[BACK]Return to mdocterm.c CVS log [TXT][DIR] Up to [cvsweb.bsd.lv] / mandoc

Annotation of mandoc/mdocterm.c, Revision 1.38

1.38    ! kristaps    1: /* $Id: mdocterm.c,v 1.37 2009/03/09 13:17:49 kristaps Exp $ */
1.1       kristaps    2: /*
                      3:  * Copyright (c) 2008 Kristaps Dzonsons <kristaps@kth.se>
                      4:  *
                      5:  * Permission to use, copy, modify, and distribute this software for any
                      6:  * purpose with or without fee is hereby granted, provided that the
                      7:  * above copyright notice and this permission notice appear in all
                      8:  * copies.
                      9:  *
                     10:  * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL
                     11:  * WARRANTIES WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED
                     12:  * WARRANTIES OF MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE
                     13:  * AUTHOR BE LIABLE FOR ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL
                     14:  * DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR
                     15:  * PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER
                     16:  * TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR
                     17:  * PERFORMANCE OF THIS SOFTWARE.
                     18:  */
1.37      kristaps   19: #include <sys/types.h>
                     20:
1.1       kristaps   21: #include <assert.h>
1.3       kristaps   22: #include <ctype.h>
1.1       kristaps   23: #include <err.h>
                     24: #include <getopt.h>
1.3       kristaps   25: #include <stdio.h>
1.1       kristaps   26: #include <stdlib.h>
1.3       kristaps   27: #include <string.h>
1.5       kristaps   28:
1.2       kristaps   29: #include "mmain.h"
1.1       kristaps   30: #include "term.h"
                     31:
1.27      kristaps   32: struct termenc {
                     33:        const char       *enc;
                     34:        int               sym;
                     35: };
                     36:
1.3       kristaps   37: static void              body(struct termp *,
1.12      kristaps   38:                                struct termpair *,
1.3       kristaps   39:                                const struct mdoc_meta *,
                     40:                                const struct mdoc_node *);
                     41: static void              header(struct termp *,
                     42:                                const struct mdoc_meta *);
                     43: static void              footer(struct termp *,
                     44:                                const struct mdoc_meta *);
                     45:
                     46: static void              pword(struct termp *, const char *, size_t);
1.25      kristaps   47: static void              pescape(struct termp *, const char *,
                     48:                                size_t *, size_t);
                     49: static void              nescape(struct termp *,
1.12      kristaps   50:                                const char *, size_t);
1.3       kristaps   51: static void              chara(struct termp *, char);
1.25      kristaps   52: static void              stringa(struct termp *,
                     53:                                const char *, size_t);
                     54: static void              symbola(struct termp *, enum tsym);
1.38    ! kristaps   55: static void              sanity(const struct mdoc_node *);
1.26      kristaps   56: static void              stylea(struct termp *, enum tstyle);
1.3       kristaps   57:
                     58: #ifdef __linux__
                     59: extern size_t            strlcat(char *, const char *, size_t);
                     60: extern size_t            strlcpy(char *, const char *, size_t);
                     61: #endif
                     62:
1.27      kristaps   63: static struct termenc    termenc1[] = {
                     64:        { "\\",           TERMSYM_SLASH },
                     65:        { "\'",           TERMSYM_RSQUOTE },
                     66:        { "`",            TERMSYM_LSQUOTE },
                     67:        { "-",            TERMSYM_HYPHEN },
                     68:        { " ",            TERMSYM_SPACE },
                     69:        { ".",            TERMSYM_PERIOD },
                     70:        { "&",            TERMSYM_BREAK },
                     71:        { "e",            TERMSYM_SLASH },
                     72:        { "q",            TERMSYM_DQUOTE },
                     73:        { NULL,           0 }
                     74: };
                     75:
                     76: static struct termenc    termenc2[] = {
                     77:        { "rB",           TERMSYM_RBRACK },
                     78:        { "lB",           TERMSYM_LBRACK },
1.31      kristaps   79:        { "ra",           TERMSYM_RANGLE },
                     80:        { "la",           TERMSYM_LANGLE },
1.27      kristaps   81:        { "Lq",           TERMSYM_LDQUOTE },
                     82:        { "lq",           TERMSYM_LDQUOTE },
                     83:        { "Rq",           TERMSYM_RDQUOTE },
                     84:        { "rq",           TERMSYM_RDQUOTE },
                     85:        { "oq",           TERMSYM_LSQUOTE },
                     86:        { "aq",           TERMSYM_RSQUOTE },
                     87:
                     88:        { "<-",           TERMSYM_LARROW },
                     89:        { "->",           TERMSYM_RARROW },
                     90:        { "ua",           TERMSYM_UARROW },
                     91:        { "da",           TERMSYM_DARROW },
                     92:
                     93:        { "bu",           TERMSYM_BULLET },
                     94:        { "Ba",           TERMSYM_BAR },
                     95:        { "ba",           TERMSYM_BAR },
                     96:        { "co",           TERMSYM_COPY },
                     97:        { "Am",           TERMSYM_AMP },
                     98:
                     99:        { "Le",           TERMSYM_LE },
                    100:        { "<=",           TERMSYM_LE },
                    101:        { "Ge",           TERMSYM_GE },
1.31      kristaps  102:        { ">=",           TERMSYM_GE },
1.27      kristaps  103:        { "==",           TERMSYM_EQ },
                    104:        { "Ne",           TERMSYM_NEQ },
                    105:        { "!=",           TERMSYM_NEQ },
                    106:        { "Pm",           TERMSYM_PLUSMINUS },
                    107:        { "+-",           TERMSYM_PLUSMINUS },
                    108:        { "If",           TERMSYM_INF2 },
