-/* $Id: term.c,v 1.131 2010/04/08 07:05:38 kristaps Exp $ */
+/* $Id: term.c,v 1.193 2011/05/17 14:38:34 kristaps Exp $ */
/*
- * Copyright (c) 2008, 2009 Kristaps Dzonsons <kristaps@kth.se>
+ * Copyright (c) 2008, 2009, 2010 Kristaps Dzonsons <kristaps@bsd.lv>
+ * Copyright (c) 2010, 2011 Ingo Schwarze <schwarze@openbsd.org>
*
* Permission to use, copy, modify, and distribute this software for any
* purpose with or without fee is hereby granted, provided that the above
#include <assert.h>
#include <ctype.h>
+#include <stdint.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
-#include <time.h>
-#include "chars.h"
+#include "mandoc.h"
#include "out.h"
#include "term.h"
-#include "man.h"
-#include "mdoc.h"
#include "main.h"
-static struct termp *term_alloc(enum termenc);
-static void term_free(struct termp *);
-static void spec(struct termp *, 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);
-
-
-void *
-ascii_alloc(void)
-{
-
- return(term_alloc(TERMENC_ASCII));
-}
-
+static void adjbuf(struct termp *p, int);
+static void bufferc(struct termp *, char);
+static void encode(struct termp *, const char *, size_t);
void
-terminal_free(void *arg)
-{
-
- term_free((struct termp *)arg);
-}
-
-
-static void
term_free(struct termp *p)
{
if (p->buf)
free(p->buf);
if (p->symtab)
- chars_free(p->symtab);
+ mchars_free(p->symtab);
free(p);
}
-static struct termp *
-term_alloc(enum termenc enc)
+void
+term_begin(struct termp *p, term_margin head,
+ term_margin foot, const void *arg)
{
- struct termp *p;
- p = calloc(1, sizeof(struct termp));
- if (NULL == p) {
- perror(NULL);
- exit(EXIT_FAILURE);
- }
- p->enc = enc;
- return(p);
+ p->headf = head;
+ p->footf = foot;
+ p->argf = arg;
+ (*p->begin)(p);
}
+void
+term_end(struct termp *p)
+{
+
+ (*p->end)(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
int i; /* current input position in p->buf */
size_t vis; /* current visual position on output */
size_t vbl; /* number of blanks to prepend to output */
- size_t vsz; /* visual characters to write to output */
+ size_t vend; /* end of word visual position on output */
size_t bp; /* visual right border position */
- size_t hyph; /* visible position of hyphen */
- int j; /* temporary loop index */
- 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.
* an indentation, but can be, for tagged lists or columns, a
* small set of values.
*/
-
- 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;
+ 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;
bp = TERMP_NOBREAK & p->flags ? mmax : maxvis;
- /*
- * FIXME: if bp is zero, we still output the first word before
- * breaking the line.
- */
-
- vis = 0;
-
/*
- * If in the standard case (left-justified), then begin with our
- * indentation, otherwise (columns, etc.) just start spitting
- * out text.
+ * Indent the first line of a paragraph.
*/
+ vbl = p->flags & TERMP_NOLPAD ? (size_t)0 : p->offset;
+
+ vis = vend = 0;
+ i = 0;
- if ( ! (p->flags & TERMP_NOLPAD))
- /* LINTED */
- for (j = 0; j < (int)p->offset; j++)
- putchar(' ');
+ while (i < p->col) {
+ /*
+ * Handle literal tab characters: collapse all
+ * subsequent tabs into a single huge set of spaces.
+ */
+ while (i < p->col && '\t' == p->buf[i]) {
+ vend = (vis / p->tabwidth + 1) * p->tabwidth;
+ vbl += vend - vis;
+ vis = vend;
+ i++;
+ }
- for (i = 0; i < (int)p->col; i++) {
/*
* Count up visible word characters. Control sequences
* (starting with the CSI) aren't counted. A space
* generates a non-printing word, which is valid (the
* space is printed according to regular spacing rules).
- * Collect the number of printable characters until the
- * first hyphen, if found. Hyphens aren't included if
- * they're the first character (so `Fl' doesn't break)
- * or second consecutive character (`Fl -').
