aboutsummaryrefslogtreecommitdiff
path: root/src/nvim/spell.c
diff options
context:
space:
mode:
Diffstat (limited to 'src/nvim/spell.c')
-rw-r--r--src/nvim/spell.c172
1 files changed, 83 insertions, 89 deletions
diff --git a/src/nvim/spell.c b/src/nvim/spell.c
index d4f49bffb2..84aeeda2bf 100644
--- a/src/nvim/spell.c
+++ b/src/nvim/spell.c
@@ -1,3 +1,6 @@
+// This is an open source non-commercial project. Dear PVS-Studio, please check
+// it. PVS-Studio Static Code Analyzer for C, C++ and C#: http://www.viva64.com
+
// spell.c: code for spell checking
//
// See spellfile.c for the Vim spell file format.
@@ -92,6 +95,7 @@
#include "nvim/func_attr.h"
#include "nvim/getchar.h"
#include "nvim/hashtab.h"
+#include "nvim/mark.h"
#include "nvim/mbyte.h"
#include "nvim/memline.h"
#include "nvim/memory.h"
@@ -1429,12 +1433,10 @@ spell_move_to (
// the cursor.
if (dir == BACKWARD
|| lnum != wp->w_cursor.lnum
- || (lnum == wp->w_cursor.lnum
- && (wrapped
- || ((colnr_T)(curline
- ? p - buf + (ptrdiff_t)len
- : p - buf)
- > wp->w_cursor.col)))) {
+ || wrapped
+ || ((colnr_T)(curline
+ ? p - buf + (ptrdiff_t)len
+ : p - buf) > wp->w_cursor.col)) {
if (has_syntax) {
col = (int)(p - buf);
(void)syn_get_id(wp, lnum, (colnr_T)col,
@@ -1482,21 +1484,23 @@ spell_move_to (
return found_len;
}
- if (curline)
+ if (curline) {
break; // only check cursor line
+ }
+
+ // If we are back at the starting line and searched it again there
+ // is no match, give up.
+ if (lnum == wp->w_cursor.lnum && wrapped) {
+ break;
+ }
// Advance to next line.
if (dir == BACKWARD) {
- // If we are back at the starting line and searched it again there
- // is no match, give up.
- if (lnum == wp->w_cursor.lnum && wrapped)
- break;
-
- if (lnum > 1)
- --lnum;
- else if (!p_ws)
+ if (lnum > 1) {
+ lnum--;
+ } else if (!p_ws) {
break; // at first line and 'nowrapscan'
- else {
+ } else {
// Wrap around to the end of the buffer. May search the
// starting line again and accept the last match.
lnum = wp->w_buffer->b_ml.ml_line_count;
@@ -1521,8 +1525,9 @@ spell_move_to (
// If we are back at the starting line and there is no match then
// give up.
- if (lnum == wp->w_cursor.lnum && (!found_one || wrapped))
+ if (lnum == wp->w_cursor.lnum && !found_one) {
break;
+ }
// Skip the characters at the start of the next line that were
// included in a match crossing line boundaries.
@@ -2066,7 +2071,7 @@ char_u *did_set_spelllang(win_T *wp)
// destroying the buffer we are using...
if (!bufref_valid(&bufref)) {
ret_msg =
- (char_u *)"E797: SpellFileMissing autocommand deleted buffer";
+ (char_u *)N_("E797: SpellFileMissing autocommand deleted buffer");
goto theend;
}
}
@@ -2526,8 +2531,7 @@ void clear_spell_chartab(spelltab_T *sp)
}
}
-// Init the chartab used for spelling. Only depends on 'encoding'.
-// Called once while starting up and when 'encoding' changes.
+// Init the chartab used for spelling. Called once while starting up.
// The default is to use isalpha(), but the spell file should define the word
// characters to make it possible that 'encoding' differs from the current
// locale. For utf-8 we don't use isalpha() but our own functions.
@@ -2537,36 +2541,17 @@ void init_spell_chartab(void)
did_set_spelltab = false;
clear_spell_chartab(&spelltab);
- if (enc_dbcs) {
- // DBCS: assume double-wide characters are word characters.
- for (i = 128; i <= 255; ++i)
- if (MB_BYTE2LEN(i) == 2)
- spelltab.st_isw[i] = true;
- } else if (enc_utf8) {
- for (i = 128; i < 256; ++i) {
- int f = utf_fold(i);
- int u = utf_toupper(i);
-
- spelltab.st_isu[i] = utf_isupper(i);
- spelltab.st_isw[i] = spelltab.st_isu[i] || utf_islower(i);
- // The folded/upper-cased value is different between latin1 and
- // utf8 for 0xb5, causing E763 for no good reason. Use the latin1
- // value for utf-8 to avoid this.
- spelltab.st_fold[i] = (f < 256) ? f : i;
- spelltab.st_upper[i] = (u < 256) ? u : i;
- }
- } else {
- // Rough guess: use locale-dependent library functions.