                    109:        { "if",           TERMSYM_INF },
                    110:        { "Na",           TERMSYM_NAN },
                    111:        { "na",           TERMSYM_NAN },
                    112:        { "**",           TERMSYM_ASTERISK },
                    113:        { "Gt",           TERMSYM_GT },
                    114:        { "Lt",           TERMSYM_LT },
                    115:
                    116:        { "aa",           TERMSYM_ACUTE },
                    117:        { "ga",           TERMSYM_GRAVE },
                    118:
                    119:        { "en",           TERMSYM_EN },
                    120:        { "em",           TERMSYM_EM },
                    121:
                    122:        { "Pi",           TERMSYM_PI },
                    123:        { NULL,           0 }
                    124: };
                    125:
1.25      kristaps  126: static struct termsym    termsym_ansi[] = {
                    127:        { "]", 1 },             /* TERMSYM_RBRACK */
                    128:        { "[", 1 },             /* TERMSYM_LBRACK */
                    129:        { "<-", 2 },            /* TERMSYM_LARROW */
                    130:        { "->", 2 },            /* TERMSYM_RARROW */
                    131:        { "^", 1 },             /* TERMSYM_UARROW */
                    132:        { "v", 1 },             /* TERMSYM_DARROW */
                    133:        { "`", 1 },             /* TERMSYM_LSQUOTE */
                    134:        { "\'", 1 },            /* TERMSYM_RSQUOTE */
                    135:        { "\'", 1 },            /* TERMSYM_SQUOTE */
                    136:        { "``", 2 },            /* TERMSYM_LDQUOTE */
                    137:        { "\'\'", 2 },          /* TERMSYM_RDQUOTE */
                    138:        { "\"", 1 },            /* TERMSYM_DQUOTE */
                    139:        { "<", 1 },             /* TERMSYM_LT */
                    140:        { ">", 1 },             /* TERMSYM_GT */
                    141:        { "<=", 2 },            /* TERMSYM_LE */
                    142:        { ">=", 2 },            /* TERMSYM_GE */
                    143:        { "==", 2 },            /* TERMSYM_EQ */
                    144:        { "!=", 2 },            /* TERMSYM_NEQ */
                    145:        { "\'", 1 },            /* TERMSYM_ACUTE */
                    146:        { "`", 1 },             /* TERMSYM_GRAVE */
                    147:        { "pi", 2 },            /* TERMSYM_PI */
                    148:        { "+=", 2 },            /* TERMSYM_PLUSMINUS */
                    149:        { "oo", 2 },            /* TERMSYM_INF */
                    150:        { "infinity", 8 },      /* TERMSYM_INF2 */
                    151:        { "NaN", 3 },           /* TERMSYM_NAN */
                    152:        { "|", 1 },             /* TERMSYM_BAR */
                    153:        { "o", 1 },             /* TERMSYM_BULLET */
1.27      kristaps  154:        { "&", 1 },             /* TERMSYM_AMP */
                    155:        { "--", 2 },            /* TERMSYM_EM */
                    156:        { "-", 1 },             /* TERMSYM_EN */
                    157:        { "(C)", 3 },           /* TERMSYM_COPY */
                    158:        { "*", 1 },             /* TERMSYM_ASTERISK */
                    159:        { "\\", 1 },            /* TERMSYM_SLASH */
                    160:        { "-", 1 },             /* TERMSYM_HYPHEN */
                    161:        { " ", 1 },             /* TERMSYM_SPACE */
                    162:        { ".", 1 },             /* TERMSYM_PERIOD */
                    163:        { "", 0 },              /* TERMSYM_BREAK */
1.31      kristaps  164:        { "<", 1 },             /* TERMSYM_LANGLE */
                    165:        { ">", 1 },             /* TERMSYM_RANGLE */
1.25      kristaps  166: };
                    167:
                    168: static const char        ansi_clear[]  = { 27, '[', '0', 'm' };
                    169: static const char        ansi_bold[]  = { 27, '[', '1', 'm' };
                    170: static const char        ansi_under[]  = { 27, '[', '4', 'm' };
                    171:
                    172: static struct termsym    termstyle_ansi[] = {
                    173:        { ansi_clear, 4 },
                    174:        { ansi_bold, 4 },
                    175:        { ansi_under, 4 }
                    176: };
                    177:
1.3       kristaps  178:
1.1       kristaps  179: int
                    180: main(int argc, char *argv[])
                    181: {
1.34      kristaps  182:        struct mmain      *p;
                    183:        int                c;
1.2       kristaps  184:        const struct mdoc *mdoc;
1.34      kristaps  185:        struct termp       termp;
1.2       kristaps  186:
                    187:        p = mmain_alloc();
1.1       kristaps  188:
1.34      kristaps  189:        c = mmain_getopt(p, argc, argv, NULL, NULL, NULL, NULL);
                    190:        if (1 != c)
                    191:                mmain_exit(p, -1 == c ? 1 : 0);
1.1       kristaps  192:
1.3       kristaps  193:        if (NULL == (mdoc = mmain_mdoc(p)))
                    194:                mmain_exit(p, 1);
                    195:
1.29      kristaps  196:        termp.maxrmargin = termp.rmargin = 78; /* XXX */
1.3       kristaps  197:        termp.maxcols = 1024;
                    198:        termp.offset = termp.col = 0;
                    199:        termp.flags = TERMP_NOSPACE;
1.25      kristaps  200:        termp.symtab = termsym_ansi;
                    201:        termp.styletab = termstyle_ansi;
1.3       kristaps  202:
                    203:        if (NULL == (termp.buf = malloc(termp.maxcols)))
                    204:                err(1, "malloc");
                    205:
1.15      kristaps  206:        header(&termp, mdoc_meta(mdoc));
1.12      kristaps  207:        body(&termp, NULL, mdoc_meta(mdoc), mdoc_node(mdoc));
1.3       kristaps  208:        footer(&termp, mdoc_meta(mdoc));
                    209:
                    210:        free(termp.buf);
                    211:
                    212:        mmain_exit(p, 0);
                    213:        /* NOTREACHED */
                    214: }
                    215:
                    216:
1.25      kristaps  217: /*
                    218:  * Flush a line of text.  A "line" is loosely defined as being something
                    219:  * that should be followed by a newline, regardless of whether it's
                    220:  * broken apart by newlines getting there.  A line can also be a
                    221:  * fragment of a columnar list.
