version 1.128, 2010/01/01 17:14:30 |
version 1.141, 2010/06/07 10:52:44 |
|
|
|
|
#include <assert.h> |
#include <assert.h> |
#include <ctype.h> |
#include <ctype.h> |
|
#include <stdint.h> |
#include <stdio.h> |
#include <stdio.h> |
#include <stdlib.h> |
#include <stdlib.h> |
#include <string.h> |
#include <string.h> |
#include <time.h> |
#include <time.h> |
|
|
|
#include "mandoc.h" |
#include "chars.h" |
#include "chars.h" |
#include "out.h" |
#include "out.h" |
#include "term.h" |
#include "term.h" |
|
|
#include "mdoc.h" |
#include "mdoc.h" |
#include "main.h" |
#include "main.h" |
|
|
static struct termp *term_alloc(enum termenc); |
static struct termp *term_alloc(char *, enum termenc); |
static void term_free(struct termp *); |
static void term_free(struct termp *); |
static void spec(struct termp *, const char *, size_t); |
static void spec(struct termp *, const char *, size_t); |
static void res(struct termp *, const char *, size_t); |
static void res(struct termp *, const char *, size_t); |
Line 45 static void encode(struct termp *, const char *, si |
|
Line 47 static void encode(struct termp *, const char *, si |
|
|
|
|
|
void * |
void * |
ascii_alloc(void) |
ascii_alloc(char *outopts) |
{ |
{ |
|
|
return(term_alloc(TERMENC_ASCII)); |
return(term_alloc(outopts, TERMENC_ASCII)); |
} |
} |
|
|
|
|
Line 74 term_free(struct termp *p) |
|
Line 76 term_free(struct termp *p) |
|
|
|
|
|
static struct termp * |
static struct termp * |
term_alloc(enum termenc enc) |
term_alloc(char *outopts, enum termenc enc) |
{ |
{ |
struct termp *p; |
struct termp *p; |
|
const char *toks[2]; |
|
char *v; |
|
size_t width; |
|
|
|
toks[0] = "width"; |
|
toks[1] = NULL; |
|
|
p = calloc(1, sizeof(struct termp)); |
p = calloc(1, sizeof(struct termp)); |
if (NULL == p) { |
if (NULL == p) { |
perror(NULL); |
perror(NULL); |
exit(EXIT_FAILURE); |
exit(EXIT_FAILURE); |
} |
} |
p->maxrmargin = 78; |
|
|
p->tabwidth = 5; |
p->enc = enc; |
p->enc = enc; |
|
width = 80; |
|
|
|
while (outopts && *outopts) |
|
switch (getsubopt(&outopts, UNCONST(toks), &v)) { |
|
case (0): |
|
width = atoi(v); |
|
break; |
|
default: |
|
break; |
|
} |
|
|
|
/* Enforce some lower boundary. */ |
|
if (width < 60) |
|
width = 60; |
|
p->defrmargin = width - 2; |
return(p); |
return(p); |
} |
} |
|
|
Line 93 term_alloc(enum termenc enc) |
|
Line 117 term_alloc(enum termenc enc) |
|
* Flush a line of text. A "line" is loosely defined as being something |
* 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 |
* that should be followed by a newline, regardless of whether it's |
* broken apart by newlines getting there. A line can also be a |
* broken apart by newlines getting there. A line can also be a |
* fragment of a columnar list. |
* fragment of a columnar list (`Bl -tag' or `Bl -column'), which does |
|
* not have a trailing newline. |
* |
* |
* Specifically, a line is whatever's in p->buf of length p->col, which |
* The following flags may be specified: |
* is zeroed after this function returns. |
|
* |
* |
* The usage of termp:flags is as follows: |
|
* |
|
