Annotation of mandoc/term_ascii.c, Revision 1.71
1.71 ! schwarze 1: /* $Id: term_ascii.c,v 1.70 2025/06/26 17:26:23 schwarze Exp $ */
1.1 kristaps 2: /*
1.71 ! schwarze 3: * Copyright (c) 2014,2015,2017-2020,2025 Ingo Schwarze <schwarze@openbsd.org>
1.18 schwarze 4: * Copyright (c) 2010, 2011 Kristaps Dzonsons <kristaps@bsd.lv>
1.1 kristaps 5: *
6: * Permission to use, copy, modify, and distribute this software for any
7: * purpose with or without fee is hereby granted, provided that the above
8: * copyright notice and this permission notice appear in all copies.
9: *
1.44 schwarze 10: * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHORS DISCLAIM ALL WARRANTIES
1.1 kristaps 11: * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
1.44 schwarze 12: * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHORS BE LIABLE FOR
1.1 kristaps 13: * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
14: * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
15: * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
16: * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
17: */
18: #include "config.h"
19:
20: #include <sys/types.h>
21:
1.39 schwarze 22: #include <assert.h>
1.31 schwarze 23: #if HAVE_WCHAR
1.60 schwarze 24: #include <langinfo.h>
1.28 schwarze 25: #include <locale.h>
1.15 kristaps 26: #endif
1.1 kristaps 27: #include <stdint.h>
28: #include <stdio.h>
29: #include <stdlib.h>
1.60 schwarze 30: #include <string.h>
1.2 kristaps 31: #include <unistd.h>
1.31 schwarze 32: #if HAVE_WCHAR
1.28 schwarze 33: #include <wchar.h>
1.15 kristaps 34: #endif
1.1 kristaps 35:
1.11 kristaps 36: #include "mandoc.h"
1.22 schwarze 37: #include "mandoc_aux.h"
1.1 kristaps 38: #include "out.h"
39: #include "term.h"
1.44 schwarze 40: #include "manconf.h"
1.1 kristaps 41: #include "main.h"
42:
1.51 schwarze 43: static struct termp *ascii_init(enum termenc, const struct manoutput *);
1.45 schwarze 44: static int ascii_hspan(const struct termp *,
1.7 kristaps 45: const struct roffsu *);
1.71 ! schwarze 46: static size_t ascii_getwidth(const struct termp *, int);
1.7 kristaps 47: static void ascii_advance(struct termp *, size_t);
48: static void ascii_begin(struct termp *);
49: static void ascii_end(struct termp *);
1.1 kristaps 50: static void ascii_endline(struct termp *);
1.13 kristaps 51: static void ascii_letter(struct termp *, int);
1.45 schwarze 52: static void ascii_setwidth(struct termp *, int, int);
1.1 kristaps 53:
1.31 schwarze 54: #if HAVE_WCHAR
1.15 kristaps 55: static void locale_advance(struct termp *, size_t);
56: static void locale_endline(struct termp *);
57: static void locale_letter(struct termp *, int);
1.71 ! schwarze 58: static size_t locale_getwidth(const struct termp *, int);
1.15 kristaps 59: #endif
60:
1.25 schwarze 61:
1.14 kristaps 62: static struct termp *
1.51 schwarze 63: ascii_init(enum termenc enc, const struct manoutput *outopts)
1.1 kristaps 64: {
1.46 schwarze 65: #if HAVE_WCHAR
1.1 kristaps 66: char *v;
1.46 schwarze 67: #endif
1.14 kristaps 68: struct termp *p;
1.1 kristaps 69:
1.57 schwarze 70: p = mandoc_calloc(1, sizeof(*p));
71: p->tcol = p->tcols = mandoc_calloc(1, sizeof(*p->tcol));
72: p->maxtcol = 1;
1.1 kristaps 73:
1.47 schwarze 74: p->line = 1;
1.69 schwarze 75: p->defindent = 5;
1.41 schwarze 76: p->fontq = mandoc_reallocarray(NULL,
1.57 schwarze 77: (p->fontsz = 8), sizeof(*p->fontq));
1.41 schwarze 78: p->fontq[0] = p->fontl = TERMFONT_NONE;
1.5 kristaps 79:
1.1 kristaps 80: p->begin = ascii_begin;
81: p->end = ascii_end;
1.15 kristaps 82: p->hspan = ascii_hspan;
83: p->type = TERMTYPE_CHAR;
84: p->enc = TERMENC_ASCII;
85: p->advance = ascii_advance;
1.1 kristaps 86: p->endline = ascii_endline;
1.7 kristaps 87: p->letter = ascii_letter;
1.23 schwarze 88: p->setwidth = ascii_setwidth;
1.71 ! schwarze 89: p->getwidth = ascii_getwidth;
1.1 kristaps 90:
1.31 schwarze 91: #if HAVE_WCHAR
1.64 schwarze 92: if (enc != TERMENC_ASCII) {
1.52 schwarze 93:
94: /*
95: * Do not change any of this to LC_ALL. It might break
96: * the formatting by subtly changing the behaviour of
97: * various functions, for example strftime(3). As a
98: * worst case, it might even cause buffer overflows.
