8 static int g_min_mapQ = 0, g_flag_on = 0, g_flag_off = 0;
9 static char *g_library, *g_rg;
11 static inline int __g_skip_aln(const bam_header_t *h, const bam1_t *b)
13 if (b->core.qual < g_min_mapQ || ((b->core.flag & g_flag_on) != g_flag_on) || (b->core.flag & g_flag_off))
15 if (g_library || g_rg) {
16 uint8_t *s = bam_aux_get(b, "RG");
18 if (g_rg && strcmp(g_rg, (char*)(s + 1)) == 0) return 0;
20 const char *p = bam_strmap_get(h->rg2lib, (char*)(s + 1));
21 return (p && strcmp(p, g_library) == 0)? 0 : 1;
27 // callback function for bam_fetch()
28 static int view_func(const bam1_t *b, void *data)
30 if (!__g_skip_aln(((samfile_t*)data)->header, b))
31 samwrite((samfile_t*)data, b);
35 static int usage(int is_long_help);
37 int main_samview(int argc, char *argv[])
39 int c, is_header = 0, is_header_only = 0, is_bamin = 1, ret = 0, is_uncompressed = 0, is_bamout = 0;
40 int of_type = BAM_OFDEC, is_long_help = 0;
41 samfile_t *in = 0, *out = 0;
42 char in_mode[5], out_mode[5], *fn_out = 0, *fn_list = 0, *fn_ref = 0;
44 /* parse command-line options */
45 strcpy(in_mode, "r"); strcpy(out_mode, "w");
46 while ((c = getopt(argc, argv, "Sbt:hHo:q:f:F:ul:r:xX?T:")) >= 0) {
48 case 'S': is_bamin = 0; break;
49 case 'b': is_bamout = 1; break;
50 case 't': fn_list = strdup(optarg); is_bamin = 0; break;
51 case 'h': is_header = 1; break;
52 case 'H': is_header_only = 1; break;
53 case 'o': fn_out = strdup(optarg); break;
54 case 'f': g_flag_on = strtol(optarg, 0, 0); break;
55 case 'F': g_flag_off = strtol(optarg, 0, 0); break;
56 case 'q': g_min_mapQ = atoi(optarg); break;
57 case 'u': is_uncompressed = 1; break;
58 case 'l': g_library = strdup(optarg); break;
59 case 'r': g_rg = strdup(optarg); break;
60 case 'x': of_type = BAM_OFHEX; break;
61 case 'X': of_type = BAM_OFSTR; break;
62 case '?': is_long_help = 1; break;
63 case 'T': fn_ref = strdup(optarg); is_bamin = 0; break;
64 default: return usage(is_long_help);
67 if (is_uncompressed) is_bamout = 1;
68 if (is_header_only) is_header = 1;
69 if (is_bamout) strcat(out_mode, "b");
71 if (of_type == BAM_OFHEX) strcat(out_mode, "x");
72 else if (of_type == BAM_OFSTR) strcat(out_mode, "X");
74 if (is_bamin) strcat(in_mode, "b");
75 if (is_header) strcat(out_mode, "h");
76 if (is_uncompressed) strcat(out_mode, "u");
77 if (argc == optind) return usage(is_long_help);
79 // generate the fn_list if necessary
80 if (fn_list == 0 && fn_ref) fn_list = samfaipath(fn_ref);
82 if ((in = samopen(argv[optind], in_mode, fn_list)) == 0) {
83 fprintf(stderr, "[main_samview] fail to open file for reading.\n");
86 if (in->header == 0) {
87 fprintf(stderr, "[main_samview] fail to read the header.\n");
90 if ((out = samopen(fn_out? fn_out : "-", out_mode, in->header)) == 0) {
91 fprintf(stderr, "[main_samview] fail to open file for writing.\n");
94 if (is_header_only) goto view_end; // no need to print alignments
96 if (argc == optind + 1) { // convert/print the entire file
97 bam1_t *b = bam_init1();
99 while ((r = samread(in, b)) >= 0) // read one alignment from `in'
100 if (!__g_skip_aln(in->header, b))
101 samwrite(out, b); // write the alignment to `out'
102 if (r < -1) fprintf(stderr, "[main_samview] truncated file.\n");
104 } else { // retrieve alignments in specified regions
106 bam_index_t *idx = 0;
107 if (is_bamin) idx = bam_index_load(argv[optind]); // load BAM index
108 if (idx == 0) { // index is unavailable
109 fprintf(stderr, "[main_samview] random alignment retrieval only works for indexed BAM files.\n");