* - TERMP_NOLPAD: when beginning to write the line, don't left-pad the |
* - TERMP_NOLPAD: when beginning to write the line, don't left-pad the |
* offset value. This is useful when doing columnar lists where the |
* offset value. This is useful when doing columnar lists where the |
* prior column has right-padded. |
* prior column has right-padded. |
Line 132 term_flushln(struct termp *p) |
|
Line 154 term_flushln(struct termp *p) |
|
int i; /* current input position in p->buf */ |
int i; /* current input position in p->buf */ |
size_t vis; /* current visual position on output */ |
size_t vis; /* current visual position on output */ |
size_t vbl; /* number of blanks to prepend to 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 bp; /* visual right border position */ |
int j; /* temporary loop index */ |
int j; /* temporary loop index */ |
|
int jhy; /* last hyphen before line overflow */ |
size_t maxvis, mmax; |
size_t maxvis, mmax; |
static int overstep = 0; |
|
|
|
/* |
/* |
* First, establish the maximum columns of "visible" content. |
* First, establish the maximum columns of "visible" content. |
Line 147 term_flushln(struct termp *p) |
|
Line 169 term_flushln(struct termp *p) |
|
|
|
assert(p->offset < p->rmargin); |
assert(p->offset < p->rmargin); |
|
|
maxvis = (int)(p->rmargin - p->offset) - overstep < 0 ? |
maxvis = (int)(p->rmargin - p->offset) - p->overstep < 0 ? |
/* LINTED */ |
/* LINTED */ |
0 : p->rmargin - p->offset - overstep; |
0 : p->rmargin - p->offset - p->overstep; |
mmax = (int)(p->maxrmargin - p->offset) - overstep < 0 ? |
mmax = (int)(p->maxrmargin - p->offset) - p->overstep < 0 ? |
/* LINTED */ |
/* LINTED */ |
0 : p->maxrmargin - p->offset - overstep; |
0 : p->maxrmargin - p->offset - p->overstep; |
|
|
bp = TERMP_NOBREAK & p->flags ? mmax : maxvis; |
bp = TERMP_NOBREAK & p->flags ? mmax : maxvis; |
|
|
|
/* |
|
* Indent the first line of a paragraph. |
|
*/ |
|
vbl = p->flags & TERMP_NOLPAD ? 0 : p->offset; |
|
|
/* |
/* |
* FIXME: if bp is zero, we still output the first word before |
* FIXME: if bp is zero, we still output the first word before |
* breaking the line. |
* breaking the line. |
*/ |
*/ |
|
|
vis = 0; |
vis = vend = i = 0; |
|
while (i < (int)p->col) { |
|
|
/* |
/* |
* If in the standard case (left-justified), then begin with our |
* Handle literal tab characters. |
* indentation, otherwise (columns, etc.) just start spitting |
*/ |
* out text. |
for (j = i; j < (int)p->col; j++) { |
*/ |
if ('\t' != p->buf[j]) |
|
break; |
|
vend = (vis/p->tabwidth+1)*p->tabwidth; |
|
vbl += vend - vis; |
|
vis = vend; |
|
} |
|
|
if ( ! (p->flags & TERMP_NOLPAD)) |
|
/* LINTED */ |
|
for (j = 0; j < (int)p->offset; j++) |
|
putchar(' '); |
|
|
|
for (i = 0; i < (int)p->col; i++) { |
|
/* |
/* |
* Count up visible word characters. Control sequences |
* Count up visible word characters. Control sequences |
* (starting with the CSI) aren't counted. A space |
* (starting with the CSI) aren't counted. A space |
Line 183 term_flushln(struct termp *p) |
|
Line 210 term_flushln(struct termp *p) |
|
*/ |
*/ |
|
|
/* LINTED */ |
/* LINTED */ |
