=================================================================== RCS file: /cvs/mandoc/term.c,v retrieving revision 1.150 retrieving revision 1.174 diff -u -p -r1.150 -r1.174 --- mandoc/term.c 2010/06/26 15:36:37 1.150 +++ mandoc/term.c 2010/10/02 15:15:55 1.174 @@ -1,6 +1,7 @@ -/* $Id: term.c,v 1.150 2010/06/26 15:36:37 kristaps Exp $ */ +/* $Id: term.c,v 1.174 2010/10/02 15:15:55 schwarze Exp $ */ /* - * Copyright (c) 2008, 2009 Kristaps Dzonsons + * Copyright (c) 2008, 2009, 2010 Kristaps Dzonsons + * Copyright (c) 2010 Ingo Schwarze * * Permission to use, copy, modify, and distribute this software for any * purpose with or without fee is hereby granted, provided that the above @@ -33,9 +34,9 @@ #include "term.h" #include "main.h" -static void spec(struct termp *, const char *, size_t); +static void spec(struct termp *, enum roffdeco, + const char *, size_t); static void res(struct termp *, const char *, size_t); -static void buffera(struct termp *, const char *, size_t); static void bufferc(struct termp *, char); static void adjbuf(struct termp *p, size_t); static void encode(struct termp *, const char *, size_t); @@ -82,7 +83,7 @@ term_alloc(enum termenc enc) p = calloc(1, sizeof(struct termp)); if (NULL == p) { perror(NULL); - exit(EXIT_FAILURE); + exit((int)MANDOCLEVEL_SYSERR); } p->enc = enc; @@ -133,9 +134,11 @@ term_flushln(struct termp *p) size_t vbl; /* number of blanks to prepend to output */ size_t vend; /* end of word visual position on output */ size_t bp; /* visual right border position */ - int j; /* temporary loop index */ - int jhy; /* last hyphen before line overflow */ - size_t maxvis, mmax; + size_t dv; /* temporary for visual pos calculations */ + int j; /* temporary loop index for p->buf */ + int jhy; /* last hyph before overflow w/r/t j */ + size_t maxvis; /* output position of visible boundary */ + size_t mmax; /* used in calculating bp */ /* * First, establish the maximum columns of "visible" content. @@ -143,40 +146,32 @@ term_flushln(struct termp *p) * an indentation, but can be, for tagged lists or columns, a * small set of values. */ + assert (p->rmargin > p->offset); + dv = p->rmargin - p->offset; + maxvis = (int)dv > p->overstep ? dv - (size_t)p->overstep : 0; + dv = p->maxrmargin - p->offset; + mmax = (int)dv > p->overstep ? dv - (size_t)p->overstep : 0; - assert(p->offset < p->rmargin); - - maxvis = (int)(p->rmargin - p->offset) - p->overstep < 0 ? - /* LINTED */ - 0 : p->rmargin - p->offset - p->overstep; - mmax = (int)(p->maxrmargin - p->offset) - p->overstep < 0 ? - /* LINTED */ - 0 : p->maxrmargin - p->offset - p->overstep; - bp = TERMP_NOBREAK & p->flags ? mmax : maxvis; /* * Indent the first line of a paragraph. */ - vbl = p->flags & TERMP_NOLPAD ? 0 : p->offset; + vbl = p->flags & TERMP_NOLPAD ? (size_t)0 : p->offset; - /* - * FIXME: if bp is zero, we still output the first word before - * breaking the line. - */ + vis = vend = 0; + i = 0; - vis = vend = i = 0; while (i < (int)p->col) { - /* - * Handle literal tab characters. + * Handle literal tab characters: collapse all + * subsequent tabs into a single huge set of spaces. */ - for (j = i; j < (int)p->col; j++) { - if ('\t' != p->buf[j]) - break; - vend = (vis/p->tabwidth+1)*p->tabwidth; + while (i < (int)p->col && '\t' == p->buf[i]) { + vend = (vis / p->tabwidth + 1) * p->tabwidth; vbl += vend - vis; vis = vend; + i++; } /* @@ -186,17 +181,24 @@ term_flushln(struct termp *p) * space is printed according to regular spacing rules). */ - /* LINTED */ - for (jhy = 0; j < (int)p->col; j++) { + for (j = i, jhy = 0; j < (int)p->col; j++) { if ((j && ' ' == p->buf[j]) || '\t' == p->buf[j]) break; - if (8 != p->buf[j]) { - if (vend > vis && vend < bp && - ASCII_HYPH == p->buf[j]) - jhy = j; - vend++; - } else - vend--; + + /* Back over the the last printed character. */ + if (8 == p->buf[j]) { + assert(j); + vend -= (*p->width)(p, p->buf[j - 1]); + continue; + } + + /* Regular word. */ + /* Break at the hyphen point if we overrun. */ + if (vend > vis && vend < bp && + ASCII_HYPH == p->buf[j]) + jhy = j; + + vend += (*p->width)(p, p->buf[j]); } /* @@ -217,17 +219,10 @@ term_flushln(struct termp *p) /* Remove the p->overstep width. */ - bp += (int)/* LINTED */ - p->overstep; + bp += (size_t)p->overstep; p->overstep = 0; } - /* - * Skip leading tabs, they were handled above. - */ - while (i < (int)p->col && '\t' == p->buf[i]) - i++; - /* Write out the [remaining] word. */ for ( ; i < (int)p->col; i++) { if (vend > bp && jhy > 0 && i > jhy) @@ -235,14 +230,16 @@ term_flushln(struct termp *p) if ('\t' == p->buf[i]) break; if (' ' == p->buf[i]) { - while (' ' == p->buf[i]) { - vbl++; + j = i; + while (' ' == p->buf[i]) i++; - } + dv = (size_t)(i - j) * (*p->width)(p, ' '); + vbl += dv; + vend += dv; break; } if (ASCII_NBRSP == p->buf[i]) { - vbl++; + vbl += (*p->width)(p, ' '); continue; } @@ -257,17 +254,23 @@ term_flushln(struct termp *p) vbl = 0; } - if (ASCII_HYPH == p->buf[i]) + if (ASCII_HYPH == p->buf[i]) { (*p->letter)(p, '-'); - else + p->viscol += (*p->width)(p, '-'); + } else { (*p->letter)(p, p->buf[i]); - - p->viscol += 1; + p->viscol += (*p->width)(p, p->buf[i]); + } } - vend += vbl; vis = vend; } + /* + * If there was trailing white space, it was not printed; + * so reset the cursor position accordingly. + */ + vis -= vbl; + p->col = 0; p->overstep = 0; @@ -279,8 +282,7 @@ term_flushln(struct termp *p) if (TERMP_HANG & p->flags) { /* We need one blank after the tag. */ - p->overstep = /* LINTED */ - vis - maxvis + 1; + p->overstep = (int)(vis - maxvis + (*p->width)(p, ' ')); /* * Behave exactly the same way as groff: @@ -294,8 +296,7 @@ term_flushln(struct termp *p) */ if (p->overstep >= -1) { assert((int)maxvis + p->overstep >= 0); - /* LINTED */ - maxvis += p->overstep; + maxvis += (size_t)p->overstep; } else p->overstep = 0; @@ -303,8 +304,8 @@ term_flushln(struct termp *p) return; /* Right-pad. */ - if (maxvis > vis + /* LINTED */ - ((TERMP_TWOSPACE & p->flags) ? 1 : 0)) { + if (maxvis > vis + + ((TERMP_TWOSPACE & p->flags) ? (*p->width)(p, ' ') : 0)) { p->viscol += maxvis - vis; (*p->advance)(p, maxvis - vis); vis += (maxvis - vis); @@ -352,14 +353,16 @@ term_vspace(struct termp *p) static void -spec(struct termp *p, const char *word, size_t len) +spec(struct termp *p, enum roffdeco d, const char *word, size_t len) { const char *rhs; size_t sz; - rhs = chars_a2ascii(p->symtab, word, len, &sz); + rhs = chars_spec2str(p->symtab, word, len, &sz); if (rhs) encode(p, rhs, sz); + else if (DECO_SSPECIAL == d) + encode(p, word, len); } @@ -369,7 +372,7 @@ res(struct termp *p, const char *word, size_t len) const char *rhs; size_t sz; - rhs = chars_a2res(p->symtab, word, len, &sz); + rhs = chars_res2str(p->symtab, word, len, &sz); if (rhs) encode(p, rhs, sz); } @@ -449,7 +452,6 @@ void term_word(struct termp *p, const char *word) { const char *sv, *seq; - int sz; size_t ssz; enum roffdeco deco; @@ -480,34 +482,42 @@ term_word(struct termp *p, const char *word) } if ( ! (TERMP_NOSPACE & p->flags)) { - bufferc(p, ' '); - if (TERMP_SENTENCE & p->flags) + if ( ! (TERMP_KEEP & p->flags)) { + if (TERMP_PREKEEP & p->flags) + p->flags |= TERMP_KEEP; bufferc(p, ' '); + if (TERMP_SENTENCE & p->flags) + bufferc(p, ' '); + } else + bufferc(p, ASCII_NBRSP); } if ( ! (p->flags & TERMP_NONOSPACE)) p->flags &= ~TERMP_NOSPACE; + else + p->flags |= TERMP_NOSPACE; - p->flags &= ~TERMP_SENTENCE; + p->flags &= ~(TERMP_SENTENCE | TERMP_IGNDELIM); - /* FIXME: use strcspn. */ - while (*word) { - if ('\\' != *word) { - encode(p, word, 1); - word++; + if ((ssz = strcspn(word, "\\")) > 0) + encode(p, word, ssz); + + word += ssz; + if ('\\' != *word) continue; - } seq = ++word; - sz = a2roffdeco(&deco, &seq, &ssz); + word += a2roffdeco(&deco, &seq, &ssz); switch (deco) { case (DECO_RESERVED): res(p, seq, ssz); break; case (DECO_SPECIAL): - spec(p, seq, ssz); + /* FALLTHROUGH */ + case (DECO_SSPECIAL): + spec(p, deco, seq, ssz); break; case (DECO_BOLD): term_fontrepl(p, TERMFONT_BOLD); @@ -525,7 +535,6 @@ term_word(struct termp *p, const char *word) break; } - word += sz; if (DECO_NOSPACE == deco && '\0' == *word) p->flags |= TERMP_NOSPACE; } @@ -534,7 +543,7 @@ term_word(struct termp *p, const char *word) * Note that we don't process the pipe: the parser sees it as * punctuation, but we don't in terms of typography. */ - if (sv[0] && 0 == sv[1]) + if (sv[0] && '\0' == sv[1]) switch (sv[0]) { case('('): /* FALLTHROUGH */ @@ -559,24 +568,12 @@ adjbuf(struct termp *p, size_t sz) p->buf = realloc(p->buf, p->maxcols); if (NULL == p->buf) { perror(NULL); - exit(EXIT_FAILURE); + exit((int)MANDOCLEVEL_SYSERR); } } static void -buffera(struct termp *p, const char *word, size_t sz) -{ - - if (p->col + sz >= p->maxcols) - adjbuf(p, p->col + sz); - - memcpy(&p->buf[(int)p->col], word, sz); - p->col += sz; -} - - -static void bufferc(struct termp *p, char c) { @@ -600,23 +597,31 @@ encode(struct termp *p, const char *word, size_t sz) */ if (TERMFONT_NONE == (f = term_fonttop(p))) { - buffera(p, word, sz); + if (p->col + sz >= p->maxcols) + adjbuf(p, p->col + sz); + memcpy(&p->buf[(int)p->col], word, sz); + p->col += sz; return; } + /* Pre-buffer, assuming worst-case. */ + + if (p->col + 1 + (sz * 3) >= p->maxcols) + adjbuf(p, p->col + 1 + (sz * 3)); + for (i = 0; i < (int)sz; i++) { if ( ! isgraph((u_char)word[i])) { - bufferc(p, word[i]); + p->buf[(int)p->col++] = word[i]; continue; } if (TERMFONT_UNDER == f) - bufferc(p, '_'); + p->buf[(int)p->col++] = '_'; else - bufferc(p, word[i]); + p->buf[(int)p->col++] = word[i]; - bufferc(p, 8); - bufferc(p, word[i]); + p->buf[(int)p->col++] = 8; + p->buf[(int)p->col++] = word[i]; } } @@ -632,15 +637,54 @@ term_len(const struct termp *p, size_t sz) size_t term_strlen(const struct termp *p, const char *cp) { - size_t sz; + size_t sz, ssz, rsz, i; + enum roffdeco d; + const char *seq, *rhs; - for (sz = 0; *cp; cp++) - sz += (*p->width)(p, *cp); + for (sz = 0; '\0' != *cp; ) + /* + * Account for escaped sequences within string length + * calculations. This follows the logic in term_word() + * as we must calculate the width of produced strings. + */ + if ('\\' == *cp) { + seq = ++cp; + cp += a2roffdeco(&d, &seq, &ssz); + switch (d) { + case (DECO_RESERVED): + rhs = chars_res2str + (p->symtab, seq, ssz, &rsz); + break; + case (DECO_SPECIAL): + /* FALLTHROUGH */ + case (DECO_SSPECIAL): + rhs = chars_spec2str + (p->symtab, seq, ssz, &rsz); + + /* Allow for one-char escapes. */ + if (DECO_SSPECIAL != d || rhs) + break; + + rhs = seq; + rsz = ssz; + break; + default: + rhs = NULL; + break; + } + + if (rhs) + for (i = 0; i < rsz; i++) + sz += (*p->width)(p, *rhs++); + } else + sz += (*p->width)(p, *cp++); + return(sz); } +/* ARGSUSED */ size_t term_vspan(const struct termp *p, const struct roffsu *su) { @@ -680,39 +724,11 @@ term_vspan(const struct termp *p, const struct roffsu size_t term_hspan(const struct termp *p, const struct roffsu *su) { - double r; + double v; - /* XXX: CM, IN, and PT are approximations. */ - - switch (su->unit) { - case (SCALE_CM): - r = 4 * su->scale; - break; - case (SCALE_IN): - /* XXX: this is an approximation. */ - r = 10 * su->scale; - break; - case (SCALE_PC): - r = (10 * su->scale) / 6; - break; - case (SCALE_PT): - r = (10 * su->scale) / 72; - break; - case (SCALE_MM): - r = su->scale / 1000; /* FIXME: double-check. */ - break; - case (SCALE_VS): - r = su->scale * 2 - 1; /* FIXME: double-check. */ - break; - default: - r = su->scale; - break; - } - - if (r < 0.0) - r = 0.0; - return((size_t)/* LINTED */ - r); + v = ((*p->hspan)(p, su)); + if (v < 0.0) + v = 0.0; + return((size_t) /* LINTED */ + v); } - -