commit 7a7405571c8f7c34bbbd2142a1efae715454d163
parent 1270657602a9061e586e58ee5f6521a8c4910e8b
Author: Benno Schulenberg <bensberg@telfort.nl>
Date: Wed, 21 May 2025 12:04:10 +0200
tweaks: rename 'use_utf8' to 'using_utf8', and make it a global variable
This elides a function call in some parts of the code.
Diffstat:
7 files changed, 33 insertions(+), 53 deletions(-)
diff --git a/src/chars.c b/src/chars.c
@@ -23,27 +23,11 @@
#include <ctype.h>
#include <string.h>
-
-static bool use_utf8 = FALSE;
- /* Whether we've enabled UTF-8 support. */
-
#ifdef ENABLE_UTF8
#include <wchar.h>
#include <wctype.h>
-
-/* Enable UTF-8 support. */
-void utf8_init(void)
-{
- use_utf8 = TRUE;
-}
#endif
-/* Is UTF-8 support enabled? */
-bool using_utf8(void)
-{
- return use_utf8;
-}
-
#ifdef ENABLE_SPELLER
/* Return TRUE when the given character is some kind of letter. */
bool is_alpha_char(const char *c)
@@ -98,7 +82,7 @@ bool is_blank_char(const char *c)
bool is_cntrl_char(const char *c)
{
#ifdef ENABLE_UTF8
- if (use_utf8)
+ if (using_utf8)
return ((c[0] & 0xE0) == 0 || c[0] == DEL_CODE ||
((signed char)c[0] == -62 && (signed char)c[1] < -96));
else
@@ -165,7 +149,7 @@ char control_mbrep(const char *c, bool isdata)
return '@';
#ifdef ENABLE_UTF8
- if (use_utf8) {
+ if (using_utf8) {
if ((unsigned char)c[0] < 128)
return control_rep(c[0]);
else
@@ -180,7 +164,7 @@ char control_mbrep(const char *c, bool isdata)
* the number of bytes in the sequence, or -1 for an invalid sequence. */
int mbtowide(wchar_t *wc, const char *c)
{
- if ((signed char)*c < 0 && use_utf8) {
+ if ((signed char)*c < 0 && using_utf8) {
unsigned char v1 = c[0];
unsigned char v2 = c[1] ^ 0x80;
@@ -229,7 +213,7 @@ bool is_doublewidth(const char *ch)
wchar_t wc;
/* Only from U+1100 can code points have double width. */
- if ((unsigned char)*ch < 0xE1 || !use_utf8)
+ if ((unsigned char)*ch < 0xE1 || !using_utf8)
return FALSE;
if (mbtowide(&wc, ch) < 0)
@@ -244,7 +228,7 @@ bool is_zerowidth(const char *ch)
wchar_t wc;
/* Only from U+0300 can code points have zero width. */
- if ((unsigned char)*ch < 0xCC || !use_utf8)
+ if ((unsigned char)*ch < 0xCC || !using_utf8)
return FALSE;
if (mbtowide(&wc, ch) < 0)
@@ -264,7 +248,7 @@ bool is_zerowidth(const char *ch)
int char_length(const char *pointer)
{
#ifdef ENABLE_UTF8
- if ((unsigned char)*pointer > 0xC1 && use_utf8) {
+ if ((unsigned char)*pointer > 0xC1 && using_utf8) {
unsigned char c1 = pointer[0];
unsigned char c2 = pointer[1];
@@ -327,7 +311,7 @@ int collect_char(const char *string, char *thechar)
int advance_over(const char *string, size_t *column)
{
#ifdef ENABLE_UTF8
- if ((signed char)*string < 0 && use_utf8) {
+ if ((signed char)*string < 0 && using_utf8) {
/* A UTF-8 upper control code has two bytes and takes two columns. */
if (((unsigned char)string[0] == 0xC2 && (signed char)string[1] < -96)) {
*column += 2;
