aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
Diffstat
-rwxr-xr-xscripts/vim-patch.sh1
-rw-r--r--src/nvim/edit.c6
-rw-r--r--src/nvim/ex_docmd.c20
-rw-r--r--src/nvim/fileio.c15
-rw-r--r--src/nvim/globals.h1
-rw-r--r--src/nvim/mbyte.c5
-rw-r--r--src/nvim/message.c12
-rw-r--r--src/nvim/ops.c3
-rw-r--r--src/nvim/screen.c409
-rw-r--r--src/nvim/search.c98
-rw-r--r--test/unit/search_spec.lua33
11 files changed, 262 insertions, 341 deletions
diff --git a/scripts/vim-patch.sh b/scripts/vim-patch.sh
index 11305421e5..7339fdc691 100755
--- a/scripts/vim-patch.sh
+++ b/scripts/vim-patch.sh
@@ -297,6 +297,7 @@ submit_pr() {
submit_fn="git_hub_pr"
else
>&2 echo "${BASENAME}: 'hub' or 'git-hub' not found in PATH or not executable."
+ >&2 echo " Get it here: https://hub.github.com/"
exit 1
fi
diff --git a/src/nvim/edit.c b/src/nvim/edit.c
index 5a21c50fa6..67ac675a14 100644
--- a/src/nvim/edit.c
+++ b/src/nvim/edit.c
@@ -5110,11 +5110,9 @@ int get_literal(void)
}
}
- if (cc == 0) /* NUL is stored as NL */
+ if (cc == 0) { // NUL is stored as NL
cc = '\n';
- if (enc_dbcs && (cc & 0xff) == 0)
- cc = '?'; /* don't accept an illegal DBCS char, the NUL in the
- second byte will cause trouble! */
+ }
--no_mapping;
if (nc)
diff --git a/src/nvim/ex_docmd.c b/src/nvim/ex_docmd.c
index c60ccabf4c..8f9af1c2ec 100644
--- a/src/nvim/ex_docmd.c
+++ b/src/nvim/ex_docmd.c
@@ -5489,24 +5489,18 @@ uc_check_code(
break;
case 1: /* Quote, but don't split */
result = STRLEN(eap->arg) + 2;
- for (p = eap->arg; *p; ++p) {
- if (enc_dbcs != 0 && (*mb_ptr2len)(p) == 2)
- /* DBCS can contain \ in a trail byte, skip the
- * double-byte character. */
- ++p;
- else if (*p == '\\' || *p == '"')
- ++result;
+ for (p = eap->arg; *p; p++) {
+ if (*p == '\\' || *p == '"') {
+ result++;
+ }
}
if (buf != NULL) {
*buf++ = '"';
- for (p = eap->arg; *p; ++p) {
- if (enc_dbcs != 0 && (*mb_ptr2len)(p) == 2)
- /* DBCS can contain \ in a trail byte, copy the
- * double-byte character to avoid escaping. */
- *buf++ = *p++;
- else if (*p == '\\' || *p == '"')
+ for (p = eap->arg; *p; p++) {
+ if (*p == '\\' || *p == '"') {
*buf++ = '\\';
+ }
*buf++ = *p;
}
*buf = '"';
diff --git a/src/nvim/fileio.c b/src/nvim/fileio.c
index ba3625bf95..b9de46efc8 100644
--- a/src/nvim/fileio.c
+++ b/src/nvim/fileio.c
@@ -5313,9 +5313,7 @@ void forward_slash(char_u *fname)
}
for (p = fname; *p != NUL; p++) {
// The Big5 encoding can have '\' in the trail byte.
- if (enc_dbcs != 0 && (*mb_ptr2len)(p) > 1) {
- p++;
- } else if (*p == '\\') {
+ if (*p == '\\') {
*p = '/';
}
}
@@ -7615,10 +7613,6 @@ char_u * file_pat_to_reg_pat(
#endif
default:
size++;
- if (enc_dbcs != 0 && (*mb_ptr2len)(p) > 1) {
- ++p;
- ++size;
- }
break;
}
}
@@ -7739,10 +7733,9 @@ char_u * file_pat_to_reg_pat(
reg_pat[i++] = ',';
break;
default:
- if (enc_dbcs != 0 && (*mb_ptr2len)(p) > 1)
- reg_pat[i++] = *p++;
- else if (allow_dirs != NULL && vim_ispathsep(*p))
- *allow_dirs = TRUE;
+ if (allow_dirs != NULL && vim_ispathsep(*p)) {
+ *allow_dirs = true;
+ }
reg_pat[i++] = *p;
break;
}
diff --git a/src/nvim/globals.h b/src/nvim/globals.h
index 1cc3d05cd3..52c5d65512 100644
--- a/src/nvim/globals.h
+++ b/src/nvim/globals.h
@@ -641,7 +641,6 @@ EXTERN int vr_lines_changed INIT(= 0); /* #Lines changed by "gR" so far */
// mbyte flags that used to depend on 'encoding'. These are now deprecated, as
// 'encoding' is always "utf-8". Code that use them can be refactored to
// remove dead code.
