Patch from Yoshida Masato to fill in the missing pieces of Windows

UTF-16 support. High Unicode characters in the terminal are now
converted back into surrogates during copy and draw operations, and
the Windows drawing code takes account of that when splitting up the
UTF-16 string for display. Meanwhile, accidental uses of wchar_t have
been replaced with 32-bit integers in parts of the cross-platform code
which were expecting not to have to deal with UTF-16.

git-svn-id: svn://svn.tartarus.org/sgt/putty@9409 cda61777-01e9-0310-a592-d414129be87e

diff --git a/ldiscucs.c b/ldiscucs.c
index c4fc726..fe0da8a 100644 (file)
--- a/ldiscucs.c
+++ b/ldiscucs.c
@@ -51,13 +51,12 @@ void luni_send(void *handle, wchar_t * widebuf, int len, int interactive)
        for (p = linebuffer, i = 0; i < len; i++) {
            unsigned long ch = widebuf[i];
 
-           if ((ch & 0xF800) == 0xD800) {
+           if (IS_SURROGATE(ch)) {
 #ifdef PLATFORM_IS_UTF16
                if (i+1 < len) {
                    unsigned long ch2 = widebuf[i+1];
-                   if ((ch & 0xFC00) == 0xD800 &&
-                       (ch2 & 0xFC00) == 0xDC00) {
-                       ch = 0x10000 + ((ch & 0x3FF) << 10) + (ch2 & 0x3FF);
+                   if (IS_SURROGATE_PAIR(ch, ch2)) {
+                       ch = FROM_SURROGATES(ch, ch2);
                        i++;
                    }
                } else

diff --git a/minibidi.c b/minibidi.c
index c0197a5..f2c68c4 100644 (file)
--- a/minibidi.c
+++ b/minibidi.c
@@ -58,7 +58,7 @@ shapetypes[(xh)-SHAPE_FIRST].type : SU) /*))*/
 #define leastGreaterEven(x) ( ((x)+2) &~ 1 )
 
 typedef struct bidi_char {
-    wchar_t origwc, wc;
+    unsigned int origwc, wc;
     unsigned short index;
 } bidi_char;
 

diff --git a/putty.h b/putty.h
index 4c338b4..93b683e 100644 (file)
--- a/putty.h
+++ b/putty.h
@@ -1126,10 +1126,10 @@ void get_unitab(int codepage, wchar_t * unitab, int ftype);
 /*
  * Exports from wcwidth.c
  */
-int mk_wcwidth(wchar_t ucs);
-int mk_wcswidth(const wchar_t *pwcs, size_t n);
-int mk_wcwidth_cjk(wchar_t ucs);
-int mk_wcswidth_cjk(const wchar_t *pwcs, size_t n);
+int mk_wcwidth(unsigned int ucs);
+int mk_wcswidth(const unsigned int *pwcs, size_t n);
+int mk_wcwidth_cjk(unsigned int ucs);
+int mk_wcswidth_cjk(const unsigned int *pwcs, size_t n);
 
 /*
  * Exports from mscrypto.c
@@ -1257,7 +1257,7 @@ void setup_config_box(struct controlbox *b, int midsession,
  * Exports from minibidi.c.
  */
 typedef struct bidi_char {
-    wchar_t origwc, wc;
+    unsigned int origwc, wc;
     unsigned short index;
 } bidi_char;
 int do_bidi(bidi_char *line, int count);
@@ -1399,4 +1399,29 @@ void timer_change_notify(long next);
 #define remove_session_from_jumplist(x) ((void)0)
 #endif
 
+/* SURROGATE PAIR */
+#ifndef IS_HIGH_SURROGATE
+#define HIGH_SURROGATE_START 0xd800
+#define HIGH_SURROGATE_END 0xdbff
+#define LOW_SURROGATE_START 0xdc00
+#define LOW_SURROGATE_END 0xdfff
+
+#define IS_HIGH_SURROGATE(wch) (((wch) >= HIGH_SURROGATE_START) && \
+                                ((wch) <= HIGH_SURROGATE_END))
+#define IS_LOW_SURROGATE(wch) (((wch) >= LOW_SURROGATE_START) && \
+                               ((wch) <= LOW_SURROGATE_END))
+#define IS_SURROGATE_PAIR(hs, ls) (IS_HIGH_SURROGATE(hs) && \
+                                   IS_LOW_SURROGATE(ls))
+#endif
+
+
+#define IS_SURROGATE(wch) (((wch) >= HIGH_SURROGATE_START) &&   \
+                           ((wch) <= LOW_SURROGATE_END))
+#define HIGH_SURROGATE_OF(codept) \
+    (HIGH_SURROGATE_START + (((codept) - 0x10000) >> 10))
+#define LOW_SURROGATE_OF(codept) \
+    (LOW_SURROGATE_START + (((codept) - 0x10000) & 0x3FF))
+#define FROM_SURROGATES(wch1, wch2) \
+    (0x10000 + (((wch1) & 0x3FF) << 10) + ((wch2) & 0x3FF))
+
 #endif

diff --git a/terminal.c b/terminal.c
index abb06e2..d23cbe9 100644 (file)
--- a/terminal.c
+++ b/terminal.c
@@ -3016,8 +3016,8 @@ static void term_out(Terminal *term)
                        width = 1;
                    if (!width)
                        width = (term->cjk_ambig_wide ?
-                                mk_wcwidth_cjk((wchar_t) c) :
-                                mk_wcwidth((wchar_t) c));
+                                mk_wcwidth_cjk((unsigned int) c) :
+                                mk_wcwidth((unsigned int) c));
 
                    if (term->wrapnext && term->wrap && width > 0) {
                        cline->lattr |= LATTR_WRAPPED;
@@ -4692,7 +4692,7 @@ static termchar *term_bidi_line(Terminal *term, struct termline *ldata,
                }
 
                term->wcFrom[it].origwc = term->wcFrom[it].wc =
-                   (wchar_t)uc;
+                   (unsigned int)uc;
                term->wcFrom[it].index = it;
            }
 