                    222:  *
                    223:  * Specifically, a line is whatever's in p->buf of length p->col, which
                    224:  * is zeroed after this function returns.
                    225:  *
                    226:  * The variables TERMP_NOLPAD, TERMP_LITERAL and TERMP_NOBREAK are of
                    227:  * critical importance here.  Their behaviour follows:
                    228:  *
                    229:  *  - TERMP_NOLPAD: when beginning to write the line, don't left-pad the
                    230:  *    offset value.  This is useful when doing columnar lists where the
                    231:  *    prior column has right-padded.
                    232:  *
                    233:  *  - TERMP_NOBREAK: this is the most important and is used when making
                    234:  *    columns.  In short: don't print a newline and instead pad to the
                    235:  *    right margin.  Used in conjunction with TERMP_NOLPAD.
                    236:  *
                    237:  *  In-line line breaking:
                    238:  *
                    239:  *  If TERMP_NOBREAK is specified and the line overruns the right
                    240:  *  margin, it will break and pad-right to the right margin after
                    241:  *  writing.  If maxrmargin is violated, it will break and continue
                    242:  *  writing from the right-margin, which will lead to the above
                    243:  *  scenario upon exit.
                    244:  *
                    245:  *  Otherwise, the line will break at the right margin.  Extremely long
                    246:  *  lines will cause the system to emit a warning (TODO: hyphenate, if
                    247:  *  possible).
                    248:  */
1.3       kristaps  249: void
                    250: flushln(struct termp *p)
                    251: {
1.25      kristaps  252:        size_t           i, j, vsz, vis, maxvis, mmax, bp;
1.3       kristaps  253:
                    254:        /*
                    255:         * First, establish the maximum columns of "visible" content.
                    256:         * This is usually the difference between the right-margin and
                    257:         * an indentation, but can be, for tagged lists or columns, a
                    258:         * small set of values.
                    259:         */
                    260:
                    261:        assert(p->offset < p->rmargin);
                    262:        maxvis = p->rmargin - p->offset;
1.25      kristaps  263:        mmax = p->maxrmargin - p->offset;
                    264:        bp = TERMP_NOBREAK & p->flags ? mmax : maxvis;
1.3       kristaps  265:        vis = 0;
                    266:
                    267:        /*
                    268:         * If in the standard case (left-justified), then begin with our
                    269:         * indentation, otherwise (columns, etc.) just start spitting
                    270:         * out text.
                    271:         */
                    272:
                    273:        if ( ! (p->flags & TERMP_NOLPAD))
                    274:                /* LINTED */
                    275:                for (j = 0; j < p->offset; j++)
                    276:                        putchar(' ');
                    277:
                    278:        for (i = 0; i < p->col; i++) {
                    279:                /*
                    280:                 * Count up visible word characters.  Control sequences
1.23      kristaps  281:                 * (starting with the CSI) aren't counted.  A space
                    282:                 * generates a non-printing word, which is valid (the
                    283:                 * space is printed according to regular spacing rules).
1.3       kristaps  284:                 */
                    285:
1.25      kristaps  286:                /* FIXME: make non-ANSI friendly. */
                    287:
1.3       kristaps  288:                /* LINTED */
                    289:                for (j = i, vsz = 0; j < p->col; j++) {
1.33      kristaps  290:                        if (isspace((u_char)p->buf[j]))
1.3       kristaps  291:                                break;
                    292:                        else if (27 == p->buf[j]) {
                    293:                                assert(j + 4 <= p->col);
                    294:                                j += 3;
                    295:                        } else
                    296:                                vsz++;
                    297:                }
                    298:
                    299:                /*
1.25      kristaps  300:                 * Do line-breaking.  If we're greater than our
                    301:                 * break-point and already in-line, break to the next
                    302:                 * line and start writing.  If we're at the line start,
                    303:                 * then write out the word (TODO: hyphenate) and break
                    304:                 * in a subsequent loop invocation.