- for (i = 128; i < 256; ++i) {
- if (vim_isupper(i)) {
- spelltab.st_isw[i] = true;
- spelltab.st_isu[i] = true;
- spelltab.st_fold[i] = vim_tolower(i);
- } else if (vim_islower(i)) {
- spelltab.st_isw[i] = true;
- spelltab.st_upper[i] = vim_toupper(i);
- }
- }
+ for (i = 128; i < 256; i++) {
+ int f = utf_fold(i);
+ int u = mb_toupper(i);
+
+ spelltab.st_isu[i] = mb_isupper(i);
+ spelltab.st_isw[i] = spelltab.st_isu[i] || mb_islower(i);
+ // The folded/upper-cased value is different between latin1 and
+ // utf8 for 0xb5, causing E763 for no good reason. Use the latin1
+ // value for utf-8 to avoid this.
+ spelltab.st_fold[i] = (f < 256) ? f : i;
+ spelltab.st_upper[i] = (u < 256) ? u : i;
}
}
@@ -2583,7 +2568,7 @@ static bool spell_iswordp(char_u *p, win_T *wp)
int c;
if (has_mbyte) {
- l = MB_BYTE2LEN(*p);
+ l = MB_PTR2LEN(p);
s = p;
if (l == 1) {
// be quick for ASCII
@@ -3154,8 +3139,13 @@ spell_find_suggest (
if (su->su_badlen >= MAXWLEN)
su->su_badlen = MAXWLEN - 1; // just in case
STRLCPY(su->su_badword, su->su_badptr, su->su_badlen + 1);
- (void)spell_casefold(su->su_badptr, su->su_badlen,
- su->su_fbadword, MAXWLEN);
+ (void)spell_casefold(su->su_badptr, su->su_badlen, su->su_fbadword, MAXWLEN);
+
+ // TODO(vim): make this work if the case-folded text is longer than the
+ // original text. Currently an illegal byte causes wrong pointer
+ // computations.
+ su->su_fbadword[su->su_badlen] = NUL;
+
// get caps flags for bad word
su->su_badflags = badword_captype(su->su_badptr,
su->su_badptr + su->su_badlen);
@@ -3231,26 +3221,25 @@ spell_find_suggest (
// Find suggestions by evaluating expression "expr".
static void spell_suggest_expr(suginfo_T *su, char_u *expr)
{
- list_T *list;
- listitem_T *li;
int score;
const char *p;
// The work is split up in a few parts to avoid having to export
// suginfo_T.
// First evaluate the expression and get the resulting list.
- list = eval_spell_expr(su->su_badword, expr);
+ list_T *const list = eval_spell_expr(su->su_badword, expr);
if (list != NULL) {
// Loop over the items in the list.
- for (li = list->lv_first; li != NULL; li = li->li_next)
- if (li->li_tv.v_type == VAR_LIST) {
+ TV_LIST_ITER(list, li, {
+ if (TV_LIST_ITEM_TV(li)->v_type == VAR_LIST) {
// Get the word and the score from the items.
- score = get_spellword(li->li_tv.vval.v_list, &p);
+ score = get_spellword(TV_LIST_ITEM_TV(li)->vval.v_list, &p);
if (score >= 0 && score <= su->su_maxscore) {
add_suggestion(su, &su->su_ga, (const char_u *)p, su->su_badlen,
score, 0, true, su->su_sallang, false);
}
}
+ });
tv_list_unref(list);
}
@@ -3651,7 +3640,7 @@ static void suggest_trie_walk(suginfo_T *su, langp_T *lp, char_u *fword, bool so
// word).
depth = 0;
sp = &stack[0];
- memset(sp, 0, sizeof(trystate_T));
+ memset(sp, 0, sizeof(trystate_T)); // -V512
sp->ts_curi = 1;
if (soundfold) {
@@ -4126,10 +4115,7 @@ static void suggest_trie_walk(suginfo_T *su, langp_T *lp, char_u *fword, bool so
&& goodword_ends) {
int l;
- if (has_mbyte)
- l = MB_BYTE2LEN(fword[sp->ts_fidx]);
- else
- l = 1;
+ l = MB_PTR2LEN(fword + sp->ts_fidx);
if (fword_ends) {
// Copy the skipped character to preword.
memmove(preword + sp->ts_prewordlen,
@@ -4275,8 +4261,7 @@ static void suggest_trie_walk(suginfo_T *su, langp_T *lp, char_u *fword, bool so
// Correct ts_fidx for the byte length of the
// character (we didn't check that before).
sp->ts_fidx = sp->ts_fcharstart
- + MB_BYTE2LEN(
- fword[sp->ts_fcharstart]);
+ + MB_PTR2LEN(fword + sp->ts_fcharstart);
// For changing a composing character adjust
// the score from SCORE_SUBST to
@@ -4382,11 +4367,12 @@ static void suggest_trie_walk(suginfo_T *su, langp_T *lp, char_u *fword, bool so
// results.
