fmt(1): reformat with indent(1).
Failed attempt to get nearer to style(9) and the format from the original OpenBSD code. At least it should be readable now. No functional change.
This commit is contained in:
parent
6dbd80fa74
commit
86aba0f293
@ -1,4 +1,4 @@
|
||||
/* $OpenBSD: fmt.c,v 1.16 2000/06/25 15:35:42 pjanzen Exp $ */
|
||||
/* $OpenBSD: fmt.c,v 1.21 2004/04/01 23:14:19 tedu Exp $ */
|
||||
|
||||
/* Sensible version of fmt
|
||||
*
|
||||
@ -170,8 +170,9 @@
|
||||
|
||||
#ifndef lint
|
||||
static const char copyright[] =
|
||||
"Copyright (c) 1997 Gareth McCaughan. All rights reserved.\n";
|
||||
#endif /* not lint */
|
||||
"Copyright (c) 1997 Gareth McCaughan. All rights reserved.\n";
|
||||
|
||||
#endif /* not lint */
|
||||
#include <sys/cdefs.h>
|
||||
__FBSDID("$FreeBSD$");
|
||||
|
||||
@ -198,57 +199,74 @@ __FBSDID("$FreeBSD$");
|
||||
* (returning 0 instead), but we do complain about bad numbers.
|
||||
*/
|
||||
static size_t
|
||||
get_positive(const char *s, const char *err_mess, int fussyP) {
|
||||
char *t;
|
||||
long result = strtol(s,&t,0);
|
||||
if (*t) { if (fussyP) goto Lose; else return 0; }
|
||||
if (result<=0) { Lose: errx(EX_USAGE, "%s", err_mess); }
|
||||
return (size_t) result;
|
||||
get_positive(const char *s, const char *err_mess, int fussyP)
|
||||
{
|
||||
char *t;
|
||||
long result = strtol(s, &t, 0);
|
||||
|
||||
if (*t) {
|
||||
if (fussyP)
|
||||
goto Lose;
|
||||
else
|
||||
return 0;
|
||||
}
|
||||
if (result <= 0) {
|
||||
Lose: errx(EX_USAGE, "%s", err_mess);
|
||||
}
|
||||
return (size_t)result;
|
||||
}
|
||||
|
||||
static size_t
|
||||
get_nonnegative(const char *s, const char *err_mess, int fussyP) {
|
||||
char *t;
|
||||
long result = strtol(s,&t,0);
|
||||
if (*t) { if (fussyP) goto Lose; else return 0; }
|
||||
if (result<0) { Lose: errx(EX_USAGE, "%s", err_mess); }
|
||||
return (size_t) result;
|
||||
get_nonnegative(const char *s, const char *err_mess, int fussyP)
|
||||
{
|
||||
char *t;
|
||||
long result = strtol(s, &t, 0);
|
||||
|
||||
if (*t) {
|
||||
if (fussyP)
|
||||
goto Lose;
|
||||
else
|
||||
return 0;
|
||||
}
|
||||
if (result < 0) {
|
||||
Lose: errx(EX_USAGE, "%s", err_mess);
|
||||
}
|
||||
return (size_t)result;
|
||||
}
|
||||
|
||||
/* Global variables */
|
||||
|
||||
static int centerP=0; /* Try to center lines? */
|
||||
static size_t goal_length=0; /* Target length for output lines */
|
||||
static size_t max_length=0; /* Maximum length for output lines */
|
||||
static int coalesce_spaces_P=0; /* Coalesce multiple whitespace -> ' ' ? */
|
||||
static int allow_indented_paragraphs=0; /* Can first line have diff. ind.? */
|
||||
static int tab_width=8; /* Number of spaces per tab stop */
|
||||
static size_t output_tab_width=8; /* Ditto, when squashing leading spaces */
|
||||
static const wchar_t *sentence_enders=L".?!"; /* Double-space after these */
|
||||
static int grok_mail_headers=0; /* treat embedded mail headers magically? */
|
||||
static int format_troff=0; /* Format troff? */
|
||||
static int centerP = 0; /* Try to center lines? */
|
||||
static size_t goal_length = 0; /* Target length for output lines */
|
||||
static size_t max_length = 0; /* Maximum length for output lines */
|
||||
static int coalesce_spaces_P = 0; /* Coalesce multiple whitespace -> ' ' ? */
|
||||
static int allow_indented_paragraphs = 0; /* Can first line have diff. ind.? */
|
||||
static int tab_width = 8; /* Number of spaces per tab stop */
|
||||
static size_t output_tab_width = 8; /* Ditto, when squashing leading spaces */
|
||||
static const wchar_t *sentence_enders = L".?!"; /* Double-space after these */
|
||||
static int grok_mail_headers = 0; /* treat embedded mail headers magically? */
|
||||
static int format_troff = 0; /* Format troff? */
|
||||
|
||||
static int n_errors=0; /* Number of failed files. Return on exit. */
|
||||
static wchar_t *output_buffer=0; /* Output line will be built here */
|
||||
static size_t x; /* Horizontal position in output line */
|
||||
static size_t x0; /* Ditto, ignoring leading whitespace */
|
||||
static int n_errors = 0; /* Number of failed files. Return on exit. */
|
||||
static wchar_t *output_buffer = 0; /* Output line will be built here */
|
||||
static size_t x; /* Horizontal position in output line */
|
||||
static size_t x0; /* Ditto, ignoring leading whitespace */
|
||||
static size_t output_buffer_length = 0;
|
||||
static size_t pending_spaces; /* Spaces to add before next word */
|
||||
static int output_in_paragraph=0; /* Any of current para written out yet? */