1.3       kristaps  305:                 */
                    306:
1.22      kristaps  307:                if ( ! (TERMP_NOBREAK & p->flags)) {
1.25      kristaps  308:                        if (vis && vis + vsz > bp) {
1.22      kristaps  309:                                putchar('\n');
                    310:                                for (j = 0; j < p->offset; j++)
                    311:                                        putchar(' ');
                    312:                                vis = 0;
1.25      kristaps  313:                        } else if (vis + vsz > bp)
                    314:                                warnx("word breaks right margin");
                    315:
                    316:                        /* TODO: hyphenate. */
                    317:
                    318:                } else {
                    319:                        if (vis && vis + vsz > bp) {
                    320:                                putchar('\n');
                    321:                                for (j = 0; j < p->rmargin; j++)
                    322:                                        putchar(' ');
1.30      kristaps  323:                                vis = p->rmargin - p->offset;
1.25      kristaps  324:                        } else if (vis + vsz > bp)
                    325:                                warnx("word breaks right margin");
                    326:
                    327:                        /* TODO: hyphenate. */
1.24      kristaps  328:                }
1.3       kristaps  329:
                    330:                /*
                    331:                 * Write out the word and a trailing space.  Omit the
1.25      kristaps  332:                 * space if we're the last word in the line or beyond
                    333:                 * our breakpoint.
1.3       kristaps  334:                 */
                    335:
                    336:                for ( ; i < p->col; i++) {
1.33      kristaps  337:                        if (isspace((u_char)p->buf[i]))
1.3       kristaps  338:                                break;
                    339:                        putchar(p->buf[i]);
                    340:                }
                    341:                vis += vsz;
1.25      kristaps  342:                if (i < p->col && vis <= bp) {
1.3       kristaps  343:                        putchar(' ');
                    344:                        vis++;
                    345:                }
                    346:        }
                    347:
1.25      kristaps  348:        /*
                    349:         * If we've overstepped our maximum visible no-break space, then
                    350:         * cause a newline and offset at the right margin.
                    351:         */
                    352:
1.22      kristaps  353:        if ((TERMP_NOBREAK & p->flags) && vis >= maxvis) {
1.32      kristaps  354:                if ( ! (TERMP_NONOBREAK & p->flags)) {
                    355:                        putchar('\n');
                    356:                        for (i = 0; i < p->rmargin; i++)
                    357:                                putchar(' ');
                    358:                }
1.22      kristaps  359:                p->col = 0;
                    360:                return;
                    361:        }
                    362:
1.3       kristaps  363:        /*
                    364:         * If we're not to right-marginalise it (newline), then instead
                    365:         * pad to the right margin and stay off.
                    366:         */
                    367:
1.32      kristaps  368:        if (p->flags & TERMP_NOBREAK) {
                    369:                if ( ! (TERMP_NONOBREAK & p->flags))
                    370:                        for ( ; vis < maxvis; vis++)
                    371:                                putchar(' ');
                    372:        } else
1.3       kristaps  373:                putchar('\n');
                    374:
                    375:        p->col = 0;
                    376: }
                    377:
                    378:
1.25      kristaps  379: /*
                    380:  * A newline only breaks an existing line; it won't assert vertical
                    381:  * space.  All data in the output buffer is flushed prior to the newline
                    382:  * assertion.
                    383:  */
1.3       kristaps  384: void
                    385: newln(struct termp *p)
                    386: {
1.1       kristaps  387:
1.3       kristaps  388:        p->flags |= TERMP_NOSPACE;
1.12      kristaps  389:        if (0 == p->col) {
                    390:                p->flags &= ~TERMP_NOLPAD;
1.3       kristaps  391:                return;
1.12      kristaps  392:        }
1.3       kristaps  393:        flushln(p);
1.11      kristaps  394:        p->flags &= ~TERMP_NOLPAD;
1.3       kristaps  395: }
                    396:
                    397:
1.25      kristaps  398: /*
                    399:  * Asserts a vertical space (a full, empty line-break between lines).
                    400:  * Note that if used twice, this will cause two blank spaces and so on.
                    401:  * All data in the output buffer is flushed prior to the newline
                    402:  * assertion.
                    403:  */
1.3       kristaps  404: void
                    405: vspace(struct termp *p)
                    406: {
                    407:
                    408:        newln(p);
                    409:        putchar('\n');
                    410: }
                    411:
                    412:
1.25      kristaps  413: /*
                    414:  * Break apart a word into "pwords" (partial-words, usually from
                    415:  * breaking up a phrase into individual words) and, eventually, put them
                    416:  * into the output buffer.  If we're a literal word, then don't break up
                    417:  * the word and put it verbatim into the output buffer.
