/*
* Copyright (C) 1984-2023 Mark Nudelman
*
* You may distribute under the terms of either the GNU General Public
* License or the Less License, as specified in the README file.
*
* For more information, see the README file.
*/
/*
* Functions to define the character set
* and do things specific to the character set.
*/
if (s != NULL)
{
while (s[0] != '\0')
{
struct wchar_range range;
wchar_range_get(&s, &range);
if (range.last == 0)
{
error("invalid hex number(s) in LESSUTFCHARDEF", NULL_PARG);
quit(QUIT_ERROR);
}
if (*s++ != ':')
{
error("missing colon in LESSUTFCHARDEF", NULL_PARG);
quit(QUIT_ERROR);
}
switch (*s++)
{
case 'b':
xbuf_add_data(&user_ubin_array, (unsigned char *) &range, sizeof(range));
break;
case 'c':
xbuf_add_data(&user_compose_array, (unsigned char *) &range, sizeof(range));
break;
case 'w':
xbuf_add_data(&user_wide_array, (unsigned char *) &range, sizeof(range));
xbuf_add_data(&user_prt_array, (unsigned char *) &range, sizeof(range));
break;
case 'p': case '.':
xbuf_add_data(&user_prt_array, (unsigned char *) &range, sizeof(range));
break;
case '\0':
s--;
break;
default:
/* Ignore unknown character attribute. */
break;
}
if (s[0] == ',') ++s;
}
}
wchar_range_table_set(&user_wide_table, &user_wide_array);
wchar_range_table_set(&user_ubin_table, &user_ubin_array);
wchar_range_table_set(&user_compose_table, &user_compose_array);
wchar_range_table_set(&user_prt_table, &user_prt_array);
}
/*
* Define a charset, given a description string.
* The string consists of 256 letters,
* one for each character in the charset.
* If the string is shorter than 256 letters, missing letters
* are taken to be identical to the last one.
* A decimal number followed by a letter is taken to be a
* repetition of the letter.
*
* Each letter is one of:
* . normal character
* b binary character
* c control character
*/
static void ichardef(char *s)
{
char *cp;
int n;
char v;
n = 0;
v = 0;
cp = chardef;
while (*s != '\0')
{
switch (*s++)
{
case '.':
v = 0;
break;
case 'c':
v = IS_CONTROL_CHAR;
break;
case 'b':
v = IS_BINARY_CHAR|IS_CONTROL_CHAR;
break;
case '0': case '1': case '2': case '3': case '4':
case '5': case '6': case '7': case '8': case '9':
if (ckd_mul(&n, n, 10) || ckd_add(&n, n, s[-1] - '0'))
goto invalid_chardef;
continue;
do
{
if (cp >= chardef + sizeof(chardef))
{
error("chardef longer than 256", NULL_PARG);
quit(QUIT_ERROR);
/*NOTREACHED*/
}
*cp++ = v;
} while (--n > 0);
n = 0;
}
while (cp < chardef + sizeof(chardef))
*cp++ = v;
}
/*
* Define a charset, given a charset name.
* The valid charset names are listed in the "charsets" array.
*/
static int icharset(char *name, int no_error)
{
struct charset *p;
struct cs_alias *a;
if (name == NULL || *name == '\0')
return (0);
/* First see if the name is an alias. */
for (a = cs_aliases; a->name != NULL; a++)
{
if (strcmp(name, a->name) == 0)
{
name = a->oname;
break;
}
}
#if HAVE_LOCALE
/*
* Define a charset, given a locale name.
*/
static void ilocale(void)
{
int c;
for (c = 0; c < (int) sizeof(chardef); c++)
{
if (isprint(c))
chardef[c] = 0;
else if (iscntrl(c))
chardef[c] = IS_CONTROL_CHAR;
else
chardef[c] = IS_BINARY_CHAR|IS_CONTROL_CHAR;
}
}
#endif
/*
* Define the printing format for control (or binary utf) chars.
*/
public void setfmt(char *s, char **fmtvarptr, int *attrptr, char *default_fmt, int for_printf)
{
if (s && utf_mode)
{
/* It would be too hard to account for width otherwise. */
char constant *t = s;
while (*t)
{
if (*t < ' ' || *t > '~')
{
s = default_fmt;
goto attr;
}
t++;
}
}
if (s == NULL || *s == '\0')
s = default_fmt;
else if (for_printf &&
((*s == '*' && (s[1] == '\0' || s[2] == '\0' || strchr(s + 2, 'n'))) ||
(*s != '*' && strchr(s, 'n'))))
/* %n is evil */
s = default_fmt;
/*
* Select the attributes if it starts with "*".
