X-Git-Url: https://git.donarmstrong.com/?p=samtools.git;a=blobdiff_plain;f=bam_tview.c;h=ede6b0c87f57d17ad25b5933da85cedd881f719b;hp=20e6f466dfdf9ab0a2ba444e4181a20115b0632e;hb=1ea60418fdfda809953490a6b5fd347b85bb583c;hpb=af97a7bdff4f0d3fec9543ce8077902f8d263ba0 diff --git a/bam_tview.c b/bam_tview.c index 20e6f46..ede6b0c 100644 --- a/bam_tview.c +++ b/bam_tview.c @@ -1,65 +1,81 @@ -#undef _HAVE_CURSES - -#if _CURSES_LIB == 0 -#elif _CURSES_LIB == 1 -#include -#ifndef NCURSES_VERSION -#warning "_CURSES_LIB=1 but NCURSES_VERSION not defined; tview is NOT compiled" -#else -#define _HAVE_CURSES -#endif -#elif _CURSES_LIB == 2 -#include -#define _HAVE_CURSES -#else -#warning "_CURSES_LIB is not 0, 1 or 2; tview is NOT compiled" -#endif - -#ifdef _HAVE_CURSES -#include #include -#include -#include -#include -#include "bam.h" -#include "faidx.h" -#include "bam2bcf.h" -#include "sam_header.h" - -char bam_aux_getCEi(bam1_t *b, int i); -char bam_aux_getCSi(bam1_t *b, int i); -char bam_aux_getCQi(bam1_t *b, int i); +#include "bam_tview.h" -#define TV_MIN_ALNROW 2 -#define TV_MAX_GOTO 40 -#define TV_LOW_MAPQ 10 +int base_tv_init(tview_t* tv,const char *fn, const char *fn_fa, const char *samples) + { + assert(tv!=NULL); + assert(fn!=NULL); + tv->mrow = 24; tv->mcol = 80; + tv->color_for = TV_COLOR_MAPQ; + tv->is_dot = 1; + + tv->fp = bam_open(fn, "r"); + if(tv->fp==0) + { + fprintf(stderr,"bam_open %s. %s\n", fn,fn_fa); + exit(EXIT_FAILURE); + } + bgzf_set_cache_size(tv->fp, 8 * 1024 *1024); + assert(tv->fp); + + tv->header = bam_header_read(tv->fp); + if(tv->header==0) + { + fprintf(stderr,"Cannot read '%s'.\n", fn); + exit(EXIT_FAILURE); + } + tv->idx = bam_index_load(fn); + if (tv->idx == 0) + { + fprintf(stderr,"Cannot read index for '%s'.\n", fn); + exit(EXIT_FAILURE); + } + tv->lplbuf = bam_lplbuf_init(tv_pl_func, tv); + if (fn_fa) tv->fai = fai_load(fn_fa); + tv->bca = bcf_call_init(0.83, 13); + tv->ins = 1; -#define TV_COLOR_MAPQ 0 -#define TV_COLOR_BASEQ 1 -#define TV_COLOR_NUCL 2 -#define TV_COLOR_COL 3 -#define TV_COLOR_COLQ 4 + if ( samples ) + { + if ( !tv->header->dict ) tv->header->dict = sam_header_parse2(tv->header->text); + void *iter = tv->header->dict; + const char *key, *val; + int n = 0; + tv->rg_hash = kh_init(kh_rg); + while ( (iter = sam_header2key_val(iter, "RG","ID","SM", &key, &val)) ) + { + if ( !strcmp(samples,key) || (val && !strcmp(samples,val)) ) + { + khiter_t k = kh_get(kh_rg, tv->rg_hash, key); + if ( k != kh_end(tv->rg_hash) ) continue; + int ret; + k = kh_put(kh_rg, tv->rg_hash, key, &ret); + kh_value(tv->rg_hash, k) = val; + n++; + } + } + if ( !n ) + { + fprintf(stderr,"The sample or read group \"%s\" not present.