99: */
100:
1.64 schwarze 101: v = enc == TERMENC_LOCALE ?
1.52 schwarze 102: setlocale(LC_CTYPE, "") :
1.54 schwarze 103: setlocale(LC_CTYPE, UTF8_LOCALE);
1.60 schwarze 104:
105: /*
106: * We only support UTF-8,
107: * so revert to ASCII for anything else.
108: */
109:
110: if (v != NULL &&
111: strcmp(nl_langinfo(CODESET), "UTF-8") != 0)
112: v = setlocale(LC_CTYPE, "C");
113:
114: if (v != NULL && MB_CUR_MAX > 1) {
1.64 schwarze 115: p->enc = TERMENC_UTF8;
1.15 kristaps 116: p->advance = locale_advance;
117: p->endline = locale_endline;
118: p->letter = locale_letter;
1.71 ! schwarze 119: p->getwidth = locale_getwidth;
1.15 kristaps 120: }
1.17 kristaps 121: }
1.15 kristaps 122: #endif
1.71 ! schwarze 123: p->defrmargin = term_len(p, outopts->width ? outopts->width : 78);
! 124: p->lastrmargin = p->defrmargin;
1.15 kristaps 125:
1.44 schwarze 126: if (outopts->indent)
127: p->defindent = outopts->indent;
128: if (outopts->synopsisonly)
129: p->synopsisonly = 1;
1.1 kristaps 130:
1.61 schwarze 131: assert(p->defindent < UINT16_MAX);
132: assert(p->defrmargin < UINT16_MAX);
1.49 schwarze 133: return p;
1.5 kristaps 134: }
135:
1.14 kristaps 136: void *
1.51 schwarze 137: ascii_alloc(const struct manoutput *outopts)
1.14 kristaps 138: {
1.51 schwarze 139: return ascii_init(TERMENC_ASCII, outopts);
1.14 kristaps 140: }
1.17 kristaps 141:
142: void *
1.51 schwarze 143: utf8_alloc(const struct manoutput *outopts)
1.17 kristaps 144: {
1.51 schwarze 145: return ascii_init(TERMENC_UTF8, outopts);
1.17 kristaps 146: }
147:
1.14 kristaps 148: void *
1.51 schwarze 149: locale_alloc(const struct manoutput *outopts)
1.14 kristaps 150: {
1.51 schwarze 151: return ascii_init(TERMENC_LOCALE, outopts);
1.23 schwarze 152: }
153:
154: static void
1.45 schwarze 155: ascii_setwidth(struct termp *p, int iop, int width)
1.23 schwarze 156: {
1.57 schwarze 157: p->tcol->rmargin = p->defrmargin;
1.40 schwarze 158: if (iop > 0)
1.24 schwarze 159: p->defrmargin += width;
1.40 schwarze 160: else if (iop == 0)
1.45 schwarze 161: p->defrmargin = width ? (size_t)width : p->lastrmargin;
162: else if (p->defrmargin > (size_t)width)
1.24 schwarze 163: p->defrmargin -= width;
164: else
1.40 schwarze 165: p->defrmargin = 0;
1.71 ! schwarze 166: if (p->defrmargin > term_len(p, 1000))
! 167: p->defrmargin = term_len(p, 1000);
1.57 schwarze 168: p->lastrmargin = p->tcol->rmargin;
169: p->tcol->rmargin = p->maxrmargin = p->defrmargin;
1.42 schwarze 170: }
171:
172: void
1.53 schwarze 173: terminal_sepline(void *arg)
1.42 schwarze 174: {
175: struct termp *p;
1.71 ! schwarze 176: size_t i; /* Printed width in basic units. */
! 177: size_t sz; /* Width of a dash in basic units. */
1.42 schwarze 178:
179: p = (struct termp *)arg;
1.53 schwarze 180: (*p->endline)(p);
1.71 ! schwarze 181: sz = (*p->getwidth)(p, '-');
! 182: for (i = 0; i < p->defrmargin; i += sz)
1.53 schwarze 183: (*p->letter)(p, '-');
184: (*p->endline)(p);
185: (*p->endline)(p);
1.34 schwarze 186: }