-#define enc_dbcs 0
#define enc_utf8 true
#define has_mbyte true
diff --git a/src/nvim/mbyte.c b/src/nvim/mbyte.c
index 1cf045d7e0..5ed2b4c564 100644
--- a/src/nvim/mbyte.c
+++ b/src/nvim/mbyte.c
@@ -4,9 +4,8 @@
/// mbyte.c: Code specifically for handling multi-byte characters.
/// Multibyte extensions partly by Sung-Hoon Baek
///
-/// The encoding used in nvim is always UTF-8. "enc_utf8" and "has_mbyte" is
-/// thus always true. "enc_dbcs" is always zero. The 'encoding' option is
-/// read-only and always reads "utf-8".
+/// Strings internal to Nvim are always encoded as UTF-8 (thus the legacy
+/// 'encoding' option is always "utf-8").
///
/// The cell width on the display needs to be determined from the character
/// value. Recognizing UTF-8 bytes is easy: 0xxx.xxxx is a single-byte char,
diff --git a/src/nvim/message.c b/src/nvim/message.c
index abe9c25bb8..b4aa333a48 100644
--- a/src/nvim/message.c
+++ b/src/nvim/message.c
@@ -281,15 +281,9 @@ msg_strtrunc (
/* Use up to 'showcmd' column. */
room = (int)(Rows - msg_row - 1) * Columns + sc_col - 1;
if (len > room && room > 0) {
- if (enc_utf8)
- /* may have up to 18 bytes per cell (6 per char, up to two
- * composing chars) */
- len = (room + 2) * 18;
- else if (enc_dbcs == DBCS_JPNU)
- /* may have up to 2 bytes per cell for euc-jp */
- len = (room + 2) * 2;
- else
- len = room + 2;
+ // may have up to 18 bytes per cell (6 per char, up to two
+ // composing chars)
+ len = (room + 2) * 18;
buf = xmalloc(len);
trunc_string(s, buf, room, len);
}
diff --git a/src/nvim/ops.c b/src/nvim/ops.c
index 674a9244f0..99dee939fc 100644
--- a/src/nvim/ops.c
+++ b/src/nvim/ops.c
@@ -1974,8 +1974,6 @@ int swapchar(int op_type, pos_T *pos)
inc(pos);
}
- if (enc_dbcs != 0 && c >= 0x100) /* No lower/uppercase letter */
- return FALSE;
nc = c;
if (mb_islower(c)) {
if (op_type == OP_ROT13) {
@@ -3610,6 +3608,7 @@ int do_join(size_t count,
int remove_comments = (use_formatoptions == TRUE)
&& has_format_option(FO_REMOVE_COMS);
bool prev_was_comment = false;
+ assert(count >= 1);
if (save_undo && u_save(curwin->w_cursor.lnum - 1,
curwin->w_cursor.lnum + (linenr_T)count) == FAIL) {
diff --git a/src/nvim/screen.c b/src/nvim/screen.c
index de353fa3f6..08bb4e4a52 100644
--- a/src/nvim/screen.c
+++ b/src/nvim/screen.c
@@ -2617,7 +2617,7 @@ win_line (
}
// Highlight one character for an empty match.
if (shl->startcol == shl->endcol) {
- if (has_mbyte && line[shl->endcol] != NUL) {
+ if (line[shl->endcol] != NUL) {
shl->endcol += (*mb_ptr2len)(line + shl->endcol);
} else {
++shl->endcol;
@@ -2963,13 +2963,8 @@ win_line (
shl->endcol = MAXCOL;
if (shl->startcol == shl->endcol) {
- /* highlight empty match, try again after
- * it */
- if (has_mbyte)
- shl->endcol += (*mb_ptr2len)(line
- + shl->endcol);
- else
- ++shl->endcol;
+ // highlight empty match, try again after it
+ shl->endcol += (*mb_ptr2len)(line + shl->endcol);
}
/* Loop to check if the match starts at the
@@ -3069,7 +3064,7 @@ win_line (
if (c_extra != NUL || (n_extra == 1 && c_final != NUL)) {
c = (n_extra == 1 && c_final != NUL) ? c_final : c_extra;
mb_c = c; // doesn't handle non-utf-8 multi-byte!