\n", samples); + exit(EXIT_FAILURE); + } + } -#define TV_BASE_NUCL 0 -#define TV_BASE_COLOR_SPACE 1 + return 0; + } -typedef struct { - int mrow, mcol; - WINDOW *wgoto, *whelp; - bam_index_t *idx; - bam_lplbuf_t *lplbuf; - bam_header_t *header; - bamFile fp; - int curr_tid, left_pos; - faidx_t *fai; - bcf_callaux_t *bca; +void base_tv_destroy(tview_t* tv) + { + bam_lplbuf_destroy(tv->lplbuf); + bcf_call_destroy(tv->bca); + bam_index_destroy(tv->idx); + if (tv->fai) fai_destroy(tv->fai); + free(tv->ref); + bam_header_destroy(tv->header); + bam_close(tv->fp); + } - int ccol, last_pos, row_shift, base_for, color_for, is_dot, l_ref, ins, no_skip, show_name; - char *ref; - char *sample; //TODO: multiple samples and read groups - void *rg2sm; -} tview_t; int tv_pl_func(uint32_t tid, uint32_t pos, int n, const bam_pileup1_t *pl, void *data) { @@ -71,11 +87,11 @@ int tv_pl_func(uint32_t tid, uint32_t pos, int n, const bam_pileup1_t *pl, void // print referece rb = (tv->ref && pos - tv->left_pos < tv->l_ref)? tv->ref[pos - tv->left_pos] : 'N'; for (i = tv->last_pos + 1; i < pos; ++i) { - if (i%10 == 0 && tv->mcol - tv->ccol >= 10) mvprintw(0, tv->ccol, "%-d", i+1); + if (i%10 == 0 && tv->mcol - tv->ccol >= 10) tv->my_mvprintw(tv,0, tv->ccol, "%-d", i+1); c = tv->ref? tv->ref[i - tv->left_pos] : 'N'; - mvaddch(1, tv->ccol++, c); + tv->my_mvaddch(tv,1, tv->ccol++, c); } - if (pos%10 == 0 && tv->mcol - tv->ccol >= 10) mvprintw(0, tv->ccol, "%-d", pos+1); + if (pos%10 == 0 && tv->mcol - tv->ccol >= 10) tv->my_mvprintw(tv,0, tv->ccol, "%-d", pos+1); { // call consensus bcf_callret1_t bcr; int qsum[4], a1, a2, tmp; @@ -93,15 +109,15 @@ int tv_pl_func(uint32_t tid, uint32_t pos, int n, const bam_pileup1_t *pl, void else if (p[2] < p[1] && p[2] < p[0]) call = (1<my_underline(tv); c = ",ACMGRSVTWYHKDBN"[call>>16&0xf]; i = (call&0xffff)/10+1; if (i > 4) i = 4; - attr |= COLOR_PAIR(i); + attr |= tv->my_colorpair(tv,i); if (c == toupper(rb)) c = '.'; - attron(attr); - mvaddch(2, tv->ccol, c); - attroff(attr); + tv->my_attron(tv,attr); + tv->my_mvaddch(tv,2, tv->ccol, c); + tv->my_attroff(tv,attr); if(tv->ins) { // calculate maximum insert for (i = 0; i < n; ++i) { @@ -151,18 +167,18 @@ int tv_pl_func(uint32_t tid, uint32_t pos, int n, const bam_pileup1_t *pl, void int x; attr = 0; if (((p->b->core.flag&BAM_FPAIRED) && !(p->b->core.flag&BAM_FPROPER_PAIR)) - || (p->b->core.flag & BAM_FSECONDARY)) attr |= A_UNDERLINE; + || (p->b->core.flag & BAM_FSECONDARY)) attr |= tv->my_underline(tv); if (tv->color_for == TV_COLOR_BASEQ) { x = bam1_qual(p->b)[p->qpos]/10 + 1; if (x > 4) x = 4; - attr |= COLOR_PAIR(x); + attr |= tv->my_colorpair(tv,x); } else if (tv->color_for == TV_COLOR_MAPQ) { x = p->b->core.qual/10 + 1; if (x > 4) x = 4; - attr |= COLOR_PAIR(x); + attr |= tv->my_colorpair(tv,x); } else if (tv->color_for == TV_COLOR_NUCL) { x = bam_nt16_nt4_table[bam1_seqi(bam1_seq(p->b), p->qpos)] + 5; - attr |= COLOR_PAIR(x); + attr |= tv->my_colorpair(tv,x); } else if(tv->color_for == TV_COLOR_COL) { x = 0; switch(bam_aux_getCSi(p->b, p->qpos)) { @@ -174,101 +190,43 @@ int tv_pl_func(uint32_t tid, uint32_t pos, int n, const bam_pileup1_t *pl, void default: x = bam_nt16_nt4_table[bam1_seqi(bam1_seq(p->b), p->qpos)]; break; } x+=5; - attr |= COLOR_PAIR(x); + attr |= tv->my_colorpair(tv,x); } else if(tv->color_for == TV_COLOR_COLQ) { x = bam_aux_getCQi(p->b, p->qpos); if(0 == x) x = bam1_qual(p->b)[p->qpos]; x = x/10 + 1; if (x > 4) x = 4; - attr |= COLOR_PAIR(x); + attr |= tv->my_colorpair(tv,x); } - attron(attr); - mvaddch(row, tv->ccol, bam1_strand(p->b)? tolower(c) : toupper(c)); - attroff(attr); + tv->my_attron(tv,attr); + tv->my_mvaddch(tv,row, tv->ccol, bam1_strand(p->b)? tolower(c) : toupper(c)); + tv->my_attroff(tv,attr); } } c = j? '*' : rb; if (c == '*') { - attr = COLOR_PAIR(8); - attron(attr); - mvaddch(1, tv->ccol++, c); - attroff(attr); - } else mvaddch(1, tv->ccol++, c); + attr = tv->my_colorpair(tv,8); + tv->my_attron(tv,attr); + tv->my_mvaddch(tv,1, tv->ccol++, c); + tv->my_attroff(tv,attr); + } else tv->my_mvaddch(tv,1, tv->ccol++, c); } tv->last_pos = pos; return 0; } -tview_t *tv_init(const char *fn, const char *fn_fa, char *samples) -{ - tview_t *tv = (tview_t*)calloc(1, sizeof(tview_t)); - tv->is_dot = 1; - tv->fp = bam_open(fn, "r"); - bgzf_set_cache_size(tv->fp, 8 * 1024 *1024); - assert(tv->fp); - tv->header = bam_header_read(tv->fp); - tv->idx = bam_index_load(fn); - if (tv->idx == 0) exit(1); - tv->lplbuf = bam_lplbuf_init(tv_pl_func, tv); - if (fn_fa) tv->fai = fai_load(fn_fa); - tv->bca = bcf_call_init(0.83, 13); - tv->ins = 1; - - if ( samples ) - { - tv->sample = samples; - tv->header->dict = sam_header_parse2(tv->header->text); - tv->rg2sm = sam_header2tbl(tv->header->dict, "RG", "ID", "SM"); - } - initscr(); - keypad(stdscr, TRUE); - clear(); - noecho(); - cbreak(); - tv->mrow = 24; tv->mcol = 80; - getmaxyx(stdscr, tv->mrow, tv->mcol); - tv->wgoto = newwin(3, TV_MAX_GOTO + 10, 10, 5); - tv->whelp = newwin(29, 40, 5, 5); - tv->color_for = TV_COLOR_MAPQ; - start_color(); - init_pair(1, COLOR_BLUE, COLOR_BLACK); - init_pair(2, COLOR_GREEN, COLOR_BLACK); - init_pair(3, COLOR_YELLOW, COLOR_BLACK); - init_pair(4, COLOR_WHITE, COLOR_BLACK); - init_pair(5, COLOR_GREEN, COLOR_BLACK); - init_pair(6, COLOR_CYAN, COLOR_BLACK); - init_pair(7, COLOR_YELLOW, COLOR_BLACK); - init_pair(8, COLOR_RED, COLOR_BLACK); - init_pair(9, COLOR_BLUE, COLOR_BLACK); - return tv; -} -void tv_destroy(tview_t *tv) -{ - delwin(tv->wgoto); delwin(tv->whelp); - endwin(); - - bam_lplbuf_destroy(tv->lplbuf); - bcf_call_destroy(tv->bca); - bam_index_destroy(tv->idx); - if (tv->fai) fai_destroy(tv->fai); - free(tv->ref); - bam_header_destroy(tv->header); - bam_close(tv->fp); - free(tv); -} int tv_fetch_func(const bam1_t *b, void *data) { tview_t *tv = (tview_t*)data; - if ( tv->sample ) + if ( tv->rg_hash ) { const uint8_t *rg = bam_aux_get(b, "RG"); if ( !rg ) return 0; - const char *sm = sam_tbl_get(tv->rg2sm, (const char*)(rg + 1)); - if ( !sm ) return 0; - if ( strcmp(sm,tv->sample) ) return 0; + khiter_t k = kh_get(kh_rg, tv->rg_hash, (const char*)(rg + 1)); + if ( k == kh_end(tv->rg_hash) ) return 0; } if (tv->no_skip) { uint32_t *cigar = bam1_cigar(b); // this is cheating... @@ -282,10 +240,11 @@ int tv_fetch_func(const bam1_t *b, void *data) return 0; } -int tv_draw_aln(tview_t *tv, int tid, int pos) -{ +int base_draw_aln(tview_t *tv, int tid, int pos) + { + assert(tv!=NULL); // reset - clear(); + tv->my_clear(tv); tv->curr_tid = tid; tv->left_pos = pos; tv->last_pos = tv->left_pos - 1; tv->ccol = 0; @@ -293,7 +252,10 @@ int tv_draw_aln(tview_t *tv, int tid, int pos) if (tv->fai) { char *str; if (tv->ref) free(tv->ref); + assert(tv->curr_tid>=0); + str = (char*)calloc(strlen(tv->header->target_name[tv->curr_tid]) + 30, 1); + assert(str!=NULL); sprintf(str, "%s:%d-%d", tv->header->target_name[tv->curr_tid], tv->left_pos + 1, tv->left_pos + tv->mcol); tv->ref = fai_fetch(tv->fai, str, &tv->l_ref); free(str); @@ -305,143 +267,26 @@ int tv_draw_aln(tview_t *tv, int tid, int pos) while (tv->ccol < tv->mcol) { int pos = tv->last_pos + 1; - if (pos%10 == 0 && tv->mcol - tv->ccol >= 10) mvprintw(0, tv->ccol, "%-d", pos+1); - mvaddch(1, tv->ccol++, (tv->ref && pos < tv->l_ref)? tv->ref[pos - tv->left_pos] : 'N'); + if (pos%10 == 0 && tv->mcol - tv->ccol >= 10) tv->my_mvprintw(tv,0, tv->ccol, "%-d", pos+1); + tv->my_mvaddch(tv,1, tv->ccol++, (tv->ref && pos < tv->l_ref)? tv->ref[pos - tv->left_pos] : 'N'); ++tv->last_pos; } return 0; } -static void tv_win_goto(tview_t *tv, int *tid, int *pos) -{ - char str[256], *p; - int i, l = 0; - wborder(tv->wgoto, '|', '|', '-', '-', '+', '+', '+', '+'); - mvwprintw(tv->wgoto, 1, 2, "Goto: "); - for (;;) { - int c = wgetch(tv->wgoto); - wrefresh(tv->wgoto); - if (c == KEY_BACKSPACE || c == '\010' || c == '\177') { - if(l > 0) --l; - } else if (c == KEY_ENTER || c == '\012' || c == '\015') { - int _tid = -1, _beg, _end; - if (str[0] == '=') { - _beg = strtol(str+1, &p, 10) - 1; - if (_beg > 0) { - *pos = _beg; - return; - } - } else { - bam_parse_region(tv->header, str, &_tid, &_beg, &_end); - if (_tid >= 0) { - *tid = _tid; *pos = _beg; - return; - } - } - } else if (isgraph(c)) { - if (l < TV_MAX_GOTO) str[l++] = c; - } else if (c == '\027') l = 0; - else if (c == '\033') return; - str[l] = '\0'; - for (i = 0; i < TV_MAX_GOTO; ++i) mvwaddch(tv->wgoto, 1, 8 + i, ' '); - mvwprintw(tv->wgoto, 1, 8, "%s", str); - } -} -static void tv_win_help(tview_t *tv) { - int r = 1; - WINDOW *win = tv->whelp; - wborder(win, '|', '|', '-', '-', '+', '+', '+', '+'); - mvwprintw(win, r++, 