Annotation of src/usr.bin/mandoc/term_ascii.c, Revision 1.55
1.55 ! schwarze 1: /* $OpenBSD: term_ascii.c,v 1.54 2022/08/16 17:44:53 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;
1.55 ! schwarze 67: p->defindent = 5;
1.12 schwarze 68: p->defrmargin = p->lastrmargin = 78;
1.28 schwarze 69: p->fontq = mandoc_reallocarray(NULL,
1.42 schwarze 70: (p->fontsz = 8), sizeof(*p->fontq));
1.28 schwarze 71: p->fontq[0] = p->fontl = TERMFONT_NONE;
1.2 schwarze 72:
1.1 schwarze 73: p->begin = ascii_begin;
74: p->end = ascii_end;
1.6 schwarze 75: p->hspan = ascii_hspan;
76: p->type = TERMTYPE_CHAR;
77:
78: p->enc = TERMENC_ASCII;
79: p->advance = ascii_advance;
1.1 schwarze 80: p->endline = ascii_endline;
1.3 schwarze 81: p->letter = ascii_letter;
1.12 schwarze 82: p->setwidth = ascii_setwidth;
1.2 schwarze 83: p->width = ascii_width;
1.1 schwarze 84:
1.49 schwarze 85: if (enc != TERMENC_ASCII) {
1.38 schwarze 86:
87: /*
88: * Do not change any of this to LC_ALL. It might break
89: * the formatting by subtly changing the behaviour of
90: * various functions, for example strftime(3). As a
91: * worst case, it might even cause buffer overflows.
92: */
93:
1.49 schwarze 94: v = enc == TERMENC_LOCALE ?
1.38 schwarze 95: setlocale(LC_CTYPE, "") :
1.50 schwarze 96: setlocale(LC_CTYPE, UTF8_LOCALE);
1.45 schwarze 97:
98: /*
99: * We only support UTF-8,
100: * so revert to ASCII for anything else.
101: */
102:
103: if (v != NULL &&
104: strcmp(nl_langinfo(CODESET), "UTF-8") != 0)
105: v = setlocale(LC_CTYPE, "C");
106:
107: if (v != NULL && MB_CUR_MAX > 1) {
1.49 schwarze 108: p->enc = TERMENC_UTF8;
1.6 schwarze 109: p->advance = locale_advance;
110: p->endline = locale_endline;
111: p->letter = locale_letter;
112: p->width = locale_width;
113: }
114: }
115:
1.55 ! schwarze 116: if (outopts->mdoc)
1.31 schwarze 117: p->mdocstyle = 1;
118: if (outopts->indent)
119: p->defindent = outopts->indent;
120: if (outopts->width)
121: p->defrmargin = outopts->width;
122: if (outopts->synopsisonly)
123: p->synopsisonly = 1;
1.1 schwarze 124:
1.46 schwarze 125: assert(p->defindent < UINT16_MAX);
126: assert(p->defrmargin < UINT16_MAX);
1.35 schwarze 127: return p;
1.2 schwarze 128: }
129:
1.6 schwarze 130: void *
1.37 schwarze 131: ascii_alloc(const struct manoutput *outopts)
1.6 schwarze 132: {
133:
1.37 schwarze 134: return ascii_init(TERMENC_ASCII, outopts);
1.6 schwarze 135: }
136:
137: void *
1.37 schwarze 138: utf8_alloc(const struct manoutput *outopts)
1.6 schwarze 139: {
140:
1.37 schwarze 141: return ascii_init(TERMENC_UTF8, outopts);
1.6 schwarze 142: }
143:
144: void *
1.37 schwarze 145: locale_alloc(const struct manoutput *outopts)
1.6 schwarze 146: {
147:
1.37 schwarze 148: return ascii_init(TERMENC_LOCALE, outopts);
1.12 schwarze 149: }
150:
151: static void
1.32 schwarze 152: ascii_setwidth(struct termp *p, int iop, int width)
1.12 schwarze 153: {
154:
1.32 schwarze 155: width /= 24;
1.42 schwarze 156: p->tcol->rmargin = p->defrmargin;