113 for (i = optind + 1; i < argc; ++i) {
115 bam_parse_region(in->header, argv[i], &tid, &beg, &end); // parse a region in the format like `chr2:100-200'
116 if (tid < 0) { // reference name is not found
117 fprintf(stderr, "[main_samview] fail to get the reference name. Continue anyway.\n");
120 bam_fetch(in->x.bam, idx, tid, beg, end, out, view_func); // fetch alignments
122 bam_index_destroy(idx); // destroy the BAM index
126 // close files, free and return
127 free(fn_list); free(fn_ref); free(fn_out); free(g_library); free(g_rg);
133 static int usage(int is_long_help)
135 fprintf(stderr, "\n");
136 fprintf(stderr, "Usage: samtools view [options] <in.bam>|<in.sam> [region1 [...]]\n\n");
137 fprintf(stderr, "Options: -b output BAM\n");
138 fprintf(stderr, " -h print header for the SAM output\n");
139 fprintf(stderr, " -H print header only (no alignments)\n");
140 fprintf(stderr, " -S input is SAM\n");
141 fprintf(stderr, " -u uncompressed BAM output (force -b)\n");
142 fprintf(stderr, " -x output FLAG in HEX (samtools-C specific)\n");
143 fprintf(stderr, " -X output FLAG in string (samtools-C specific)\n");
144 fprintf(stderr, " -t FILE list of reference names and lengths (force -S) [null]\n");
145 fprintf(stderr, " -T FILE reference sequence file (force -S) [null]\n");
146 fprintf(stderr, " -o FILE output file name [stdout]\n");
147 fprintf(stderr, " -f INT required flag, 0 for unset [0]\n");
148 fprintf(stderr, " -F INT filtering flag, 0 for unset [0]\n");
149 fprintf(stderr, " -q INT minimum mapping quality [0]\n");
150 fprintf(stderr, " -l STR only output reads in library STR [null]\n");
151 fprintf(stderr, " -r STR only output reads in read group STR [null]\n");
152 fprintf(stderr, " -? longer help\n");
153 fprintf(stderr, "\n");
155 fprintf(stderr, "Notes:\n\
157 1. By default, this command assumes the file on the command line is in\n\
158 the BAM format and it prints the alignments in SAM. If `-t' is\n\
159 applied, the input file is assumed to be in the SAM format. The\n\
160 file supplied with `-t' is SPACE/TAB delimited with the first two\n\
161 fields of each line consisting of the reference name and the\n\
162 corresponding sequence length. The `.fai' file generated by `faidx'\n\
163 can be used here. This file may be empty if reads are unaligned.\n\
165 2. SAM->BAM conversion: `samtools view -bT ref.fa in.sam.gz'.\n\
167 3. BAM->SAM conversion: `samtools view in.bam'.\n\
169 4. A region should be presented in one of the following formats:\n\
170 `chr1', `chr2:1,000' and `chr3:1000-2,000'. When a region is\n\
171 specified, the input alignment file must be an indexed BAM file.\n\
173 5. Option `-u' is preferred over `-b' when the output is piped to\n\
174 another samtools command.\n\
176 6. In a string FLAG, each character represents one bit with\n\
177 p=0x1 (paired), P=0x2 (properly paired), u=0x4 (unmapped),\n\
178 U=0x8 (mate unmapped), r=0x10 (reverse), R=0x20 (mate reverse)\n\
179 1=0x40 (first), 2=0x80 (second), s=0x100 (not primary), \n\
180 f=0x200 (failure) and d=0x400 (duplicate). Note that `-x' and\n\
181 `-X' are samtools-C specific. Picard and older samtools do not\n\
182 support HEX or string flags.\n\
187 int main_import(int argc, char *argv[])
192 fprintf(stderr, "Usage: bamtk import <in.ref_list> <in.sam> <out.bam>\n");
196 argv2 = calloc(6, sizeof(char*));
197 argv2[0] = "import", argv2[1] = "-o", argv2[2] = argv[3], argv2[3] = "-bt", argv2[4] = argv[1], argv2[5] = argv[2];
198 ret = main_samview(argc2, argv2);