Annotation of src/usr.bin/mandoc/term_ascii.c, Revision 1.53
1.53 ! schwarze 1: /* $OpenBSD: term_ascii.c,v 1.52 2020/09/09 13:40:24 schwarze Exp $ */
1.1 schwarze 2: /*
1.6 schwarze 3: * Copyright (c) 2010, 2011 Kristaps Dzonsons <kristaps@bsd.lv>
1.51 schwarze 4: * Copyright (c) 2014,2015,2017,2018,2020 Ingo Schwarze <schwarze@openbsd.org>
1.1 schwarze 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: *
1.31 schwarze 10: * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHORS DISCLAIM ALL WARRANTIES
1.1 schwarze 11: * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
1.31 schwarze 12: * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHORS BE LIABLE FOR
1.1 schwarze 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 <sys/types.h>
19:
1.26 schwarze 20: #include <assert.h>
1.45 schwarze 21: #include <langinfo.h>
1.6 schwarze 22: #include <locale.h>
1.1 schwarze 23: #include <stdint.h>
24: #include <stdio.h>
25: #include <stdlib.h>
1.45 schwarze 26: #include <string.h>
1.1 schwarze 27: #include <unistd.h>
1.6 schwarze 28: #include <wchar.h>
1.1 schwarze 29:
1.4 schwarze 30: #include "mandoc.h"
1.11 schwarze 31: #include "mandoc_aux.h"
1.1 schwarze 32: #include "out.h"
33: #include "term.h"
1.31 schwarze 34: #include "manconf.h"
1.1 schwarze 35: #include "main.h"
36:
1.50 schwarze 37: #define UTF8_LOCALE "en_US.UTF-8"
38:
1.37 schwarze 39: static struct termp *ascii_init(enum termenc, const struct manoutput *);
1.32 schwarze 40: static int ascii_hspan(const struct termp *,
1.3 schwarze 41: const struct roffsu *);
1.6 schwarze 42: static size_t ascii_width(const struct termp *, int);
1.3 schwarze 43: static void ascii_advance(struct termp *, size_t);
44: static void ascii_begin(struct termp *);
45: static void ascii_end(struct termp *);
1.1 schwarze 46: static void ascii_endline(struct termp *);
1.6 schwarze 47: static void ascii_letter(struct termp *, int);
1.32 schwarze 48: static void ascii_setwidth(struct termp *, int, int);
1.1 schwarze 49:
1.6 schwarze 50: static void locale_advance(struct termp *, size_t);
51: static void locale_endline(struct termp *);
52: static void locale_letter(struct termp *, int);
53: static size_t locale_width(const struct termp *, int);
1.1 schwarze 54:
1.14 schwarze 55:
1.6 schwarze 56: static struct termp *
1.37 schwarze 57: ascii_init(enum termenc enc, const struct manoutput *outopts)
1.1 schwarze 58: {
59: char *v;
1.6 schwarze 60: struct termp *p;
1.1 schwarze 61:
1.42 schwarze 62: p = mandoc_calloc(1, sizeof(*p));
63: p->tcol = p->tcols = mandoc_calloc(1, sizeof(*p->tcol));
64: p->maxtcol = 1;
1.1 schwarze 65:
1.33 schwarze 66: p->line = 1;
1.12 schwarze 67: p->defrmargin = p->lastrmargin = 78;
1.28 schwarze 68: p->fontq = mandoc_reallocarray(NULL,
1.42 schwarze 69: (p->fontsz = 8), sizeof(*p->fontq));
1.28 schwarze 70: p->fontq[0] = p->fontl = TERMFONT_NONE;
1.2 schwarze 71:
1.1 schwarze 72: p->begin = ascii_begin;
73: p->end = ascii_end;
1.6 schwarze 74: p->hspan = ascii_hspan;
75: p->type = TERMTYPE_CHAR;
76:
77: p->enc = TERMENC_ASCII;
78: p->advance = ascii_advance;
1.1 schwarze 79: p->endline = ascii_endline;
1.3 schwarze 80: p->letter = ascii_letter;
1.12 schwarze 81: p->setwidth = ascii_setwidth;
1.2 schwarze 82: p->width = ascii_width;
1.1 schwarze 83:
1.49 schwarze 84: if (enc != TERMENC_ASCII) {
1.38 schwarze 85:
86: /*
87: * Do not change any of this to LC_ALL. It might break
88: * the formatting by subtly changing the behaviour of
89: * various functions, for example strftime(3). As a
90: * worst case, it might even cause buffer overflows.