1.27 schwarze 157: if (iop > 0)
1.13 schwarze 158: p->defrmargin += width;
1.27 schwarze 159: else if (iop == 0)
1.32 schwarze 160: p->defrmargin = width ? (size_t)width : p->lastrmargin;
161: else if (p->defrmargin > (size_t)width)
1.13 schwarze 162: p->defrmargin -= width;
163: else
1.27 schwarze 164: p->defrmargin = 0;
1.46 schwarze 165: if (p->defrmargin > 1000)
166: p->defrmargin = 1000;
1.42 schwarze 167: p->lastrmargin = p->tcol->rmargin;
168: p->tcol->rmargin = p->maxrmargin = p->defrmargin;
1.29 schwarze 169: }
170:
171: void
1.39 schwarze 172: terminal_sepline(void *arg)
1.29 schwarze 173: {
174: struct termp *p;
175: size_t i;
176:
177: p = (struct termp *)arg;
1.39 schwarze 178: (*p->endline)(p);
1.29 schwarze 179: for (i = 0; i < p->defrmargin; i++)
1.39 schwarze 180: (*p->letter)(p, '-');
181: (*p->endline)(p);
182: (*p->endline)(p);
1.6 schwarze 183: }
1.2 schwarze 184:
185: static size_t
1.6 schwarze 186: ascii_width(const struct termp *p, int c)
1.2 schwarze 187: {
1.54 schwarze 188: return c != ASCII_BREAK && c != ASCII_NBRZW && c != ASCII_TABREF;
1.1 schwarze 189: }
190:
191: void
192: ascii_free(void *arg)
193: {
194:
195: term_free((struct termp *)arg);
196: }
197:
198: static void
1.6 schwarze 199: ascii_letter(struct termp *p, int c)
1.1 schwarze 200: {
1.14 schwarze 201:
1.1 schwarze 202: putchar(c);
203: }
204:
205: static void
206: ascii_begin(struct termp *p)
207: {
208:
209: (*p->headf)(p, p->argf);
210: }
211:
212: static void
213: ascii_end(struct termp *p)
214: {
215:
216: (*p->footf)(p, p->argf);
217: }
218:
219: static void
220: ascii_endline(struct termp *p)
221: {
222:
1.33 schwarze 223: p->line++;
1.51 schwarze 224: if ((int)p->tcol->offset > p->ti)
225: p->tcol->offset -= p->ti;
226: else
227: p->tcol->offset = 0;
1.41 schwarze 228: p->ti = 0;
1.1 schwarze 229: putchar('\n');
230: }
231:
232: static void
233: ascii_advance(struct termp *p, size_t len)
234: {
1.14 schwarze 235: size_t i;
1.1 schwarze 236:
1.52 schwarze 237: /*
238: * XXX We used to have "assert(len < UINT16_MAX)" here.
239: * that is not quite right because the input document
240: * can trigger that by merely providing large input.
241: * For now, simply truncate.
242: */
243: if (len > 256)
244: len = 256;
1.1 schwarze 245: for (i = 0; i < len; i++)
246: putchar(' ');
247: }
1.3 schwarze 248:
1.32 schwarze 249: static int
1.3 schwarze 250: ascii_hspan(const struct termp *p, const struct roffsu *su)
251: {
252: double r;
253:
254: switch (su->unit) {
1.17 schwarze 255: case SCALE_BU:
1.32 schwarze 256: r = su->scale;
1.17 schwarze 257: break;
1.14 schwarze 258: case SCALE_CM:
1.32 schwarze 259: r = su->scale * 240.0 / 2.54;
1.17 schwarze 260: break;
261: case SCALE_FS:
1.32 schwarze 262: r = su->scale * 65536.0;
1.3 schwarze 263: break;
1.14 schwarze 264: case SCALE_IN:
1.32 schwarze 265: r = su->scale * 240.0;
1.3 schwarze 266: break;
1.17 schwarze 267: case SCALE_MM:
1.32 schwarze 268: r = su->scale * 0.24;