|
||||
static size_t pending_spaces; /* Spaces to add before next word */
|
||||
static int output_in_paragraph = 0; /* Any of current para written out yet? */
|
||||
|
||||
/* Prototypes */
|
||||
|
||||
static void process_named_file (const char *);
|
||||
static void process_stream (FILE *, const char *);
|
||||
static size_t indent_length (const wchar_t *, size_t);
|
||||
static int might_be_header (const wchar_t *);
|
||||
static void new_paragraph (size_t, size_t);
|
||||
static void output_word (size_t, size_t, const wchar_t *, size_t,
|
||||
size_t);
|
||||
static void output_indent (size_t);
|
||||
static void center_stream (FILE *, const char *);
|
||||
static wchar_t * get_line (FILE *, size_t *);
|
||||
static void * xrealloc (void *, size_t);
|
||||
static void process_named_file(const char *);
|
||||
static void process_stream(FILE *, const char *);
|
||||
static size_t indent_length(const wchar_t *, size_t);
|
||||
static int might_be_header(const wchar_t *);
|
||||
static void new_paragraph(size_t, size_t);
|
||||
static void output_word(size_t, size_t, const wchar_t *, size_t, size_t);
|
||||
static void output_indent(size_t);
|
||||
static void center_stream(FILE *, const char *);
|
||||
static wchar_t *get_line(FILE *, size_t *);
|
||||
static void *xrealloc(void *, size_t);
|
||||
|
||||
#define XMALLOC(x) xrealloc(0,x)
|
||||
|
||||
@ -256,234 +274,277 @@ static void * xrealloc (void *, size_t);
|
||||
* all in top-down order. Hence, |main| comes first.
|
||||
*/
|
||||
int
|
||||
main(int argc, char *argv[]) {
|
||||
int ch; /* used for |getopt| processing */
|
||||
wchar_t *tmp;
|
||||
size_t len;
|
||||
const char *src;
|
||||
main(int argc, char *argv[])
|
||||
{
|
||||
int ch; /* used for |getopt| processing */
|
||||
wchar_t *tmp;
|
||||
size_t len;
|
||||
const char *src;
|
||||
|
||||
(void) setlocale(LC_CTYPE, "");
|
||||
(void)setlocale(LC_CTYPE, "");
|
||||
|
||||
/* 1. Grok parameters. */
|
||||
/* 1. Grok parameters. */
|
||||
|
||||
while ((ch = getopt(argc, argv, "0123456789cd:hl:mnpst:w:")) != -1)
|
||||
switch(ch) {
|
||||
case 'c':
|
||||
centerP = 1;
|
||||
format_troff = 1;
|
||||
continue;
|
||||
case 'd':
|
||||
src = optarg;
|
||||
len = mbsrtowcs(NULL, &src, 0, NULL);
|
||||
if (len == (size_t)-1)
|
||||
err(EX_USAGE, "bad sentence-ending character set");
|
||||
tmp = XMALLOC((len + 1) * sizeof(wchar_t));
|
||||
mbsrtowcs(tmp, &src, len + 1, NULL);
|
||||
sentence_enders = tmp;
|
||||
continue;
|
||||
case 'l':
|
||||
output_tab_width
|
||||
= get_nonnegative(optarg, "output tab width must be non-negative", 1);
|
||||
continue;
|
||||
case 'm':
|
||||
grok_mail_headers = 1;
|
||||
continue;
|
||||
case 'n':
|
||||
format_troff = 1;
|
||||
continue;
|
||||
case 'p':
|
||||
allow_indented_paragraphs = 1;
|
||||
continue;
|
||||
case 's':
|
||||
coalesce_spaces_P = 1;
|
||||
continue;
|
||||
case 't':
|
||||
tab_width = get_positive(optarg, "tab width must be positive", 1);
|
||||
continue;
|
||||
case 'w':
|
||||
goal_length = get_positive(optarg, "width must be positive", 1);
|
||||
max_length = goal_length;
|
||||
continue;
|
||||
case '0': case '1': case '2': case '3': case '4': case '5':
|
||||
case '6': case '7': case '8': case '9':
|
||||
/* XXX this is not a stylistically approved use of getopt() */
|
||||
if (goal_length==0) {
|
||||
char *p;
|
||||
p = argv[optind - 1];
|
||||
if (p[0] == '-' && p[1] == ch && !p[2])
|
||||
goal_length = get_positive(++p, "width must be nonzero", 1);
|
||||
else
|
||||
goal_length = get_positive(argv[optind]+1,
|
||||
"width must be nonzero", 1);
|
||||
max_length = goal_length;
|
||||
}
|
||||
continue;
|
||||
case 'h': default:
|
||||
fprintf(stderr,
|
||||
"usage: fmt [-cmps] [-d chars] [-l num] [-t num]\n"
|
||||
" [-w width | -width | goal [maximum]] [file ...]\n"
|
||||
"Options: -c center each line instead of formatting\n"
|
||||
" -d <chars> double-space after <chars> at line end\n"
|
||||
" -l <n> turn each <n> spaces at start of line into a tab\n"
|
||||
" -m try to make sure mail header lines stay separate\n"
|
||||
" -n format lines beginning with a dot\n"
|
||||
" -p allow indented paragraphs\n"
|
||||
" -s coalesce whitespace inside lines\n"
|
||||
" -t <n> have tabs every <n> columns\n"
|
||||
" -w <n> set maximum width to <n>\n"
|
||||
" goal set target width to goal\n");
|
||||
exit(ch=='h' ? 0 : EX_USAGE);
|
||||
}
|
||||
argc -= optind; argv += optind;
|
||||
while ((ch = getopt(argc, argv, "0123456789cd:hl:mnpst:w:")) != -1)
|
||||
