Rationalise null pointer checks in both decode_codepage functions, so

[u/mdw/putty] / unix / uxucs.c

diff --git a/unix/uxucs.c b/unix/uxucs.c
index 8731239..5d3d5af 100644 (file)
--- a/unix/uxucs.c
+++ b/unix/uxucs.c
@@ -8,6 +8,7 @@
 #include <time.h>
 
 #include "putty.h"
+#include "charset.h"
 #include "terminal.h"
 #include "misc.h"
 
@@ -20,14 +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;
 
-       setlocale(LC_CTYPE, "");
+       memset(&state, 0, sizeof state);
 
        while (mblen > 0) {
            size_t i = mbrtowc(wcstr+n, mbstr, (size_t)mblen, &state);
@@ -38,8 +39,6 @@ int mb_to_wc(int codepage, int flags, char *mbstr, int mblen,
            mblen -= i;
        }
 
-       setlocale(LC_CTYPE, "C");
-
        return n;
     } else if (codepage == CS_NONE) {
        int n = 0;
@@ -57,7 +56,7 @@ 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)
 {
@@ -67,10 +66,10 @@ 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;
 
-       setlocale(LC_CTYPE, "");
+       memset(&state, 0, sizeof state);
 
        while (wclen > 0) {
            int i = wcrtomb(output, wcstr[0], &state);
@@ -82,8 +81,6 @@ int wc_to_mb(int codepage, int flags, wchar_t *wcstr, int wclen,
            wclen--;
        }
 
-       setlocale(LC_CTYPE, "C");
-
        return n;
     } else if (codepage == CS_NONE) {
        int n = 0;
@@ -105,8 +102,8 @@ 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 init_ucs(struct unicode_data *ucsdata, char *linecharset,
+            int utf8_override, int font_charset, int vtmode)
 {
     int i, ret = 0;
 
@@ -119,12 +116,27 @@ 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 = 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.
      */
-    ucsdata->line_codepage = charset_from_mimeenc(linecharset);
     if (ucsdata->line_codepage == CS_NONE)
-       ucsdata->line_codepage = charset_from_xenc(linecharset);
+       ucsdata->line_codepage = decode_codepage(linecharset);
 
     /*
      * If line_codepage is _still_ CS_NONE, we assume we're using
@@ -144,7 +156,8 @@ 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;
@@ -177,8 +190,18 @@ int init_ucs(struct unicode_data *ucsdata,
            0x23bb, 0x2500, 0x23bc, 0x23bd, 0x251c, 0x2524, 0x2534, 0x252c,
            0x2502, 0x2264, 0x2265, 0x03c0, 0x2260, 0x00a3, 0x00b7, 0x0020
        };
+       static const wchar_t unitab_xterm_poorman[32] =
+           L"*#****o~**+++++-----++++|****L. ";
+
+       const wchar_t *ptr;
+
+       if (vtmode == VT_POORMAN)
+           ptr = unitab_xterm_poorman;
+       else
+           ptr = unitab_xterm_std;
+
        if (i >= 0x5F && i < 0x7F)
-           ucsdata->unitab_xterm[i] = unitab_xterm_std[i & 0x1F];
+           ucsdata->unitab_xterm[i] = ptr[i & 0x1F];
        else
            ucsdata->unitab_xterm[i] = ucsdata->unitab_line[i];
     }
@@ -188,7 +211,8 @@ 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;
@@ -218,3 +242,30 @@ int init_ucs(struct unicode_data *ucsdata,
 
     return ret;
 }
+
+const char *cp_name(int codepage)
+{
+    if (codepage == CS_NONE)
+       return "Use font encoding";
+    return charset_to_localenc(codepage);
+}
+
+const char *cp_enumerate(int index)
+{
+    int charset;
+    charset = charset_localenc_nth(index);
+    if (charset == CS_NONE) {
+        /* "Use font encoding" comes after all the named charsets */
+        if (charset_localenc_nth(index-1) != CS_NONE)
+            return "Use font encoding";
+       return NULL;
+    }
+    return charset_to_localenc(charset);
+}
+
+int decode_codepage(char *cp_name)
+{
+    if (!cp_name || !*cp_name)
+       return CS_UTF8;
+    return charset_from_localenc(cp_name);
+}