- if (enc_utf8 && utf_char2len(c) > 1) {
+ if (utf_char2len(c) > 1) {
mb_utf8 = true;
u8cc[0] = 0;
c = 0xc0;
@@ -3078,43 +3073,38 @@ win_line (
}
} else {
c = *p_extra;
- if (has_mbyte) {
+ mb_c = c;
+ // If the UTF-8 character is more than one byte:
+ // Decode it into "mb_c".
+ mb_l = utfc_ptr2len(p_extra);
+ mb_utf8 = false;
+ if (mb_l > n_extra) {
+ mb_l = 1;
+ } else if (mb_l > 1) {
+ mb_c = utfc_ptr2char(p_extra, u8cc);
+ mb_utf8 = true;
+ c = 0xc0;
+ }
+ if (mb_l == 0) { // at the NUL at end-of-line
+ mb_l = 1;
+ }
+
+ // If a double-width char doesn't fit display a '>' in the last column.
+ if ((wp->w_p_rl ? (col <= 0) : (col >= grid->Columns - 1))
+ && (*mb_char2cells)(mb_c) == 2) {
+ c = '>';
mb_c = c;
- if (enc_utf8) {
- // If the UTF-8 character is more than one byte:
- // Decode it into "mb_c".
- mb_l = utfc_ptr2len(p_extra);
- mb_utf8 = false;
- if (mb_l > n_extra) {
- mb_l = 1;
- } else if (mb_l > 1) {
- mb_c = utfc_ptr2char(p_extra, u8cc);
- mb_utf8 = true;
- c = 0xc0;
- }
- }
- if (mb_l == 0) /* at the NUL at end-of-line */
- mb_l = 1;
-
- /* If a double-width char doesn't fit display a '>' in the
- * last column. */
- if ((wp->w_p_rl ? (col <= 0) :
- (col >= grid->Columns - 1))
- && (*mb_char2cells)(mb_c) == 2) {
- c = '>';
- mb_c = c;
- mb_l = 1;
- mb_utf8 = false;
- multi_attr = win_hl_attr(wp, HLF_AT);
+ mb_l = 1;
+ mb_utf8 = false;
+ multi_attr = win_hl_attr(wp, HLF_AT);
- // put the pointer back to output the double-width
- // character at the start of the next line.
- n_extra++;
- p_extra--;
- } else {
- n_extra -= mb_l - 1;
- p_extra += mb_l - 1;
- }
+ // put the pointer back to output the double-width
+ // character at the start of the next line.
+ n_extra++;
+ p_extra--;
+ } else {
+ n_extra -= mb_l - 1;
+ p_extra += mb_l - 1;
}
++p_extra;
}
@@ -3129,151 +3119,113 @@ win_line (
// Get a character from the line itself.
c0 = c = *ptr;
- if (has_mbyte) {
- mb_c = c;
- if (enc_utf8) {
- // If the UTF-8 character is more than one byte: Decode it
- // into "mb_c".
- mb_l = utfc_ptr2len(ptr);
- mb_utf8 = false;
- if (mb_l > 1) {
- mb_c = utfc_ptr2char(ptr, u8cc);
- // Overlong encoded ASCII or ASCII with composing char
- // is displayed normally, except a NUL.
- if (mb_c < 0x80) {
- c0 = c = mb_c;
- }
- mb_utf8 = true;
+ mb_c = c;
+ // If the UTF-8 character is more than one byte: Decode it
+ // into "mb_c".
+ mb_l = utfc_ptr2len(ptr);
+ mb_utf8 = false;
+ if (mb_l > 1) {
+ mb_c = utfc_ptr2char(ptr, u8cc);
+ // Overlong encoded ASCII or ASCII with composing char
+ // is displayed normally, except a NUL.
+ if (mb_c < 0x80) {
+ c0 = c = mb_c;
+ }
+ mb_utf8 = true;
- /* At start of the line we can have a composing char.
- * Draw it as a space with a composing char. */
- if (utf_iscomposing(mb_c)) {
- int i;
+ // At start of the line we can have a composing char.
+ // Draw it as a space with a composing char.