91: */
92:
1.49 schwarze 93: v = enc == TERMENC_LOCALE ?
1.38 schwarze 94: setlocale(LC_CTYPE, "") :
1.50 schwarze 95: setlocale(LC_CTYPE, UTF8_LOCALE);
1.45 schwarze 96:
97: /*
98: * We only support UTF-8,
99: * so revert to ASCII for anything else.
100: */
101:
102: if (v != NULL &&
103: strcmp(nl_langinfo(CODESET), "UTF-8") != 0)
104: v = setlocale(LC_CTYPE, "C");
105:
106: if (v != NULL && MB_CUR_MAX > 1) {
1.49 schwarze 107: p->enc = TERMENC_UTF8;
1.6 schwarze 108: p->advance = locale_advance;
109: p->endline = locale_endline;
110: p->letter = locale_letter;
111: p->width = locale_width;
112: }
113: }
114:
1.31 schwarze 115: if (outopts->mdoc) {
116: p->mdocstyle = 1;
117: p->defindent = 5;
118: }
119: if (outopts->indent)
120: p->defindent = outopts->indent;
121: if (outopts->width)
122: p->defrmargin = outopts->width;
123: if (outopts->synopsisonly)
124: p->synopsisonly = 1;
1.1 schwarze 125:
1.46 schwarze 126: assert(p->defindent < UINT16_MAX);
127: assert(p->defrmargin < UINT16_MAX);
1.35 schwarze 128: return p;
1.2 schwarze 129: }
130:
1.6 schwarze 131: void *
1.37 schwarze 132: ascii_alloc(const struct manoutput *outopts)
1.6 schwarze 133: {
134:
1.37 schwarze 135: return ascii_init(TERMENC_ASCII, outopts);
1.6 schwarze 136: }
137:
138: void *
1.37 schwarze 139: utf8_alloc(const struct manoutput *outopts)
1.6 schwarze 140: {
141:
1.37 schwarze 142: return ascii_init(TERMENC_UTF8, outopts);
1.6 schwarze 143: }
144:
145: void *
1.37 schwarze 146: locale_alloc(const struct manoutput *outopts)
1.6 schwarze 147: {
148:
1.37 schwarze 149: return ascii_init(TERMENC_LOCALE, outopts);
1.12 schwarze 150: }
151:
152: static void
1.32 schwarze 153: ascii_setwidth(struct termp *p, int iop, int width)
1.12 schwarze 154: {
155:
1.32 schwarze 156: width /= 24;
1.42 schwarze 157: p->tcol->rmargin = p->defrmargin;
1.27 schwarze 158: if (iop > 0)
1.13 schwarze 159: p->defrmargin += width;
1.27 schwarze 160: else if (iop == 0)
1.32 schwarze 161: p->defrmargin = width ? (size_t)width : p->lastrmargin;
162: else if (p->defrmargin > (size_t)width)
1.13 schwarze 163: p->defrmargin -= width;
164: else
1.27 schwarze 165: p->defrmargin = 0;
1.46 schwarze 166: if (p->defrmargin > 1000)
167: p->defrmargin = 1000;
1.42 schwarze 168: p->lastrmargin = p->tcol->rmargin;
169: p->tcol->rmargin = p->maxrmargin = p->defrmargin;
1.29 schwarze 170: }
171:
172: void
1.39 schwarze 173: terminal_sepline(void *arg)
1.29 schwarze 174: {
175: struct termp *p;
176: size_t i;
177:
178: p = (struct termp *)arg;
1.39 schwarze 179: (*p->endline)(p);
1.29 schwarze 180: for (i = 0; i < p->defrmargin; i++)
1.39 schwarze 181: (*p->letter)(p, '-');
182: (*p->endline)(p);
183: (*p->endline)(p);
1.6 schwarze 184: }