187:
1.5 kristaps 188: static size_t
1.71 ! schwarze 189: ascii_getwidth(const struct termp *p, int c)
1.5 kristaps 190: {
1.71 ! schwarze 191: switch (c) {
! 192: case ASCII_BREAK:
! 193: case ASCII_NBRZW:
! 194: case ASCII_TABREF:
! 195: return 0;
! 196: default:
! 197: return 24;
! 198: }
1.1 kristaps 199: }
200:
201: void
202: ascii_free(void *arg)
203: {
204: term_free((struct termp *)arg);
205: }
206:
207: static void
1.13 kristaps 208: ascii_letter(struct termp *p, int c)
1.1 kristaps 209: {
210: putchar(c);
211: }
212:
213: static void
214: ascii_begin(struct termp *p)
215: {
216: (*p->headf)(p, p->argf);
217: }
218:
219: static void
220: ascii_end(struct termp *p)
221: {
222: (*p->footf)(p, p->argf);
223: }
224:
225: static void
226: ascii_endline(struct termp *p)
227: {
1.47 schwarze 228: p->line++;
1.65 schwarze 229: if ((int)p->tcol->offset > p->ti)
230: p->tcol->offset -= p->ti;
231: else
232: p->tcol->offset = 0;
1.56 schwarze 233: p->ti = 0;
1.71 ! schwarze 234: p->minbl = 0;
! 235: p->viscol = 0;
1.1 kristaps 236: putchar('\n');
237: }
238:
239: static void
240: ascii_advance(struct termp *p, size_t len)
241: {
1.71 ! schwarze 242: size_t dst; /* Destination column in basic units. */
! 243: size_t sz; /* Width of a space in basic units. */
! 244:
! 245: sz = (*p->getwidth)(p, ' ');
1.1 kristaps 246:
1.66 schwarze 247: /*
248: * XXX We used to have "assert(len < UINT16_MAX)" here.
249: * that is not quite right because the input document
250: * can trigger that by merely providing large input.
251: * For now, simply truncate.
252: */
1.71 ! schwarze 253: if (len > 256 * sz)
! 254: len = 256 * sz;
! 255:
! 256: dst = p->viscol + len;
! 257: while (p->viscol + sz / 2 < dst) {
1.1 kristaps 258: putchar(' ');
1.71 ! schwarze 259: p->viscol += sz;
! 260: }
1.1 kristaps 261: }
1.7 kristaps 262:
1.45 schwarze 263: static int
1.7 kristaps 264: ascii_hspan(const struct termp *p, const struct roffsu *su)
265: {
266: double r;
267:
268: switch (su->unit) {
1.29 kristaps 269: case SCALE_BU:
1.45 schwarze 270: r = su->scale;
1.29 kristaps 271: break;
1.25 schwarze 272: case SCALE_CM:
1.45 schwarze 273: r = su->scale * 240.0 / 2.54;
1.29 kristaps 274: break;
275: case SCALE_FS:
1.45 schwarze 276: r = su->scale * 65536.0;
1.7 kristaps 277: break;
1.25 schwarze 278: case SCALE_IN:
1.45 schwarze 279: r = su->scale * 240.0;
1.7 kristaps 280: break;
1.29 kristaps 281: case SCALE_MM:
1.45 schwarze 282: r = su->scale * 0.24;
1.29 kristaps 283: break;
1.45 schwarze 284: case SCALE_VS:
1.25 schwarze 285: case SCALE_PC:
1.45 schwarze 286: r = su->scale * 40.0;
1.7 kristaps 287: break;
1.25 schwarze 288: case SCALE_PT:
1.45 schwarze 289: r = su->scale * 10.0 / 3.0;
1.7 kristaps 290: break;
1.29 kristaps 291: case SCALE_EN:
292: case SCALE_EM:
1.45 schwarze 293: r = su->scale * 24.0;
1.29 kristaps 294: break;
1.32 schwarze 295: default:
1.29 kristaps 296: abort();
1.7 kristaps 297: }
1.49 schwarze 298: return r > 0.0 ? r + 0.01 : r - 0.01;
