char out_mode[5], out_un_mode[5], *out_format = "";
    char *fn_in = 0, *fn_out = 0, *fn_list = 0, *q, *fn_un_out = 0;
    sam_global_args ga = SAM_GLOBAL_ARGS_INIT;

    samview_settings_t settings = {
        .rghash = NULL,
        .min_mapQ = 0,
        .flag_on = 0,
        .flag_off = 0,
        .min_qlen = 0,
        .remove_B = 0,
        .subsam_seed = 0,
        .subsam_frac = -1.,
        .library = NULL,
        .bed = NULL,
    };

    static const struct option lopts[] = {
        SAM_OPT_GLOBAL_OPTIONS('-', 0, 'O', 0, 'T'),
        { "threads", required_argument, NULL, '@' },
        { NULL, 0, NULL, 0 }
    };

    /* parse command-line options */
    strcpy(out_mode, "w");
    strcpy(out_un_mode, "w");
    while ((c = getopt_long(argc, argv,
                            "SbBcCt:h1Ho:O:q:f:F:ul:r:?T:R:L:s:@:m:x:U:",
                            lopts, NULL)) >= 0) {
        switch (c) {
        case 's':
            if ((settings.subsam_seed = strtol(optarg, &q, 10)) != 0) {
                srand(settings.subsam_seed);
                settings.subsam_seed = rand();
            }
            settings.subsam_frac = strtod(q, &q);
            break;
        case 'm': settings.min_qlen = atoi(optarg); break;
        case 'c': is_count = 1; break;
        case 'S': break;
        case 'b': out_format = "b"; break;
        case 'C': out_format = "c"; break;
        case 't': fn_list = strdup(optarg); break;
        case 'h': is_header = 1; break;
        case 'H': is_header_only = 1; break;
        case 'o': fn_out = strdup(optarg); break;
        case 'U': fn_un_out = strdup(optarg); break;
        case 'f': settings.flag_on |= strtol(optarg, 0, 0); break;
        case 'F': settings.flag_off |= strtol(optarg, 0, 0); break;
        case 'q': settings.min_mapQ = atoi(optarg); break;
        case 'u': compress_level = 0; break;
        case '1': compress_level = 1; break;
        case 'l': settings.library = strdup(optarg); break;
        case 'L':
            if ((settings.bed = bed_read(optarg)) == NULL) {
                print_error_errno("view", "Could not read file \"%s\"", optarg);
                ret = 1;
                goto view_end;
            }
            break;
        case 'r':
            if (add_read_group_single("view", &settings, optarg) != 0) {
                ret = 1;
                goto view_end;
            }
            break;
        case 'R':
            if (add_read_groups_file("view", &settings, optarg) != 0) {
                ret = 1;
                goto view_end;
            }
            break;
                /* REMOVED as htslib doesn't support this
        //case 'x': out_format = "x"; break;
        //case 'X': out_format = "X"; break;
                 */
        case '?': is_long_help = 1; break;
        case 'B': settings.remove_B = 1; break;
        case '@': n_threads = strtol(optarg, 0, 0); break;
        case 'x':
            {
                if (strlen(optarg) != 2) {
                    fprintf(stderr, "main_samview: Error parsing -x auxiliary tags should be exactly two characters long.\n");
                    return usage(stderr, EXIT_FAILURE, is_long_help);
                }
                settings.remove_aux = (char**)realloc(settings.remove_aux, sizeof(char*) * (++settings.remove_aux_len));
                settings.remove_aux[settings.remove_aux_len-1] = optarg;
            }
            break;

        default:
            if (parse_sam_global_opt(c, optarg, lopts, &ga) != 0)
                return usage(stderr, EXIT_FAILURE, is_long_help);
            break;
        }
    }
    if (compress_level >= 0 && !*out_format) out_format = "b";
    if (is_header_only) is_header = 1;
    // File format auto-detection first
    if (fn_out)    sam_open_mode(out_mode+1,    fn_out,    NULL);
    if (fn_un_out) sam_open_mode(out_un_mode+1, fn_un_out, NULL);
    // Overridden by manual -b, -C
    if (*out_format)
        out_mode[1] = out_un_mode[1] = *out_format;
    out_mode[2] = out_un_mode[2] = '\0';
    // out_(un_)mode now 1 or 2 bytes long, followed by nul.
    if (compress_level >= 0) {
        char tmp[2];
        tmp[0] = compress_level + '0'; tmp[1] = '\0';
        strcat(out_mode, tmp);
        strcat(out_un_mode, tmp);
    }
    if (argc == optind && isatty(STDIN_FILENO)) return usage(stdout, EXIT_SUCCESS, is_long_help); // potential memory leak...