1.2 schwarze 185:
186: static size_t
1.6 schwarze 187: ascii_width(const struct termp *p, int c)
1.2 schwarze 188: {
1.53 ! schwarze 189: return c != ASCII_BREAK && c != ASCII_NBRZW;
1.1 schwarze 190: }
191:
192: void
193: ascii_free(void *arg)
194: {
195:
196: term_free((struct termp *)arg);
197: }
198:
199: static void
1.6 schwarze 200: ascii_letter(struct termp *p, int c)
1.1 schwarze 201: {
1.14 schwarze 202:
1.1 schwarze 203: putchar(c);
204: }
205:
206: static void
207: ascii_begin(struct termp *p)
208: {
209:
210: (*p->headf)(p, p->argf);
211: }
212:
213: static void
214: ascii_end(struct termp *p)
215: {
216:
217: (*p->footf)(p, p->argf);
218: }
219:
220: static void
221: ascii_endline(struct termp *p)
222: {
223:
1.33 schwarze 224: p->line++;
1.51 schwarze 225: if ((int)p->tcol->offset > p->ti)
226: p->tcol->offset -= p->ti;
227: else
228: p->tcol->offset = 0;
1.41 schwarze 229: p->ti = 0;
1.1 schwarze 230: putchar('\n');
231: }
232:
233: static void
234: ascii_advance(struct termp *p, size_t len)
235: {
1.14 schwarze 236: size_t i;
1.1 schwarze 237:
1.52 schwarze 238: /*
239: * XXX We used to have "assert(len < UINT16_MAX)" here.
240: * that is not quite right because the input document
241: * can trigger that by merely providing large input.
242: * For now, simply truncate.
243: */
244: if (len > 256)
245: len = 256;
1.1 schwarze 246: for (i = 0; i < len; i++)
247: putchar(' ');
248: }
1.3 schwarze 249:
1.32 schwarze 250: static int
1.3 schwarze 251: ascii_hspan(const struct termp *p, const struct roffsu *su)
252: {
253: double r;
254:
255: switch (su->unit) {
1.17 schwarze 256: case SCALE_BU:
1.32 schwarze 257: r = su->scale;
1.17 schwarze 258: break;
1.14 schwarze 259: case SCALE_CM:
1.32 schwarze 260: r = su->scale * 240.0 / 2.54;
1.17 schwarze 261: break;
262: case SCALE_FS:
1.32 schwarze 263: r = su->scale * 65536.0;
1.3 schwarze 264: break;
1.14 schwarze 265: case SCALE_IN:
1.32 schwarze 266: r = su->scale * 240.0;
1.3 schwarze 267: break;
1.17 schwarze 268: case SCALE_MM:
1.32 schwarze 269: r = su->scale * 0.24;
1.17 schwarze 270: break;
1.32 schwarze 271: case SCALE_VS:
1.14 schwarze 272: case SCALE_PC:
1.32 schwarze 273: r = su->scale * 40.0;
1.3 schwarze 274: break;
1.14 schwarze 275: case SCALE_PT:
1.32 schwarze 276: r = su->scale * 10.0 / 3.0;
1.3 schwarze 277: break;
1.17 schwarze 278: case SCALE_EN:
279: case SCALE_EM:
1.32 schwarze 280: r = su->scale * 24.0;
1.17 schwarze 281: break;
1.20 schwarze 282: default:
1.17 schwarze 283: abort();
1.3 schwarze 284: }
1.35 schwarze 285: return r > 0.0 ? r + 0.01 : r - 0.01;
1.22 schwarze 286: }
287:
288: const char *
289: ascii_uc2str(int uc)
290: {
291: static const char nbrsp[2] = { ASCII_NBRSP, '\0' };