+ if (utf_iscomposing(mb_c)) {
+ int i;
- for (i = MAX_MCO - 1; i > 0; i--) {
- u8cc[i] = u8cc[i - 1];
- }
- u8cc[0] = mb_c;
- mb_c = ' ';
- }
+ for (i = MAX_MCO - 1; i > 0; i--) {
+ u8cc[i] = u8cc[i - 1];
}
+ u8cc[0] = mb_c;
+ mb_c = ' ';
+ }
+ }
- if ((mb_l == 1 && c >= 0x80)
- || (mb_l >= 1 && mb_c == 0)
- || (mb_l > 1 && (!vim_isprintc(mb_c)))) {
- // Illegal UTF-8 byte: display as <xx>.
- // Non-BMP character : display as ? or fullwidth ?.
- transchar_hex((char *)extra, mb_c);
- if (wp->w_p_rl) { // reverse
- rl_mirror(extra);
- }
-
- p_extra = extra;
- c = *p_extra;
- mb_c = mb_ptr2char_adv((const char_u **)&p_extra);
- mb_utf8 = (c >= 0x80);
- n_extra = (int)STRLEN(p_extra);
- c_extra = NUL;
- c_final = NUL;
- if (area_attr == 0 && search_attr == 0) {
- n_attr = n_extra + 1;
- extra_attr = win_hl_attr(wp, HLF_8);
- saved_attr2 = char_attr; // save current attr
- }
- } else if (mb_l == 0) /* at the NUL at end-of-line */
- mb_l = 1;
- else if (p_arshape && !p_tbidi && arabic_char(mb_c)) {
- /* Do Arabic shaping. */
- int pc, pc1, nc;
- int pcc[MAX_MCO];
-
- /* The idea of what is the previous and next
- * character depends on 'rightleft'. */
- if (wp->w_p_rl) {
- pc = prev_c;
- pc1 = prev_c1;
- nc = utf_ptr2char(ptr + mb_l);
- prev_c1 = u8cc[0];
- } else {
- pc = utfc_ptr2char(ptr + mb_l, pcc);
- nc = prev_c;
- pc1 = pcc[0];
- }
- prev_c = mb_c;
+ if ((mb_l == 1 && c >= 0x80)
+ || (mb_l >= 1 && mb_c == 0)
+ || (mb_l > 1 && (!vim_isprintc(mb_c)))) {
+ // Illegal UTF-8 byte: display as <xx>.
+ // Non-BMP character : display as ? or fullwidth ?.
+ transchar_hex((char *)extra, mb_c);
+ if (wp->w_p_rl) { // reverse
+ rl_mirror(extra);
+ }
- mb_c = arabic_shape(mb_c, &c, &u8cc[0], pc, pc1, nc);
- } else
- prev_c = mb_c;
- } else { /* enc_dbcs */
- mb_l = MB_BYTE2LEN(c);
- if (mb_l == 0) /* at the NUL at end-of-line */
- mb_l = 1;
- else if (mb_l > 1) {
- /* We assume a second byte below 32 is illegal.
- * Hopefully this is OK for all double-byte encodings!
- */
- if (ptr[1] >= 32)
- mb_c = (c << 8) + ptr[1];
- else {
- if (ptr[1] == NUL) {
- /* head byte at end of line */
- mb_l = 1;
- transchar_nonprint(extra, c);
- } else {
- /* illegal tail byte */
- mb_l = 2;
- STRCPY(extra, "XX");
- }
- p_extra = extra;
- n_extra = (int)STRLEN(extra) - 1;
- c_extra = NUL;
- c_final = NUL;
- c = *p_extra++;
- if (area_attr == 0 && search_attr == 0) {
- n_attr = n_extra + 1;
- extra_attr = win_hl_attr(wp, HLF_8);
- saved_attr2 = char_attr; // save current attr
- }
- mb_c = c;
- }
- }
+ p_extra = extra;
+ c = *p_extra;
+ mb_c = mb_ptr2char_adv((const char_u **)&p_extra);
+ mb_utf8 = (c >= 0x80);
+ n_extra = (int)STRLEN(p_extra);
+ c_extra = NUL;
+ c_final = NUL;
+ if (area_attr == 0 && search_attr == 0) {
+ n_attr = n_extra + 1;
+ extra_attr = win_hl_attr(wp, HLF_8);
+ saved_attr2 = char_attr; // save current attr
}
- /* If a double-width char doesn't fit display a '>' in the
- * last column; the character is displayed at the start of the
- * next line. */
- if ((wp->w_p_rl ? (col <= 0) :
- (col >= grid->Columns - 1))
- && (*mb_char2cells)(mb_c) == 2) {
- c = '>';
- mb_c = c;
- mb_utf8 = false;
- mb_l = 1;
- multi_attr = win_hl_attr(wp, HLF_AT);
- // Put pointer back so that the character will be
- // displayed at the start of the next line.