switch (ch) {
|
||||
case 'c':
|
||||
centerP = 1;
|
||||
format_troff = 1;
|
||||
continue;
|
||||
case 'd':
|
||||
src = optarg;
|
||||
len = mbsrtowcs(NULL, &src, 0, NULL);
|
||||
if (len == (size_t)-1)
|
||||
err(EX_USAGE, "bad sentence-ending character set");
|
||||
tmp = XMALLOC((len + 1) * sizeof(wchar_t));
|
||||
mbsrtowcs(tmp, &src, len + 1, NULL);
|
||||
sentence_enders = tmp;
|
||||
continue;
|
||||
case 'l':
|
||||
output_tab_width
|
||||
= get_nonnegative(optarg, "output tab width must be non-negative", 1);
|
||||
continue;
|
||||
case 'm':
|
||||
grok_mail_headers = 1;
|
||||
continue;
|
||||
case 'n':
|
||||
format_troff = 1;
|
||||
continue;
|
||||
case 'p':
|
||||
allow_indented_paragraphs = 1;
|
||||
continue;
|
||||
case 's':
|
||||
coalesce_spaces_P = 1;
|
||||
continue;
|
||||
case 't':
|
||||
tab_width = get_positive(optarg, "tab width must be positive", 1);
|
||||
continue;
|
||||
case 'w':
|
||||
goal_length = get_positive(optarg, "width must be positive", 1);
|
||||
max_length = goal_length;
|
||||
continue;
|
||||
case '0': case '1': case '2': case '3': case '4': case '5':
|
||||
case '6': case '7': case '8': case '9':
|
||||
/*
|
||||
* XXX this is not a stylistically approved use of
|
||||
* getopt()
|
||||
*/
|
||||
if (goal_length == 0) {
|
||||
char *p;
|
||||
|
||||
/* [ goal [ maximum ] ] */
|
||||
p = argv[optind - 1];
|
||||
if (p[0] == '-' && p[1] == ch && !p[2])
|
||||
goal_length = get_positive(++p, "width must be nonzero", 1);
|
||||
else
|
||||
goal_length = get_positive(argv[optind] + 1,
|
||||
"width must be nonzero", 1);
|
||||
max_length = goal_length;
|
||||
}
|
||||
continue;
|
||||
case 'h':
|
||||
default:
|
||||
fprintf(stderr,
|
||||
"usage: fmt [-cmps] [-d chars] [-l num] [-t num]\n"
|
||||
" [-w width | -width | goal [maximum]] [file ...]\n"
|
||||
"Options: -c center each line instead of formatting\n"
|
||||
" -d <chars> double-space after <chars> at line end\n"
|
||||
" -l <n> turn each <n> spaces at start of line into a tab\n"
|
||||
" -m try to make sure mail header lines stay separate\n"
|
||||
" -n format lines beginning with a dot\n"
|
||||
" -p allow indented paragraphs\n"
|
||||
" -s coalesce whitespace inside lines\n"
|
||||
" -t <n> have tabs every <n> columns\n"
|
||||
" -w <n> set maximum width to <n>\n"
|
||||
" goal set target width to goal\n");
|
||||
exit(ch == 'h' ? 0 : EX_USAGE);
|
||||
}
|
||||
argc -= optind;
|
||||
argv += optind;
|
||||
|
||||
if (argc>0 && goal_length==0
|
||||
&& (goal_length=get_positive(*argv,"goal length must be positive", 0))
|
||||
!= 0) {
|
||||
--argc; ++argv;
|
||||
if (argc>0
|
||||
&& (max_length=get_positive(*argv,"max length must be positive", 0))
|
||||
!= 0) {
|
||||
--argc; ++argv;
|
||||
if (max_length<goal_length)
|
||||
errx(EX_USAGE, "max length must be >= goal length");
|
||||
}
|
||||
}
|
||||
if (goal_length==0) goal_length = 65;
|
||||
if (max_length==0) max_length = goal_length+10;
|
||||
if (max_length >= SIZE_T_MAX / sizeof (wchar_t)) errx(EX_USAGE, "max length too large");
|
||||
/* really needn't be longer */
|
||||
output_buffer = XMALLOC((max_length+1) * sizeof(wchar_t));
|
||||
/* [ goal [ maximum ] ] */
|
||||
|
||||
/* 2. Process files. */
|
||||
if (argc > 0 && goal_length == 0
|
||||
&& (goal_length = get_positive(*argv, "goal length must be positive", 0))
|
||||
!= 0) {
|
||||
--argc;
|
||||
++argv;
|
||||
if (argc > 0
|
||||
&& (max_length = get_positive(*argv, "max length must be positive", 0))
|
||||
!= 0) {
|
||||
--argc;
|
||||
++argv;
|
||||
if (max_length < goal_length)
|
||||
errx(EX_USAGE, "max length must be >= goal length");
|
||||
}
|
||||
}
|
||||
if (goal_length == 0)
|
||||
goal_length = 65;
|
||||
if (max_length == 0)
|
||||
max_length = goal_length + 10;
|
||||
if (max_length >= SIZE_T_MAX / sizeof(wchar_t))
|
||||
errx(EX_USAGE, "max length too large");
|
||||
/* really needn't be longer */
|
||||
output_buffer = XMALLOC((max_length + 1) * sizeof(wchar_t));
|
||||
|
||||
if (argc>0) {
|
||||
while (argc-->0) process_named_file(*argv++);
|
||||
}
|
||||
else {
|
||||
process_stream(stdin, "standard input");
|
||||
}
|
||||
/* 2. Process files. */
|
||||
|
||||
/* We're done. */
|
||||
if (argc > 0) {
|
||||
while (argc-- > 0)
|
||||
process_named_file(*argv++);
|
||||
} else {
|
||||
process_stream(stdin, "standard input");
|
||||
}
|
||||
|
||||
return n_errors ? EX_NOINPUT : 0;
|
||||
/* We're done. */
|
||||
|
||||
return n_errors ? EX_NOINPUT : 0;
|
||||
|
||||
}
|
||||
|
||||
/* Process a single file, given its name.
