=================================================================== RCS file: /cvs/mandoc/term.c,v retrieving revision 1.174 retrieving revision 1.218 diff -u -p -r1.174 -r1.218 --- mandoc/term.c 2010/10/02 15:15:55 1.174 +++ mandoc/term.c 2014/03/23 11:25:26 1.218 @@ -1,7 +1,7 @@ -/* $Id: term.c,v 1.174 2010/10/02 15:15:55 schwarze Exp $ */ +/* $Id: term.c,v 1.218 2014/03/23 11:25:26 schwarze Exp $ */ /* - * Copyright (c) 2008, 2009, 2010 Kristaps Dzonsons - * Copyright (c) 2010 Ingo Schwarze + * Copyright (c) 2008, 2009, 2010, 2011 Kristaps Dzonsons + * Copyright (c) 2010-2014 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 @@ -23,25 +23,22 @@ #include #include -#include #include #include #include #include "mandoc.h" -#include "chars.h" +#include "mandoc_aux.h" #include "out.h" #include "term.h" #include "main.h" -static void spec(struct termp *, enum roffdeco, - const char *, size_t); -static void res(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); +static size_t cond_width(const struct termp *, int, int *); +static void adjbuf(struct termp *p, size_t); +static void bufferc(struct termp *, char); +static void encode(struct termp *, const char *, size_t); +static void encode1(struct termp *, int); - void term_free(struct termp *p) { @@ -49,7 +46,7 @@ term_free(struct termp *p) if (p->buf) free(p->buf); if (p->symtab) - chars_free(p->symtab); + mchars_free(p->symtab); free(p); } @@ -74,23 +71,6 @@ term_end(struct termp *p) (*p->end)(p); } - -struct termp * -term_alloc(enum termenc enc) -{ - struct termp *p; - - p = calloc(1, sizeof(struct termp)); - if (NULL == p) { - perror(NULL); - exit((int)MANDOCLEVEL_SYSERR); - } - - p->enc = enc; - return(p); -} - - /* * Flush a line of text. A "line" is loosely defined as being something * that should be followed by a newline, regardless of whether it's @@ -100,22 +80,17 @@ term_alloc(enum termenc enc) * * The following flags may be specified: * - * - TERMP_NOLPAD: when beginning to write the line, don't left-pad the - * offset value. This is useful when doing columnar lists where the - * prior column has right-padded. - * * - TERMP_NOBREAK: this is the most important and is used when making - * columns. In short: don't print a newline and instead pad to the - * right margin. Used in conjunction with TERMP_NOLPAD. + * columns. In short: don't print a newline and instead expect the + * next call to do the padding up to the start of the next column. + * p->trailspace may be set to 0, 1, or 2, depending on how many + * space characters are required at the end of the column. * - * - TERMP_TWOSPACE: when padding, make sure there are at least two - * space characters of padding. Otherwise, rather break the line. - * * - TERMP_DANGLE: don't newline when TERMP_NOBREAK is specified and * the line is overrun, and don't pad-right if it's underrun. * * - TERMP_HANG: like TERMP_DANGLE, but doesn't newline when - * overruning, instead save the position and continue at that point + * overrunning, instead save the position and continue at that point * when the next invocation. * * In-line line breaking: @@ -129,14 +104,15 @@ term_alloc(enum termenc enc) void term_flushln(struct