- ptr--;
- } else if (*ptr != NUL) {
- ptr += mb_l - 1;
+ } else if (mb_l == 0) { // at the NUL at end-of-line
+ mb_l = 1;
+ } else if (p_arshape && !p_tbidi && arabic_char(mb_c)) {
+ // Do Arabic shaping.
+ int pc, pc1, nc;
+ int pcc[MAX_MCO];
+
+ // The idea of what is the previous and next
+ // character depends on 'rightleft'.
+ if (wp->w_p_rl) {
+ pc = prev_c;
+ pc1 = prev_c1;
+ nc = utf_ptr2char(ptr + mb_l);
+ prev_c1 = u8cc[0];
+ } else {
+ pc = utfc_ptr2char(ptr + mb_l, pcc);
+ nc = prev_c;
+ pc1 = pcc[0];
}
+ prev_c = mb_c;
- /* If a double-width char doesn't fit at the left side display
- * a '<' in the first column. Don't do this for unprintable
- * characters. */
- if (n_skip > 0 && mb_l > 1 && n_extra == 0) {
- n_extra = 1;
- c_extra = MB_FILLER_CHAR;
- c_final = NUL;
- c = ' ';
- if (area_attr == 0 && search_attr == 0) {
- n_attr = n_extra + 1;
- extra_attr = win_hl_attr(wp, HLF_AT);
- saved_attr2 = char_attr; // save current attr
- }
- mb_c = c;
- mb_utf8 = false;
- mb_l = 1;
+ mb_c = arabic_shape(mb_c, &c, &u8cc[0], pc, pc1, nc);
+ } else {
+ prev_c = mb_c;
+ }
+ // If a double-width char doesn't fit display a '>' in the
+ // last column; the character is displayed at the start of the
+ // next line.
+ if ((wp->w_p_rl ? (col <= 0) :
+ (col >= grid->Columns - 1))
+ && (*mb_char2cells)(mb_c) == 2) {
+ c = '>';
+ mb_c = c;
+ mb_utf8 = false;
+ mb_l = 1;
+ multi_attr = win_hl_attr(wp, HLF_AT);
+ // Put pointer back so that the character will be
+ // displayed at the start of the next line.
+ ptr--;
+ } else if (*ptr != NUL) {
+ ptr += mb_l - 1;
+ }
+
+ // If a double-width char doesn't fit at the left side display a '<' in
+ // the first column. Don't do this for unprintable characters.
+ if (n_skip > 0 && mb_l > 1 && n_extra == 0) {
+ n_extra = 1;
+ c_extra = MB_FILLER_CHAR;
+ c_final = NUL;
+ c = ' ';
+ if (area_attr == 0 && search_attr == 0) {
+ n_attr = n_extra + 1;
+ extra_attr = win_hl_attr(wp, HLF_AT);
+ saved_attr2 = char_attr; // save current attr
}
-
+ mb_c = c;
+ mb_utf8 = false;
+ mb_l = 1;
}
ptr++;
@@ -3331,11 +3283,8 @@ win_line (
char_u *p;
int len;
hlf_T spell_hlf = HLF_COUNT;
- if (has_mbyte) {
- prev_ptr = ptr - mb_l;
- v -= mb_l - 1;
- } else
- prev_ptr = ptr - 1;
+ prev_ptr = ptr - mb_l;
+ v -= mb_l - 1;
/* Use nextline[] if possible, it has the start of the
* next line concatenated. */
@@ -3445,7 +3394,7 @@ win_line (
extra_attr = win_hl_attr(wp, HLF_0);
saved_attr2 = char_attr; // save current attr
mb_c = c;
- if (enc_utf8 && utf_char2len(c) > 1) {
+ if (utf_char2len(c) > 1) {
mb_utf8 = true;
u8cc[0] = 0;
c = 0xc0;