|
||||
*/
|
||||
static void
|
||||
process_named_file(const char *name) {
|
||||
FILE *f=fopen(name, "r");
|
||||
if (!f) { warn("%s", name); ++n_errors; }
|
||||
else {
|
||||
process_stream(f, name);
|
||||
if (ferror(f)) { warn("%s", name); ++n_errors; }
|
||||
fclose(f);
|
||||
}
|
||||
process_named_file(const char *name)
|
||||
{
|
||||
FILE *f = fopen(name, "r");
|
||||
|
||||
if (!f) {
|
||||
warn("%s", name);
|
||||
++n_errors;
|
||||
} else {
|
||||
process_stream(f, name);
|
||||
if (ferror(f)) {
|
||||
warn("%s", name);
|
||||
++n_errors;
|
||||
}
|
||||
fclose(f);
|
||||
}
|
||||
}
|
||||
|
||||
/* Types of mail header continuation lines:
|
||||
*/
|
||||
typedef enum {
|
||||
hdr_ParagraphStart = -1,
|
||||
hdr_NonHeader = 0,
|
||||
hdr_Header = 1,
|
||||
hdr_Continuation = 2
|
||||
hdr_ParagraphStart = -1,
|
||||
hdr_NonHeader = 0,
|
||||
hdr_Header = 1,
|
||||
hdr_Continuation = 2
|
||||
} HdrType;
|
||||
|
||||
/* Process a stream. This is where the real work happens,
|
||||
* except that centering is handled separately.
|
||||
*/
|
||||
static void
|
||||
process_stream(FILE *stream, const char *name) {
|
||||
size_t last_indent=SILLY; /* how many spaces in last indent? */
|
||||
size_t para_line_number=0; /* how many lines already read in this para? */
|
||||
size_t first_indent=SILLY; /* indentation of line 0 of paragraph */
|
||||
HdrType prev_header_type=hdr_ParagraphStart;
|
||||
process_stream(FILE *stream, const char *name)
|
||||
{
|
||||
size_t last_indent = SILLY; /* how many spaces in last indent? */
|
||||
size_t para_line_number = 0; /* how many lines already read in this para? */
|
||||
size_t first_indent = SILLY; /* indentation of line 0 of paragraph */
|
||||
HdrType prev_header_type = hdr_ParagraphStart;
|
||||
|
||||
/* ^-- header_type of previous line; -1 at para start */
|
||||
wchar_t *line;
|
||||
size_t length;
|
||||
wchar_t *line;
|
||||
size_t length;
|
||||
|
||||
if (centerP) { center_stream(stream, name); return; }
|
||||
while ((line=get_line(stream,&length)) != NULL) {
|
||||
size_t np=indent_length(line, length);
|
||||
{ HdrType header_type=hdr_NonHeader;
|
||||
if (grok_mail_headers && prev_header_type!=hdr_NonHeader) {
|
||||
if (np==0 && might_be_header(line))
|
||||
header_type = hdr_Header;
|
||||
else if (np>0 && prev_header_type>hdr_NonHeader)
|
||||
header_type = hdr_Continuation;
|
||||
}
|
||||
/* We need a new paragraph if and only if:
|
||||
* this line is blank,
|
||||
* OR it's a troff request (and we don't format troff),
|
||||
* OR it's a mail header,
|
||||
* OR it's not a mail header AND the last line was one,
|
||||
* OR the indentation has changed
|
||||
* AND the line isn't a mail header continuation line
|
||||
* AND this isn't the second line of an indented paragraph.
|
||||
*/
|
||||
if ( length==0
|
||||
|| (line[0]=='.' && !format_troff)
|
||||
|| header_type==hdr_Header
|
||||
|| (header_type==hdr_NonHeader && prev_header_type>hdr_NonHeader)
|
||||
|| (np!=last_indent
|
||||
&& header_type != hdr_Continuation
|
||||
&& (!allow_indented_paragraphs || para_line_number != 1)) ) {
|
||||
new_paragraph(output_in_paragraph ? last_indent : first_indent, np);
|
||||
para_line_number = 0;
|
||||
first_indent = np;
|
||||
last_indent = np;
|
||||
if (header_type==hdr_Header) last_indent=2; /* for cont. lines */
|
||||
if (length==0 || (line[0]=='.' && !format_troff)) {
|
||||
if (length==0)
|
||||
putwchar('\n');
|
||||
else
|
||||
wprintf(L"%.*ls\n", (int)length, line);
|
||||
prev_header_type=hdr_ParagraphStart;
|
||||
continue;
|
||||
}
|
||||
}
|
||||
else {
|
||||
/* If this is an indented paragraph other than a mail header
|
||||
* continuation, set |last_indent|.