292: static const char *tab[] = {
293: "<NUL>","<SOH>","<STX>","<ETX>","<EOT>","<ENQ>","<ACK>","<BEL>",
294: "<BS>", "\t", "<LF>", "<VT>", "<FF>", "<CR>", "<SO>", "<SI>",
295: "<DLE>","<DC1>","<DC2>","<DC3>","<DC4>","<NAK>","<SYN>","<ETB>",
296: "<CAN>","<EM>", "<SUB>","<ESC>","<FS>", "<GS>", "<RS>", "<US>",
297: " ", "!", "\"", "#", "$", "%", "&", "'",
298: "(", ")", "*", "+", ",", "-", ".", "/",
299: "0", "1", "2", "3", "4", "5", "6", "7",
300: "8", "9", ":", ";", "<", "=", ">", "?",
301: "@", "A", "B", "C", "D", "E", "F", "G",
302: "H", "I", "J", "K", "L", "M", "N", "O",
303: "P", "Q", "R", "S", "T", "U", "V", "W",
304: "X", "Y", "Z", "[", "\\", "]", "^", "_",
305: "`", "a", "b", "c", "d", "e", "f", "g",
306: "h", "i", "j", "k", "l", "m", "n", "o",
307: "p", "q", "r", "s", "t", "u", "v", "w",
308: "x", "y", "z", "{", "|", "}", "~", "<DEL>",
309: "<80>", "<81>", "<82>", "<83>", "<84>", "<85>", "<86>", "<87>",
310: "<88>", "<89>", "<8A>", "<8B>", "<8C>", "<8D>", "<8E>", "<8F>",
311: "<90>", "<91>", "<92>", "<93>", "<94>", "<95>", "<96>", "<97>",
1.43 schwarze 312: "<98>", "<99>", "<9A>", "<9B>", "<9C>", "<9D>", "<9E>", "<9F>",
1.48 schwarze 313: nbrsp, "!", "/\bc", "-\bL", "o\bx", "=\bY", "|", "<section>",
1.24 schwarze 314: "\"", "(C)", "_\ba", "<<", "~", "", "(R)", "-",
1.44 schwarze 315: "<degree>","+-","^2", "^3", "'","<micro>","<paragraph>",".",
316: ",", "^1", "_\bo", ">>", "1/4", "1/2", "3/4", "?",
1.24 schwarze 317: "`\bA", "'\bA", "^\bA", "~\bA", "\"\bA","o\bA", "AE", ",\bC",
318: "`\bE", "'\bE", "^\bE", "\"\bE","`\bI", "'\bI", "^\bI", "\"\bI",
1.44 schwarze 319: "Dh", "~\bN", "`\bO", "'\bO", "^\bO", "~\bO", "\"\bO","x",
1.24 schwarze 320: "/\bO", "`\bU", "'\bU", "^\bU", "\"\bU","'\bY", "Th", "ss",
321: "`\ba", "'\ba", "^\ba", "~\ba", "\"\ba","o\ba", "ae", ",\bc",
322: "`\be", "'\be", "^\be", "\"\be","`\bi", "'\bi", "^\bi", "\"\bi",
1.44 schwarze 323: "dh", "~\bn", "`\bo", "'\bo", "^\bo", "~\bo", "\"\bo","/",
1.24 schwarze 324: "/\bo", "`\bu", "'\bu", "^\bu", "\"\bu","'\by", "th", "\"\by",
325: "A", "a", "A", "a", "A", "a", "'\bC", "'\bc",
326: "^\bC", "^\bc", "C", "c", "C", "c", "D", "d",
327: "/\bD", "/\bd", "E", "e", "E", "e", "E", "e",
328: "E", "e", "E", "e", "^\bG", "^\bg", "G", "g",
329: "G", "g", ",\bG", ",\bg", "^\bH", "^\bh", "/\bH", "/\bh",
330: "~\bI", "~\bi", "I", "i", "I", "i", "I", "i",
331: "I", "i", "IJ", "ij", "^\bJ", "^\bj", ",\bK", ",\bk",
332: "q", "'\bL", "'\bl", ",\bL", ",\bl", "L", "l", "L",
333: "l", "/\bL", "/\bl", "'\bN", "'\bn", ",\bN", ",\bn", "N",
1.22 schwarze 334: "n", "'n", "Ng", "ng", "O", "o", "O", "o",
1.24 schwarze 335: "O", "o", "OE", "oe", "'\bR", "'\br", ",\bR", ",\br",