1.17 schwarze 269: break;
1.32 schwarze 270: case SCALE_VS:
1.14 schwarze 271: case SCALE_PC:
1.32 schwarze 272: r = su->scale * 40.0;
1.3 schwarze 273: break;
1.14 schwarze 274: case SCALE_PT:
1.32 schwarze 275: r = su->scale * 10.0 / 3.0;
1.3 schwarze 276: break;
1.17 schwarze 277: case SCALE_EN:
278: case SCALE_EM:
1.32 schwarze 279: r = su->scale * 24.0;
1.17 schwarze 280: break;
1.20 schwarze 281: default:
1.17 schwarze 282: abort();
1.3 schwarze 283: }
1.35 schwarze 284: return r > 0.0 ? r + 0.01 : r - 0.01;
1.22 schwarze 285: }
286:
287: const char *
288: ascii_uc2str(int uc)
289: {
290: static const char nbrsp[2] = { ASCII_NBRSP, '\0' };
291: static const char *tab[] = {
292: "<NUL>","<SOH>","<STX>","<ETX>","<EOT>","<ENQ>","<ACK>","<BEL>",
293: "<BS>", "\t", "<LF>", "<VT>", "<FF>", "<CR>", "<SO>", "<SI>",
294: "<DLE>","<DC1>","<DC2>","<DC3>","<DC4>","<NAK>","<SYN>","<ETB>",
295: "<CAN>","<EM>", "<SUB>","<ESC>","<FS>", "<GS>", "<RS>", "<US>",
296: " ", "!", "\"", "#", "$", "%", "&", "'",
297: "(", ")", "*", "+", ",", "-", ".", "/",
298: "0", "1", "2", "3", "4", "5", "6", "7",
299: "8", "9", ":", ";", "<", "=", ">", "?",
300: "@", "A", "B", "C", "D", "E", "F", "G",
301: "H", "I", "J", "K", "L", "M", "N", "O",
302: "P", "Q", "R", "S", "T", "U", "V", "W",
303: "X", "Y", "Z", "[", "\\", "]", "^", "_",
304: "`", "a", "b", "c", "d", "e", "f", "g",
305: "h", "i", "j", "k", "l", "m", "n", "o",
306: "p", "q", "r", "s", "t", "u", "v", "w",
307: "x", "y", "z", "{", "|", "}", "~", "<DEL>",
308: "<80>", "<81>", "<82>", "<83>", "<84>", "<85>", "<86>", "<87>",
309: "<88>", "<89>", "<8A>", "<8B>", "<8C>", "<8D>", "<8E>", "<8F>",
310: "<90>", "<91>", "<92>", "<93>", "<94>", "<95>", "<96>", "<97>",
1.43 schwarze 311: "<98>", "<99>", "<9A>", "<9B>", "<9C>", "<9D>", "<9E>", "<9F>",
1.48 schwarze 312: nbrsp, "!", "/\bc", "-\bL", "o\bx", "=\bY", "|", "<section>",
1.24 schwarze 313: "\"", "(C)", "_\ba", "<<", "~", "", "(R)", "-",
1.44 schwarze 314: "<degree>","+-","^2", "^3", "'","<micro>","<paragraph>",".",
315: ",", "^1", "_\bo", ">>", "1/4", "1/2", "3/4", "?",
1.24 schwarze 316: "`\bA", "'\bA", "^\bA", "~\bA", "\"\bA","o\bA", "AE", ",\bC",
317: "`\bE", "'\bE", "^\bE", "\"\bE","`\bI", "'\bI", "^\bI", "\"\bI",
1.44 schwarze 318: "Dh", "~\bN", "`\bO", "'\bO", "^\bO", "~\bO", "\"\bO","x",
1.24 schwarze 319: "/\bO", "`\bU", "'\bU", "^\bU", "\"\bU","'\bY", "Th", "ss",
320: "`\ba", "'\ba", "^\ba", "~\ba", "\"\ba","o\ba", "ae", ",\bc",
321: "`\be", "'\be", "^\be", "\"\be","`\bi", "'\bi", "^\bi", "\"\bi",
1.44 schwarze 322: "dh", "~\bn", "`\bo", "'\bo", "^\bo", "~\bo", "\"\bo","/",
1.24 schwarze 323: "/\bo", "`\bu", "'\bu", "^\bu", "\"\bu","'\by", "th", "\"\by",
324: "A", "a", "A", "a", "A", "a", "'\bC", "'\bc",
325: "^\bC", "^\bc", "C", "c", "C", "c", "D", "d",
326: "/\bD", "/\bd", "E", "e", "E", "e", "E", "e",
327: "E", "e", "E", "e", "^\bG", "^\bg", "G", "g",
328: "G", "g", ",\bG", ",\bg", "^\bH", "^\bh", "/\bH", "/\bh",
