/// Convert the Farsi 3342 standard into Farsi VIM.
void conv_to_pvim(void)
{
char_u *ptr;
int lnum, llen, i;
for (lnum = 1; lnum <= curbuf->b_ml.ml_line_count; ++lnum) {
ptr = ml_get((linenr_T)lnum);
llen = (int)STRLEN(ptr);
for (i = 0; i < llen - 1; i++) {
if (canF_Ljoin(ptr[i]) && canF_Rjoin(ptr[i + 1])) {
ptr[i] = toF_leading(ptr[i]);
i++;
while (canF_Rjoin(ptr[i]) && i < llen) {
ptr[i] = toF_Rjoin(ptr[i]);
if (F_isterm(ptr[i]) || !F_isalpha(ptr[i])) {
break;
}
i++;
}
if (!F_isalpha(ptr[i]) || !canF_Rjoin(ptr[i])) {
ptr[i - 1] = toF_ending(ptr[i - 1]);
}
} else {
ptr[i] = toF_TyA(ptr[i]);
}
}
}
// Following lines contains Farsi encoded character.
do_cmdline_cmd((char_u *)"%s/\202\231/\232/g");
do_cmdline_cmd((char_u *)"%s/\201\231/\370\334/g");
// Assume the screen has been messed up: clear it and redraw.
redraw_later(CLEAR);
MSG_ATTR(farsi_text_1, hl_attr(HLF_S));
}
/*
* swap all the Farsi characters in reverse direction
*/
char_u *lrFswap(char_u *cmdbuf, int len)
{
int i, cnt;
if (cmdbuf == NULL)
return cmdbuf;
if (len == 0 && (len = (int)STRLEN(cmdbuf)) == 0)
return cmdbuf;
for (i = 0; i < len; i++) {
for (cnt = 0; i + cnt < len
&& (F_isalpha(cmdbuf[i + cnt])
|| F_isdigit(cmdbuf[i + cnt])
|| cmdbuf[i + cnt] == ' '); ++cnt)
;
lrswapbuf(cmdbuf + i, cnt);
i += cnt;
}
return cmdbuf;
}
/// Helper for init_chartab
///
/// @param global FALSE: only set buf->b_chartab[]
///
/// @return FAIL if 'iskeyword', 'isident', 'isfname' or 'isprint' option has
/// an error, OK otherwise.
int buf_init_chartab(buf_T *buf, int global)
{
int c;
int c2;
char_u *p;
int i;
int tilde;
int do_isalpha;
if (global) {
// Set the default size for printable characters:
// From <Space> to '~' is 1 (printable), others are 2 (not printable).
// This also inits all 'isident' and 'isfname' flags to FALSE.
//
// EBCDIC: all chars below ' ' are not printable, all others are
// printable.
c = 0;
while (c < ' ') {
chartab[c++] = (dy_flags & DY_UHEX) ? 4 : 2;
}
while (c <= '~') {
chartab[c++] = 1 + CT_PRINT_CHAR;
}
if (p_altkeymap) {
while (c < YE) {
chartab[c++] = 1 + CT_PRINT_CHAR;
}
}
while (c < 256) {
if (enc_utf8 && (c >= 0xa0)) {
// UTF-8: bytes 0xa0 - 0xff are printable (latin1)
chartab[c++] = CT_PRINT_CHAR + 1;
} else if ((enc_dbcs == DBCS_JPNU) && (c == 0x8e)) {
// euc-jp characters starting with 0x8e are single width
chartab[c++] = CT_PRINT_CHAR + 1;
} else if ((enc_dbcs != 0) && (MB_BYTE2LEN(c) == 2)) {
// other double-byte chars can be printable AND double-width
chartab[c++] = CT_PRINT_CHAR + 2;
} else {
// the rest is unprintable by default
chartab[c++] = (dy_flags & DY_UHEX) ? 4 : 2;
}
}
// Assume that every multi-byte char is a filename character.
for (c = 1; c < 256; ++c) {
if (((enc_dbcs != 0) && (MB_BYTE2LEN(c) > 1))
|| ((enc_dbcs == DBCS_JPNU) && (c == 0x8e))
|| (enc_utf8 && (c >= 0xa0))) {
chartab[c] |= CT_FNAME_CHAR;
}
}
}
// Init word char flags all to FALSE
memset(buf->b_chartab, 0, (size_t)32);
if (enc_dbcs != 0) {
for (c = 0; c < 256; ++c) {
// double-byte characters are probably word characters
if (MB_BYTE2LEN(c) == 2) {
SET_CHARTAB(buf, c);
}
}
}
// In lisp mode the '-' character is included in keywords.