@@ -371,7 +355,7 @@ int advance_over(const char *string, size_t *column)
size_t step_left(const char *buf, size_t pos)
{
#ifdef ENABLE_UTF8
- if (use_utf8) {
+ if (using_utf8) {
size_t before, charlen = 0;
if (pos < 4)
@@ -422,7 +406,7 @@ int mbstrcasecmp(const char *s1, const char *s2)
int mbstrncasecmp(const char *s1, const char *s2, size_t n)
{
#ifdef ENABLE_UTF8
- if (use_utf8) {
+ if (using_utf8) {
wchar_t wc1, wc2;
while (*s1 != '\0' && *s2 != '\0' && n > 0) {
@@ -473,7 +457,7 @@ int mbstrncasecmp(const char *s1, const char *s2, size_t n)
char *mbstrcasestr(const char *haystack, const char *needle)
{
#ifdef ENABLE_UTF8
- if (use_utf8) {
+ if (using_utf8) {
size_t needle_len = mbstrlen(needle);
while (*haystack != '\0') {
@@ -535,7 +519,7 @@ char *mbrevstrcasestr(const char *haystack, const char *needle,
const char *pointer)
{
#ifdef ENABLE_UTF8
- if (use_utf8) {
+ if (using_utf8) {
size_t needle_len = mbstrlen(needle);
size_t tail_len = mbstrlen(pointer);
@@ -564,7 +548,7 @@ char *mbrevstrcasestr(const char *haystack, const char *needle,
char *mbstrchr(const char *string, const char *chr)
{
#ifdef ENABLE_UTF8
- if (use_utf8) {
+ if (using_utf8) {
bool bad_s = FALSE, bad_c = FALSE;
wchar_t ws, wc;
diff --git a/src/global.c b/src/global.c
@@ -34,6 +34,8 @@ volatile sig_atomic_t the_window_resized = FALSE;
bool on_a_vt = FALSE;
/* Whether we're running on a Linux console (a VT). */
+bool using_utf8 = FALSE;
+ /* Whether we're in a UTF-8 locale. */
bool shifted_metas = FALSE;
/* Whether any Sh-M-<letter> combo has been bound. */
@@ -1354,7 +1356,7 @@ void shortcut_init(void)
add_to_sclist(MMOST & ~MMAIN, "^B", 0, do_left, 0);
add_to_sclist(MMOST & ~MMAIN, "^F", 0, do_right, 0);
#ifdef ENABLE_UTF8
- if (using_utf8()) {
+ if (using_utf8) {
add_to_sclist(MMOST|MBROWSER|MHELP, "\xE2\x97\x82", KEY_LEFT, do_left, 0);
add_to_sclist(MMOST|MBROWSER|MHELP, "\xE2\x96\xb8", KEY_RIGHT, do_right, 0);
add_to_sclist(MSOME, "^\xE2\x97\x82", CONTROL_LEFT, to_prev_word, 0);
@@ -1384,7 +1386,7 @@ void shortcut_init(void)
add_to_sclist(MMOST, "Home", KEY_HOME, do_home, 0);
add_to_sclist(MMOST, "End", KEY_END, do_end, 0);
#ifdef ENABLE_UTF8
- if (using_utf8()) {
+ if (using_utf8) {
add_to_sclist(MMAIN|MBROWSER|MHELP, "\xE2\x96\xb4", KEY_UP, do_up, 0);
add_to_sclist(MMAIN|MBROWSER|MHELP, "\xE2\x96\xbe", KEY_DOWN, do_down, 0);
add_to_sclist(MMAIN|MBROWSER|MLINTER, "^\xE2\x96\xb4", CONTROL_UP, to_prev_block, 0);
@@ -1407,7 +1409,7 @@ void shortcut_init(void)
#endif
#ifndef NANO_TINY
#ifdef ENABLE_UTF8
- if (using_utf8()) {
+ if (using_utf8) {
add_to_sclist(MMAIN|MHELP, "M-\xE2\x96\xb4", ALT_UP, do_scroll_up, 0);
add_to_sclist(MMAIN|MHELP, "M-\xE2\x96\xbe", ALT_DOWN, do_scroll_down, 0);
} else
@@ -1492,7 +1494,7 @@ void shortcut_init(void)
add_to_sclist(MWHEREIS|MREPLACE|MREPLACEWITH|MWHEREISFILE|MFINDINHELP|MEXECUTE, "^P", 0, get_older_item, 0);