*/
attr:
if (*s == '*' && s[1] != '\0')
{
switch (s[1])
{
case 'd': *attrptr = AT_BOLD; break;
case 'k': *attrptr = AT_BLINK; break;
case 's': *attrptr = AT_STANDOUT; break;
case 'u': *attrptr = AT_UNDERLINE; break;
default: *attrptr = AT_NORMAL; break;
}
s += 2;
}
*fmtvarptr = s;
}
/*
*
*/
static void set_charset(void)
{
char *s;
#if MSDOS_COMPILER==WIN32C
/*
* If the Windows console is using UTF-8, we'll use it too.
*/
if (GetConsoleOutputCP() == CP_UTF8)
if (icharset("utf-8", 1))
return;
#endif
ichardef_utf(lgetenv("LESSUTFCHARDEF"));
/*
* See if environment variable LESSCHARSET is defined.
*/
s = lgetenv("LESSCHARSET");
if (icharset(s, 0))
return;
/*
* LESSCHARSET is not defined: try LESSCHARDEF.
*/
s = lgetenv("LESSCHARDEF");
if (!isnullenv(s))
{
ichardef(s);
return;
}
#if HAVE_LOCALE
#ifdef CODESET
/*
* Try using the codeset name as the charset name.
*/
s = nl_langinfo(CODESET);
if (icharset(s, 1))
return;
#endif
#endif
#if HAVE_STRSTR
/*
* Check whether LC_ALL, LC_CTYPE or LANG look like UTF-8 is used.
*/
if ((s = lgetenv("LC_ALL")) != NULL ||
(s = lgetenv("LC_CTYPE")) != NULL ||
(s = lgetenv("LANG")) != NULL)
{
if ( strstr(s, "UTF-8") != NULL || strstr(s, "utf-8") != NULL
|| strstr(s, "UTF8") != NULL || strstr(s, "utf8") != NULL)
if (icharset("utf-8", 1))
return;
}
#endif
#if HAVE_LOCALE
/*
* Get character definitions from locale functions,
* rather than from predefined charset entry.
*/
ilocale();
#else
#if MSDOS_COMPILER
/*
* Default to "dos".
*/
(void) icharset("dos", 1);
#else
/*
* Default to "latin1".
*/
(void) icharset("latin1", 1);
#endif
#endif
}
/*
* Initialize charset data structures.
*/
public void init_charset(void)
{
char *s;
#if HAVE_LOCALE
setlocale(LC_ALL, "");
#endif
set_charset();
s = lgetenv("LESSBINFMT");
setfmt(s, &binfmt, &binattr, "*s<%02X>", TRUE);
s = lgetenv("LESSUTFBINFMT");
setfmt(s, &utfbinfmt, &binattr, "<U+%04lX>", TRUE);
}
/*
* Is a given character a "binary" character?
*/
public int binary_char(LWCHAR c)
{
if (utf_mode)
return (is_ubin_char(c));
c &= 0377;
return (chardef[c] & IS_BINARY_CHAR);
}
/*
* Is a given character a "control" character?
*/
public int control_char(LWCHAR c)
{
c &= 0377;
return (chardef[c] & IS_CONTROL_CHAR);
}
/*
* Return the printable form of a character.
* For example, in the "ascii" charset '\3' is printed as "^C".
*/
public char * prchar(LWCHAR c)
{
/* {{ This buffer can be overrun if LESSBINFMT is a long string. }} */
static char buf[MAX_PRCHAR_LEN+1];
c &= 0377;
if ((c < 128 || !utf_mode) && !control_char(c))
SNPRINTF1(buf, sizeof(buf), "%c", (int) c);
else if (c == ESC)
strcpy(buf, "ESC");
#if IS_EBCDIC_HOST
else if (!binary_char(c) && c < 64)
SNPRINTF1(buf, sizeof(buf), "^%c",
/*
* This array roughly inverts CONTROL() #defined in less.h,
* and should be kept in sync with CONTROL() and IBM-1047.
*/
"@ABC.I.?...KLMNO"
"PQRS.JH.XY.."
"\\]^_"
"......W[.....EFG"
"..V....D....TU.Z"[c]);
#else
else if (c < 128 && !control_char(c ^ 0100))
SNPRINTF1(buf, sizeof(buf), "^%c", (int) (c ^ 0100));
#endif
else
SNPRINTF1(buf, sizeof(buf), binfmt, c);
return (buf);
}
/*
* Return the printable form of a UTF-8 character.
*/
public char * prutfchar(LWCHAR ch)
{
static char buf[MAX_PRCHAR_LEN+1];
/*
* Get the length of a UTF-8 character in bytes.