|
||||
*/
|
||||
if (np != last_indent && header_type != hdr_Continuation)
|
||||
last_indent=np;
|
||||
}
|
||||
prev_header_type = header_type;
|
||||
}
|
||||
if (centerP) {
|
||||
center_stream(stream, name);
|
||||
return;
|
||||
}
|
||||
while ((line = get_line(stream, &length)) != NULL) {
|
||||
size_t np = indent_length(line, length);
|
||||
|
||||
{ size_t n=np;
|
||||
while (n<length) {
|
||||
/* Find word end and count spaces after it */
|
||||
size_t word_length=0, space_length=0;
|
||||
while (n+word_length < length && line[n+word_length] != ' ')
|
||||
++word_length;
|
||||
space_length = word_length;
|
||||
while (n+space_length < length && line[n+space_length] == ' ')
|
||||
++space_length;
|
||||
/* Send the word to the output machinery. */
|
||||
output_word(first_indent, last_indent,
|
||||
line+n, word_length, space_length-word_length);
|
||||
n += space_length;
|
||||
}
|
||||
}
|
||||
++para_line_number;
|
||||
}
|
||||
new_paragraph(output_in_paragraph ? last_indent : first_indent, 0);
|
||||
if (ferror(stream)) { warn("%s", name); ++n_errors; }
|
||||
{
|
||||
HdrType header_type = hdr_NonHeader;
|
||||
|
||||
if (grok_mail_headers && prev_header_type != hdr_NonHeader) {
|
||||
if (np == 0 && might_be_header(line))
|
||||
header_type = hdr_Header;
|
||||
else if (np > 0 && prev_header_type > hdr_NonHeader)
|
||||
header_type = hdr_Continuation;
|
||||
}
|
||||
/*
|
||||
* We need a new paragraph if and only if: this line
|
||||
* is blank, OR it's a troff request (and we don't
|
||||
* format troff), OR it's a mail header, OR it's not
|
||||
* a mail header AND the last line was one, OR the
|
||||
* indentation has changed AND the line isn't a mail
|
||||
* header continuation line AND this isn't the
|
||||
* second line of an indented paragraph.
|
||||
*/
|
||||
if (length == 0
|
||||
|| (line[0] == '.' && !format_troff)
|
||||
|| header_type == hdr_Header
|
||||
|| (header_type == hdr_NonHeader && prev_header_type > hdr_NonHeader)
|
||||
|| (np != last_indent
|
||||
&& header_type != hdr_Continuation
|
||||
&& (!allow_indented_paragraphs || para_line_number != 1))) {
|
||||
new_paragraph(output_in_paragraph ? last_indent : first_indent, np);
|
||||
para_line_number = 0;
|
||||
first_indent = np;
|
||||
last_indent = np;
|
||||
if (header_type == hdr_Header)
|
||||
last_indent = 2; /* for cont. lines */
|
||||
if (length == 0 || (line[0] == '.' && !format_troff)) {
|
||||
if (length == 0)
|
||||
putwchar('\n');
|
||||
else
|
||||
wprintf(L"%.*ls\n", (int)length,
|
||||
line);
|
||||
prev_header_type = hdr_ParagraphStart;
|
||||
continue;
|
||||
}
|
||||
} else {
|
||||
/*
|
||||
* If this is an indented paragraph other
|
||||
* than a mail header continuation, set
|
||||
* |last_indent|.
|
||||
*/
|
||||
if (np != last_indent &&
|
||||
header_type != hdr_Continuation)
|
||||
last_indent = np;
|
||||
}
|
||||
prev_header_type = header_type;
|
||||
}
|
||||
|
||||
{
|
||||
size_t n = np;
|
||||
|
||||
while (n < length) {
|
||||
/* Find word end and count spaces after it */
|
||||
size_t word_length = 0, space_length = 0;
|
||||
|
||||
while (n + word_length < length &&
|
||||
line[n + word_length] != ' ')
|
||||
++word_length;
|
||||
space_length = word_length;
|
||||
while (n + space_length < length &&
|
||||
line[n + space_length] == ' ')
|
||||
++space_length;
|
||||
/* Send the word to the output machinery. */
|
||||
output_word(first_indent, last_indent,
|
||||
line + n, word_length,
|
||||
space_length - word_length);
|
||||
n += space_length;
|
||||
}
|
||||
}
|
||||
++para_line_number;
|
||||
}
|
||||
new_paragraph(output_in_paragraph ? last_indent : first_indent, 0);
|
||||
if (ferror(stream)) {
|
||||
warn("%s", name);
|
||||
++n_errors;
|
||||
}
|
||||
}
|
||||
|
||||
/* How long is the indent on this line?