add_to_sclist(MWHEREIS|MREPLACE|MREPLACEWITH|MWHEREISFILE|MFINDINHELP|MEXECUTE, "^N", 0, get_newer_item, 0);
#ifdef ENABLE_UTF8
- if (using_utf8()) {
+ if (using_utf8) {
add_to_sclist(MWHEREIS|MREPLACE|MREPLACEWITH|MWHEREISFILE|MFINDINHELP|MEXECUTE, "\xE2\x96\xb4", KEY_UP, get_older_item, 0);
add_to_sclist(MWHEREIS|MREPLACE|MREPLACEWITH|MWHEREISFILE|MFINDINHELP|MEXECUTE, "\xE2\x96\xbe", KEY_DOWN, get_newer_item, 0);
} else
diff --git a/src/nano.c b/src/nano.c
@@ -1865,7 +1865,7 @@ int main(int argc, char **argv)
/* If setting the locale is successful and it uses UTF-8, we will
* need to use the multibyte functions for text processing. */
if (setlocale(LC_ALL, "") && strcmp(nl_langinfo(CODESET), "UTF-8") == 0)
- utf8_init();
+ using_utf8 = TRUE;
#else
setlocale(LC_ALL, "");
#endif
@@ -2374,7 +2374,7 @@ int main(int argc, char **argv)
/* If the whitespace option wasn't specified, set its default value. */
if (whitespace == NULL) {
#ifdef ENABLE_UTF8
- if (using_utf8()) {
+ if (using_utf8) {
/* A tab is shown as a Right-Pointing Double Angle Quotation Mark
* (U+00BB), and a space as a Middle Dot (U+00B7). */
whitespace = copy_of("\xC2\xBB\xC2\xB7");
@@ -2687,7 +2687,7 @@ int main(int argc, char **argv)
#define byte(n) (unsigned char)openfile->current->data[n]
/* Tell the user when the cursor sits on a BOM. */
if (openfile->current_x == 0 && byte(0) == 0xEF && byte(1) == 0xBB &&
- byte(2) == 0xBF && using_utf8()) {
+ byte(2) == 0xBF && using_utf8) {
statusline(NOTICE, _("Byte Order Mark"));
set_blankdelay_to_one();
}
diff --git a/src/prompt.c b/src/prompt.c
@@ -741,7 +741,7 @@ int ask_user(bool withall, const char *question)
#ifdef ENABLE_UTF8
/* If the received code is a UTF-8 starter byte, get also the
* continuation bytes and assemble them into one letter. */
- if (using_utf8() && 0xC0 <= kbinput && kbinput <= 0xF7) {
+ if (0xC0 <= kbinput && kbinput <= 0xF7 && using_utf8) {
int extras = (kbinput / 16) % 4 + (kbinput <= 0xCF ? 1 : 0);
while (extras <= waiting_keycodes() && extras-- > 0)
diff --git a/src/prototypes.h b/src/prototypes.h
@@ -27,6 +27,7 @@ extern volatile sig_atomic_t the_window_resized;
#endif
extern bool on_a_vt;
+extern bool using_utf8;
extern bool shifted_metas;
extern bool meta_key;
@@ -206,10 +207,6 @@ void to_last_file(void);
#endif
/* Most functions in chars.c. */
-#ifdef ENABLE_UTF8
-void utf8_init(void);
-#endif
-bool using_utf8(void);
bool is_alpha_char(const char *c);
bool is_blank_char(const char *c);
bool is_cntrl_char(const char *c);
diff --git a/src/rcfile.c b/src/rcfile.c
@@ -1585,7 +1585,7 @@ void parse_rcfile(FILE *rcstream, bool just_syntax, bool intros_only)
#ifdef ENABLE_UTF8
/* When in a UTF-8 locale, ignore arguments with invalid sequences. */
- if (using_utf8() && mbstowcs(NULL, argument, 0) == (size_t)-1) {
+ if (using_utf8 && mbstowcs(NULL, argument, 0) == (size_t)-1) {
jot_error(N_("Argument is not a valid multibyte string"));