1.36 schwarze 299: }
300:
301: const char *
302: ascii_uc2str(int uc)
303: {
304: static const char nbrsp[2] = { ASCII_NBRSP, '\0' };
305: static const char *tab[] = {
306: "<NUL>","<SOH>","<STX>","<ETX>","<EOT>","<ENQ>","<ACK>","<BEL>",
307: "<BS>", "\t", "<LF>", "<VT>", "<FF>", "<CR>", "<SO>", "<SI>",
308: "<DLE>","<DC1>","<DC2>","<DC3>","<DC4>","<NAK>","<SYN>","<ETB>",
309: "<CAN>","<EM>", "<SUB>","<ESC>","<FS>", "<GS>", "<RS>", "<US>",
310: " ", "!", "\"", "#", "$", "%", "&", "'",
311: "(", ")", "*", "+", ",", "-", ".", "/",
312: "0", "1", "2", "3", "4", "5", "6", "7",
313: "8", "9", ":", ";", "<", "=", ">", "?",
314: "@", "A", "B", "C", "D", "E", "F", "G",
315: "H", "I", "J", "K", "L", "M", "N", "O",
316: "P", "Q", "R", "S", "T", "U", "V", "W",
317: "X", "Y", "Z", "[", "\\", "]", "^", "_",
318: "`", "a", "b", "c", "d", "e", "f", "g",
319: "h", "i", "j", "k", "l", "m", "n", "o",
320: "p", "q", "r", "s", "t", "u", "v", "w",
321: "x", "y", "z", "{", "|", "}", "~", "<DEL>",
322: "<80>", "<81>", "<82>", "<83>", "<84>", "<85>", "<86>", "<87>",
323: "<88>", "<89>", "<8A>", "<8B>", "<8C>", "<8D>", "<8E>", "<8F>",
324: "<90>", "<91>", "<92>", "<93>", "<94>", "<95>", "<96>", "<97>",
1.58 schwarze 325: "<98>", "<99>", "<9A>", "<9B>", "<9C>", "<9D>", "<9E>", "<9F>",
1.63 schwarze 326: nbrsp, "!", "/\bc", "-\bL", "o\bx", "=\bY", "|", "<section>",
1.37 schwarze 327: "\"", "(C)", "_\ba", "<<", "~", "", "(R)", "-",
1.59 schwarze 328: "<degree>","+-","^2", "^3", "'","<micro>","<paragraph>",".",
329: ",", "^1", "_\bo", ">>", "1/4", "1/2", "3/4", "?",
1.37 schwarze 330: "`\bA", "'\bA", "^\bA", "~\bA", "\"\bA","o\bA", "AE", ",\bC",
331: "`\bE", "'\bE", "^\bE", "\"\bE","`\bI", "'\bI", "^\bI", "\"\bI",
1.59 schwarze 332: "Dh", "~\bN", "`\bO", "'\bO", "^\bO", "~\bO", "\"\bO","x",
1.37 schwarze 333: "/\bO", "`\bU", "'\bU", "^\bU", "\"\bU","'\bY", "Th", "ss",
334: "`\ba", "'\ba", "^\ba", "~\ba", "\"\ba","o\ba", "ae", ",\bc",
335: "`\be", "'\be", "^\be", "\"\be","`\bi", "'\bi", "^\bi", "\"\bi",
1.59 schwarze 336: "dh", "~\bn", "`\bo", "'\bo", "^\bo", "~\bo", "\"\bo","/",
1.37 schwarze 337: "/\bo", "`\bu", "'\bu", "^\bu", "\"\bu","'\by", "th", "\"\by",
338: "A", "a", "A", "a", "A", "a", "'\bC", "'\bc",
339: "^\bC", "^\bc", "C", "c", "C", "c", "D", "d",
340: "/\bD", "/\bd", "E", "e", "E", "e", "E", "e",
341: "E", "e", "E", "e", "^\bG", "^\bg", "G", "g",
342: "G", "g", ",\bG", ",\bg", "^\bH", "^\bh", "/\bH", "/\bh",
343: "~\bI", "~\bi", "I", "i", "I", "i", "I", "i",
344: "I", "i", "IJ", "ij", "^\bJ", "^\bj", ",\bK", ",\bk",
345: "q", "'\bL", "'\bl", ",\bL", ",\bl", "L", "l", "L",
346: "l", "/\bL", "/\bl", "'\bN", "'\bn", ",\bN", ",\bn", "N",
1.36 schwarze 347: "n", "'n", "Ng", "ng", "O", "o", "O", "o",
1.37 schwarze 348: "O", "o", "OE", "oe", "'\bR", "'\br", ",\bR", ",\br",
349: "R", "r", "'\bS", "'\bs", "^\bS", "^\bs", ",\bS", ",\bs",
350: "S", "s", ",\bT", ",\bt", "T", "t", "/\bT", "/\bt",
351: "~\bU", "~\bu", "U", "u", "U", "u", "U", "u",