336: "R", "r", "'\bS", "'\bs", "^\bS", "^\bs", ",\bS", ",\bs",
337: "S", "s", ",\bT", ",\bt", "T", "t", "/\bT", "/\bt",
338: "~\bU", "~\bu", "U", "u", "U", "u", "U", "u",
339: "U", "u", "U", "u", "^\bW", "^\bw", "^\bY", "^\by",
340: "\"\bY","'\bZ", "'\bz", "Z", "z", "Z", "z", "s",
1.22 schwarze 341: "b", "B", "B", "b", "6", "6", "O", "C",
342: "c", "D", "D", "D", "d", "d", "3", "@",
1.24 schwarze 343: "E", "F", ",\bf", "G", "G", "hv", "I", "/\bI",
344: "K", "k", "/\bl", "l", "W", "N", "n", "~\bO",
1.22 schwarze 345: "O", "o", "OI", "oi", "P", "p", "YR", "2",
346: "2", "SH", "sh", "t", "T", "t", "T", "U",
1.24 schwarze 347: "u", "Y", "V", "Y", "y", "/\bZ", "/\bz", "ZH",
348: "ZH", "zh", "zh", "/\b2", "5", "5", "ts", "w",
1.22 schwarze 349: "|", "||", "|=", "!", "DZ", "Dz", "dz", "LJ",
350: "Lj", "lj", "NJ", "Nj", "nj", "A", "a", "I",
351: "i", "O", "o", "U", "u", "U", "u", "U",
352: "u", "U", "u", "U", "u", "@", "A", "a",
1.24 schwarze 353: "A", "a", "AE", "ae", "/\bG", "/\bg", "G", "g",
1.22 schwarze 354: "K", "k", "O", "o", "O", "o", "ZH", "zh",
1.24 schwarze 355: "j", "DZ", "Dz", "dz", "'\bG", "'\bg", "HV", "W",
356: "`\bN", "`\bn", "A", "a", "'\bAE","'\bae","O", "o"};
1.22 schwarze 357:
1.26 schwarze 358: assert(uc >= 0);
1.23 schwarze 359: if ((size_t)uc < sizeof(tab)/sizeof(tab[0]))
1.35 schwarze 360: return tab[uc];
361: return mchars_uc2str(uc);
1.3 schwarze 362: }
363:
1.6 schwarze 364: static size_t
365: locale_width(const struct termp *p, int c)
366: {
367: int rc;
368:
1.15 schwarze 369: if (c == ASCII_NBRSP)
370: c = ' ';
371: rc = wcwidth(c);
372: if (rc < 0)
373: rc = 0;
1.35 schwarze 374: return rc;
1.6 schwarze 375: }
376:
377: static void
378: locale_advance(struct termp *p, size_t len)
379: {
1.14 schwarze 380: size_t i;
1.6 schwarze 381:
1.52 schwarze 382: /*
383: * XXX We used to have "assert(len < UINT16_MAX)" here.
384: * that is not quite right because the input document
385: * can trigger that by merely providing large input.
386: * For now, simply truncate.
387: */
388: if (len > 256)
389: len = 256;
1.6 schwarze 390: for (i = 0; i < len; i++)
391: putwchar(L' ');
392: }
393:
394: static void
395: locale_endline(struct termp *p)
396: {
397:
1.33 schwarze 398: p->line++;
1.51 schwarze 399: if ((int)p->tcol->offset > p->ti)
400: p->tcol->offset -= p->ti;
401: else
402: p->tcol->offset = 0;
1.41 schwarze 403: p->ti = 0;
1.6 schwarze 404: putwchar(L'\n');
405: }
406:
407: static void
408: locale_letter(struct termp *p, int c)
409: {
1.14 schwarze 410:
1.6 schwarze 411: putwchar(c);
412: }
![[BACK]](/icons/back.gif){kind=icon}
Return to term_ascii.c CVS log ![[TXT]](/icons/text.gif){kind=icon}
![[DIR]](/icons/dir.gif){kind=icon}
Up to [local] / src / usr.bin / mandoc