for (j = i, vsz = 0; j < (int)p->col; j++) { |
for (jhy = 0; j < (int)p->col; j++) { |
if (j && ' ' == p->buf[j]) |
if ((j && ' ' == p->buf[j]) || '\t' == p->buf[j]) |
break; |
break; |
else if (8 == p->buf[j]) |
if (8 != p->buf[j]) { |
vsz--; |
if (vend > vis && vend < bp && |
else |
ASCII_HYPH == p->buf[j]) |
vsz++; |
jhy = j; |
|
vend++; |
|
} else |
|
vend--; |
} |
} |
|
|
/* |
/* |
* 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. |
|
*/ |
|
vbl = (size_t)(0 == vis ? 0 : 1); |
|
|
|
/* |
|
* Find out whether we would exceed the right margin. |
* Find out whether we would exceed the right margin. |
* If so, break to the next line. (TODO: hyphenate) |
* If so, break to the next line. |
* Otherwise, write the chosen number of blanks now. |
|
*/ |
*/ |
if (vis && vis + vbl + vsz > bp) { |
if (vend > bp && 0 == jhy && vis > 0) { |
|
vend -= vis; |
putchar('\n'); |
putchar('\n'); |
if (TERMP_NOBREAK & p->flags) { |
if (TERMP_NOBREAK & p->flags) { |
|
p->viscol = p->rmargin; |
for (j = 0; j < (int)p->rmargin; j++) |
for (j = 0; j < (int)p->rmargin; j++) |
putchar(' '); |
putchar(' '); |
vis = p->rmargin - p->offset; |
vend += p->rmargin - p->offset; |
} else { |
} else { |
for (j = 0; j < (int)p->offset; j++) |
p->viscol = 0; |
putchar(' '); |
vbl = p->offset; |
vis = 0; |
|
} |
} |
/* Remove the overstep width. */ |
|
|
/* Remove the p->overstep width. */ |
|
|
bp += (int)/* LINTED */ |
bp += (int)/* LINTED */ |
overstep; |
p->overstep; |
overstep = 0; |
p->overstep = 0; |
} else { |
|
for (j = 0; j < (int)vbl; j++) |
|
putchar(' '); |
|
vis += vbl; |
|
} |
} |
|
|
/* |
/* |
* Finally, write out the word. |
* 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++) { |
for ( ; i < (int)p->col; i++) { |
if (' ' == p->buf[i]) |
if (vend > bp && jhy > 0 && i > jhy) |
break; |
break; |
|
if ('\t' == p->buf[i]) |
|
break; |
|
if (' ' == p->buf[i]) { |
|
while (' ' == p->buf[i]) { |
|
vbl++; |
|
i++; |
|
} |
|
break; |
|
} |
|
if (ASCII_NBRSP == p->buf[i]) { |
|
vbl++; |
|
continue; |
|
} |
|
|
/* The unit sep. is a non-breaking space. */ |
/* |
if (31 == p->buf[i]) |
* Now we definitely know there will be |
putchar(' '); |
* printable characters to output, |
|
* so write preceding white space now. |
|
*/ |
|
if (vbl) { |
|
for (j = 0; j < (int)vbl; j++) |
|
putchar(' '); |
|
p->viscol += vbl; |
|
vbl = 0; |
|
} |
|
|
|
if (ASCII_HYPH == p->buf[i]) |
|
putchar('-'); |
else |
else |
putchar(p->buf[i]); |
putchar(p->buf[i]); |
|
|
|
p->viscol += 1; |
} |
} |
vis += vsz; |
vend += vbl; |
|
vis = vend; |
} |
} |
|
|
p->col = 0; |
p->col = 0; |
overstep = 0; |
p->overstep = 0; |
|
|
if ( ! (TERMP_NOBREAK & p->flags)) { |
if ( ! (TERMP_NOBREAK & p->flags)) { |
|
p->viscol = 0; |
putchar('\n'); |
putchar('\n'); |
return; |
return; |
} |
} |
|
|
if (TERMP_HANG & p->flags) { |
if (TERMP_HANG & p->flags) { |
/* We need one blank after the tag. */ |
/* We need one blank after the tag. */ |
overstep = /* LINTED */ |
p->overstep = /* LINTED */ |
vis - maxvis + 1; |
vis - maxvis + 1; |
|
|
/* |
/* |
Line 264 term_flushln(struct termp *p) |
|