if (buf->b_p_lisp) {
SET_CHARTAB(buf, '-');
}
// Walk through the 'isident', 'iskeyword', 'isfname' and 'isprint'
// options Each option is a list of characters, character numbers or
// ranges, separated by commas, e.g.: "200-210,x,#-178,-"
for (i = global ? 0 : 3; i <= 3; ++i) {
if (i == 0) {
// first round: 'isident'
p = p_isi;
} else if (i == 1) {
// second round: 'isprint'
p = p_isp;
} else if (i == 2) {
// third round: 'isfname'
p = p_isf;
} else { // i == 3
// fourth round: 'iskeyword'
p = buf->b_p_isk;
}
while (*p) {
tilde = FALSE;
do_isalpha = FALSE;
if ((*p == '^') && (p[1] != NUL)) {
tilde = TRUE;
++p;
}
if (VIM_ISDIGIT(*p)) {
c = getdigits(&p);
} else if (has_mbyte) {
c = mb_ptr2char_adv(&p);
} else {
c = *p++;
}
c2 = -1;
if ((*p == '-') && (p[1] != NUL)) {
++p;
if (VIM_ISDIGIT(*p)) {
c2 = getdigits(&p);
} else if (has_mbyte) {
c2 = mb_ptr2char_adv(&p);
} else {
c2 = *p++;
}
}
if ((c <= 0)
|| (c >= 256)
|| ((c2 < c) && (c2 != -1))
|| (c2 >= 256)
|| !((*p == NUL) || (*p == ','))) {
return FAIL;
}
if (c2 == -1) { // not a range
// A single '@' (not "@-@"):
// Decide on letters being ID/printable/keyword chars with
// standard function isalpha(). This takes care of locale for
// single-byte characters).
if (c == '@') {
do_isalpha = TRUE;
c = 1;
c2 = 255;
} else {
c2 = c;
}
}
while (c <= c2) {
// Use the MB_ functions here, because isalpha() doesn't
// work properly when 'encoding' is "latin1" and the locale is
// "C".
if (!do_isalpha
|| MB_ISLOWER(c)
|| MB_ISUPPER(c)
|| (p_altkeymap && (F_isalpha(c) || F_isdigit(c)))) {
if (i == 0) {
// (re)set ID flag
if (tilde) {
chartab[c] &= ~CT_ID_CHAR;
} else {
chartab[c] |= CT_ID_CHAR;
}
} else if (i == 1) {
// (re)set printable
// For double-byte we keep the cell width, so
// that we can detect it from the first byte.
if (((c < ' ')
|| (c > '~')
|| (p_altkeymap && (F_isalpha(c) || F_isdigit(c))))
&& !(enc_dbcs && (MB_BYTE2LEN(c) == 2))) {
if (tilde) {
chartab[c] = (chartab[c] & ~CT_CELL_MASK)
+ ((dy_flags & DY_UHEX) ? 4 : 2);
chartab[c] &= ~CT_PRINT_CHAR;
} else {
chartab[c] = (chartab[c] & ~CT_CELL_MASK) + 1;
chartab[c] |= CT_PRINT_CHAR;
}
}
} else if (i == 2) {
// (re)set fname flag
if (tilde) {
chartab[c] &= ~CT_FNAME_CHAR;
} else {
chartab[c] |= CT_FNAME_CHAR;
}
} else { // i == 3
// (re)set keyword flag
if (tilde) {
RESET_CHARTAB(buf, c);
} else {
SET_CHARTAB(buf, c);
}
}
}
++c;
}
c = *p;
p = skip_to_option_part(p);
if ((c == ',') && (*p == NUL)) {
// Trailing comma is not allowed.
return FAIL;
}
}
}
chartab_initialized = TRUE;
return OK;
}
int
buf_init_chartab (
buf_T *buf,
int global /* FALSE: only set buf->b_chartab[] */
)
{
int c;
int c2;
char_u *p;
int i;
int tilde;
int do_isalpha;
if (global) {
/*
* Set the default size for printable characters:
* From <Space> to '~' is 1 (printable), others are 2 (not printable).
* This also inits all 'isident' and 'isfname' flags to FALSE.
*
* EBCDIC: all chars below ' ' are not printable, all others are
* printable.