*/
- /* LINTED */
- for (j = i, vsz = 0, hyph = 0; j < (int)p->col; j++) {
- if (j && ' ' == p->buf[j])
+ for (j = i, jhy = 0; j < p->col; j++) {
+ if ((j && ' ' == p->buf[j]) || '\t' == p->buf[j])
break;
- if (8 == p->buf[j])
- vsz--;
- else
- vsz++;
- if (j > i && '-' == p->buf[j] && 0 == hyph)
- if ('-' != p->buf[j - 1])
- hyph = vsz;
- }
- /*
- * Choose the number of blanks to prepend: no blank at the
- * beginning of a line, one between words -- but do not
- * actually write them yet.
- */
+ /* Back over the the last printed character. */
+ if (8 == p->buf[j]) {
+ assert(j);
+ vend -= (*p->width)(p, p->buf[j - 1]);
+ continue;
+ }
- vbl = (size_t)(0 == vis ? 0 : 1);
+ /* 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]);
+ }
/*
* Find out whether we would exceed the right margin.
- * If so, break to the next line, possibly after
- * emittign character up to a hyphen. Otherwise, write
- * the chosen number of blanks.
+ * If so, break to the next line.
*/
-
- if (vis && vis + vbl + vsz > bp) {
- /*
- * Has a hyphen been found before the breakpoint
- * that we can use?
- */
- if (hyph && vis + vbl + hyph <= bp) {
- /* First prepend blanks. */
- for (j = 0; j < (int)vbl; j++)
- putchar(' ');
-
- /* Emit up to the character. */
- do {
- if (31 == p->buf[i])
- putchar(' ');
- else
- putchar(p->buf[i]);
- if (8 != p->buf[i])
- vsz--;
- } while ('-' != p->buf[i++]);
-
- /* Emit trailing decoration. */
- if (8 == p->buf[i]) {
- putchar(p->buf[i]);
- putchar(p->buf[i + 1]);
- }
- }
-
- putchar('\n');
+ if (vend > bp && 0 == jhy && vis > 0) {
+ vend -= vis;
+ (*p->endline)(p);
if (TERMP_NOBREAK & p->flags) {
- for (j = 0; j < (int)p->rmargin; j++)
- putchar(' ');
- vis = p->rmargin - p->offset;
+ p->viscol = p->rmargin;
+ (*p->advance)(p, p->rmargin);
+ vend += p->rmargin - p->offset;
} else {
- for (j = 0; j < (int)p->offset; j++)
- putchar(' ');
- vis = 0;
+ p->viscol = 0;
+ vbl = p->offset;
}
/* Remove the p->overstep width. */
- bp += (int)/* LINTED */
- p->overstep;
+ bp += (size_t)p->overstep;
p->overstep = 0;
- } else {
- for (j = 0; j < (int)vbl; j++)
- putchar(' ');
- vis += vbl;
}
/* Write out the [remaining] word. */
- for ( ; i < (int)p->col; i++)
- if (' ' == p->buf[i])
+ for ( ; i < p->col; i++) {
+ if (vend > bp && jhy > 0 && i > jhy)
+ break;
+ if ('\t' == p->buf[i])
break;
- else if (31 == p->buf[i])
- putchar(' ');
- else
- putchar(p->buf[i]);
+ if (' ' == p->buf[i]) {
+ 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 += (*p->width)(p, ' ');
+ continue;
+ }
+
+ /*
+ * Now we definitely know there will be
+ * printable characters to output,
+ * so write preceding white space now.
