X-Git-Url: https://git.distorted.org.uk/u/mdw/putty/blobdiff_plain/3900c2d642c2960fdac52a9afea854a375441fcc..863c5362e83ad7d153bc39fdbc526f9d0674a0c2:/unix/uxucs.c

diff --git a/unix/uxucs.c b/unix/uxucs.c
index 601b35da..20f25d4e 100644
--- a/unix/uxucs.c
+++ b/unix/uxucs.c
@@ -21,13 +21,14 @@ int is_dbcs_leadbyte(int codepage, char byte)
     return 0;			       /* we don't do DBCS */
 }
 
-int mb_to_wc(int codepage, int flags, char *mbstr, int mblen,
+int mb_to_wc(int codepage, int flags, const char *mbstr, int mblen,
 	     wchar_t *wcstr, int wclen)
 {
     if (codepage == DEFAULT_CODEPAGE) {
 	int n = 0;
-	mbstate_t state = { 0 };
+	mbstate_t state;
 
+	memset(&state, 0, sizeof state);
 	setlocale(LC_CTYPE, "");
 
 	while (mblen > 0) {
@@ -58,7 +59,7 @@ int mb_to_wc(int codepage, int flags, char *mbstr, int mblen,
 				  NULL, NULL, 0);
 }
 
-int wc_to_mb(int codepage, int flags, wchar_t *wcstr, int wclen,
+int wc_to_mb(int codepage, int flags, const wchar_t *wcstr, int wclen,
 	     char *mbstr, int mblen, char *defchr, int *defused,
 	     struct unicode_data *ucsdata)
 {
@@ -68,9 +69,10 @@ int wc_to_mb(int codepage, int flags, wchar_t *wcstr, int wclen,
 
     if (codepage == DEFAULT_CODEPAGE) {
 	char output[MB_LEN_MAX];
-	mbstate_t state = { 0 };
+	mbstate_t state;
 	int n = 0;
 
+	memset(&state, 0, sizeof state);
 	setlocale(LC_CTYPE, "");
 
 	while (wclen > 0) {
@@ -106,8 +108,8 @@ int wc_to_mb(int codepage, int flags, wchar_t *wcstr, int wclen,
 /*
  * Return value is TRUE if pterm is to run in direct-to-font mode.
  */
-int init_ucs(struct unicode_data *ucsdata, 
-	     char *linecharset, int font_charset, int vtmode)
+int init_ucs(struct unicode_data *ucsdata, char *linecharset,
+	     int utf8_override, int font_charset, int vtmode)
 {
     int i, ret = 0;
 
@@ -120,10 +122,27 @@ int init_ucs(struct unicode_data *ucsdata,
     ucsdata->font_codepage = -1;
 
     /*
-     * line_codepage should be decoded from the specification in
-     * cfg.
+     * If utf8_override is set and the POSIX locale settings
+     * dictate a UTF-8 character set, then just go straight for
+     * UTF-8.
      */
-    ucsdata->line_codepage = decode_codepage(linecharset);
+    ucsdata->line_codepage = CS_NONE;
+    if (utf8_override) {
+	const char *s;
+	if (((s = getenv("LC_ALL"))   && *s) ||
+	    ((s = getenv("LC_CTYPE")) && *s) ||
+	    ((s = getenv("LANG"))     && *s)) {
+	    if (strstr(s, "UTF-8"))
+		ucsdata->line_codepage = CS_UTF8;
+	}
+    }
+
+    /*
+     * Failing that, line_codepage should be decoded from the
+     * specification in conf.
+     */
+    if (ucsdata->line_codepage == CS_NONE)
+	ucsdata->line_codepage = decode_codepage(linecharset);
 
     /*
      * If line_codepage is _still_ CS_NONE, we assume we're using
@@ -143,7 +162,8 @@ int init_ucs(struct unicode_data *ucsdata,
      * in the line codepage into Unicode.
      */
     for (i = 0; i < 256; i++) {
-	char c[1], *p;
+	char c[1];
+        const char *p;
 	wchar_t wc[1];
 	int len;
 	c[0] = i;
@@ -197,7 +217,8 @@ int init_ucs(struct unicode_data *ucsdata,
      * simply CP437.
      */
     for (i = 0; i < 256; i++) {
-	char c[1], *p;
+	char c[1];
+        const char *p;
 	wchar_t wc[1];
 	int len;
 	c[0] = i;