termp *p) { - int i; /* current input position in p->buf */ + size_t i; /* current input position in p->buf */ + int ntab; /* number of tabs to prepend */ size_t vis; /* current visual position on output */ 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 */ 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 j; /* temporary loop index for p->buf */ + size_t jhy; /* last hyph before overflow w/r/t j */ size_t maxvis; /* output position of visible boundary */ size_t mmax; /* used in calculating bp */ @@ -144,9 +120,14 @@ term_flushln(struct termp *p) * First, establish the maximum columns of "visible" content. * This is usually the difference between the right-margin and * an indentation, but can be, for tagged lists or columns, a - * small set of values. + * small set of values. + * + * The following unsigned-signed subtractions look strange, + * but they are actually correct. If the int p->overstep + * is negative, it gets sign extended. Subtracting that + * very large size_t effectively adds a small number to dv. */ - assert (p->rmargin > p->offset); + 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; @@ -155,22 +136,25 @@ term_flushln(struct termp *p) bp = TERMP_NOBREAK & p->flags ? mmax : maxvis; /* - * Indent the first line of a paragraph. + * Calculate the required amount of padding. */ - vbl = p->flags & TERMP_NOLPAD ? (size_t)0 : p->offset; + vbl = p->offset + p->overstep > p->viscol ? + p->offset + p->overstep - p->viscol : 0; vis = vend = 0; i = 0; - while (i < (int)p->col) { + while (i < p->col) { /* * Handle literal tab characters: collapse all * subsequent tabs into a single huge set of spaces. */ - while (i < (int)p->col && '\t' == p->buf[i]) { + ntab = 0; + while (i < p->col && '\t' == p->buf[i]) { vend = (vis / p->tabwidth + 1) * p->tabwidth; vbl += vend - vis; vis = vend; + ntab++; i++; } @@ -181,8 +165,8 @@ term_flushln(struct termp *p) * space is printed according to regular spacing rules). */ - for (j = i, jhy = 0; j < (int)p->col; j++) { - if ((j && ' ' == p->buf[j]) || '\t' == p->buf[j]) + for (j = i, jhy = 0; j < p->col; j++) { + if (' ' == p->buf[j] || '\t' == p->buf[j]) break; /* Back over the the last printed character. */ @@ -195,9 +179,17 @@ term_flushln(struct termp *p) /* Regular word. */ /* Break at the hyphen point if we overrun. */ if (vend > vis && vend < bp && - ASCII_HYPH == p->buf[j]) + (ASCII_HYPH == p->buf[j] || + ASCII_BREAK == p->buf[j])) jhy = j; + /* + * Hyphenation now decided, put back a real + * hyphen such that we get the correct width. + */ + if (ASCII_HYPH == p->buf[j]) + p->buf[j] = '-'; + vend += (*p->width)(p, p->buf[j]); } @@ -208,23 +200,30 @@ term_flushln(struct termp *p) if (vend > bp && 0 == jhy && vis > 0) { vend -= vis; (*p->endline)(p); + p->viscol = 0; if (TERMP_NOBREAK & p->flags) { - p->viscol = p->rmargin; - (*p->advance)(p, p->rmargin); + vbl = p->rmargin; vend += p->rmargin - p->offset; - } else { - p->viscol = 0; + } else vbl = p->offset; - } - /* Remove the p->overstep width. */ + /* use pending tabs on the new line */ + if (0 < ntab) + vbl += ntab * p->tabwidth; + + /* + * Remove the p->overstep width. + * Again, if