2, " -=- Help -=- "); - r++; - mvwprintw(win, r++, 2, "? This window"); - mvwprintw(win, r++, 2, "Arrows Small scroll movement"); - mvwprintw(win, r++, 2, "h,j,k,l Small scroll movement"); - mvwprintw(win, r++, 2, "H,J,K,L Large scroll movement"); - mvwprintw(win, r++, 2, "ctrl-H Scroll 1k left"); - mvwprintw(win, r++, 2, "ctrl-L Scroll 1k right"); - mvwprintw(win, r++, 2, "space Scroll one screen"); - mvwprintw(win, r++, 2, "backspace Scroll back one screen"); - mvwprintw(win, r++, 2, "g Go to specific location"); - mvwprintw(win, r++, 2, "m Color for mapping qual"); - mvwprintw(win, r++, 2, "n Color for nucleotide"); - mvwprintw(win, r++, 2, "b Color for base quality"); - mvwprintw(win, r++, 2, "c Color for cs color"); - mvwprintw(win, r++, 2, "z Color for cs qual"); - mvwprintw(win, r++, 2, ". Toggle on/off dot view"); - mvwprintw(win, r++, 2, "s Toggle on/off ref skip"); - mvwprintw(win, r++, 2, "r Toggle on/off rd name"); - mvwprintw(win, r++, 2, "N Turn on nt view"); - mvwprintw(win, r++, 2, "C Turn on cs view"); - mvwprintw(win, r++, 2, "i Toggle on/off ins"); - mvwprintw(win, r++, 2, "q Exit"); - r++; - mvwprintw(win, r++, 2, "Underline: Secondary or orphan"); - mvwprintw(win, r++, 2, "Blue: 0-9 Green: 10-19"); - mvwprintw(win, r++, 2, "Yellow: 20-29 White: >=30"); - wrefresh(win); - wgetch(win); -} -void tv_loop(tview_t *tv) -{ - int tid, pos; - tid = tv->curr_tid; pos = tv->left_pos; - while (1) { - int c = getch(); - switch (c) { - case '?': tv_win_help(tv); break; - case '\033': - case 'q': goto end_loop; - case '/': - case 'g': tv_win_goto(tv, &tid, &pos); break; - case 'm': tv->color_for = TV_COLOR_MAPQ; break; - case 'b': tv->color_for = TV_COLOR_BASEQ; break; - case 'n': tv->color_for = TV_COLOR_NUCL; break; - case 'c': tv->color_for = TV_COLOR_COL; break; - case 'z': tv->color_for = TV_COLOR_COLQ; break; - case 's': tv->no_skip = !tv->no_skip; break; - case 'r': tv->show_name = !tv->show_name; break; - case KEY_LEFT: - case 'h': --pos; break; - case KEY_RIGHT: - case 'l': ++pos; break; - case KEY_SLEFT: - case 'H': pos -= 20; break; - case KEY_SRIGHT: - case 'L': pos += 20; break; - case '.': tv->is_dot = !tv->is_dot; break; - case 'N': tv->base_for = TV_BASE_NUCL; break; - case 'C': tv->base_for = TV_BASE_COLOR_SPACE; break; - case 'i': tv->ins = !tv->ins; break; - case '\010': pos -= 1000; break; - case '\014': pos += 1000; break; - case ' ': pos += tv->mcol; break; - case KEY_UP: - case 'j': --tv->row_shift; break; - case KEY_DOWN: - case 'k': ++tv->row_shift; break; - case KEY_BACKSPACE: - case '\177': pos -= tv->mcol; break; - case KEY_RESIZE: getmaxyx(stdscr, tv->mrow, tv->mcol); break; - default: continue; - } - if (pos < 0) pos = 0; - if (tv->row_shift < 0) tv->row_shift = 0; - tv_draw_aln(tv, tid, pos); - } -end_loop: - return; -} -void