*/
c = 0;
while (c < ' ')
chartab[c++] = (dy_flags & DY_UHEX) ? 4 : 2;
while (c <= '~')
chartab[c++] = 1 + CT_PRINT_CHAR;
if (p_altkeymap) {
while (c < YE)
chartab[c++] = 1 + CT_PRINT_CHAR;
}
while (c < 256) {
/* UTF-8: bytes 0xa0 - 0xff are printable (latin1) */
if (enc_utf8 && c >= 0xa0)
chartab[c++] = CT_PRINT_CHAR + 1;
/* euc-jp characters starting with 0x8e are single width */
else if (enc_dbcs == DBCS_JPNU && c == 0x8e)
chartab[c++] = CT_PRINT_CHAR + 1;
/* other double-byte chars can be printable AND double-width */
else if (enc_dbcs != 0 && MB_BYTE2LEN(c) == 2)
chartab[c++] = CT_PRINT_CHAR + 2;
else
/* the rest is unprintable by default */
chartab[c++] = (dy_flags & DY_UHEX) ? 4 : 2;
}
/* Assume that every multi-byte char is a filename character. */
for (c = 1; c < 256; ++c)
if ((enc_dbcs != 0 && MB_BYTE2LEN(c) > 1)
|| (enc_dbcs == DBCS_JPNU && c == 0x8e)
|| (enc_utf8 && c >= 0xa0))
chartab[c] |= CT_FNAME_CHAR;
}
/*
* Init word char flags all to FALSE
*/
vim_memset(buf->b_chartab, 0, (size_t)32);
if (enc_dbcs != 0)
for (c = 0; c < 256; ++c) {
/* double-byte characters are probably word characters */
if (MB_BYTE2LEN(c) == 2)
SET_CHARTAB(buf, c);
}
/*
* In lisp mode the '-' character is included in keywords.
*/
if (buf->b_p_lisp)
SET_CHARTAB(buf, '-');
/* Walk through the 'isident', 'iskeyword', 'isfname' and 'isprint'
* options Each option is a list of characters, character numbers or
* ranges, separated by commas, e.g.: "200-210,x,#-178,-"
*/
for (i = global ? 0 : 3; i <= 3; ++i) {
if (i == 0)
p = p_isi; /* first round: 'isident' */
else if (i == 1)
p = p_isp; /* second round: 'isprint' */
else if (i == 2)
p = p_isf; /* third round: 'isfname' */
else /* i == 3 */
p = buf->b_p_isk; /* fourth round: 'iskeyword' */
while (*p) {
tilde = FALSE;
do_isalpha = FALSE;
if (*p == '^' && p[1] != NUL) {
tilde = TRUE;
++p;
}
if (VIM_ISDIGIT(*p))
c = getdigits(&p);
else if (has_mbyte)
c = mb_ptr2char_adv(&p);
else
c = *p++;
c2 = -1;
if (*p == '-' && p[1] != NUL) {
++p;
if (VIM_ISDIGIT(*p))
c2 = getdigits(&p);
else if (has_mbyte)
c2 = mb_ptr2char_adv(&p);
else
c2 = *p++;
}
if (c <= 0 || c >= 256 || (c2 < c && c2 != -1) || c2 >= 256
|| !(*p == NUL || *p == ','))
return FAIL;
if (c2 == -1) { /* not a range */
/*
* A single '@' (not "@-@"):
* Decide on letters being ID/printable/keyword chars with
* standard function isalpha(). This takes care of locale for
* single-byte characters).
*/
if (c == '@') {
do_isalpha = TRUE;
c = 1;
c2 = 255;
} else
c2 = c;
}
while (c <= c2) {
/* Use the MB_ functions here, because isalpha() doesn't
* work properly when 'encoding' is "latin1" and the locale is
* "C". */
if (!do_isalpha || MB_ISLOWER(c) || MB_ISUPPER(c)
|| (p_altkeymap && (F_isalpha(c) || F_isdigit(c)))
) {
if (i == 0) { /* (re)set ID flag */
if (tilde)
chartab[c] &= ~CT_ID_CHAR;
else
chartab[c] |= CT_ID_CHAR;
} else if (i == 1) { /* (re)set printable */
if ((c < ' '
|| c > '~'
|| (p_altkeymap
&& (F_isalpha(c) || F_isdigit(c)))
)
/* For double-byte we keep the cell width, so
* that we can detect it from the first byte. */
&& !(enc_dbcs && MB_BYTE2LEN(c) == 2)
) {
if (tilde) {
chartab[c] = (chartab[c] & ~CT_CELL_MASK)
+ ((dy_flags & DY_UHEX) ? 4 : 2);
chartab[c] &= ~CT_PRINT_CHAR;
} else {
chartab[c] = (chartab[c] & ~CT_CELL_MASK) + 1;
chartab[c] |= CT_PRINT_CHAR;
}
}
} else if (i == 2) { /* (re)set fname flag */
if (tilde)
chartab[c] &= ~CT_FNAME_CHAR;
else
chartab[c] |= CT_FNAME_CHAR;
} else { /* i == 3 */ /* (re)set keyword flag */
if (tilde)
RESET_CHARTAB(buf, c);
else
SET_CHARTAB(buf, c);
}
}
++c;
}
c = *p;
p = skip_to_option_part(p);
if (c == ',' && *p == NUL)
/* Trailing comma is not allowed. */
return FAIL;
}
}
chartab_initialized = TRUE;
return OK;
}
/// Map Farsi keyboard when in fkmap mode.