@@ -3460,7 +3409,7 @@ win_line (
extra_attr = win_hl_attr(wp, HLF_0);
saved_attr2 = char_attr; // save current attr
mb_c = c;
- if (enc_utf8 && utf_char2len(c) > 1) {
+ if (utf_char2len(c) > 1) {
mb_utf8 = true;
u8cc[0] = 0;
c = 0xc0;
@@ -3567,7 +3516,7 @@ win_line (
extra_attr = win_hl_attr(wp, HLF_0);
saved_attr2 = char_attr; // save current attr
mb_c = c;
- if (enc_utf8 && utf_char2len(c) > 1) {
+ if (utf_char2len(c) > 1) {
mb_utf8 = true;
u8cc[0] = 0;
c = 0xc0;
@@ -3614,7 +3563,7 @@ win_line (
extra_attr = win_hl_attr(wp, HLF_AT);
n_attr = 1;
mb_c = c;
- if (enc_utf8 && utf_char2len(c) > 1) {
+ if (utf_char2len(c) > 1) {
mb_utf8 = true;
u8cc[0] = 0;
c = 0xc0;
@@ -3703,7 +3652,7 @@ win_line (
n_skip = 1;
}
mb_c = c;
- if (enc_utf8 && utf_char2len(c) > 1) {
+ if (utf_char2len(c) > 1) {
mb_utf8 = true;
u8cc[0] = 0;
c = 0xc0;
@@ -3749,9 +3698,9 @@ win_line (
&& c != NUL) {
c = wp->w_p_lcs_chars.prec;
lcs_prec_todo = NUL;
- if (has_mbyte && (*mb_char2cells)(mb_c) > 1) {
- /* Double-width character being overwritten by the "precedes"
- * character, need to fill up half the character. */
+ if ((*mb_char2cells)(mb_c) > 1) {
+ // Double-width character being overwritten by the "precedes"
+ // character, need to fill up half the character.
c_extra = MB_FILLER_CHAR;
c_final = NUL;
n_extra = 1;
@@ -3759,7 +3708,7 @@ win_line (
extra_attr = win_hl_attr(wp, HLF_AT);
}
mb_c = c;
- if (enc_utf8 && utf_char2len(c) > 1) {
+ if (utf_char2len(c) > 1) {
mb_utf8 = true;
u8cc[0] = 0;
c = 0xc0;
@@ -4033,7 +3982,7 @@ win_line (
c = wp->w_p_lcs_chars.ext;
char_attr = win_hl_attr(wp, HLF_AT);
mb_c = c;
- if (enc_utf8 && utf_char2len(c) > 1) {
+ if (utf_char2len(c) > 1) {
mb_utf8 = true;
u8cc[0] = 0;
c = 0xc0;
@@ -4074,13 +4023,13 @@ win_line (
*/
vcol_prev = vcol;
if (draw_state < WL_LINE || n_skip <= 0) {
- /*
- * Store the character.
- */
- if (has_mbyte && wp->w_p_rl && (*mb_char2cells)(mb_c) > 1) {
- /* A double-wide character is: put first halve in left cell. */
- --off;
- --col;
+ //
+ // Store the character.
+ //
+ if (wp->w_p_rl && (*mb_char2cells)(mb_c) > 1) {
+ // A double-wide character is: put first halve in left cell.
+ off--;
+ col--;
}
if (mb_utf8) {
schar_from_cc(linebuf_char[off], mb_c, u8cc);
@@ -4094,7 +4043,7 @@ win_line (
linebuf_attr[off] = char_attr;
}
- if (has_mbyte && (*mb_char2cells)(mb_c) > 1) {
+ if ((*mb_char2cells)(mb_c) > 1) {
// Need to fill two screen columns.
off++;
col++;
@@ -4154,8 +4103,8 @@ win_line (
}
- if (has_mbyte && (*mb_char2cells)(mb_c) > 1) {
- /* Need to fill two screen columns. */
+ if ((*mb_char2cells)(mb_c) > 1) {
+ // Need to fill two screen columns.