if (has_mbyte) {
c = mb_ptr2char(fword + sp->ts_fidx);
- stack[depth].ts_fidx += MB_BYTE2LEN(fword[sp->ts_fidx]);
- if (enc_utf8 && utf_iscomposing(c))
+ stack[depth].ts_fidx += MB_PTR2LEN(fword + sp->ts_fidx);
+ if (enc_utf8 && utf_iscomposing(c)) {
stack[depth].ts_score -= SCORE_DEL - SCORE_DELCOMP;
- else if (c == mb_ptr2char(fword + stack[depth].ts_fidx))
+ } else if (c == mb_ptr2char(fword + stack[depth].ts_fidx)) {
stack[depth].ts_score -= SCORE_DEL - SCORE_DELDUP;
+ }
} else {
++stack[depth].ts_fidx;
if (fword[sp->ts_fidx] == fword[sp->ts_fidx + 1])
@@ -4568,9 +4554,9 @@ static void suggest_trie_walk(suginfo_T *su, langp_T *lp, char_u *fword, bool so
// Undo the STATE_SWAP swap: "21" -> "12".
p = fword + sp->ts_fidx;
if (has_mbyte) {
- n = MB_BYTE2LEN(*p);
+ n = MB_PTR2LEN(p);
c = mb_ptr2char(p + n);
- memmove(p + MB_BYTE2LEN(p[n]), p, n);
+ memmove(p + MB_PTR2LEN(p + n), p, n);
mb_char2bytes(c, p);
} else {
c = *p;
@@ -4643,11 +4629,11 @@ static void suggest_trie_walk(suginfo_T *su, langp_T *lp, char_u *fword, bool so
// Undo STATE_SWAP3: "321" -> "123"
p = fword + sp->ts_fidx;
if (has_mbyte) {
- n = MB_BYTE2LEN(*p);
+ n = MB_PTR2LEN(p);
c2 = mb_ptr2char(p + n);
- fl = MB_BYTE2LEN(p[n]);
+ fl = MB_PTR2LEN(p + n);
c = mb_ptr2char(p + n + fl);
- tl = MB_BYTE2LEN(p[n + fl]);
+ tl = MB_PTR2LEN(p + n + fl);
memmove(p + fl + tl, p, n);
mb_char2bytes(c, p);
mb_char2bytes(c2, p + tl);
@@ -4706,10 +4692,10 @@ static void suggest_trie_walk(suginfo_T *su, langp_T *lp, char_u *fword, bool so
// Undo ROT3L: "231" -> "123"
p = fword + sp->ts_fidx;
if (has_mbyte) {
- n = MB_BYTE2LEN(*p);
- n += MB_BYTE2LEN(p[n]);
+ n = MB_PTR2LEN(p);
+ n += MB_PTR2LEN(p + n);
c = mb_ptr2char(p + n);
- tl = MB_BYTE2LEN(p[n]);
+ tl = MB_PTR2LEN(p + n);
memmove(p + tl, p, n);
mb_char2bytes(c, p);
} else {
@@ -4759,9 +4745,9 @@ static void suggest_trie_walk(suginfo_T *su, langp_T *lp, char_u *fword, bool so
p = fword + sp->ts_fidx;
if (has_mbyte) {
c = mb_ptr2char(p);
- tl = MB_BYTE2LEN(*p);
- n = MB_BYTE2LEN(p[tl]);
- n += MB_BYTE2LEN(p[tl + n]);
+ tl = MB_PTR2LEN(p);
+ n = MB_PTR2LEN(p + tl);
+ n += MB_PTR2LEN(p + tl + n);
memmove(p, p + tl, n);
mb_char2bytes(c, p + n);
} else {
@@ -5373,7 +5359,7 @@ add_sound_suggest (
// Find the word nr in the soundfold tree.
sfwordnr = soundfold_find(slang, goodword);
if (sfwordnr < 0) {
- EMSG2(_(e_intern2), "add_sound_suggest()");
+ internal_error("add_sound_suggest()");
return;
}
@@ -7161,7 +7147,7 @@ void ex_spelldump(exarg_T *eap)
set_option_value("spl", dummy, (char *)spl, OPT_LOCAL);
xfree(spl);
- if (!bufempty()) {
+ if (!BUFEMPTY()) {
return;
}
@@ -7382,16 +7368,24 @@ static void dump_word(slang_T *slang, char_u *word, char_u *pat, int *dir, int d
if ((flags & (WF_BANNED | WF_RARE | WF_REGION)) || keepcap) {
STRCPY(badword, p);
STRCAT(badword, "/");
- if (keepcap)
+ if (keepcap) {
STRCAT(badword, "=");
- if (flags & WF_BANNED)
+ }
+ if (flags & WF_BANNED) {
STRCAT(badword, "!");
- else if (flags & WF_RARE)
+ } else if (flags & WF_RARE) {
STRCAT(badword, "?");
- if (flags & WF_REGION)
- for (i = 0; i < 7; ++i)
- if (flags & (0x10000 << i))
- sprintf((char *)badword + STRLEN(badword), "%d", i + 1);
+ }
+ if (flags & WF_REGION) {
+ for (i = 0; i < 7; i++) {
+ if (flags & (0x10000 << i)) {
+ const size_t badword_len = STRLEN(badword);
+ snprintf((char *)badword + badword_len,
+ sizeof(badword) - badword_len,
+ "%d", i + 1);
+ }
+ }
+ }
p = badword;
}
generated by cgit (git 2.34.1) at 2025-11-01 06:47:07 +0000