deepend/zoitechat
1
0
Fork 0
mirror of https://github.com/ZoiteChat/zoitechat.git synced 2026-03-10 07:50:19 +00:00
Code Issues Packages Projects Releases Wiki Activity

Updated emoji-token normalization to first apply Unicode composition normalization (G_NORMALIZE_ALL_COMPOSE) before matching, so canonically-equivalent sequences compare reliably in highlight checks.

Browse Source 
Expanded ignored codepoints during highlight token comparison to include zero-width joiner (U+200D) in addition to variation selectors (U+FE0E/U+FE0F), which helps emoji entered from different input methods still match configured “extra highlight words.”

This logic is used by the alert/highlight matching path (alert_match_word) that compares configured extra words against extracted message tokens.
This commit is contained in:
deepend
2026-02-24 11:18:46 -07:00
parent fac15811c2
commit 0935799f7d
1 changed files with 54 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files

61
src/common/inbound.c
View File

@@ -222,15 +222,51 @@ inbound_privmsg (server *serv, char *from, char *ip, char *text, int id,
/* used for Alerts section. Masks can be separated by commas and spaces. */
static char *
alert_normalize_word (const char *text)
{
GString *normalized;
char *composed;
const char *p;
composed = g_utf8_normalize (text, -1, G_NORMALIZE_ALL_COMPOSE);
if (!composed)
composed = g_strdup (text);
normalized = g_string_sized_new (strlen (composed));
p = composed;
while (*p)
{
gunichar ch = g_utf8_get_char ((const guchar *)p);
/* Ignore selector/joiner codepoints that vary by input method. */
if (ch != 0x200D && ch != 0xFE0E && ch != 0xFE0F)
g_string_append_unichar (normalized, ch);
p = g_utf8_next_char (p);
}
g_free (composed);
return g_string_free (normalized, FALSE);
}
gboolean
alert_match_word (char *word, char *masks)
{
char *p = masks;
char endchar;
char *word_normalized;
char *mask_normalized;
int res;
word_normalized = alert_normalize_word (word);
if (masks[0] == 0)
{
g_free (word_normalized);
return FALSE;
}
while (1)
{
@@ -239,15 +275,23 @@ alert_match_word (char *word, char *masks)
{
endchar = *p;
*p = 0;
res = match (g_strchug (masks), word);
mask_normalized = alert_normalize_word (g_strchug (masks));
res = match (mask_normalized, word_normalized);
g_free (mask_normalized);
*p = endchar;
if (res)
{
g_free (word_normalized);
return TRUE; /* yes, matched! */
}
masks = p + 1;
if (*p == 0)
{
g_free (word_normalized);
return FALSE;
}
}
p++;
}
@@ -259,6 +303,7 @@ alert_match_text (char *text, char *masks)
unsigned char *p = text;
unsigned char endchar;
gunichar ch;
GUnicodeType ch_type;
int res;
if (masks[0] == 0)
@@ -267,6 +312,7 @@ alert_match_text (char *text, char *masks)
while (1)
{
ch = g_utf8_get_char (p);
ch_type = g_unichar_type (ch);
if (g_unichar_isdigit (ch) || g_unichar_isalpha (ch))
{
@@ -285,17 +331,18 @@ alert_match_text (char *text, char *masks)
}
/* Symbols (including emoji) can be part of highlighted words. */
if (!g_unichar_isspace (ch) && !g_unichar_ispunct (ch) &&
!g_unichar_iscntrl (ch))
if (ch_type == G_UNICODE_MATH_SYMBOL ||
ch_type == G_UNICODE_CURRENCY_SYMBOL ||
ch_type == G_UNICODE_MODIFIER_SYMBOL ||
ch_type == G_UNICODE_OTHER_SYMBOL)
{
p += g_utf8_skip [p[0]];
continue;
}
/* if it's a 0, space or comma, the word has ended. */
if (*p == 0 || *p == ' ' || *p == ',' ||
/* if it's anything BUT a letter, the word has ended. */
(!g_unichar_isalpha (ch)))
/* Delimiters end the word. */
if (*p == 0 || g_unichar_isspace (ch) || g_unichar_ispunct (ch) ||
g_unichar_iscntrl (ch))
{
endchar = *p;
*p = 0;