if (wp->w_p_rl) {
--boguscols;
--col;
@@ -4755,8 +4704,8 @@ win_redr_status_matches (
for (; *s != NUL; ++s) {
s += skip_status_match_char(xp, s);
clen += ptr2cells(s);
- if (has_mbyte && (l = (*mb_ptr2len)(s)) > 1) {
- STRNCPY(buf + len, s, l);
+ if ((l = (*mb_ptr2len)(s)) > 1) {
+ STRNCPY(buf + len, s, l); // NOLINT(runtime/printf)
s += l - 1;
len += l;
} else {
@@ -5024,10 +4973,11 @@ get_keymap_str (
curbuf = old_curbuf;
curwin = old_curwin;
if (p == NULL || *p == NUL) {
- if (wp->w_buffer->b_kmap_state & KEYMAP_LOADED)
+ if (wp->w_buffer->b_kmap_state & KEYMAP_LOADED) {
p = wp->w_buffer->b_p_keymap;
- else
+ } else {
p = (char_u *)"lang";
+ }
}
if (vim_snprintf((char *)buf, len, (char *)fmt, p) > len - 1) {
buf[0] = NUL;
@@ -5688,12 +5638,10 @@ next_search_hl (
shl->lnum = 0;
break;
}
- if (has_mbyte)
- matchcol += mb_ptr2len(ml);
- else
- ++matchcol;
- } else
+ matchcol += mb_ptr2len(ml);
+ } else {
matchcol = shl->rm.endpos[0].col;
+ }
shl->lnum = lnum;
if (shl->rm.regprog != NULL) {
@@ -5828,18 +5776,16 @@ void grid_fill(ScreenGrid *grid, int start_row, int end_row, int start_col,
}
for (int row = start_row; row < end_row; row++) {
- if (has_mbyte) {
- // When drawing over the right halve of a double-wide char clear
- // out the left halve. When drawing over the left halve of a
- // double wide-char clear out the right halve. Only needed in a
- // terminal.
- if (start_col > 0 && grid_fix_col(grid, start_col, row) != start_col) {
- grid_puts_len(grid, (char_u *)" ", 1, row, start_col - 1, 0);
- }
- if (end_col < grid->Columns
- && grid_fix_col(grid, end_col, row) != end_col) {
- grid_puts_len(grid, (char_u *)" ", 1, row, end_col, 0);
- }
+ // When drawing over the right halve of a double-wide char clear
+ // out the left halve. When drawing over the left halve of a
+ // double wide-char clear out the right halve. Only needed in a
+ // terminal.
+ if (start_col > 0 && grid_fix_col(grid, start_col, row) != start_col) {
+ grid_puts_len(grid, (char_u *)" ", 1, row, start_col - 1, 0);
+ }
+ if (end_col < grid->Columns
+ && grid_fix_col(grid, end_col, row) != end_col) {
+ grid_puts_len(grid, (char_u *)" ", 1, row, end_col, 0);
}
// if grid was resized (in ext_multigrid mode), the UI has no redraw updates
@@ -6770,14 +6716,9 @@ static void draw_tabline(void)
(void)shorten_dir(NameBuff);
len = vim_strsize(NameBuff);
p = NameBuff;
- if (has_mbyte)
- while (len > room) {
- len -= ptr2cells(p);
- MB_PTR_ADV(p);
- }
- else if (len > room) {
- p += len - room;
- len = room;
+ while (len > room) {
+ len -= ptr2cells(p);
+ MB_PTR_ADV(p);
}
if (len > Columns - col - 1) {
len = Columns - col - 1;
diff --git a/src/nvim/search.c b/src/nvim/search.c
index d635763acc..b09931b337 100644
--- a/src/nvim/search.c
+++ b/src/nvim/search.c
@@ -360,17 +360,16 @@ int ignorecase_opt(char_u *pat, int ic_in, int scs)
return ic;
}
-/*
- * Return TRUE if pattern "pat" has an uppercase character.
- */
-int pat_has_uppercase(char_u *pat)
+/// Returns true if pattern `pat` has an uppercase character.
+bool pat_has_uppercase(char_u *pat)
+ FUNC_ATTR_NONNULL_ALL
{
char_u *p = pat;
while (*p != NUL) {
- int l;
+ const int l = mb_ptr2len(p);
- if ((l = mb_ptr2len(p)) > 1) {
+ if (l > 1) {
if (mb_isupper(utf_ptr2char(p))) {
return true;
}
@@ -391,7 +390,7 @@ int pat_has_uppercase(char_u *pat)
p++;
}
}
- return FALSE;
+ return false;
}
const char *last_csearch(void)
@@ -744,37 +743,29 @@ int searchit(
} else
break;
- /*
- * We found a valid match, now check if there is
- * another one after it.
- * If vi-compatible searching, continue at the end
- * of the match, otherwise continue one position
- * forward.
- */
+ // We found a valid match, now check if there is
+ // another one after it.
+ // If vi-compatible searching, continue at the end
+ // of the match, otherwise continue one position
+ // forward.
if (vim_strchr(p_cpo, CPO_SEARCH) != NULL) {
- if (nmatched > 1)
+ if (nmatched > 1) {
break;
+ }
matchcol = endpos.col;
- /* for empty match: advance one char */
+ // for empty match: advance one char
if (matchcol == matchpos.col
&& ptr[matchcol] != NUL) {
- if (has_mbyte)
- matchcol +=
- (*mb_ptr2len)(ptr + matchcol);
- else
- ++matchcol;
+ matchcol += mb_ptr2len(ptr + matchcol);
}
} else {
- /* Stop when the match is in a next line. */
- if (matchpos.lnum > 0)
+ // Stop when the match is in a next line.
