more charset improvements.

[apps/madmutt.git] / charset.c

diff --git a/charset.c b/charset.c
index a1541f0..8cea280 100644 (file)
--- a/charset.c
+++ b/charset.c
@@ -37,10 +37,10 @@
 #  define EILSEQ EINVAL
 #endif
 
-char *Charset;
+char *Charset = NULL;
 int Charset_is_utf8 = 0;
 
-void mutt_set_langinfo_charset(void)
+void charset_initialize(void)
 {
 #ifdef HAVE_LANGINFO_CODESET
     char buff[LONG_STRING];
@@ -50,32 +50,43 @@ void mutt_set_langinfo_charset(void)
     mutt_canonical_charset(buff2, sizeof(buff2), buff);
 
     /* finally, set $charset */
-    if (!(Charset = m_strdup(buff2)))
+    if (!m_strisempty(buff2)) {
+        m_strreplace(&Charset, buff2);
+    } else
+#endif
+        m_strreplace(&Charset, "iso-8859-1");
+
+    Charset_is_utf8 = !strcmp(Charset, "utf-8");
+#ifdef HAVE_BIND_TEXTDOMAIN_CODESET
+    bind_textdomain_codeset(PACKAGE, Charset);
 #endif
-        Charset = m_strdup("iso-8859-1");
 }
 
+
 #include "charset.gperf"
 
 void mutt_canonical_charset(char *dest, ssize_t dlen, const char *name)
 {
     const struct cset_pair *cp;
     char scratch[LONG_STRING];
-    int i;
+    const char *p;
+    int i = 0;
 
     // canonize name: only keep a-z0-9 and dots, put into lowercase
-    for (i = 0; i < ssizeof(scratch); i++) {
-        if (isalnum(*name) || *name == '.') {
-            scratch[i] = tolower((unsigned char)*name);
-        } else
-        if (!*name || *name == ':' || i + 1 == ssizeof(scratch)) {
-            scratch[i] = '\0';
-            break;
+    for (p = name; *p && *p != ':' && i < ssizeof(scratch) - 1; p++) {
+        if (isalnum(*p) || *p== '.') {
+            scratch[i++] = tolower((unsigned char)*p);
         }
     }
+    scratch[i] = '\0';
 
     cp = mutt_canonical_charset_aux(scratch, strlen(scratch));
-    m_strcpy(dest, dlen, cp ? cp->pref : name);
+    if (cp) {
+        m_strcpy(dest, dlen, cp->pref);
+    } else {
+        m_strcpy(dest, dlen, name);
+        m_strtolower(dest);
+    }
 }
 
 static int mutt_chscmp(const char *s, const char *chs)
@@ -89,12 +100,12 @@ static int mutt_chscmp(const char *s, const char *chs)
     return !strcmp(buffer, chs);
 }
 
-int mutt_is_utf8(const char *s)
+int charset_is_utf8(const char *s)
 {
     return mutt_chscmp(s, "utf-8");
 }
 
-int mutt_is_us_ascii(const char *s)
+int charset_is_us_ascii(const char *s)
 {
     return mutt_chscmp(s, "us-ascii");
 }
@@ -226,9 +237,9 @@ int mutt_convert_string (char **ps, const char *from, const char *to,
     const char **inrepls = NULL;
     const char *outrepl = NULL;
 
-    if (mutt_is_utf8 (to))
+    if (charset_is_utf8 (to))
       outrepl = "\357\277\275";
-    else if (mutt_is_utf8 (from))
+    else if (charset_is_utf8 (from))
       inrepls = repls;
     else
       outrepl = "?";
@@ -289,7 +300,7 @@ FGETCONV *fgetconv_open (FILE * file, const char *from, const char *to,
     fc->p = fc->ob = fc->bufo;
     fc->ib = fc->bufi;
     fc->ibl = 0;
-    fc->inrepls = mutt_is_utf8 (to) ? repls : repls + 1;
+    fc->inrepls = charset_is_utf8 (to) ? repls : repls + 1;
   }
   else
     fc = p_new(struct fgetconv_s, 1);
@@ -459,18 +470,6 @@ int mutt_convert_nonmime_string (char **ps)
   return -1;
 }
 
-void mutt_set_charset (char *charset)
-{
-    char buffer[STRING];
-
-    mutt_canonical_charset (buffer, sizeof (buffer), charset);
-    Charset_is_utf8 = !strcmp(buffer, "utf-8");
-
-#ifdef HAVE_BIND_TEXTDOMAIN_CODESET
-    bind_textdomain_codeset (PACKAGE, buffer);
-#endif
-}
-
 wchar_t replacement_char(void)
 {
     return Charset_is_utf8 ? 0xfffd : '?';