continue;
}
diff --git a/src/winio.c b/src/winio.c
@@ -1036,7 +1036,7 @@ int parse_kbinput(WINDOW *frame)
return CONTROL_SHIFT_DELETE;
#endif
#ifdef ENABLE_UTF8
- else if (0xC0 <= keycode && keycode <= 0xFF && using_utf8()) {
+ else if (0xC0 <= keycode && keycode <= 0xFF && using_utf8) {
while (waiting_codes && 0x80 <= nextcodes[0] && nextcodes[0] <= 0xBF)
get_input(NULL);
return FOREIGN_SEQUENCE;
@@ -1085,7 +1085,7 @@ int parse_kbinput(WINDOW *frame)
return ERR;
}
#ifdef ENABLE_UTF8
- else if (byte > 0x7F && using_utf8()) {
+ else if (byte > 0x7F && using_utf8) {
/* Convert the code to the corresponding Unicode, and
* put the second byte back into the keyboard buffer. */
if (byte < 0xC0) {
@@ -1442,7 +1442,7 @@ int *parse_verbatim_kbinput(WINDOW *frame, size_t *count)
#ifdef ENABLE_UTF8
/* If the key code is a hexadecimal digit, commence Unicode input. */
- if (using_utf8() && isxdigit(keycode)) {
+ if (using_utf8 && isxdigit(keycode)) {
long unicode = assemble_unicode(keycode);
char multibyte[MB_CUR_MAX];
@@ -2230,16 +2230,13 @@ void minibar(void)
if (*this_position == '\0')
sprintf(hexadecimal, openfile->current->next ?
-#ifdef ENABLE_UTF8
- using_utf8() ? "U+000A" :
-#endif
- " 0x0A" : " ----");
+ (using_utf8 ? "U+000A" : " 0x0A") : " ----");
else if (*this_position == '\n')
sprintf(hexadecimal, " 0x00");
#ifdef ENABLE_UTF8
- else if ((unsigned char)*this_position < 0x80 && using_utf8())
+ else if ((unsigned char)*this_position < 0x80 && using_utf8)
sprintf(hexadecimal, "U+%04X", (unsigned char)*this_position);
- else if (using_utf8() && mbtowide(&widecode, this_position) > 0)
+ else if (using_utf8 && mbtowide(&widecode, this_position) > 0)
sprintf(hexadecimal, "U+%04X", (int)widecode);
#endif
else
@@ -2274,7 +2271,7 @@ void minibar(void)
/* Indicate it when the line has an anchor. */
if (openfile->current->has_anchor && namewidth + 7 < COLS)
- mvwaddstr(footwin, 0, COLS - 5 - padding, using_utf8() ? "\xE2\x80\xA0" : "+");
+ mvwaddstr(footwin, 0, COLS - 5 - padding, using_utf8 ? "\xE2\x80\xA0" : "+");
/* Display how many percent the current line is into the file. */
if (namewidth + 6 < COLS) {
@@ -2560,7 +2557,7 @@ void draw_row(int row, const char *converted, linestruct *line, size_t from_col)
wattroff(midwin, interface_color_pair[LINE_NUMBER]);
#ifndef NANO_TINY
if (line->has_anchor && (from_col == 0 || !ISSET(SOFTWRAP)))
- wprintw(midwin, using_utf8() ? "\xE2\x80\xA0" : "+");
+ wprintw(midwin, using_utf8 ? "\xE2\x80\xA0" : "+");
else
#endif
wprintw(midwin, " ");
@@ -2756,7 +2753,7 @@ void draw_row(int row, const char *converted, linestruct *line, size_t from_col)
} else if (target_column + 1 == editwincols) {
/* Defeat a VTE bug -- see https://sv.gnu.org/bugs/?55896. */
#ifdef ENABLE_UTF8
- if (using_utf8()) {
+ if (using_utf8) {
striped_char[0] = '\xC2';
striped_char[1] = '\xA0';
charlen = 2;