int fkmap(int c)
{
int tempc;
static int revins;
if (IS_SPECIAL(c)) {
return c;
}
if (VIM_ISDIGIT(c)
|| (((c == '.')
|| (c == '+')
|| (c == '-')
|| (c == '^')
|| (c == '%')
|| (c == '#')
|| (c == '='))
&& revins)) {
if (!revins) {
if (curwin->w_cursor.col) {
if (!p_ri) {
dec_cursor();
}
chg_c_toX_orX();
chg_l_toXor_X();
if (!p_ri) {
inc_cursor();
}
}
}
arrow_used = TRUE;
(void)stop_arrow();
if (!curwin->w_p_rl && revins) {
inc_cursor();
}
revins++;
p_ri = 1;
} else {
if (revins) {
arrow_used = TRUE;
(void)stop_arrow();
revins = 0;
if (curwin->w_p_rl) {
while ((F_isdigit(gchar_cursor())
|| (gchar_cursor() == F_PERIOD
|| gchar_cursor() == F_PLUS
|| gchar_cursor() == F_MINUS
|| gchar_cursor() == F_MUL
|| gchar_cursor() == F_DIVIDE
|| gchar_cursor() == F_PERCENT
|| gchar_cursor() == F_EQUALS))
&& gchar_cursor() != '\0') {
curwin->w_cursor.col++;
}
} else {
if (curwin->w_cursor.col) {
while ((F_isdigit(gchar_cursor())
|| (gchar_cursor() == F_PERIOD
|| gchar_cursor() == F_PLUS
|| gchar_cursor() == F_MINUS
|| gchar_cursor() == F_MUL
|| gchar_cursor() == F_DIVIDE
|| gchar_cursor() == F_PERCENT
|| gchar_cursor() == F_EQUALS))
&& --curwin->w_cursor.col) {
}
}
if (!F_isdigit(gchar_cursor())) {
++curwin->w_cursor.col;
}
}
}
}
if (!revins) {
if (curwin->w_p_rl) {
p_ri = 0;
}
if (!curwin->w_p_rl) {
p_ri = 1;
}
}
if ((c < 0x100) &&
(isalpha(c) ||
(c == '&') ||
(c == '^') ||
(c == ';') ||
(c == '\'') ||
(c == ',') ||
(c == '[') ||
(c == ']') ||
(c == '{') ||
(c == '}'))) {
chg_r_to_Xor_X_();
}
tempc = 0;
switch (c) {
case '`':
case ' ':
case '.':
case '!':
case '"':
case '$':
case '%':
case '^':
case '&':
case '/':
case '(':
case ')':
case '=':
case '\\':
case '?':
case '+':
case '-':
case '_':
case '*':
case ':':
case '#':
case '~':
case '@':
case '<':
case '>':
case '{':
case '}':
case '|':
case '0':
case '1':
case '2':
case '3':
case '4':
case '5':
case '6':
case '7':
case '8':
case '9':
case 'B':
case 'E':
case 'F':
case 'H':
case 'I':
case 'K':
case 'L':
case 'M':
case 'O':
case 'P':
case 'Q':
case 'R':
case 'T':
case 'U':
case 'W':
case 'Y':
case NL:
case TAB:
if (p_ri && (c == NL) && curwin->w_cursor.col) {
// If the char before the cursor is _X_ or X_ do not change
// the one under the cursor with X type.