Line 317 term_flushln(struct termp *p) |
|
* move it one step LEFT and flag the rest of the line |
* move it one step LEFT and flag the rest of the line |
* to be longer. |
* to be longer. |
*/ |
*/ |
if (overstep >= -1) { |
if (p->overstep >= -1) { |
assert((int)maxvis + overstep >= 0); |
assert((int)maxvis + p->overstep >= 0); |
/* LINTED */ |
/* LINTED */ |
maxvis += overstep; |
maxvis += p->overstep; |
} else |
} else |
overstep = 0; |
p->overstep = 0; |
|
|
} else if (TERMP_DANGLE & p->flags) |
} else if (TERMP_DANGLE & p->flags) |
return; |
return; |
|
|
/* Right-pad. */ |
/* Right-pad. */ |
if (maxvis > vis + /* LINTED */ |
if (maxvis > vis + /* LINTED */ |
((TERMP_TWOSPACE & p->flags) ? 1 : 0)) |
((TERMP_TWOSPACE & p->flags) ? 1 : 0)) { |
|
p->viscol += maxvis - vis; |
for ( ; vis < maxvis; vis++) |
for ( ; vis < maxvis; vis++) |
putchar(' '); |
putchar(' '); |
else { /* ...or newline break. */ |
} else { /* ...or newline break. */ |
putchar('\n'); |
putchar('\n'); |
|
p->viscol = p->rmargin; |
for (i = 0; i < (int)p->rmargin; i++) |
for (i = 0; i < (int)p->rmargin; i++) |
putchar(' '); |
putchar(' '); |
} |
} |
Line 297 term_newln(struct termp *p) |
|
Line 352 term_newln(struct termp *p) |
|
{ |
{ |
|
|
p->flags |= TERMP_NOSPACE; |
p->flags |= TERMP_NOSPACE; |
if (0 == p->col) { |
if (0 == p->col && 0 == p->viscol) { |
p->flags &= ~TERMP_NOLPAD; |
p->flags &= ~TERMP_NOLPAD; |
return; |
return; |
} |
} |
Line 317 term_vspace(struct termp *p) |
|
Line 372 term_vspace(struct termp *p) |
|
{ |
{ |
|
|
term_newln(p); |
term_newln(p); |
|
p->viscol = 0; |
putchar('\n'); |
putchar('\n'); |
} |
} |
|
|
Line 442 term_word(struct termp *p, const char *word) |
|
Line 498 term_word(struct termp *p, const char *word) |
|
case(')'): |
case(')'): |
/* FALLTHROUGH */ |
/* FALLTHROUGH */ |
case(']'): |
case(']'): |
/* FALLTHROUGH */ |
|
case('}'): |
|
if ( ! (TERMP_IGNDELIM & p->flags)) |
if ( ! (TERMP_IGNDELIM & p->flags)) |
p->flags |= TERMP_NOSPACE; |
p->flags |= TERMP_NOSPACE; |
break; |
break; |
Line 451 term_word(struct termp *p, const char *word) |
|
Line 505 term_word(struct termp *p, const char *word) |
|
break; |
break; |
} |
} |
|
|
if ( ! (TERMP_NOSPACE & p->flags)) |
if ( ! (TERMP_NOSPACE & p->flags)) { |
bufferc(p, ' '); |
bufferc(p, ' '); |
|
if (TERMP_SENTENCE & p->flags) |
|
bufferc(p, ' '); |
|
} |
|
|
if ( ! (p->flags & TERMP_NONOSPACE)) |
if ( ! (p->flags & TERMP_NONOSPACE)) |
p->flags &= ~TERMP_NOSPACE; |
p->flags &= ~TERMP_NOSPACE; |
|
|
|
p->flags &= ~TERMP_SENTENCE; |
|
|
/* FIXME: use strcspn. */ |
/* FIXME: use strcspn. */ |
|
|
while (*word) { |
while (*word) { |
Line 497 term_word(struct termp *p, const char *word) |
|
Line 556 term_word(struct termp *p, const char *word) |
|
p->flags |= TERMP_NOSPACE; |
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]) |
if (sv[0] && 0 == sv[1]) |
switch (sv[0]) { |
switch (sv[0]) { |
case('('): |
case('('): |
/* FALLTHROUGH */ |
/* FALLTHROUGH */ |
case('['): |
case('['): |
/* FALLTHROUGH */ |
|
case('{'): |
|
p->flags |= TERMP_NOSPACE; |
p->flags |= TERMP_NOSPACE; |
break; |
break; |
default: |
default: |