352: "U", "u", "U", "u", "^\bW", "^\bw", "^\bY", "^\by",
353: "\"\bY","'\bZ", "'\bz", "Z", "z", "Z", "z", "s",
1.36 schwarze 354: "b", "B", "B", "b", "6", "6", "O", "C",
355: "c", "D", "D", "D", "d", "d", "3", "@",
1.37 schwarze 356: "E", "F", ",\bf", "G", "G", "hv", "I", "/\bI",
357: "K", "k", "/\bl", "l", "W", "N", "n", "~\bO",
1.36 schwarze 358: "O", "o", "OI", "oi", "P", "p", "YR", "2",
359: "2", "SH", "sh", "t", "T", "t", "T", "U",
1.37 schwarze 360: "u", "Y", "V", "Y", "y", "/\bZ", "/\bz", "ZH",
361: "ZH", "zh", "zh", "/\b2", "5", "5", "ts", "w",
1.36 schwarze 362: "|", "||", "|=", "!", "DZ", "Dz", "dz", "LJ",
363: "Lj", "lj", "NJ", "Nj", "nj", "A", "a", "I",
364: "i", "O", "o", "U", "u", "U", "u", "U",
365: "u", "U", "u", "U", "u", "@", "A", "a",
1.37 schwarze 366: "A", "a", "AE", "ae", "/\bG", "/\bg", "G", "g",
1.36 schwarze 367: "K", "k", "O", "o", "O", "o", "ZH", "zh",
1.37 schwarze 368: "j", "DZ", "Dz", "dz", "'\bG", "'\bg", "HV", "W",
369: "`\bN", "`\bn", "A", "a", "'\bAE","'\bae","O", "o"};
1.36 schwarze 370:
1.39 schwarze 371: assert(uc >= 0);
1.36 schwarze 372: if ((size_t)uc < sizeof(tab)/sizeof(tab[0]))
1.49 schwarze 373: return tab[uc];
374: return mchars_uc2str(uc);
1.7 kristaps 375: }
376:
1.31 schwarze 377: #if HAVE_WCHAR
1.15 kristaps 378: static size_t
1.71 ! schwarze 379: locale_getwidth(const struct termp *p, int c)
1.15 kristaps 380: {
381: int rc;
382:
1.26 schwarze 383: if (c == ASCII_NBRSP)
384: c = ' ';
385: rc = wcwidth(c);
386: if (rc < 0)
387: rc = 0;
1.71 ! schwarze 388: return rc * 24;
1.15 kristaps 389: }
390:
391: static void
392: locale_advance(struct termp *p, size_t len)
393: {
1.71 ! schwarze 394: size_t dst; /* Destination column in basic units. */
! 395: size_t sz; /* Width of a space in basic units. */
! 396:
! 397: sz = (*p->getwidth)(p, ' ');
1.15 kristaps 398:
1.66 schwarze 399: /*
400: * XXX We used to have "assert(len < UINT16_MAX)" here.
401: * that is not quite right because the input document
402: * can trigger that by merely providing large input.
403: * For now, simply truncate.
404: */
1.71 ! schwarze 405: if (len > 256 * sz)
! 406: len = 256 * sz;
! 407:
! 408: dst = p->viscol + len;
! 409: while (p->viscol + sz / 2 < dst) {
1.15 kristaps 410: putwchar(L' ');
1.71 ! schwarze 411: p->viscol += sz;
! 412: }
1.15 kristaps 413: }
414:
415: static void
416: locale_endline(struct termp *p)
417: {
1.47 schwarze 418: p->line++;
1.65 schwarze 419: if ((int)p->tcol->offset > p->ti)
420: p->tcol->offset -= p->ti;
421: else
422: p->tcol->offset = 0;
1.56 schwarze 423: p->ti = 0;
1.71 ! schwarze 424: p->minbl = 0;
! 425: p->viscol = 0;
1.15 kristaps 426: putwchar(L'\n');
427: }
428:
429: static void
430: locale_letter(struct termp *p, int c)
431: {
432: putwchar(c);
433: }
434: #endif
CVSweb
![[BACK]](/cvsweb/icons/back.gif){kind=icon}
Return to term_ascii.c CVS log ![[TXT]](/cvsweb/icons/text.gif){kind=icon}
![[DIR]](/cvsweb/icons/dir.gif){kind=icon}
Up to [cvsweb.bsd.lv] / mandoc