dec_cursor();
if (F_isalpha(gchar_cursor())) {
inc_cursor();
return NL;
}
inc_cursor();
}
if (!p_ri) {
if (!curwin->w_cursor.col) {
switch (c) {
case '0':
return FARSI_0;
case '1':
return FARSI_1;
case '2':
return FARSI_2;
case '3':
return FARSI_3;
case '4':
return FARSI_4;
case '5':
return FARSI_5;
case '6':
return FARSI_6;
case '7':
return FARSI_7;
case '8':
return FARSI_8;
case '9':
return FARSI_9;
case 'B':
return F_PSP;
case 'E':
return JAZR_N;
case 'F':
return ALEF_D_H;
case 'H':
return ALEF_A;
case 'I':
return TASH;
case 'K':
return F_LQUOT;
case 'L':
return F_RQUOT;
case 'M':
return HAMZE;
case 'O':
return '[';
case 'P':
return ']';
case 'Q':
return OO;
case 'R':
return MAD_N;
case 'T':
return OW;
case 'U':
return MAD;
case 'W':
return OW_OW;
case 'Y':
return JAZR;
case '`':
return F_PCN;
case '!':
return F_EXCL;
case '@':
return F_COMMA;
case '#':
return F_DIVIDE;
case '$':
return F_CURRENCY;
case '%':
return F_PERCENT;
case '^':
return F_MUL;
case '&':
return F_BCOMMA;
case '*':
return F_STAR;
case '(':
return F_LPARENT;
case ')':
return F_RPARENT;
case '-':
return F_MINUS;
case '_':
return F_UNDERLINE;
case '=':
return F_EQUALS;
case '+':
return F_PLUS;
case '\\':
return F_BSLASH;
case '|':
return F_PIPE;
case ':':
return F_DCOLON;
case '"':
return F_SEMICOLON;
case '.':
return F_PERIOD;
case '/':
return F_SLASH;
case '<':
return F_LESS;
case '>':
return F_GREATER;
case '?':
return F_QUESTION;
case ' ':
return F_BLANK;
}
break;
}
}
if (!p_ri) {
dec_cursor();
}
switch ((tempc = gchar_cursor())) {
case _BE:
case _PE:
case _TE:
case _SE:
case _JIM:
case _CHE:
case _HE_J:
case _XE:
case _SIN:
case _SHIN:
case _SAD:
case _ZAD:
case _FE:
case _GHAF:
case _KAF:
case _KAF_H:
case _GAF:
case _LAM:
case _MIM:
case _NOON:
case _HE:
case _HE_:
case _TA:
case _ZA:
put_curr_and_l_to_X(toF_TyA(tempc));
break;
case _AYN:
case _AYN_:
if (!p_ri) {
if (!curwin->w_cursor.col) {
put_curr_and_l_to_X(AYN);
break;
}
}
if (p_ri) {
inc_cursor();
} else {
dec_cursor();
}
if (F_is_TyB_TyC_TyD(SRC_EDT, AT_CURSOR)) {
tempc = AYN_;
} else {
tempc = AYN;
}
if (p_ri) {
dec_cursor();
} else {
inc_cursor();
}
put_curr_and_l_to_X(tempc);
break;
case _GHAYN:
case _GHAYN_:
if (!p_ri) {
if (!curwin->w_cursor.col) {
put_curr_and_l_to_X(GHAYN);
break;
}
}
if (p_ri) {
inc_cursor();
} else {
dec_cursor();
}
if (F_is_TyB_TyC_TyD(SRC_EDT, AT_CURSOR)) {
tempc = GHAYN_;
} else {
tempc = GHAYN;
}
if (p_ri) {
dec_cursor();
} else {
inc_cursor();
}
put_curr_and_l_to_X(tempc);
break;
case _YE:
case _IE:
case _YEE:
if (!p_ri) {
if (!curwin->w_cursor.col) {
put_curr_and_l_to_X((tempc == _YE ? YE :
(tempc == _IE ? IE : YEE)));
break;
}
}
if (p_ri) {
inc_cursor();
} else {
dec_cursor();
}
if (F_is_TyB_TyC_TyD(SRC_EDT, AT_CURSOR)) {
tempc = (tempc == _YE ? YE_ :
(tempc == _IE ? IE_ : YEE_));
} else {
tempc = (tempc == _YE ? YE :