                    418:  */
1.3       kristaps  419: void
                    420: word(struct termp *p, const char *word)
                    421: {
                    422:        size_t           i, j, len;
                    423:
                    424:        if (p->flags & TERMP_LITERAL) {
                    425:                pword(p, word, strlen(word));
                    426:                return;
                    427:        }
                    428:
                    429:        len = strlen(word);
                    430:        assert(len > 0);
                    431:
                    432:        if (mdoc_isdelim(word)) {
                    433:                if ( ! (p->flags & TERMP_IGNDELIM))
                    434:                        p->flags |= TERMP_NOSPACE;
                    435:                p->flags &= ~TERMP_IGNDELIM;
                    436:        }
                    437:
                    438:        /* LINTED */
                    439:        for (j = i = 0; i < len; i++) {
1.33      kristaps  440:                if ( ! isspace((u_char)word[i])) {
1.3       kristaps  441:                        j++;
                    442:                        continue;
1.20      kristaps  443:                }
                    444:
                    445:                /* Escaped spaces don't delimit... */
1.33      kristaps  446:                if (i > 0 && isspace((u_char)word[i]) &&
1.25      kristaps  447:                                '\\' == word[i - 1]) {
1.20      kristaps  448:                        j++;
                    449:                        continue;
1.1       kristaps  450:                }
1.20      kristaps  451:
1.3       kristaps  452:                if (0 == j)
                    453:                        continue;
                    454:                assert(i >= j);
                    455:                pword(p, &word[i - j], j);
                    456:                j = 0;
                    457:        }
                    458:        if (j > 0) {
                    459:                assert(i >= j);
                    460:                pword(p, &word[i - j], j);
                    461:        }
                    462: }
                    463:
                    464:
1.25      kristaps  465: /*
                    466:  * This is the main function for printing out nodes.  It's constituted
                    467:  * of PRE and POST functions, which correspond to prefix and infix
                    468:  * processing.  The termpair structure allows data to persist between
                    469:  * prefix and postfix invocations.
                    470:  */
1.3       kristaps  471: static void
1.12      kristaps  472: body(struct termp *p, struct termpair *ppair,
                    473:                const struct mdoc_meta *meta,
1.3       kristaps  474:                const struct mdoc_node *node)
                    475: {
                    476:        int              dochild;
1.9       kristaps  477:        struct termpair  pair;
1.3       kristaps  478:
1.38    ! kristaps  479:        /* Some quick sanity-checking. */
        !           480:
        !           481:        sanity(node);
        !           482:
1.3       kristaps  483:        /* Pre-processing. */
                    484:
                    485:        dochild = 1;
1.12      kristaps  486:        pair.ppair = ppair;
1.9       kristaps  487:        pair.type = 0;
1.11      kristaps  488:        pair.offset = pair.rmargin = 0;
1.10      kristaps  489:        pair.flag = 0;
1.12      kristaps  490:        pair.count = 0;
1.3       kristaps  491:
                    492:        if (MDOC_TEXT != node->type) {
                    493:                if (termacts[node->tok].pre)
1.9       kristaps  494:                        if ( ! (*termacts[node->tok].pre)(p, &pair, meta, node))
1.3       kristaps  495:                                dochild = 0;
                    496:        } else /* MDOC_TEXT == node->type */
1.35      kristaps  497:                word(p, node->string);
1.3       kristaps  498:
                    499:        /* Children. */
                    500:
1.10      kristaps  501:        if (TERMPAIR_FLAG & pair.type)
                    502:                p->flags |= pair.flag;
1.9       kristaps  503:
1.3       kristaps  504:        if (dochild && node->child)
1.12      kristaps  505:                body(p, &pair, meta, node->child);
1.3       kristaps  506:
1.10      kristaps  507:        if (TERMPAIR_FLAG & pair.type)
                    508:                p->flags &= ~pair.flag;
1.9       kristaps  509:
1.3       kristaps  510:        /* Post-processing. */
                    511:
                    512:        if (MDOC_TEXT != node->type)
                    513:                if (termacts[node->tok].post)
1.9       kristaps  514:                        (*termacts[node->tok].post)(p, &pair, meta, node);
1.3       kristaps  515:
                    516:        /* Siblings. */
1.1       kristaps  517:
1.3       kristaps  518:        if (node->next)
1.12      kristaps  519:                body(p, ppair, meta, node->next);
1.3       kristaps  520: }
                    521:
                    522:
                    523: static void
                    524: footer(struct termp *p, const struct mdoc_meta *meta)
                    525: {
                    526:        struct tm       *tm;
                    527:        char            *buf, *os;
                    528:
                    529:        if (NULL == (buf = malloc(p->rmargin)))
                    530:                err(1, "malloc");
                    531:        if (NULL == (os = malloc(p->rmargin)))
                    532:                err(1, "malloc");
                    533:
                    534:        tm = localtime(&meta->date);
                    535:
1.7       kristaps  536: #ifdef __OpenBSD__
                    537:        if (NULL == strftime(buf, p->rmargin, "%B %d, %Y", tm))
                    538: #else
1.3       kristaps  539:        if (0 == strftime(buf, p->rmargin, "%B %d, %Y", tm))
                    540: #endif
                    541:                err(1, "strftime");
                    542:
1.15      kristaps  543:        (void)strlcpy(os, meta->os, p->rmargin);
1.3       kristaps  544:
1.16      kristaps  545:        /*
                    546:         * This is /slightly/ different from regular groff output
                    547:         * because we don't have page numbers.  Print the following:
                    548:         *
                    549:         * OS                                            MDOCDATE
                    550:         */
                    551:
1.15      kristaps  552:        vspace(p);
1.3       kristaps  553:
1.15      kristaps  554:        p->flags |= TERMP_NOSPACE | TERMP_NOBREAK;
                    555:        p->rmargin = p->maxrmargin - strlen(buf);
                    556:        p->offset = 0;
1.3       kristaps  557:
1.15      kristaps  558:        word(p, os);
                    559:        flushln(p);
1.3       kristaps  560:
1.15      kristaps  561:        p->flags |= TERMP_NOLPAD | TERMP_NOSPACE;
                    562:        p->offset = p->rmargin;
                    563:        p->rmargin = p->maxrmargin;
                    564:        p->flags &= ~TERMP_NOBREAK;
                    565:
                    566:        word(p, buf);
                    567:        flushln(p);
1.1       kristaps  568:
1.3       kristaps  569:        free(buf);
                    570:        free(os);
1.1       kristaps  571: }
                    572:
                    573:
1.3       kristaps  574: static void
                    575: header(struct termp *p, const struct mdoc_meta *meta)
                    576: {
1.34      kristaps  577:        char            *buf, *title, *bufp;
1.18      kristaps  578:
                    579:        p->rmargin = p->maxrmargin;
                    580:        p->offset = 0;
1.3       kristaps  581:
                    582:        if (NULL == (buf = malloc(p->rmargin)))
                    583:                err(1, "malloc");
                    584:        if (NULL == (title = malloc(p->rmargin)))
                    585:                err(1, "malloc");
                    586:
1.16      kristaps  587:        /*
                    588:         * The header is strange.  It has three components, which are
                    589:         * really two with the first duplicated.  It goes like this:
                    590:         *
                    591:         * IDENTIFIER              TITLE                   IDENTIFIER
                    592:         *
                    593:         * The IDENTIFIER is NAME(SECTION), which is the command-name
                    594:         * (if given, or "unknown" if not) followed by the manual page
                    595:         * section.  These are given in `Dt'.  The TITLE is a free-form
                    596:         * string depending on the manual volume.  If not specified, it
                    597:         * switches on the manual section.