p->overstep is negative, + * sign extension does the right thing. + */ + bp += (size_t)p->overstep; p->overstep = 0; } /* Write out the [remaining] word. */ - for ( ; i < (int)p->col; i++) { + for ( ; i < p->col; i++) { if (vend > bp && jhy > 0 && i > jhy) break; if ('\t' == p->buf[i]) @@ -233,7 +232,7 @@ term_flushln(struct termp *p) j = i; while (' ' == p->buf[i]) i++; - dv = (size_t)(i - j) * (*p->width)(p, ' '); + dv = (i - j) * (*p->width)(p, ' '); vbl += dv; vend += dv; break; @@ -242,6 +241,8 @@ term_flushln(struct termp *p) vbl += (*p->width)(p, ' '); continue; } + if (ASCII_BREAK == p->buf[i]) + continue; /* * Now we definitely know there will be @@ -254,13 +255,11 @@ term_flushln(struct termp *p) vbl = 0; } - if (ASCII_HYPH == p->buf[i]) { - (*p->letter)(p, '-'); - p->viscol += (*p->width)(p, '-'); - } else { - (*p->letter)(p, p->buf[i]); + (*p->letter)(p, p->buf[i]); + if (8 == p->buf[i]) + p->viscol -= (*p->width)(p, p->buf[i-1]); + else p->viscol += (*p->width)(p, p->buf[i]); - } } vis = vend; } @@ -269,7 +268,8 @@ term_flushln(struct termp *p) * If there was trailing white space, it was not printed; * so reset the cursor position accordingly. */ - vis -= vbl; + if (vis) + vis -= vbl; p->col = 0; p->overstep = 0; @@ -281,38 +281,27 @@ term_flushln(struct termp *p) } if (TERMP_HANG & p->flags) { - /* We need one blank after the tag. */ - p->overstep = (int)(vis - maxvis + (*p->width)(p, ' ')); + p->overstep = (int)(vis - maxvis + + p->trailspace * (*p->width)(p, ' ')); /* - * Behave exactly the same way as groff: * If we have overstepped the margin, temporarily move * it to the right and flag the rest of the line to be * shorter. - * If we landed right at the margin, be happy. - * If we are one step before the margin, temporarily - * move it one step LEFT and flag the rest of the line - * to be longer. + * If there is a request to keep the columns together, + * allow negative overstep when the column is not full. */ - if (p->overstep >= -1) { - assert((int)maxvis + p->overstep >= 0); - maxvis += (size_t)p->overstep; - } else + if (p->trailspace && p->overstep < 0) p->overstep = 0; + return; } else if (TERMP_DANGLE & p->flags) return; - /* Right-pad. */ - if (maxvis > vis + - ((TERMP_TWOSPACE & p->flags) ? (*p->width)(p, ' ') : 0)) { - p->viscol += maxvis - vis; - (*p->advance)(p, maxvis - vis); - vis += (maxvis - vis); - } else { /* ...or newline break. */ + /* If the column was overrun, break the line. */ + if (maxvis < vis + p->trailspace * (*p->width)(p, ' ')) { (*p->endline)(p); - p->viscol = p->rmargin; - (*p->advance)(p, p->rmargin); + p->viscol = 0; } } @@ -327,12 +316,8 @@ term_newln(struct termp *p) { p->flags |= TERMP_NOSPACE; - if (0 == p->col && 0 == p->viscol) { - p->flags &= ~TERMP_NOLPAD; - return; - } - term_flushln(p); - p->flags &= ~TERMP_NOLPAD; + if (p->col || p->viscol) + term_flushln(p); } @@ -348,36 +333,12 @@ term_vspace(struct termp *p) term_newln(p); p->viscol = 0; - (*p->endline)(p); + if (0 < p->skipvsp) + p->skipvsp--; + else + (*p->endline)(p); } - -static void -spec(struct termp *p, enum roffdeco d, const char *word, size_t len) -{ - const char *rhs; - size_t sz; - - rhs = chars_spec2str(p->symtab, word, len, &sz); - if (rhs) - encode(p, rhs, sz); - else if (DECO_SSPECIAL == d) - encode(p, word, len); -} - - -static void -res(struct termp *p, const char *word, size_t len) -{ - const char *rhs; - size_t sz; - - rhs = chars_res2str(p->symtab, word, len, &sz); - if (rhs) - encode(p, rhs, sz); -} - - void term_fontlast(struct termp *p) { @@ -428,7 +389,7 @@ void term_fontpopq(struct termp *p, const void *key) { - while (p->fonti >= 0 && key != &p->fontq[p->fonti]) + while (p->fonti >= 0 && key < (void *)(p->fontq + p->fonti)) p->fonti--; assert(p->fonti >= 0); } @@ -442,7 +403,6 @@ term_fontpop(struct termp *p) p->fonti--; } - /* * Handle pwords, partial words, which may be either a single word or a * phrase that cannot be broken down (such as a literal string). This @@ -451,111 +411,124 @@ term_fontpop(struct termp *p) void term_word(struct termp *p, const char *word) { - const char *sv, *seq; + const char nbrsp[2] = { ASCII_NBRSP, 0 }; + const char *seq, *cp; + char c; + int sz, uc; size_t ssz; - enum roffdeco deco; + enum mandoc_esc esc; - sv = word; - - if (word[0] && '\0' == word[1]) - switch (word[0]) { - case('.'): - /* FALLTHROUGH */ - case(','): - /* FALLTHROUGH */ - case(';'): - /* FALLTHROUGH */ - case(':'): - /* FALLTHROUGH */ - case('?'): - /* FALLTHROUGH */ - case('!'): - /* FALLTHROUGH */ - case(')'): - /* FALLTHROUGH */ - case(']'): - if ( ! (TERMP_IGNDELIM & p->flags)) - p->flags |= TERMP_NOSPACE; - break; - default: - break; - } - if ( ! (TERMP_NOSPACE & 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 (TERMP_PREKEEP & p->flags) + p->flags |= TERMP_KEEP; if ( ! (p->flags & TERMP_NONOSPACE)) p->flags &= ~TERMP_NOSPACE; else p->flags |= TERMP_NOSPACE; - p->flags &= ~(TERMP_SENTENCE | TERMP_IGNDELIM); + p->flags &= ~TERMP_SENTENCE; - while (*word) { - if ((ssz = strcspn(word, "\\")) > 0) + while ('\0' != *word) { + if ('\\' != *word) { + if (TERMP_SKIPCHAR & p->flags) { + p->flags &= ~TERMP_SKIPCHAR; + word++; + continue; + } + if (TERMP_NBRWORD & p->flags) { + if (' ' == *word) { + encode(p, nbrsp, 1); + word++; + continue; + } + ssz = strcspn(word, "\\ "); + } else + ssz = strcspn(word, "\\"); encode(p, word, ssz); - - word += ssz; - if ('\\' != *word) + word += (int)ssz; continue; + } - seq = ++word; - word += a2roffdeco(&deco, &seq, &ssz); + word++; + esc = mandoc_escape(&word, &seq, &sz); + if (ESCAPE_ERROR == esc) + break; - switch (deco) { - case (DECO_RESERVED): - res(p, seq, ssz); + if (TERMENC_ASCII != p->enc) + switch (esc) { + case (ESCAPE_UNICODE): + uc = mchars_num2uc(seq + 1, sz - 1); + if ('\0' == uc) + break; + encode1(p, uc); + continue; + case (ESCAPE_SPECIAL): + uc = mchars_spec2cp(p->symtab, seq, sz); + if (uc <= 0) + break; + encode1(p, uc); + continue; + default: + break; + } + + switch (esc) { + case (ESCAPE_UNICODE): + encode1(p, '?'); break; - case (DECO_SPECIAL): - /* FALLTHROUGH */ - case (DECO_SSPECIAL): - spec(p, deco, seq, ssz); + case (ESCAPE_NUMBERED): + c = mchars_num2char(seq, sz); + if ('\0' != c) + encode(p, &c, 1); break; - case (DECO_BOLD): + case (ESCAPE_SPECIAL): + cp = mchars_spec2str(p->symtab, seq, sz, &ssz); + if (NULL != cp) + encode(p, cp, ssz); + else if (1 == ssz) + encode(p, seq, sz); + break; + case (ESCAPE_FONTBOLD): term_fontrepl(p, TERMFONT_BOLD); break; - case (DECO_ITALIC): + case (ESCAPE_FONTITALIC): term_fontrepl(p, TERMFONT_UNDER); break; - case (DECO_ROMAN): + case (ESCAPE_FONTBI): + term_fontrepl(p, TERMFONT_BI); + break; + case (ESCAPE_FONT): + /* FALLTHROUGH */ + case (ESCAPE_FONTROMAN): term_fontrepl(p, TERMFONT_NONE); break; - case (DECO_PREVIOUS): + case (ESCAPE_FONTPREV): term_fontlast(p); break; - default: + case (ESCAPE_NOSPACE): + if (TERMP_SKIPCHAR & p->flags) + p->flags &= ~TERMP_SKIPCHAR; + else if ('\0' == *word) + p->flags |= TERMP_NOSPACE; break; - } - - if (DECO_NOSPACE == deco && '\0' == *word) - p->flags |= TERMP_NOSPACE; - } - - /* - * 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]) - switch (sv[0]) { - case('('): - /* FALLTHROUGH */ - case('['): - p->flags |= TERMP_NOSPACE; + case (ESCAPE_SKIPCHAR): + p->flags |= TERMP_SKIPCHAR; break; default: break; } + } + p->flags &= ~TERMP_NBRWORD; } - static void adjbuf(struct termp *p, size_t sz) { @@ -565,14 +538,9 @@ adjbuf(struct termp *p, size_t sz) while (sz >= p->maxcols) p->maxcols <<= 2; - p->buf = realloc(p->buf, p->maxcols); - if (NULL == p->buf) { - perror(NULL); - exit((int)MANDOCLEVEL_SYSERR); - } + p->buf = mandoc_realloc(p->buf, sizeof(int) * p->maxcols); } - static void bufferc(struct termp *p, char c) { @@ -580,52 +548,81 @@ bufferc(struct termp *p, char c) if (p->col + 1 >= p->maxcols) adjbuf(p, p->col + 1); - p->buf[(int)p->col++] = c; + p->buf[p->col++] = c; } +/* + * See encode(). + * Do this for a single (probably unicode) value. + * Does not check for non-decorated glyphs. + */ +static void +encode1(struct termp *p, int c) +{ + enum termfont f; + if (TERMP_SKIPCHAR & p->flags) { + p->flags &= ~TERMP_SKIPCHAR; + return; + } + + if (p->col + 6 >= p->maxcols) + adjbuf(p, p->col + 6); + + f = term_fonttop(p); + + if (TERMFONT_UNDER == f || TERMFONT_BI == f) { + p->buf[p->col++] = '_'; + p->buf[p->col++] = 8; + } + if (TERMFONT_BOLD == f || TERMFONT_BI == f) { + if (ASCII_HYPH == c) + p->buf[p->col++] = '-'; + else + p->buf[p->col++] = c; + p->buf[p->col++] = 8; + } + p->buf[p->col++] = c; +} + static void encode(struct termp *p, const char *word, size_t sz) { - enum termfont f; - int i; + size_t i; + if (TERMP_SKIPCHAR & p->flags) { + p->flags &= ~TERMP_SKIPCHAR; + return; + } + /* * Encode and buffer a string of characters. If the current * font mode is unset, buffer directly, else encode then buffer * character by character. */ - if (TERMFONT_NONE == (f = term_fonttop(p))) { + if (TERMFONT_NONE == term_fonttop(p)) { if (p->col + sz >= p->maxcols) adjbuf(p, p->col + sz); - memcpy(&p->buf[(int)p->col], word, sz); - p->col += sz; + for (i = 0; i < sz; i++) + p->buf[p->col++] = word[i]; return; } /* Pre-buffer, assuming worst-case. */ - if (p->col + 1 + (sz * 3) >= p->maxcols) - adjbuf(p, p->col + 1 + (sz * 3)); + if (p->col + 1 + (sz * 5) >= p->maxcols) + adjbuf(p, p->col + 1 + (sz * 5)); - for (i = 0; i < (int)sz; i++) { - if ( ! isgraph((u_char)word[i])) { - p->buf[(int)p->col++] = word[i]; - continue; - } - - if (TERMFONT_UNDER == f) - p->buf[(int)p->col++] = '_'; + for (i = 0; i < sz; i++) { + if (ASCII_HYPH == word[i] || + isgraph((unsigned char)word[i])) + encode1(p, word[i]); else - p->buf[(int)p->col++] = word[i]; - - p->buf[(int)p->col++] = 8; - p->buf[(int)p->col++] = word[i]; + p->buf[p->col++] = word[i]; } } - size_t term_len(const struct termp *p, size_t sz) { @@ -633,57 +630,124 @@ term_len(const struct termp *p, size_t sz) return((*p->width)(p, ' ') * sz); } +static size_t +cond_width(const struct termp *p, int c, int *skip) +{ + if (*skip) { + (*skip) = 0; + return(0); + } else + return((*p->width)(p, c)); +} + size_t term_strlen(const struct termp *p, const char *cp) { - size_t sz, ssz, rsz, i; - enum roffdeco d; + size_t sz, rsz, i; + int ssz, skip, c; const char *seq, *rhs; + enum mandoc_esc esc; + static const char rej[] = { '\\', ASCII_NBRSP, ASCII_HYPH, + ASCII_BREAK, '\0' }; - 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); + /* + * Account for escaped sequences within string length + * calculations. This follows the logic in term_word() as we + * must calculate the width of produced strings. + */ - switch (d) { - case (DECO_RESERVED): - rhs = chars_res2str - (p->symtab, seq, ssz, &rsz); + sz = 0; + skip = 0; + while ('\0' != *cp) { + rsz = strcspn(cp, rej); + for (i = 0; i < rsz; i++) + sz += cond_width(p, *cp++, &skip); + + switch (*cp) { + case ('\\'): + cp++; + esc = mandoc_escape(&cp, &seq, &ssz); + if (ESCAPE_ERROR == esc) + return(sz); + + if (TERMENC_ASCII != p->enc) + switch (esc) { + case (ESCAPE_UNICODE): + c = mchars_num2uc + (seq + 1, ssz - 1); + if ('\0' == c) + break; + sz += cond_width(p, c, &skip); + continue; + case (ESCAPE_SPECIAL): + c = mchars_spec2cp + (p->symtab, seq, ssz); + if (c <= 0) + break; + sz += cond_width(p, c, &skip); + continue; + default: + break; + } + + rhs = NULL; + + switch (esc) { + case (ESCAPE_UNICODE): + sz += cond_width(p, '?', &skip); break; - case (DECO_SPECIAL): - /* FALLTHROUGH */ - case (DECO_SSPECIAL): - rhs = chars_spec2str + case (ESCAPE_NUMBERED): + c = mchars_num2char(seq, ssz); + if ('\0' != c) + sz += cond_width(p, c, &skip); + break; + case (ESCAPE_SPECIAL): + rhs = mchars_spec2str (p->symtab, seq, ssz, &rsz); - /* Allow for one-char escapes. */ - if (DECO_SSPECIAL != d || rhs) + if (ssz != 1 || rhs) break; rhs = seq; rsz = ssz; break; + case (ESCAPE_SKIPCHAR): + skip = 1; + break; default: - rhs = NULL; break; } - if (rhs) - for (i = 0; i < rsz; i++) - sz += (*p->width)(p, *rhs++); - } else - sz += (*p->width)(p, *cp++); + if (NULL == rhs) + break; + if (skip) { + skip = 0; + break; + } + + for (i = 0; i < rsz; i++) + sz += (*p->width)(p, *rhs++); + break; + case (ASCII_NBRSP): + sz += cond_width(p, ' ', &skip); + cp++; + break; + case (ASCII_HYPH): + sz += cond_width(p, '-', &skip); + cp++; + /* FALLTHROUGH */ + case (ASCII_BREAK): + break; + default: + break; + } + } + return(sz); } - /* ARGSUSED */ size_t term_vspan(const struct termp *p, const struct roffsu *su) @@ -719,7 +783,6 @@ term_vspan(const struct termp *p, const struct roffsu return(/* LINTED */(size_t) r); } - size_t term_hspan(const struct termp *p, const struct roffsu *su)