+ */
+ if (vbl) {
+ (*p->advance)(p, vbl);
+ p->viscol += vbl;
+ vbl = 0;
+ }
- vis += vsz;
+ if (ASCII_HYPH == p->buf[i]) {
+ (*p->letter)(p, '-');
+ p->viscol += (*p->width)(p, '-');
+ } else {
+ (*p->letter)(p, p->buf[i]);
+ p->viscol += (*p->width)(p, p->buf[i]);
+ }
+ }
+ 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;
if ( ! (TERMP_NOBREAK & p->flags)) {
- putchar('\n');
+ p->viscol = 0;
+ (*p->endline)(p);
return;
}
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:
*/
if (p->overstep >= -1) {
assert((int)maxvis + p->overstep >= 0);
- /* LINTED */
- maxvis += p->overstep;
+ maxvis += (size_t)p->overstep;
} else
p->overstep = 0;
return;
/* Right-pad. */
- if (maxvis > vis + /* LINTED */
- ((TERMP_TWOSPACE & p->flags) ? 1 : 0))
- for ( ; vis < maxvis; vis++)
- putchar(' ');
- else { /* ...or newline break. */
- putchar('\n');
- for (i = 0; i < (int)p->rmargin; i++)
- putchar(' ');
+ 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. */
+ (*p->endline)(p);
+ p->viscol = p->rmargin;
+ (*p->advance)(p, p->rmargin);
}
}
{
p->flags |= TERMP_NOSPACE;
- if (0 == p->col) {
+ if (0 == p->col && 0 == p->viscol) {
p->flags &= ~TERMP_NOLPAD;
return;
}
{
term_newln(p);
- putchar('\n');
-}
-
-
-static void
-spec(struct termp *p, const char *word, size_t len)
-{
- const char *rhs;
- size_t sz;
-
- rhs = chars_a2ascii(p->symtab, word, len, &sz);
- if (rhs)
- encode(p, rhs, sz);
+ p->viscol = 0;
+ (*p->endline)(p);
}
-
-static void
-res(struct termp *p, const char *word, size_t len)
-{
- const char *rhs;
- size_t sz;
-
- rhs = chars_a2res(p->symtab, word, len, &sz);
- if (rhs)
- encode(p, rhs, sz);
-}
-
-
void
term_fontlast(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
void
term_word(struct termp *p, const char *word)
{
- const char *sv, *seq;
+ const char *seq, *cp;
+ char c;
int sz;
size_t ssz;
- enum roffdeco deco;
-
- 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))
- bufferc(p, ' ');
+ enum mandoc_esc esc;
+
+ 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 ( ! (p->flags & TERMP_NONOSPACE))
p->flags &= ~TERMP_NOSPACE;
+ else
+ p->flags |= TERMP_NOSPACE;
- /* FIXME: use strcspn. */
+ p->flags &= ~(TERMP_SENTENCE | TERMP_IGNDELIM);
- while (*word) {
- if ('\\' != *word) {
- encode(p, word, 1);
- word++;
+ while ('\0' != *word) {
+ if ((ssz = strcspn(word, "\\")) > 0)
+ encode(p, word, ssz);
+
+ word += (int)ssz;
+ if ('\\' != *word)
continue;
- }
- seq = ++word;
- sz = 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);
+ switch (esc) {
+ case (ESCAPE_UNICODE):
+ encode(p, "?", 1);
+ break;
+ case (ESCAPE_NUMBERED):
+ if ('\0' != (c = mchars_num2char(seq, sz)))
+ encode(p, &c, 1);
break;
- case (DECO_SPECIAL):
- spec(p, seq, ssz);
+ case (ESCAPE_PREDEF):
+ cp = mchars_res2str(p->symtab, seq, sz, &ssz);
+ if (NULL != cp)
+ encode(p, cp, ssz);
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_FONTROMAN):
term_fontrepl(p, TERMFONT_NONE);
break;
- case (DECO_PREVIOUS):
+ case (ESCAPE_FONTPREV):
term_fontlast(p);
break;
- default:
- break;
- }
-
- word += sz;
- 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_NOSPACE):
+ if ('\0' == *word)
+ p->flags |= TERMP_NOSPACE;
break;
default:
break;
}