*/
public int utf_len(int ch)
{
if ((ch & 0x80) == 0)
return 1;
if ((ch & 0xE0) == 0xC0)
return 2;
if ((ch & 0xF0) == 0xE0)
return 3;
if ((ch & 0xF8) == 0xF0)
return 4;
if ((ch & 0xFC) == 0xF8)
return 5;
if ((ch & 0xFE) == 0xFC)
return 6;
/* Invalid UTF-8 encoding. */
return 1;
}
/*
* Does the parameter point to the lead byte of a well-formed UTF-8 character?
*/
public int is_utf8_well_formed(char *ss, int slen)
{
int i;
int len;
unsigned char *s = (unsigned char *) ss;
if (IS_UTF8_INVALID(s[0]))
return (0);
len = utf_len(s[0]);
if (len > slen)
return (0);
if (len == 1)
return (1);
if (len == 2)
{
if (s[0] < 0xC2)
return (0);
} else
{
unsigned char mask;
mask = (~((1 << (8-len)) - 1)) & 0xFF;
if (s[0] == mask && (s[1] & mask) == 0x80)
return (0);
}
for (i = 1; i < len; i++)
if (!IS_UTF8_TRAIL(s[i]))
return (0);
return (1);
}
/*
* Skip bytes until a UTF-8 lead byte (11xxxxxx) or ASCII byte (0xxxxxxx) is found.
*/
public void utf_skip_to_lead(char **pp, char *limit)
{
do {
++(*pp);
} while (*pp < limit && !IS_UTF8_LEAD((*pp)[0] & 0377) && !IS_ASCII_OCTET((*pp)[0]));
}
/*
* Step forward or backward one character in a string.
*/
public LWCHAR step_char(char **pp, signed int dir, constant char *limit)
{
LWCHAR ch;
int len;
char *p = *pp;
if (!utf_mode)
{
/* It's easy if chars are one byte. */
if (dir > 0)
ch = (LWCHAR) (unsigned char) ((p < limit) ? *p++ : 0);
else
ch = (LWCHAR) (unsigned char) ((p > limit) ? *--p : 0);
} else if (dir > 0)
{
len = utf_len(*p);
if (p + len > limit)
{
ch = 0;
p = (char *) limit;
} else
{
ch = get_wchar(p);
p += len;
}
} else
{
while (p > limit && IS_UTF8_TRAIL(p[-1]))
p--;
if (p > limit)
ch = get_wchar(--p);
else
ch = 0;
}
*pp = p;
return ch;
}
/*
* Unicode characters data
* Actual data is in the generated *.uni files.
*/
/* comb_table is special pairs, not ranges. */
static struct wchar_range comb_table[] = {
{0x0644,0x0622}, {0x0644,0x0623}, {0x0644,0x0625}, {0x0644,0x0627},
};
static int is_in_table(LWCHAR ch, struct wchar_range_table *table)
{
int hi;
int lo;
/* Binary search in the table. */
if (table->table == NULL || table->count == 0 || ch < table->table[0].first)
return 0;
lo = 0;
hi = table->count - 1;
while (lo <= hi)
{
int mid = (lo + hi) / 2;
if (ch > table->table[mid].last)
lo = mid + 1;
else if (ch < table->table[mid].first)
hi = mid - 1;
else
return 1;
}
return 0;
}
/*
* Is a character a UTF-8 composing character?
* If a composing character follows any char, the two combine into one glyph.
*/
public int is_composing_char(LWCHAR ch)
{
if (is_in_table(ch, &user_prt_table)) return 0;
return is_in_table(ch, &user_compose_table) ||
is_in_table(ch, &compose_table) ||
(bs_mode != BS_CONTROL && is_in_table(ch, &fmt_table));
}
/*
* Should this UTF-8 character be treated as binary?
*/
public int is_ubin_char(LWCHAR ch)
{
if (is_in_table(ch, &user_prt_table)) return 0;
return is_in_table(ch, &user_ubin_table) ||
is_in_table(ch, &ubin_table) ||
(bs_mode == BS_CONTROL && is_in_table(ch, &fmt_table));
}
/*
* Is this a double width UTF-8 character?
*/
public int is_wide_char(LWCHAR ch)
{
return is_in_table(ch, &user_wide_table) ||
is_in_table(ch, &wide_table);
}
/*
* Is a character a UTF-8 combining character?
* A combining char acts like an ordinary char, but if it follows
* a specific char (not any char), the two combine into one glyph.
*/
public int is_combining_char(LWCHAR ch1, LWCHAR ch2)
{
/* The table is small; use linear search. */
int i;
for (i = 0; i < sizeof(comb_table)/sizeof(*comb_table); i++)
{
if (ch1 == comb_table[i].first &&
ch2 == comb_table[i].last)
return 1;
}
return 0;
}