                    598:         */
                    599:
1.34      kristaps  600:        assert(meta->vol);
                    601:        (void)strlcpy(buf, meta->vol, p->rmargin);
1.13      kristaps  602:
1.34      kristaps  603:        if (meta->arch) {
                    604:                (void)strlcat(buf, " (", p->rmargin);
                    605:                (void)strlcat(buf, meta->arch, p->rmargin);
                    606:                (void)strlcat(buf, ")", p->rmargin);
                    607:        }
1.13      kristaps  608:
1.34      kristaps  609:        (void)snprintf(title, p->rmargin, "%s(%d)",
                    610:                        meta->title, meta->msec);
1.13      kristaps  611:
1.16      kristaps  612:        for (bufp = title; *bufp; bufp++)
1.33      kristaps  613:                *bufp = toupper((u_char)*bufp);
1.16      kristaps  614:
1.13      kristaps  615:        p->offset = 0;
                    616:        p->rmargin = (p->maxrmargin - strlen(buf)) / 2;
1.15      kristaps  617:        p->flags |= TERMP_NOBREAK | TERMP_NOSPACE;
1.3       kristaps  618:
1.13      kristaps  619:        word(p, title);
                    620:        flushln(p);
1.3       kristaps  621:
1.15      kristaps  622:        p->flags |= TERMP_NOLPAD | TERMP_NOSPACE;
1.13      kristaps  623:        p->offset = p->rmargin;
1.15      kristaps  624:        p->rmargin = p->maxrmargin - strlen(title);
1.3       kristaps  625:
1.13      kristaps  626:        word(p, buf);
                    627:        flushln(p);
1.3       kristaps  628:
1.13      kristaps  629:        p->offset = p->rmargin;
                    630:        p->rmargin = p->maxrmargin;
                    631:        p->flags &= ~TERMP_NOBREAK;
1.15      kristaps  632:        p->flags |= TERMP_NOLPAD | TERMP_NOSPACE;
1.3       kristaps  633:
1.13      kristaps  634:        word(p, title);
                    635:        flushln(p);
1.3       kristaps  636:
1.13      kristaps  637:        p->rmargin = p->maxrmargin;
                    638:        p->offset = 0;
                    639:        p->flags &= ~TERMP_NOSPACE;
1.3       kristaps  640:
                    641:        free(title);
                    642:        free(buf);
                    643: }
1.25      kristaps  644:
                    645:
                    646: /*
                    647:  * Determine the symbol indicated by an escape sequences, that is, one
                    648:  * starting with a backslash.  Once done, we pass this value into the
                    649:  * output buffer by way of the symbol table.
                    650:  */
                    651: static void
                    652: nescape(struct termp *p, const char *word, size_t len)
                    653: {
1.27      kristaps  654:        struct termenc  *enc;
1.25      kristaps  655:
                    656:        switch (len) {
                    657:        case (1):
1.27      kristaps  658:                enc = termenc1;
1.25      kristaps  659:                break;
                    660:        case (2):
1.27      kristaps  661:                enc = termenc2;
1.25      kristaps  662:                break;
                    663:        default:
1.27      kristaps  664:                warnx("unsupported %zu-byte escape sequence", len);
                    665:                return;
1.25      kristaps  666:        }
1.27      kristaps  667:
                    668:        for ( ; enc->enc; enc++)
                    669:                if (0 == memcmp(enc->enc, word, len)) {
                    670:                        symbola(p, enc->sym);
                    671:                        return;
                    672:                }
                    673:
                    674:        warnx("unsupported %zu-byte escape sequence", len);
1.25      kristaps  675: }
                    676:
                    677:
                    678: /*
                    679:  * Handle an escape sequence: determine its length and pass it to the
                    680:  * escape-symbol look table.  Note that we assume mdoc(3) has validated
                    681:  * the escape sequence (we assert upon badly-formed escape sequences).