+ if (matchpos.lnum > 0) {
break;
+ }
matchcol = matchpos.col;
if (ptr[matchcol] != NUL) {
- if (has_mbyte)
- matchcol +=
- (*mb_ptr2len)(ptr + matchcol);
- else
- ++matchcol;
+ matchcol += mb_ptr2len(ptr + matchcol);
}
}
if (ptr[matchcol] == NUL
@@ -3229,37 +3220,18 @@ static int in_html_tag(int end_tag)
int lc = NUL;
pos_T pos;
- if (enc_dbcs) {
- char_u *lp = NULL;
-
- /* We search forward until the cursor, because searching backwards is
- * very slow for DBCS encodings. */
- for (p = line; p < line + curwin->w_cursor.col; MB_PTR_ADV(p)) {
- if (*p == '>' || *p == '<') {
- lc = *p;
- lp = p;
- }
- }
- if (*p != '<') { // check for '<' under cursor
- if (lc != '<') {
- return false;
- }
- p = lp;
- }
- } else {
- for (p = line + curwin->w_cursor.col; p > line; ) {
- if (*p == '<') { // find '<' under/before cursor
- break;
- }
- MB_PTR_BACK(line, p);
- if (*p == '>') { // find '>' before cursor
- break;
- }
+ for (p = line + curwin->w_cursor.col; p > line; ) {
+ if (*p == '<') { // find '<' under/before cursor
+ break;
}
- if (*p != '<') {
- return false;
+ MB_PTR_BACK(line, p);
+ if (*p == '>') { // find '>' before cursor
+ break;
}
}
+ if (*p != '<') {
+ return false;
+ }
pos.lnum = curwin->w_cursor.lnum;
pos.col = (colnr_T)(p - line);
@@ -3655,16 +3627,14 @@ find_next_quote(
for (;; ) {
c = line[col];
- if (c == NUL)
+ if (c == NUL) {
return -1;
- else if (escape != NULL && vim_strchr(escape, c))
- ++col;
- else if (c == quotechar)
+ } else if (escape != NULL && vim_strchr(escape, c)) {
+ col++;
+ } else if (c == quotechar) {
break;
- if (has_mbyte)
- col += (*mb_ptr2len)(line + col);
- else
- ++col;
+ }
+ col += mb_ptr2len(line + col);
}
return col;
}
diff --git a/test/unit/search_spec.lua b/test/unit/search_spec.lua
new file mode 100644
index 0000000000..3c2d485e0e
--- /dev/null
+++ b/test/unit/search_spec.lua
@@ -0,0 +1,33 @@
+local helpers = require("test.unit.helpers")(after_each)
+local itp = helpers.gen_itp(it)
+
+local to_cstr = helpers.to_cstr
+local eq = helpers.eq
+
+local search = helpers.cimport("./src/nvim/search.h")
+
+itp('pat_has_uppercase', function()
+ -- works on empty string
+ eq(false, search.pat_has_uppercase(to_cstr("")))
+
+ -- works with utf uppercase
+ eq(false, search.pat_has_uppercase(to_cstr("ä")))
+ eq(true, search.pat_has_uppercase(to_cstr("Ä")))
+ eq(true, search.pat_has_uppercase(to_cstr("äaÅ")))
+
+ -- works when pat ends with backslash
+ eq(false, search.pat_has_uppercase(to_cstr("\\")))
+ eq(false, search.pat_has_uppercase(to_cstr("ab$\\")))
+
+ -- skips escaped characters
+ eq(false, search.pat_has_uppercase(to_cstr("\\Ab")))
+ eq(true, search.pat_has_uppercase(to_cstr("\\AU")))
+
+ -- skips _X escaped characters
+ eq(false, search.pat_has_uppercase(to_cstr("\\_Ab")))
+ eq(true, search.pat_has_uppercase(to_cstr("\\_AU")))
+
+ -- skips %X escaped characters
+ eq(false, search.pat_has_uppercase(to_cstr("aa\\%Ab")))
+ eq(true, search.pat_has_uppercase(to_cstr("aab\\%AU")))
+end)
generated by cgit (git 2.34.1) at 2025-06-02 13:03:28 +0000