    fn_in = (optind < argc)? argv[optind] : "-";
    // generate the fn_list if necessary
    if (fn_list == 0 && ga.reference) fn_list = samfaipath(ga.reference);
    // open file handlers
    if ((in = sam_open_format(fn_in, "r", &ga.in)) == 0) {
        print_error_errno("view", "failed to open \"%s\" for reading", fn_in);
        ret = 1;
        goto view_end;
    }

    if (fn_list) {
        if (hts_set_fai_filename(in, fn_list) != 0) {
            fprintf(stderr, "[main_samview] failed to use reference \"%s\".\n", fn_list);
            ret = 1;
            goto view_end;
        }
    }
    if ((header = sam_hdr_read(in)) == 0) {
        fprintf(stderr, "[main_samview] fail to read the header from \"%s\".\n", fn_in);
        ret = 1;
        goto view_end;
    }
    if (settings.rghash) { // FIXME: I do not know what "bam_header_t::n_text" is for...
        char *tmp;
        int l;
        tmp = drop_rg(header->text, settings.rghash, &l);
        free(header->text);
        header->text = tmp;
        header->l_text = l;
    }
    if (!is_count) {
        if ((out = sam_open_format(fn_out? fn_out : "-", out_mode, &ga.out)) == 0) {
            print_error_errno("view", "failed to open \"%s\" for writing", fn_out? fn_out : "standard output");
            ret = 1;
            goto view_end;
        }
        if (fn_list) {
            if (hts_set_fai_filename(out, fn_list) != 0) {
                fprintf(stderr, "[main_samview] failed to use reference \"%s\".\n", fn_list);
                ret = 1;
                goto view_end;
            }
        }
        if (*out_format || is_header ||
            out_mode[1] == 'b' || out_mode[1] == 'c' ||
            (ga.out.format != sam && ga.out.format != unknown_format))  {
            if (sam_hdr_write(out, header) != 0) {
                fprintf(stderr, "[main_samview] failed to write the SAM header\n");
                ret = 1;
                goto view_end;
            }
        }
        if (fn_un_out) {
            if ((un_out = sam_open_format(fn_un_out, out_un_mode, &ga.out)) == 0) {
                print_error_errno("view", "failed to open \"%s\" for writing", fn_un_out);
                ret = 1;
                goto view_end;
            }
            if (fn_list) {
                if (hts_set_fai_filename(un_out, fn_list) != 0) {
                    fprintf(stderr, "[main_samview] failed to use reference \"%s\".\n", fn_list);
                    ret = 1;
                    goto view_end;
                }
            }
            if (*out_format || is_header ||
                out_un_mode[1] == 'b' || out_un_mode[1] == 'c' ||
                (ga.out.format != sam && ga.out.format != unknown_format))  {
                if (sam_hdr_write(un_out, header) != 0) {
                    fprintf(stderr, "[main_samview] failed to write the SAM header\n");
                    ret = 1;
                    goto view_end;
                }
            }
        }
    }

    if (n_threads > 1) { if (out) hts_set_threads(out, n_threads); }
    if (is_header_only) goto view_end; // no need to print alignments

    if (optind + 1 >= argc) { // convert/print the entire file
        bam1_t *b = bam_init1();
        int r;
        while ((r = sam_read1(in, header, b)) >= 0) { // read one alignment from `in'
            if (!process_aln(header, b, &settings)) {
                if (!is_count) { if (check_sam_write1(out, header, b, fn_out, &ret) < 0) break; }
                count++;
            } else {
                if (un_out) { if (check_sam_write1(un_out, header, b, fn_un_out, &ret) < 0) break; }
            }
        }
        if (r < -1) {
            fprintf(stderr, "[main_samview] truncated file.\n");
            ret = 1;
        }
        bam_destroy1(b);
    } else { // retrieve alignments in specified regions
        int i;
        bam1_t *b;
        hts_idx_t *idx = sam_index_load(in, fn_in); // load index
        if (idx == 0) { // index is unavailable
            fprintf(stderr, "[main_samview] random alignment retrieval only works for indexed BAM or CRAM files.\n");
            ret = 1;
            goto view_end;
        }
        b = bam_init1();
        for (i = optind + 1; i < argc; ++i) {
            int result;
            hts_itr_t *iter = sam_itr_querys(idx, header, argv[i]); // parse a region in the format like `chr2:100-200'
            if (iter == NULL) { // region invalid or reference name not found
                int beg, end;
                if (hts_parse_reg(argv[i], &beg, &end))
                    fprintf(stderr, "[main_samview] region \"%s\" specifies an unknown reference name. Continue anyway.\n", argv[i]);
                else
                    fprintf(stderr, "[main_samview] region \"%s\" could not be parsed. Continue anyway.\n", argv[i]);
                continue;
            }
            // fetch alignments
            while ((result = sam_itr_next(in, iter, b)) >= 0) {
                if (!process_aln(header, b, &settings)) {
                    if (!is_count) { if (check_sam_write1(out, header, b, fn_out, &ret) < 0) break; }
                    count++;
                } else {
                    if (un_out) { if (check_sam_write1(un_out, header, b, fn_un_out, &ret) < 0) break; }
                }
            }
            hts_itr_destroy(iter);
            if (result < -1) {
                fprintf(stderr, "[main_samview] retrieval of region \"%s\" failed due to truncated file or corrupt BAM index file\n", argv[i]);
                ret = 1;
                break;
            }
        }
        bam_destroy1(b);
        hts_idx_destroy(idx); // destroy the BAM index
    }