                    682:  */
                    683: static void
                    684: pescape(struct termp *p, const char *word, size_t *i, size_t len)
                    685: {
                    686:        size_t           j;
                    687:
1.36      kristaps  688:        if (++(*i) >= len) {
                    689:                warnx("ignoring bad escape sequence");
                    690:                return;
                    691:        }
1.25      kristaps  692:
                    693:        if ('(' == word[*i]) {
                    694:                (*i)++;
1.36      kristaps  695:                if (*i + 1 >= len) {
                    696:                        warnx("ignoring bad escape sequence");
                    697:                        return;
                    698:                }
1.25      kristaps  699:                nescape(p, &word[*i], 2);
                    700:                (*i)++;
                    701:                return;
                    702:
                    703:        } else if ('*' == word[*i]) {
                    704:                (*i)++;
1.36      kristaps  705:                if (*i >= len) {
                    706:                        warnx("ignoring bad escape sequence");
                    707:                        return;
                    708:                }
1.25      kristaps  709:                switch (word[*i]) {
                    710:                case ('('):
                    711:                        (*i)++;
1.36      kristaps  712:                        if (*i + 1 >= len) {
                    713:                                warnx("ignoring bad escape sequence");
                    714:                                return;
                    715:                        }
1.25      kristaps  716:                        nescape(p, &word[*i], 2);
                    717:                        (*i)++;
                    718:                        return;
                    719:                case ('['):
                    720:                        break;
                    721:                default:
                    722:                        nescape(p, &word[*i], 1);
                    723:                        return;
                    724:                }
                    725:
                    726:        } else if ('[' != word[*i]) {
                    727:                nescape(p, &word[*i], 1);
                    728:                return;
                    729:        }
                    730:
                    731:        (*i)++;
                    732:        for (j = 0; word[*i] && ']' != word[*i]; (*i)++, j++)
                    733:                /* Loop... */ ;
                    734:
1.36      kristaps  735:        if (0 == word[*i]) {
                    736:                warnx("ignoring bad escape sequence");
                    737:                return;
                    738:        }
1.25      kristaps  739:        nescape(p, &word[*i - j], j);
                    740: }
                    741:
                    742:
                    743: /*
                    744:  * Handle pwords, partial words, which may be either a single word or a
                    745:  * phrase that cannot be broken down (such as a literal string).  This
                    746:  * handles word styling.
                    747:  */
                    748: static void
                    749: pword(struct termp *p, const char *word, size_t len)
                    750: {
                    751:        size_t           i;
                    752:
                    753:        if ( ! (TERMP_NOSPACE & p->flags) &&
                    754:                        ! (TERMP_LITERAL & p->flags))
                    755:                chara(p, ' ');
                    756:
                    757:        if ( ! (p->flags & TERMP_NONOSPACE))
                    758:                p->flags &= ~TERMP_NOSPACE;
                    759:
                    760:        /*
                    761:         * XXX - if literal and underlining, this will underline the
                    762:         * spaces between literal words.
                    763:         */
                    764:
                    765:        if (p->flags & TERMP_BOLD)
1.26      kristaps  766:                stylea(p, TERMSTYLE_BOLD);
1.25      kristaps  767:        if (p->flags & TERMP_UNDERLINE)
1.26      kristaps  768:                stylea(p, TERMSTYLE_UNDER);
1.25      kristaps  769:
                    770:        for (i = 0; i < len; i++) {
                    771:                if ('\\' == word[i]) {
                    772:                        pescape(p, word, &i, len);
                    773:                        continue;
                    774:                }
                    775:                chara(p, word[i]);
                    776:        }
                    777:
                    778:        if (p->flags & TERMP_BOLD ||
                    779:                        p->flags & TERMP_UNDERLINE)
1.26      kristaps  780:                stylea(p, TERMSTYLE_CLEAR);
1.25      kristaps  781: }
                    782:
                    783:
                    784: /*
                    785:  * Add a symbol to the output line buffer.
                    786:  */
                    787: static void
                    788: symbola(struct termp *p, enum tsym sym)
                    789: {
                    790:
                    791:        assert(p->symtab[sym].sym);
                    792:        stringa(p, p->symtab[sym].sym, p->symtab[sym].sz);
                    793: }
                    794:
                    795:
                    796: /*
1.26      kristaps  797:  * Add a style to the output line buffer.
                    798:  */
                    799: static void
                    800: stylea(struct termp *p, enum tstyle style)
                    801: {
                    802:
                    803:        assert(p->styletab[style].sym);
                    804:        stringa(p, p->styletab[style].sym, p->styletab[style].sz);
                    805: }
                    806:
                    807:
                    808: /*
1.25      kristaps  809:  * Like chara() but for arbitrary-length buffers.  Resize the buffer by
                    810:  * a factor of two (if the buffer is less than that) or the buffer's
                    811:  * size.
                    812:  */
                    813: static void
                    814: stringa(struct termp *p, const char *c, size_t sz)
                    815: {
                    816:        size_t           s;
                    817:
1.27      kristaps  818:        if (0 == sz)
                    819:                return;
                    820:
1.25      kristaps  821:        s = sz > p->maxcols * 2 ? sz : p->maxcols * 2;
                    822:
                    823:        assert(c);
                    824:        if (p->col + sz >= p->maxcols) {
                    825:                p->buf = realloc(p->buf, s);
                    826:                if (NULL == p->buf)
                    827:                        err(1, "realloc");
                    828:                p->maxcols = s;
                    829:        }
                    830:
                    831:        (void)memcpy(&p->buf[p->col], c, sz);
                    832:        p->col += sz;
                    833: }
                    834:
                    835:
                    836: /*
                    837:  * Insert a single character into the line-buffer.  If the buffer's
                    838:  * space is exceeded, then allocate more space by doubling the buffer
                    839:  * size.