(tempc == _IE ? IE : YEE));
}
if (p_ri) {
dec_cursor();
} else {
inc_cursor();
}
put_curr_and_l_to_X(tempc);
break;
}
if (!p_ri) {
inc_cursor();
}
tempc = 0;
switch (c) {
case '0':
return FARSI_0;
case '1':
return FARSI_1;
case '2':
return FARSI_2;
case '3':
return FARSI_3;
case '4':
return FARSI_4;
case '5':
return FARSI_5;
case '6':
return FARSI_6;
case '7':
return FARSI_7;
case '8':
return FARSI_8;
case '9':
return FARSI_9;
case 'B':
return F_PSP;
case 'E':
return JAZR_N;
case 'F':
return ALEF_D_H;
case 'H':
return ALEF_A;
case 'I':
return TASH;
case 'K':
return F_LQUOT;
case 'L':
return F_RQUOT;
case 'M':
return HAMZE;
case 'O':
return '[';
case 'P':
return ']';
case 'Q':
return OO;
case 'R':
return MAD_N;
case 'T':
return OW;
case 'U':
return MAD;
case 'W':
return OW_OW;
case 'Y':
return JAZR;
case '`':
return F_PCN;
case '!':
return F_EXCL;
case '@':
return F_COMMA;
case '#':
return F_DIVIDE;
case '$':
return F_CURRENCY;
case '%':
return F_PERCENT;
case '^':
return F_MUL;
case '&':
return F_BCOMMA;
case '*':
return F_STAR;
case '(':
return F_LPARENT;
case ')':
return F_RPARENT;
case '-':
return F_MINUS;
case '_':
return F_UNDERLINE;
case '=':
return F_EQUALS;
case '+':
return F_PLUS;
case '\\':
return F_BSLASH;
case '|':
return F_PIPE;
case ':':
return F_DCOLON;
case '"':
return F_SEMICOLON;
case '.':
return F_PERIOD;
case '/':
return F_SLASH;
case '<':
return F_LESS;
case '>':
return F_GREATER;
case '?':
return F_QUESTION;
case ' ':
return F_BLANK;
}
break;
case 'a':
tempc = _SHIN;
break;
case 'A':
tempc = WAW_H;
break;
case 'b':
tempc = ZAL;
break;
case 'c':
tempc = ZE;
break;
case 'C':
tempc = JE;
break;
case 'd':
tempc = _YE;
break;
case 'D':
tempc = _YEE;
break;
case 'e':
tempc = _SE;
break;
case 'f':
tempc = _BE;
break;
case 'g':
tempc = _LAM;
break;
case 'G':
if (!curwin->w_cursor.col && STRLEN(ml_get_curline())) {
if (gchar_cursor() == _LAM) {
chg_c_toX_orX();
} else if (p_ri) {
chg_c_to_X_or_X();
}
}
if (!p_ri) {
if (!curwin->w_cursor.col) {
return ALEF_U_H;
}
}
if (!p_ri) {
dec_cursor();
}
if (gchar_cursor() == _LAM) {
chg_c_toX_orX();
chg_l_toXor_X();
tempc = ALEF_U_H;
} else if (F_is_TyB_TyC_TyD(SRC_EDT, AT_CURSOR)) {
tempc = ALEF_U_H_;
chg_l_toXor_X();
} else {
tempc = ALEF_U_H;
}
if (!p_ri) {
inc_cursor();
}
return tempc;
case 'h':
if (!curwin->w_cursor.col && STRLEN(ml_get_curline())) {
if (p_ri) {
chg_c_to_X_or_X();
}
}
if (!p_ri) {
if (!curwin->w_cursor.col) {
return ALEF;
}
}
if (!p_ri) {
dec_cursor();
}
if (gchar_cursor() == _LAM) {
chg_l_toXor_X();
del_char(FALSE);
AppendCharToRedobuff(K_BS);
if (!p_ri) {
dec_cursor();
}
tempc = LA;
} else {
if (F_is_TyB_TyC_TyD(SRC_EDT, AT_CURSOR)) {
tempc = ALEF_;
chg_l_toXor_X();
} else {
tempc = ALEF;
}
}
if (!p_ri) {
inc_cursor();
}
return tempc;
case 'i':
if (!curwin->w_cursor.col && STRLEN(ml_get_curline())) {
if (!p_ri && !F_is_TyE(tempc)) {