+ }
}
-
static void
-adjbuf(struct termp *p, size_t sz)
+adjbuf(struct termp *p, int sz)
{
if (0 == p->maxcols)
while (sz >= p->maxcols)
p->maxcols <<= 2;
- p->buf = realloc(p->buf, p->maxcols);
- if (NULL == p->buf) {
- perror(NULL);
- exit(EXIT_FAILURE);
- }
-}
-
-
-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;
+ p->buf = mandoc_realloc
+ (p->buf, sizeof(int) * (size_t)p->maxcols);
}
-
static void
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;
}
-
static void
encode(struct termp *p, const char *word, size_t sz)
{
enum termfont f;
- int i;
+ int i, len;
+
+ /* LINTED */
+ len = sz;
/*
* Encode and buffer a string of characters. If the current
*/
if (TERMFONT_NONE == (f = term_fonttop(p))) {
- buffera(p, word, sz);
+ if (p->col + len >= p->maxcols)
+ adjbuf(p, p->col + len);
+ for (i = 0; i < len; i++)
+ p->buf[p->col++] = word[i];
return;
}
- for (i = 0; i < (int)sz; i++) {
- if ( ! isgraph((u_char)word[i])) {
- bufferc(p, word[i]);
+ /* Pre-buffer, assuming worst-case. */
+
+ if (p->col + 1 + (len * 3) >= p->maxcols)
+ adjbuf(p, p->col + 1 + (len * 3));
+
+ for (i = 0; i < len; i++) {
+ if ( ! isgraph((unsigned char)word[i])) {
+ p->buf[p->col++] = word[i];
continue;
}
if (TERMFONT_UNDER == f)
- bufferc(p, '_');
+ p->buf[p->col++] = '_';
else
- bufferc(p, word[i]);
+ p->buf[p->col++] = word[i];
- bufferc(p, 8);
- bufferc(p, word[i]);
+ p->buf[p->col++] = 8;
+ p->buf[p->col++] = word[i];
}
}
+size_t
+term_len(const struct termp *p, size_t sz)
+{
+
+ return((*p->width)(p, ' ') * sz);
+}
+
size_t
-term_vspan(const struct roffsu *su)
+term_strlen(const struct termp *p, const char *cp)
+{
+ size_t sz, rsz, i;
+ int ssz, c;
+ const char *seq, *rhs;
+ static const char rej[] = { '\\', ASCII_HYPH, ASCII_NBRSP, '\0' };
+
+ /*
+ * Account for escaped sequences within string length
+ * calculations. This follows the logic in term_word() as we
+ * must calculate the width of produced strings.
+ */
+
+ sz = 0;
+ while ('\0' != *cp) {
+ rsz = strcspn(cp, rej);
+ for (i = 0; i < rsz; i++)
+ sz += (*p->width)(p, *cp++);
+
+ c = 0;
+ switch (*cp) {
+ case ('\\'):
+ cp++;
+ rhs = NULL;
+ switch (mandoc_escape(&cp, &seq, &ssz)) {
+ case (ESCAPE_ERROR):
+ return(sz);
+ case (ESCAPE_UNICODE):
+ c = '?';
+ /* FALLTHROUGH */
+ case (ESCAPE_NUMBERED):
+ if ('\0' != c)
+ c = mchars_num2char(seq, ssz);
+ if ('\0' != c)
+ sz += (*p->width)(p, c);
+ break;
+ case (ESCAPE_PREDEF):
+ rhs = mchars_res2str
+ (p->symtab, seq, ssz, &rsz);
+ break;
+ case (ESCAPE_SPECIAL):
+ rhs = mchars_spec2str
+ (p->symtab, seq, ssz, &rsz);
+
+ if (ssz != 1 || rhs)
+ break;
+
+ rhs = seq;
+ rsz = ssz;
+ break;
+ default:
+ break;
+ }
+
+ if (NULL == rhs)
+ break;
+
+ for (i = 0; i < rsz; i++)
+ sz += (*p->width)(p, *rhs++);
+ break;
+ case (ASCII_NBRSP):
+ sz += (*p->width)(p, ' ');
+ cp++;
+ break;
+ case (ASCII_HYPH):
+ sz += (*p->width)(p, '-');
+ cp++;
+ break;
+ default:
+ break;
+ }
+ }
+
+ return(sz);
+}
+
+/* ARGSUSED */
+size_t
+term_vspan(const struct termp *p, const struct roffsu *su)
{
double r;
r);
}
-
size_t
-term_hspan(const struct roffsu *su)
+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);
}
-
-
git.ckatri.com / mandoc.git / blobdiff