|
||||
*/
|
||||
static size_t
|
||||
indent_length(const wchar_t *line, size_t length) {
|
||||
size_t n=0;
|
||||
while (n<length && *line++ == ' ') ++n;
|
||||
return n;
|
||||
indent_length(const wchar_t *line, size_t length)
|
||||
{
|
||||
size_t n = 0;
|
||||
|
||||
while (n < length && *line++ == ' ')
|
||||
++n;
|
||||
return n;
|
||||
}
|
||||
|
||||
/* Might this line be a mail header?
|
||||
@ -493,35 +554,45 @@ indent_length(const wchar_t *line, size_t length) {
|
||||
* conservative to avoid mangling ordinary civilised text.
|
||||
*/
|
||||
static int
|
||||
might_be_header(const wchar_t *line) {
|
||||
if (!iswupper(*line++)) return 0;
|
||||
while (*line && (iswalnum(*line) || *line=='-')) ++line;
|
||||
return (*line==':' && iswspace(line[1]));
|
||||
might_be_header(const wchar_t *line)
|
||||
{
|
||||
if (!iswupper(*line++))
|
||||
return 0;
|
||||
while (*line && (iswalnum(*line) || *line == '-'))
|
||||
++line;
|
||||
return (*line == ':' && iswspace(line[1]));
|
||||
}
|
||||
|
||||
/* Begin a new paragraph with an indent of |indent| spaces.
|
||||
*/
|
||||
static void
|
||||
new_paragraph(size_t old_indent, size_t indent) {
|
||||
if (output_buffer_length) {
|
||||
if (old_indent>0) output_indent(old_indent);
|
||||
wprintf(L"%.*ls\n", (int)output_buffer_length, output_buffer);
|
||||
}
|
||||
x=indent; x0=0; output_buffer_length=0; pending_spaces=0;
|
||||
output_in_paragraph = 0;
|
||||
new_paragraph(size_t old_indent, size_t indent)
|
||||
{
|
||||
if (output_buffer_length) {
|
||||
if (old_indent > 0)
|
||||
output_indent(old_indent);
|
||||
wprintf(L"%.*ls\n", (int)output_buffer_length, output_buffer);
|
||||
}
|
||||
x = indent;
|
||||
x0 = 0;
|
||||
output_buffer_length = 0;
|
||||
pending_spaces = 0;
|
||||
output_in_paragraph = 0;
|
||||
}
|
||||
|
||||
/* Output spaces or tabs for leading indentation.
|
||||
*/
|
||||
static void
|
||||
output_indent(size_t n_spaces) {
|
||||
if (output_tab_width) {
|
||||
while (n_spaces >= output_tab_width) {
|
||||
putwchar('\t');
|
||||
n_spaces -= output_tab_width;
|
||||
}
|
||||
}
|
||||
while (n_spaces-- > 0) putwchar(' ');
|
||||
output_indent(size_t n_spaces)
|
||||
{
|
||||
if (output_tab_width) {
|
||||
while (n_spaces >= output_tab_width) {
|
||||
putwchar('\t');
|
||||
n_spaces -= output_tab_width;
|
||||
}
|
||||
}
|
||||
while (n_spaces-- > 0)
|
||||
putwchar(' ');
|
||||
}
|
||||
|
||||
/* Output a single word, or add it to the buffer.
|
||||
@ -529,93 +600,120 @@ output_indent(size_t n_spaces) {
|
||||
* lines of a paragraph. They'll often be the same, of course.
|
||||
*/
|
||||
static void
|
||||
output_word(size_t indent0, size_t indent1, const wchar_t *word, size_t length, size_t spaces) {
|
||||
size_t new_x;
|
||||
size_t indent = output_in_paragraph ? indent1 : indent0;
|
||||
size_t width;
|
||||
const wchar_t *p;
|
||||
int cwidth;
|
||||
output_word(size_t indent0, size_t indent1, const wchar_t *word, size_t length, size_t spaces)
|
||||
{
|
||||
size_t new_x;
|
||||
size_t indent = output_in_paragraph ? indent1 : indent0;
|
||||
size_t width;
|
||||
const wchar_t *p;
|
||||
int cwidth;
|
||||
|
||||
for (p = word, width = 0; p < &word[length]; p++)
|
||||
width += (cwidth = wcwidth(*p)) > 0 ? cwidth : 1;
|
||||
for (p = word, width = 0; p < &word[length]; p++)
|
||||
width += (cwidth = wcwidth(*p)) > 0 ? cwidth : 1;
|
||||
|
||||
new_x = x + pending_spaces + width;
|
||||
new_x = x + pending_spaces + width;
|
||||
|
||||
/* If either |spaces==0| (at end of line) or |coalesce_spaces_P|
|
||||
* (squashing internal whitespace), then add just one space;
|
||||
* except that if the last character was a sentence-ender we
|
||||
* actually add two spaces.
|
||||
*/
|
||||
if (coalesce_spaces_P || spaces==0)
|
||||
spaces = wcschr(sentence_enders, word[length-1]) ? 2 : 1;
|
||||
/*
|
||||
* If either |spaces==0| (at end of line) or |coalesce_spaces_P|
|
||||
* (squashing internal whitespace), then add just one space; except
|
||||
* that if the last character was a sentence-ender we actually add
|
||||
* two spaces.
|
||||
*/
|
||||
if (coalesce_spaces_P || spaces == 0)
|
||||
spaces = wcschr(sentence_enders, word[length - 1]) ? 2 : 1;
|
||||
|
||||
if (new_x<=goal_length) {
|
||||
/* After adding the word we still aren't at the goal length,
|
||||
* so clearly we add it to the buffer rather than outputing it.