error(const char *format, ...) +static void error(const char *format, ...) { if ( !format ) { fprintf(stderr, "\n"); fprintf(stderr, "Usage: bamtk tview [options] [ref.fasta]\n"); fprintf(stderr, "Options:\n"); - fprintf(stderr, " -s STR display only reads from this sample\n"); + fprintf(stderr, " -p chr:pos go directly to this position\n"); + fprintf(stderr, " -s STR display only reads from this sample or group\n"); + fprintf(stderr, " -d display (H)tml or (C)urses or (T)ext \n"); fprintf(stderr, "\n\n"); } else @@ -454,31 +299,70 @@ void error(const char *format, ...) exit(-1); } +enum dipsay_mode {display_ncurses,display_html,display_text}; +extern tview_t* curses_tv_init(const char *fn, const char *fn_fa, const char *samples); +extern tview_t* html_tv_init(const char *fn, const char *fn_fa, const char *samples); +extern tview_t* text_tv_init(const char *fn, const char *fn_fa, const char *samples); int bam_tview_main(int argc, char *argv[]) -{ - tview_t *tv; - char *samples=NULL; + { + int view_mode=display_ncurses; + tview_t* tv=NULL; + char *samples=NULL, *position=NULL; int c; - while ((c = getopt(argc, argv, "s:")) >= 0) { + while ((c = getopt(argc, argv, "s:p:d:")) >= 0) { switch (c) { case 's': samples=optarg; break; + case 'p': position=optarg; break; + case 'd': + { + switch(optarg[0]) + { + case 'H': case 'h': view_mode=display_html;break; + case 'T': case 't': view_mode=display_text;break; + case 'C': case 'c': view_mode=display_ncurses;break; + default: view_mode=display_ncurses;break; + } + break; + } default: error(NULL); } } if (argc==optind) error(NULL); - tv = tv_init(argv[optind], (optind+1>=argc)? 0 : argv[optind+1], samples); - tv_draw_aln(tv, 0, 0); - tv_loop(tv); - tv_destroy(tv); - return 0; -} -#else // #ifdef _HAVE_CURSES -#include -#warning "No curses library is available; tview is disabled." -int bam_tview_main(int argc, char *argv[]) -{ - fprintf(stderr, "[bam_tview_main] The ncurses library is unavailable; tview is not compiled.\n"); - return 1; -} -#endif // #ifdef _HAVE_CURSES + + switch(view_mode) + { + case display_ncurses: + { + tv = curses_tv_init(argv[optind], (optind+1>=argc)? 0 : argv[optind+1], samples); + break; + } + case display_text: + { + tv = text_tv_init(argv[optind], (optind+1>=argc)? 0 : argv[optind+1], samples); + break; + } + case display_html: + { + tv = html_tv_init(argv[optind], (optind+1>=argc)? 0 : argv[optind+1], samples); + break; + } + } + if(tv==NULL) + { + error("cannot create view"); + return EXIT_FAILURE; + } + + if ( position ) + { + int _tid = -1, _beg, _end; + bam_parse_region(tv->header, position, &_tid, &_beg, &_end); + if (_tid >= 0) { tv->curr_tid = _tid; tv->left_pos = _beg; } + } + tv->my_drawaln(tv, tv->curr_tid, tv->left_pos); + tv->my_loop(tv); + tv->my_destroy(tv); + + return EXIT_SUCCESS; + }