JasonWoof Got questions, comments, patches, etc.? Contact Jason Woofenden
Add support for multiple charset definitions
[st.git] / st.c
diff --git a/st.c b/st.c
index 84f9abb..77ea0c8 100644 (file)
--- a/st.c
+++ b/st.c
@@ -137,6 +137,16 @@ enum term_mode {
                          |MODE_MOUSEMANY,
 };
 
+enum charset {
+       CS_GRAPHIC0,
+       CS_GRAPHIC1,
+       CS_UK,
+       CS_USA,
+       CS_MULTI,
+       CS_GER,
+       CS_FIN
+};
+
 enum escape_state {
        ESC_START      = 1,
        ESC_CSI        = 2,
@@ -216,6 +226,9 @@ typedef struct {
        int bot;      /* bottom scroll limit */
        int mode;     /* terminal mode flags */
        int esc;      /* escape state flags */
+       char trantbl[4]; /* charset table translation */
+       int charset;  /* current charset */
+       int icharset; /* selected charset for sequence */
        bool numlock; /* lock numbers in keyboard */
        bool *tabs;
 } Term;
@@ -367,6 +380,8 @@ static void tsetmode(bool, bool, int *, int);
 static void tfulldirt(void);
 static void techo(char *, int);
 static long tdefcolor(int *, int *, int);
+static void tselcs(void);
+static void tdeftran(char);
 static inline bool match(uint, uint);
 static void ttynew(void);
 static void ttyread(void);
@@ -1369,6 +1384,8 @@ treset(void) {
        term.top = 0;
        term.bot = term.row - 1;
        term.mode = MODE_WRAP;
+       memset(term.trantbl, sizeof(term.trantbl), CS_USA);
+       term.charset = 0;
 
        tclearregion(0, 0, term.col-1, term.row-1);
        tmoveto(0, 0);
@@ -2260,6 +2277,33 @@ techo(char *buf, int len) {
 }
 
 void
+tdeftran(char ascii) {
+       char c, (*bp)[2];
+       static char tbl[][2] = {
+               {'0', CS_GRAPHIC0}, {'1', CS_GRAPHIC1}, {'A', CS_UK},
+               {'B', CS_USA},      {'<', CS_MULTI},    {'K', CS_GER},
+               {'5', CS_FIN},      {'C', CS_FIN},
+               {0, 0}
+       };
+
+       for (bp = &tbl[0]; (c = (*bp)[0]) && c != ascii; ++bp)
+               /* nothing */;
+
+       if (c == 0)
+               fprintf(stderr, "esc unhandled charset: ESC ( %c\n", ascii);
+       else
+               term.trantbl[term.icharset] = (*bp)[1];
+}
+
+void
+tselcs(void) {
+       if (term.trantbl[term.charset] == CS_GRAPHIC0)
+               term.c.attr.mode |= ATTR_GFX;
+       else
+               term.c.attr.mode &= ~ATTR_GFX;
+}
+
+void
 tputc(char *c, int len) {
        uchar ascii = *c;
        bool control = ascii < '\x20' || ascii == 0177;
@@ -2351,13 +2395,12 @@ tputc(char *c, int len) {
                        term.esc = ESC_START;
                        return;
                case '\016': /* SO */
+                       term.charset = 0;
+                       tselcs();
+                       return;
                case '\017': /* SI */
-                       /*
-                        * Different charsets are hard to handle. Applications
-                        * should use the right alt charset escapes for the
-                        * only reason they still exist: line drawing. The
-                        * rest is incompatible history st should not support.
-                        */
+                       term.charset = 1;
+                       tselcs();
                        return;
                case '\032': /* SUB */
                case '\030': /* CAN */
@@ -2385,22 +2428,8 @@ tputc(char *c, int len) {
                        if(ascii == '\\')
                                strhandle();
                } else if(term.esc & ESC_ALTCHARSET) {
-                       switch(ascii) {
-                       case '0': /* Line drawing set */
-                               term.c.attr.mode |= ATTR_GFX;
-                               break;
-                       case 'B': /* USASCII */
-                               term.c.attr.mode &= ~ATTR_GFX;
-                               break;
-                       case 'A': /* UK (IGNORED) */
-                       case '<': /* multinational charset (IGNORED) */
-                       case '5': /* Finnish (IGNORED) */
-                       case 'C': /* Finnish (IGNORED) */
-                       case 'K': /* German (IGNORED) */
-                               break;
-                       default:
-                               fprintf(stderr, "esc unhandled charset: ESC ( %c\n", ascii);
-                       }
+                       tdeftran(ascii);
+                       tselcs();
                        term.esc = 0;
                } else if(term.esc & ESC_TEST) {
                        if(ascii == '8') { /* DEC screen alignment test. */
@@ -2431,13 +2460,12 @@ tputc(char *c, int len) {
                                term.esc |= ESC_STR;
                                break;
                        case '(': /* set primary charset G0 */
+                       case ')': /* set secondary charset G1 */
+                       case '*': /* set tertiary charset G2 */
+                       case '+': /* set quaternary charset G3 */
+                               term.icharset = ascii - '(';
                                term.esc |= ESC_ALTCHARSET;
                                break;
-                       case ')': /* set secondary charset G1 (IGNORED) */
-                       case '*': /* set tertiary charset G2 (IGNORED) */
-                       case '+': /* set quaternary charset G3 (IGNORED) */
-                               term.esc = 0;
-                               break;
                        case 'D': /* IND -- Linefeed */
                                if(term.c.y == term.bot) {
                                        tscrollup(term.top, 1);