view_end:
    if (is_count && ret == 0)
        printf("%" PRId64 "\n", count);

    // close files, free and return
    if (in) check_sam_close("view", in, fn_in, "standard input", &ret);
    if (out) check_sam_close("view", out, fn_out, "standard output", &ret);
    if (un_out) check_sam_close("view", un_out, fn_un_out, "file", &ret);

    free(fn_list); free(fn_out); free(settings.library);  free(fn_un_out);
    sam_global_args_free(&ga);
    if ( header ) bam_hdr_destroy(header);
    if (settings.bed) bed_destroy(settings.bed);
    if (settings.rghash) {
        khint_t k;
        for (k = 0; k < kh_end(settings.rghash); ++k)
            if (kh_exist(settings.rghash, k)) free((char*)kh_key(settings.rghash, k));
        kh_destroy(rg, settings.rghash);
    }
    if (settings.remove_aux_len) {
        free(settings.remove_aux);
    }
    return ret;
}

static int usage(FILE *fp, int exit_status, int is_long_help)
{
    fprintf(fp,
"\n"
"Usage: samtools view [options] <in.bam>|<in.sam>|<in.cram> [region ...]\n"
"\n"
"Options:\n"
// output options
"  -b       output BAM\n"
"  -C       output CRAM (requires -T)\n"
"  -1       use fast BAM compression (implies -b)\n"
"  -u       uncompressed BAM output (implies -b)\n"
"  -h       include header in SAM output\n"
"  -H       print SAM header only (no alignments)\n"
"  -c       print only the count of matching records\n"
"  -o FILE  output file name [stdout]\n"
"  -U FILE  output reads not selected by filters to FILE [null]\n"
// extra input
"  -t FILE  FILE listing reference names and lengths (see long help) [null]\n"
// read filters
"  -L FILE  only include reads overlapping this BED FILE [null]\n"
"  -r STR   only include reads in read group STR [null]\n"
"  -R FILE  only include reads with read group listed in FILE [null]\n"
"  -q INT   only include reads with mapping quality >= INT [0]\n"
"  -l STR   only include reads in library STR [null]\n"
"  -m INT   only include reads with number of CIGAR operations consuming\n"
"           query sequence >= INT [0]\n"
"  -f INT   only include reads with all bits set in INT set in FLAG [0]\n"
"  -F INT   only include reads with none of the bits set in INT set in FLAG [0]\n"
// read processing
"  -x STR   read tag to strip (repeatable) [null]\n"
"  -B       collapse the backward CIGAR operation\n"
"  -s FLOAT integer part sets seed of random number generator [0];\n"
"           rest sets fraction of templates to subsample [no subsampling]\n"
// general options
"  -@, --threads INT\n"
"           number of BAM/CRAM compression threads [0]\n"
"  -?       print long help, including note about region specification\n"
"  -S       ignored (input format is auto-detected)\n");

    sam_global_opt_help(fp, "-.O.T");
    fprintf(fp, "\n");

    if (is_long_help)
        fprintf(fp,
"Notes:\n"
"\n"
"1. This command now auto-detects the input format (BAM/CRAM/SAM).\n"
"   Further control over the CRAM format can be specified by using the\n"
"   --output-fmt-option, e.g. to specify the number of sequences per slice\n"
"   and to use avoid reference based compression:\n"
"\n"
"\tsamtools view -C --output-fmt-option seqs_per_slice=5000 \\\n"
"\t   --output-fmt-option no_ref -o out.cram in.bam\n"
"\n"
"   Options can also be specified as a comma separated list within the\n"
"   --output-fmt value too.  For example this is equivalent to the above\n"
"\n"
"\tsamtools view --output-fmt cram,seqs_per_slice=5000,no_ref \\\n"
"\t   -o out.cram in.bam\n"
"\n"
"2. The file supplied with `-t' is SPACE/TAB delimited with the first\n"
"   two fields of each line consisting of the reference name and the\n"
"   corresponding sequence length. The `.fai' file generated by \n"
"   `samtools faidx' is suitable for use as this file. This may be an\n"
"   empty file if reads are unaligned.\n"
"\n"
"3. SAM->BAM conversion:  samtools view -bT ref.fa in.sam.gz\n"
"\n"
"4. BAM->SAM conversion:  samtools view -h in.bam\n"
"\n"
"5. A region should be presented in one of the following formats:\n"
"   `chr1', `chr2:1,000' and `chr3:1000-2,000'. When a region is\n"
"   specified, the input alignment file must be a sorted and indexed\n"
"   alignment (BAM/CRAM) file.\n"
"\n"
"6. Option `-u' is preferred over `-b' when the output is piped to\n"
"   another samtools command.\n"
"\n");