|
||||
*/
|
||||
wmemset(output_buffer+output_buffer_length, L' ', pending_spaces);
|
||||
x0 += pending_spaces; x += pending_spaces;
|
||||
output_buffer_length += pending_spaces;
|
||||
wmemcpy(output_buffer+output_buffer_length, word, length);
|
||||
x0 += width; x += width; output_buffer_length += length;
|
||||
pending_spaces = spaces;
|
||||
}
|
||||
else {
|
||||
/* Adding the word takes us past the goal. Print the line-so-far,
|
||||
* and the word too iff either (1) the lsf is empty or (2) that
|
||||
* makes us nearer the goal but doesn't take us over the limit,
|
||||
* or (3) the word on its own takes us over the limit.
|
||||
* In case (3) we put a newline in between.
|
||||
*/
|
||||
if (indent>0) output_indent(indent);
|
||||
wprintf(L"%.*ls", (int)output_buffer_length, output_buffer);
|
||||
if (x0==0 || (new_x <= max_length && new_x-goal_length <= goal_length-x)) {
|
||||
wprintf(L"%*ls", (int)pending_spaces, L"");
|
||||
goto write_out_word;
|
||||
}
|
||||
else {
|
||||
/* If the word takes us over the limit on its own, just
|
||||
* spit it out and don't bother buffering it.
|
||||
*/
|
||||
if (indent+width > max_length) {
|
||||
putwchar('\n');
|
||||
if (indent>0) output_indent(indent);
|
||||
write_out_word:
|
||||
wprintf(L"%.*ls", (int)length, word);
|
||||
x0 = 0; x = indent1; pending_spaces = 0;
|
||||
output_buffer_length = 0;
|
||||
}
|
||||
else {
|
||||
wmemcpy(output_buffer, word, length);
|
||||
x0 = width; x = width+indent1; pending_spaces = spaces;
|
||||
output_buffer_length = length;
|
||||
}
|
||||
}
|
||||
putwchar('\n');
|
||||
output_in_paragraph = 1;
|
||||
}
|
||||
if (new_x <= goal_length) {
|
||||
/*
|
||||
* After adding the word we still aren't at the goal length,
|
||||
* so clearly we add it to the buffer rather than outputing
|
||||
* it.
|
||||
*/
|
||||
wmemset(output_buffer + output_buffer_length, L' ',
|
||||
pending_spaces);
|
||||
x0 += pending_spaces;
|
||||
x += pending_spaces;
|
||||
output_buffer_length += pending_spaces;
|
||||
wmemcpy(output_buffer + output_buffer_length, word, length);
|
||||
x0 += width;
|
||||
x += width;
|
||||
output_buffer_length += length;
|
||||
pending_spaces = spaces;
|
||||
} else {
|
||||
/*
|
||||
* Adding the word takes us past the goal. Print the
|
||||
* line-so-far, and the word too iff either (1) the lsf is
|
||||
* empty or (2) that makes us nearer the goal but doesn't
|
||||
* take us over the limit, or (3) the word on its own takes
|
||||
* us over the limit. In case (3) we put a newline in
|
||||
* between.
|
||||
*/
|
||||
if (indent > 0)
|
||||
output_indent(indent);
|
||||
wprintf(L"%.*ls", (int)output_buffer_length, output_buffer);
|
||||
if (x0 == 0 || (new_x <= max_length &&
|
||||
new_x - goal_length <= goal_length - x)) {
|
||||
wprintf(L"%*ls", (int)pending_spaces, L"");
|
||||
goto write_out_word;
|
||||
} else {
|
||||
/*
|
||||
* If the word takes us over the limit on its own,
|
||||
* just spit it out and don't bother buffering it.
|
||||
*/
|
||||
if (indent + width > max_length) {
|
||||
putwchar('\n');
|
||||
if (indent > 0)
|
||||
output_indent(indent);
|
||||
write_out_word:
|
||||
wprintf(L"%.*ls", (int)length, word);
|
||||
x0 = 0;
|
||||
x = indent1;
|
||||
pending_spaces = 0;
|
||||
output_buffer_length = 0;
|
||||
} else {
|
||||
wmemcpy(output_buffer, word, length);
|
||||
x0 = width;
|
||||
x = width + indent1;
|
||||
pending_spaces = spaces;
|
||||
output_buffer_length = length;
|
||||
}
|
||||
}
|
||||
putwchar('\n');
|
||||
output_in_paragraph = 1;
|
||||
}
|
||||
}
|
||||
|
||||
/* Process a stream, but just center its lines rather than trying to
|
||||
* format them neatly.