329: "~\bI", "~\bi", "I", "i", "I", "i", "I", "i",
330: "I", "i", "IJ", "ij", "^\bJ", "^\bj", ",\bK", ",\bk",
331: "q", "'\bL", "'\bl", ",\bL", ",\bl", "L", "l", "L",
332: "l", "/\bL", "/\bl", "'\bN", "'\bn", ",\bN", ",\bn", "N",
1.22 schwarze 333: "n", "'n", "Ng", "ng", "O", "o", "O", "o",
1.24 schwarze 334: "O", "o", "OE", "oe", "'\bR", "'\br", ",\bR", ",\br",
335: "R", "r", "'\bS", "'\bs", "^\bS", "^\bs", ",\bS", ",\bs",
336: "S", "s", ",\bT", ",\bt", "T", "t", "/\bT", "/\bt",
337: "~\bU", "~\bu", "U", "u", "U", "u", "U", "u",
338: "U", "u", "U", "u", "^\bW", "^\bw", "^\bY", "^\by",
339: "\"\bY","'\bZ", "'\bz", "Z", "z", "Z", "z", "s",
1.22 schwarze 340: "b", "B", "B", "b", "6", "6", "O", "C",
341: "c", "D", "D", "D", "d", "d", "3", "@",
1.24 schwarze 342: "E", "F", ",\bf", "G", "G", "hv", "I", "/\bI",
343: "K", "k", "/\bl", "l", "W", "N", "n", "~\bO",
1.22 schwarze 344: "O", "o", "OI", "oi", "P", "p", "YR", "2",
345: "2", "SH", "sh", "t", "T", "t", "T", "U",
1.24 schwarze 346: "u", "Y", "V", "Y", "y", "/\bZ", "/\bz", "ZH",
347: "ZH", "zh", "zh", "/\b2", "5", "5", "ts", "w",
1.22 schwarze 348: "|", "||", "|=", "!", "DZ", "Dz", "dz", "LJ",
349: "Lj", "lj", "NJ", "Nj", "nj", "A", "a", "I",
350: "i", "O", "o", "U", "u", "U", "u", "U",
351: "u", "U", "u", "U", "u", "@", "A", "a",
1.24 schwarze 352: "A", "a", "AE", "ae", "/\bG", "/\bg", "G", "g",
1.22 schwarze 353: "K", "k", "O", "o", "O", "o", "ZH", "zh",
1.24 schwarze 354: "j", "DZ", "Dz", "dz", "'\bG", "'\bg", "HV", "W",
355: "`\bN", "`\bn", "A", "a", "'\bAE","'\bae","O", "o"};
1.22 schwarze 356:
1.26 schwarze 357: assert(uc >= 0);
1.23 schwarze 358: if ((size_t)uc < sizeof(tab)/sizeof(tab[0]))
1.35 schwarze 359: return tab[uc];
360: return mchars_uc2str(uc);
1.3 schwarze 361: }
362:
1.6 schwarze 363: static size_t
364: locale_width(const struct termp *p, int c)
365: {
366: int rc;
367:
1.15 schwarze 368: if (c == ASCII_NBRSP)
369: c = ' ';
370: rc = wcwidth(c);
371: if (rc < 0)
372: rc = 0;
1.35 schwarze 373: return rc;
1.6 schwarze 374: }
375:
376: static void
377: locale_advance(struct termp *p, size_t len)
378: {
1.14 schwarze 379: size_t i;
1.6 schwarze 380:
1.52 schwarze 381: /*
382: * XXX We used to have "assert(len < UINT16_MAX)" here.
383: * that is not quite right because the input document
384: * can trigger that by merely providing large input.
385: * For now, simply truncate.
386: */
387: if (len > 256)
388: len = 256;
1.6 schwarze 389: for (i = 0; i < len; i++)
390: putwchar(L' ');
391: }
392:
393: static void
394: locale_endline(struct termp *p)
395: {
396:
1.33 schwarze 397: p->line++;
1.51 schwarze 398: if ((int)p->tcol->offset > p->ti)
399: p->tcol->offset -= p->ti;
400: else
401: p->tcol->offset = 0;
1.41 schwarze 402: p->ti = 0;
1.6 schwarze 403: putwchar(L'\n');
404: }
405:
406: static void
407: locale_letter(struct termp *p, int c)
408: {
1.14 schwarze 409:
1.6 schwarze 410: putwchar(c);
411: }
![[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