chg_c_to_X_orX_();
}
if (p_ri) {
chg_c_to_X_or_X();
}
}
if (!p_ri && !curwin->w_cursor.col) {
return _HE;
}
if (!p_ri) {
dec_cursor();
}
if (F_is_TyB_TyC_TyD(SRC_EDT, AT_CURSOR)) {
tempc = _HE_;
} else {
tempc = _HE;
}
if (!p_ri) {
inc_cursor();
}
break;
case 'j':
tempc = _TE;
break;
case 'J':
if (!curwin->w_cursor.col && STRLEN(ml_get_curline())) {
if (p_ri) {
chg_c_to_X_or_X();
}
}
if (!p_ri) {
if (!curwin->w_cursor.col) {
return TEE;
}
}
if (!p_ri) {
dec_cursor();
}
if (F_is_TyB_TyC_TyD(SRC_EDT, AT_CURSOR)) {
tempc = TEE_;
chg_l_toXor_X();
} else {
tempc = TEE;
}
if (!p_ri) {
inc_cursor();
}
return tempc;
case 'k':
tempc = _NOON;
break;
case 'l':
tempc = _MIM;
break;
case 'm':
tempc = _PE;
break;
case 'n':
case 'N':
tempc = DAL;
break;
case 'o':
tempc = _XE;
break;
case 'p':
tempc = _HE_J;
break;
case 'q':
tempc = _ZAD;
break;
case 'r':
tempc = _GHAF;
break;
case 's':
tempc = _SIN;
break;
case 'S':
tempc = _IE;
break;
case 't':
tempc = _FE;
break;
case 'u':
if (!curwin->w_cursor.col && STRLEN(ml_get_curline())) {
if (!p_ri && !F_is_TyE(tempc)) {
chg_c_to_X_orX_();
}
if (p_ri) {
chg_c_to_X_or_X();
}
}
if (!p_ri && !curwin->w_cursor.col) {
return _AYN;
}
if (!p_ri) {
dec_cursor();
}
if (F_is_TyB_TyC_TyD(SRC_EDT, AT_CURSOR)) {
tempc = _AYN_;
} else {
tempc = _AYN;
}
if (!p_ri) {
inc_cursor();
}
break;
case 'v':
case 'V':
tempc = RE;
break;
case 'w':
tempc = _SAD;
break;
case 'x':
case 'X':
tempc = _TA;
break;
case 'y':
if (!curwin->w_cursor.col && STRLEN(ml_get_curline())) {
if (!p_ri && !F_is_TyE(tempc)) {
chg_c_to_X_orX_();
}
if (p_ri) {
chg_c_to_X_or_X();
}
}
if (!p_ri && !curwin->w_cursor.col) {
return _GHAYN;
}
if (!p_ri) {
dec_cursor();
}
if (F_is_TyB_TyC_TyD(SRC_EDT, AT_CURSOR)) {
tempc = _GHAYN_;
} else {
tempc = _GHAYN;
}
if (!p_ri) {
inc_cursor();
}
break;
case 'z':
tempc = _ZA;
break;
case 'Z':
tempc = _KAF_H;
break;
case ';':
tempc = _KAF;
break;
case '\'':
tempc = _GAF;
break;
case ',':
tempc = WAW;
break;
case '[':
tempc = _JIM;
break;
case ']':
tempc = _CHE;
break;
}
if ((F_isalpha(tempc) || F_isdigit(tempc))) {
if (!curwin->w_cursor.col && STRLEN(ml_get_curline())) {
if (!p_ri && !F_is_TyE(tempc)) {
chg_c_to_X_orX_();
}
if (p_ri) {
chg_c_to_X_or_X();
}
}
if (curwin->w_cursor.col) {
if (!p_ri) {
dec_cursor();
}
if (F_is_TyE(tempc)) {
chg_l_toXor_X();
} else {
chg_l_to_X_orX_();
}
if (!p_ri) {
inc_cursor();
}
}
}
if (tempc) {
return tempc;
}
return c;
}
/// Helper for init_chartab
///
/// @param global false: only set buf->b_chartab[]
///
/// @return FAIL if 'iskeyword', 'isident', 'isfname' or 'isprint' option has
/// an error, OK otherwise.
int buf_init_chartab(buf_T *buf, int global)
{
int c;
int c2;
int i;
bool tilde;
bool do_isalpha;
if (global) {
// Set the default size for printable characters:
// From <Space> to '~' is 1 (printable), others are 2 (not printable).