                    840:  */
                    841: static void
                    842: chara(struct termp *p, char c)
                    843: {
                    844:
                    845:        if (p->col + 1 >= p->maxcols) {
                    846:                p->buf = realloc(p->buf, p->maxcols * 2);
                    847:                if (NULL == p->buf)
                    848:                        err(1, "malloc");
                    849:                p->maxcols *= 2;
                    850:        }
                    851:        p->buf[(p->col)++] = c;
                    852: }
1.38    ! kristaps  853:
        !           854:
        !           855: static void
        !           856: sanity(const struct mdoc_node *n)
        !           857: {
        !           858:
        !           859:        switch (n->type) {
        !           860:        case (MDOC_TEXT):
        !           861:                if (n->child)
        !           862:                        errx(1, "regular form violated (1)");
        !           863:                if (NULL == n->parent)
        !           864:                        errx(1, "regular form violated (2)");
        !           865:                if (NULL == n->string)
        !           866:                        errx(1, "regular form violated (3)");
        !           867:                switch (n->parent->type) {
        !           868:                case (MDOC_TEXT):
        !           869:                        /* FALLTHROUGH */
        !           870:                case (MDOC_ROOT):
        !           871:                        errx(1, "regular form violated (4)");
        !           872:                        /* NOTREACHED */
        !           873:                default:
        !           874:                        break;
        !           875:                }
        !           876:                break;
        !           877:        case (MDOC_ELEM):
        !           878:                if (NULL == n->parent)
        !           879:                        errx(1, "regular form violated (5)");
        !           880:                switch (n->parent->type) {
        !           881:                case (MDOC_TAIL):
        !           882:                        /* FALLTHROUGH */
        !           883:                case (MDOC_BODY):
        !           884:                        /* FALLTHROUGH */
        !           885:                case (MDOC_HEAD):
        !           886:                        break;
        !           887:                default:
        !           888:                        errx(1, "regular form violated (6)");
        !           889:                        /* NOTREACHED */
        !           890:                }
        !           891:                if (n->child) switch (n->child->type) {
        !           892:                case (MDOC_TEXT):
        !           893:                        break;
        !           894:                default:
        !           895:                        errx(1, "regular form violated (7(");
        !           896:                        /* NOTREACHED */
        !           897:                }
        !           898:                break;
        !           899:        case (MDOC_HEAD):
        !           900:                /* FALLTHROUGH */
        !           901:        case (MDOC_BODY):
        !           902:                /* FALLTHROUGH */
        !           903:        case (MDOC_TAIL):
        !           904:                if (NULL == n->parent)
        !           905:                        errx(1, "regular form violated (8)");
        !           906:                if (MDOC_BLOCK != n->parent->type)
        !           907:                        errx(1, "regular form violated (9)");
        !           908:                if (n->child) switch (n->child->type) {
        !           909:                case (MDOC_BLOCK):
        !           910:                        /* FALLTHROUGH */
        !           911:                case (MDOC_ELEM):
        !           912:                        /* FALLTHROUGH */
        !           913:                case (MDOC_TEXT):
        !           914:                        break;
        !           915:                default:
        !           916:                        errx(1, "regular form violated (a)");
        !           917:                        /* NOTREACHED */
        !           918:                }
        !           919:                break;
        !           920:        case (MDOC_BLOCK):
        !           921:                if (NULL == n->parent)
        !           922:                        errx(1, "regular form violated (b)");
        !           923:                if (NULL == n->child)
        !           924:                        errx(1, "regular form violated (c)");
        !           925:                switch (n->parent->type) {
        !           926:                case (MDOC_ROOT):
        !           927:                        /* FALLTHROUGH */
        !           928:                case (MDOC_HEAD):
        !           929:                        /* FALLTHROUGH */
        !           930:                case (MDOC_BODY):
        !           931:                        /* FALLTHROUGH */
        !           932:                case (MDOC_TAIL):
        !           933:                        break;
        !           934:                default:
        !           935:                        errx(1, "regular form violated (d)");
        !           936:                        /* NOTREACHED */
        !           937:                }
        !           938:                switch (n->child->type) {
        !           939:                case (MDOC_ROOT):
        !           940:                        /* FALLTHROUGH */
        !           941:                case (MDOC_ELEM):
        !           942:                        errx(1, "regular form violated (e)");
        !           943:                        /* NOTREACHED */
        !           944:                default:
        !           945:                        break;
        !           946:                }
        !           947:                break;
        !           948:        case (MDOC_ROOT):
        !           949:                if (n->parent)
        !           950:                        errx(1, "regular form violated (f)");
        !           951:                if (NULL == n->child)
        !           952:                        errx(1, "regular form violated (10)");
        !           953:                switch (n->child->type) {
        !           954:                case (MDOC_BLOCK):
        !           955:                        break;
        !           956:                default:
        !           957:                        errx(1, "regular form violated (11)");
        !           958:                        /* NOTREACHED */
        !           959:                }
        !           960:                break;
        !           961:        }
        !           962: }
        !           963:

CVSweb