    return exit_status;
}
Ejemplo n.º 2
0
int main_pad2unpad(int argc, char *argv[])
{
    samFile *in = 0, *out = 0;
    bam_hdr_t *h = 0, *h_fix = 0;
    faidx_t *fai = 0;
    int c, compress_level = -1, is_long_help = 0;
    char in_mode[5], out_mode[6], *fn_out = 0, *fn_list = 0;
    int ret=0;
    sam_global_args ga = SAM_GLOBAL_ARGS_INIT;

    static const struct option lopts[] = {
        SAM_OPT_GLOBAL_OPTIONS('-', 0, 0, 0, 'T'),
        { NULL, 0, NULL, 0 }
    };

    /* parse command-line options */
    strcpy(in_mode, "r"); strcpy(out_mode, "w");
    while ((c = getopt_long(argc, argv, "SCso:u1T:?", lopts, NULL)) >= 0) {
        switch (c) {
        case 'S': break;
        case 'C': hts_parse_format(&ga.out, "cram"); break;
        case 's': assert(compress_level == -1); hts_parse_format(&ga.out, "sam"); break;
        case 'o': fn_out = strdup(optarg); break;
        case 'u':
            compress_level = 0;
            if (ga.out.format == unknown_format)
                hts_parse_format(&ga.out, "bam");
            break;
        case '1':
            compress_level = 1;
            if (ga.out.format == unknown_format)
                hts_parse_format(&ga.out, "bam");
            break;
        case '?': is_long_help = 1; break;
        default:  if (parse_sam_global_opt(c, optarg, lopts, &ga) == 0) break;
            fprintf(stderr, "[bam_fillmd] unrecognized option '-%c'\n\n", c);
            return usage(is_long_help);
        }
    }
    if (argc == optind) return usage(is_long_help);

    strcat(out_mode, "h");
    if (compress_level >= 0) {
        char tmp[2];
        tmp[0] = compress_level + '0'; tmp[1] = '\0';
        strcat(out_mode, tmp);
    }

    // Load FASTA reference (also needed for SAM -> BAM if missing header)
    if (ga.reference) {
        fn_list = samfaipath(ga.reference);
        fai = fai_load(ga.reference);
    }
    // open file handlers
    if ((in = sam_open_format(argv[optind], in_mode, &ga.in)) == 0) {
        fprintf(stderr, "[depad] failed to open \"%s\" for reading.\n", argv[optind]);
        ret = 1;
        goto depad_end;
    }
    if (fn_list && hts_set_fai_filename(in, fn_list) != 0) {
        fprintf(stderr, "[depad] failed to load reference file \"%s\".\n", fn_list);
        ret = 1;
        goto depad_end;
    }
    if ((h = sam_hdr_read(in)) == 0) {
        fprintf(stderr, "[depad] failed to read the header from \"%s\".\n", argv[optind]);
        ret = 1;
        goto depad_end;
    }
    if (fai) {
        h_fix = fix_header(h, fai);
    } else {
        fprintf(stderr, "[depad] Warning - reference lengths will not be corrected without FASTA reference\n");
        h_fix = h;
    }
    char wmode[2];
    strcat(out_mode, sam_open_mode(wmode, fn_out, NULL)==0 ? wmode : "b");
    if ((out = sam_open_format(fn_out? fn_out : "-", out_mode, &ga.out)) == 0) {
        fprintf(stderr, "[depad] failed to open \"%s\" for writing.\n", fn_out? fn_out : "standard output");
        ret = 1;
        goto depad_end;
    }

    // Reference-based CRAM won't work unless we also create a new reference.
    // We could embed this, but for now we take the easy option.
    if (ga.out.format == cram)
        hts_set_opt(out, CRAM_OPT_NO_REF, 1);

    if (sam_hdr_write(out, h_fix) != 0) {
        fprintf(stderr, "[depad] failed to write header.\n");
        ret = 1;
        goto depad_end;
    }

    // Do the depad
    ret = bam_pad2unpad(in, out, h, fai);

depad_end:
    // close files, free and return
    if (fai) fai_destroy(fai);
    if (h) bam_hdr_destroy(h);
    sam_close(in);
    sam_close(out);
    free(fn_list); free(fn_out);
    return ret;
}