// This also inits all 'isident' and 'isfname' flags to false.
c = 0;
while (c < ' ') {
g_chartab[c++] = (dy_flags & DY_UHEX) ? 4 : 2;
}
while (c <= '~') {
g_chartab[c++] = 1 + CT_PRINT_CHAR;
}
if (p_altkeymap) {
while (c < YE) {
g_chartab[c++] = 1 + CT_PRINT_CHAR;
}
}
while (c < 256) {
if (c >= 0xa0) {
// UTF-8: bytes 0xa0 - 0xff are printable (latin1)
g_chartab[c++] = CT_PRINT_CHAR + 1;
} else {
// the rest is unprintable by default
g_chartab[c++] = (dy_flags & DY_UHEX) ? 4 : 2;
}
}
// Assume that every multi-byte char is a filename character.
for (c = 1; c < 256; c++) {
if (c >= 0xa0) {
g_chartab[c] |= CT_FNAME_CHAR;
}
}
}
// Init word char flags all to false
memset(buf->b_chartab, 0, (size_t)32);
// In lisp mode the '-' character is included in keywords.
if (buf->b_p_lisp) {
SET_CHARTAB(buf, '-');
}
// Walk through the 'isident', 'iskeyword', 'isfname' and 'isprint'
// options Each option is a list of characters, character numbers or
// ranges, separated by commas, e.g.: "200-210,x,#-178,-"
for (i = global ? 0 : 3; i <= 3; i++) {
const char_u *p;
if (i == 0) {
// first round: 'isident'
p = p_isi;
} else if (i == 1) {
// second round: 'isprint'
p = p_isp;
} else if (i == 2) {
// third round: 'isfname'
p = p_isf;
} else { // i == 3
// fourth round: 'iskeyword'
p = buf->b_p_isk;
}
while (*p) {
tilde = false;
do_isalpha = false;
if ((*p == '^') && (p[1] != NUL)) {
tilde = true;
++p;
}
if (ascii_isdigit(*p)) {
c = getdigits_int((char_u **)&p);
} else {
c = mb_ptr2char_adv(&p);
}
c2 = -1;
if ((*p == '-') && (p[1] != NUL)) {
++p;
if (ascii_isdigit(*p)) {
c2 = getdigits_int((char_u **)&p);
} else {
c2 = mb_ptr2char_adv(&p);
}
}
if ((c <= 0)
|| (c >= 256)
|| ((c2 < c) && (c2 != -1))
|| (c2 >= 256)
|| !((*p == NUL) || (*p == ','))) {
return FAIL;
}
if (c2 == -1) { // not a range
// A single '@' (not "@-@"):
// Decide on letters being ID/printable/keyword chars with
// standard function isalpha(). This takes care of locale for
// single-byte characters).
if (c == '@') {
do_isalpha = true;
c = 1;
c2 = 255;
} else {
c2 = c;
}
}
while (c <= c2) {
// Use the MB_ functions here, because isalpha() doesn't
// work properly when 'encoding' is "latin1" and the locale is
// "C".
if (!do_isalpha
|| mb_islower(c)
|| mb_isupper(c)
|| (p_altkeymap && (F_isalpha(c) || F_isdigit(c)))) {
if (i == 0) {
// (re)set ID flag
if (tilde) {
g_chartab[c] &= (uint8_t)~CT_ID_CHAR;
} else {
g_chartab[c] |= CT_ID_CHAR;
}
} else if (i == 1) {
// (re)set printable
// For double-byte we keep the cell width, so
// that we can detect it from the first byte.
if (((c < ' ')
|| (c > '~')
|| (p_altkeymap && (F_isalpha(c) || F_isdigit(c))))) {
if (tilde) {
g_chartab[c] = (uint8_t)((g_chartab[c] & ~CT_CELL_MASK)
+ ((dy_flags & DY_UHEX) ? 4 : 2));
g_chartab[c] &= (uint8_t)~CT_PRINT_CHAR;
} else {
g_chartab[c] = (uint8_t)((g_chartab[c] & ~CT_CELL_MASK) + 1);
g_chartab[c] |= CT_PRINT_CHAR;
}
}
} else if (i == 2) {
// (re)set fname flag
if (tilde) {
g_chartab[c] &= (uint8_t)~CT_FNAME_CHAR;
} else {
g_chartab[c] |= CT_FNAME_CHAR;
}
} else { // i == 3
// (re)set keyword flag
if (tilde) {
RESET_CHARTAB(buf, c);
} else {
SET_CHARTAB(buf, c);
}
}
}
++c;
}
c = *p;
p = skip_to_option_part(p);
if ((c == ',') && (*p == NUL)) {
// Trailing comma is not allowed.
return FAIL;
}
}
}
chartab_initialized = true;
return OK;
}