|
||||
*/
|
||||
static void
|
||||
center_stream(FILE *stream, const char *name) {
|
||||
wchar_t *line, *p;
|
||||
size_t length;
|
||||
size_t width;
|
||||
int cwidth;
|
||||
while ((line=get_line(stream, &length)) != 0) {
|
||||
size_t l=length;
|
||||
while (l>0 && iswspace(*line)) { ++line; --l; }
|
||||
length=l;
|
||||
for (p = line, width = 0; p < &line[length]; p++)
|
||||
width += (cwidth = wcwidth(*p)) > 0 ? cwidth : 1;
|
||||
l = width;
|
||||
while (l<goal_length) { putwchar(' '); l+=2; }
|
||||
wprintf(L"%.*ls\n", (int)length, line);
|
||||
}
|
||||
if (ferror(stream)) { warn("%s", name); ++n_errors; }
|
||||
center_stream(FILE *stream, const char *name)
|
||||
{
|
||||
wchar_t *line, *p;
|
||||
size_t length;
|
||||
size_t width;
|
||||
int cwidth;
|
||||
|
||||
while ((line = get_line(stream, &length)) != 0) {
|
||||
size_t l = length;
|
||||
|
||||
while (l > 0 && iswspace(*line)) {
|
||||
++line;
|
||||
--l;
|
||||
}
|
||||
length = l;
|
||||
for (p = line, width = 0; p < &line[length]; p++)
|
||||
width += (cwidth = wcwidth(*p)) > 0 ? cwidth : 1;
|
||||
l = width;
|
||||
while (l < goal_length) {
|
||||
putwchar(' ');
|
||||
l += 2;
|
||||
}
|
||||
wprintf(L"%.*ls\n", (int)length, line);
|
||||
}
|
||||
if (ferror(stream)) {
|
||||
warn("%s", name);
|
||||
++n_errors;
|
||||
}
|
||||
}
|
||||
|
||||
/* Get a single line from a stream. Expand tabs, strip control
|
||||
@ -630,41 +728,60 @@ center_stream(FILE *stream, const char *name) {
|
||||
* |pending_spaces|.
|
||||
*/
|
||||
static wchar_t *
|
||||
get_line(FILE *stream, size_t *lengthp) {
|
||||
static wchar_t *buf=NULL;
|
||||
static size_t length=0;
|
||||
size_t len=0;
|
||||
wint_t ch;
|
||||
size_t spaces_pending=0;
|
||||
int troff=0;
|
||||
size_t col=0;
|
||||
int cwidth;
|
||||
get_line(FILE *stream, size_t *lengthp)
|
||||
{
|
||||
static wchar_t *buf = NULL;
|
||||
static size_t length = 0;
|
||||
size_t len = 0;
|
||||
wint_t ch;
|
||||
size_t spaces_pending = 0;
|
||||
int troff = 0;
|
||||
size_t col = 0;
|
||||
int cwidth;
|
||||
|
||||
if (buf==NULL) { length=100; buf=XMALLOC(length * sizeof(wchar_t)); }
|
||||
while ((ch=getwc(stream)) != '\n' && ch != WEOF) {
|
||||
if (len+spaces_pending==0 && ch=='.' && !format_troff) troff=1;
|
||||
if (ch==' ') ++spaces_pending;
|
||||
else if (troff || iswprint(ch)) {
|
||||
while (len+spaces_pending >= length) {
|
||||
length*=2; buf=xrealloc(buf, length * sizeof(wchar_t));
|
||||
}
|
||||
while (spaces_pending > 0) { --spaces_pending; buf[len++]=' '; col++; }
|
||||
buf[len++] = ch;
|
||||
col += (cwidth = wcwidth(ch)) > 0 ? cwidth : 1;
|
||||
}
|
||||
else if (ch=='\t')
|
||||
spaces_pending += tab_width - (col+spaces_pending)%tab_width;
|
||||
else if (ch=='\b') { if (len) --len; if (col) --col; }
|
||||
}
|
||||
*lengthp=len;
|
||||
return (len>0 || ch!=WEOF) ? buf : 0;
|
||||
if (buf == NULL) {
|
||||
length = 100;
|
||||
buf = XMALLOC(length * sizeof(wchar_t));
|
||||
}
|
||||
while ((ch = getwc(stream)) != '\n' && ch != WEOF) {
|
||||
if (len + spaces_pending == 0 && ch == '.' && !format_troff)
|
||||
troff = 1;
|
||||
if (ch == ' ')
|
||||
++spaces_pending;
|
||||
else if (troff || iswprint(ch)) {
|
||||
while (len + spaces_pending >= length) {
|
||||
length *= 2;
|
||||
buf = xrealloc(buf, length * sizeof(wchar_t));
|
||||
}
|
||||
while (spaces_pending > 0) {
|
||||
--spaces_pending;
|
||||
buf[len++] = ' ';
|
||||
col++;
|
||||
}
|
||||
buf[len++] = ch;
|
||||
col += (cwidth = wcwidth(ch)) > 0 ? cwidth : 1;
|
||||
} else if (ch == '\t')
|
||||
spaces_pending += tab_width -
|
||||
(col + spaces_pending) % tab_width;
|
||||
else if (ch == '\b') {
|
||||
if (len)
|
||||
--len;
|
||||
if (col)
|
||||
--col;
|
||||
}
|
||||
}
|
||||
*lengthp = len;
|
||||
return (len > 0 || ch != WEOF) ? buf : 0;
|
||||
}
|
||||
|
||||
/* (Re)allocate some memory, exiting with an error if we can't.
|
||||
*/
|
||||
static void *
|
||||
xrealloc(void *ptr, size_t nbytes) {
|
||||
void *p = realloc(ptr, nbytes);
|
||||
if (p == NULL) errx(EX_OSERR, "out of memory");
|
||||
return p;
|
||||
xrealloc(void *ptr, size_t nbytes)
|
||||
{
|
||||
void *p = realloc(ptr, nbytes);
|
||||
|
||||
if (p == NULL)
|
||||
errx(EX_OSERR, "out of memory");
|
||||
return p;
|
||||
}
|
||||
|
Loading…
Reference in New Issue
Block a user