@@ -5067,10 +5067,17 @@ static void do_paint(Terminal *term, Context ctx, int may_optimise)
                dirty_run = TRUE;
            }
 
-           if (ccount >= chlen) {
+           if (ccount+2 > chlen) {
                chlen = ccount + 256;
                ch = sresize(ch, chlen, wchar_t);
            }
+
+#ifdef PLATFORM_IS_UTF16
+           if (tchar > 0x10000 && tchar < 0x110000) {
+               ch[ccount++] = (wchar_t) HIGH_SURROGATE_OF(tchar);
+               ch[ccount++] = (wchar_t) LOW_SURROGATE_OF(tchar);
+           } else
+#endif /* PLATFORM_IS_UTF16 */
            ch[ccount++] = (wchar_t) tchar;
 
            if (d->cc_next) {
@@ -5094,10 +5101,17 @@ static void do_paint(Terminal *term, Context ctx, int may_optimise)
                        break;
                    }
 
-                   if (ccount >= chlen) {
+                   if (ccount+2 > chlen) {
                        chlen = ccount + 256;
                        ch = sresize(ch, chlen, wchar_t);
                    }
+
+#ifdef PLATFORM_IS_UTF16
+                   if (schar > 0x10000 && schar < 0x110000) {
+                       ch[ccount++] = (wchar_t) HIGH_SURROGATE_OF(schar);
+                       ch[ccount++] = (wchar_t) LOW_SURROGATE_OF(schar);
+                   } else
+#endif /* PLATFORM_IS_UTF16 */
                    ch[ccount++] = (wchar_t) schar;
                }
 

diff --git a/wcwidth.c b/wcwidth.c
index 94ecb54..46ac1e3 100644 (file)
--- a/wcwidth.c
+++ b/wcwidth.c
@@ -69,7 +69,7 @@ struct interval {
 };
 
 /* auxiliary function for binary search in interval table */
-static int bisearch(wchar_t ucs, const struct interval *table, int max) {
+static int bisearch(unsigned int ucs, const struct interval *table, int max) {
   int min = 0;
   int mid;
 
@@ -121,7 +121,7 @@ static int bisearch(wchar_t ucs, const struct interval *table, int max) {
  * in ISO 10646.
  */
 
-int mk_wcwidth(wchar_t ucs)
+int mk_wcwidth(unsigned int ucs)
 {
   /* sorted list of non-overlapping intervals of non-spacing characters */
   /* generated by "uniset +cat=Me +cat=Mn +cat=Cf -00AD +1160-11FF +200B c" */
@@ -198,7 +198,7 @@ int mk_wcwidth(wchar_t ucs)
 }
 
 
-int mk_wcswidth(const wchar_t *pwcs, size_t n)
+int mk_wcswidth(const unsigned int *pwcs, size_t n)
 {
   int w, width = 0;
 
@@ -221,7 +221,7 @@ int mk_wcswidth(const wchar_t *pwcs, size_t n)
  * the traditional terminal character-width behaviour. It is not
  * otherwise recommended for general use.
  */
-int mk_wcwidth_cjk(wchar_t ucs)
+int mk_wcwidth_cjk(unsigned int ucs)
 {
   /* sorted list of non-overlapping intervals of East Asian Ambiguous
    * characters, generated by "uniset +WIDTH-A -cat=Me -cat=Mn -cat=Cf c" */
@@ -289,7 +289,7 @@ int mk_wcwidth_cjk(wchar_t ucs)
 }
 
 
-int mk_wcswidth_cjk(const wchar_t *pwcs, size_t n)
+int mk_wcswidth_cjk(const unsigned int *pwcs, size_t n)
 {
   int w, width = 0;
 

diff --git a/windows/window.c b/windows/window.c
index a1d99ba..80ca795 100644 (file)
--- a/windows/window.c
+++ b/windows/window.c
@@ -206,6 +206,12 @@ static int compose_state = 0;
 
 static UINT wm_mousewheel = WM_MOUSEWHEEL;
 
+#define IS_HIGH_VARSEL(wch1, wch2) \
+    ((wch1) == 0xDB40 && ((wch2) >= 0xDD00 && (wch2) <= 0xDDEF))
+#define IS_LOW_VARSEL(wch) \
+    (((wch) >= 0x180B && (wch) <= 0x180D) || /* MONGOLIAN FREE VARIATION SELECTOR */ \
+     ((wch) >= 0xFE00 && (wch) <= 0xFE0F)) /* VARIATION SELECTOR 1-16 */
+
 /* Dummy routine, only required in plink. */
 void ldisc_update(void *frontend, int echo, int edit)
 {
@@ -3105,9 +3111,20 @@ static LRESULT CALLBACK WndProc(HWND hwnd, UINT message,
                 * instead we luni_send the characters one by one.
                 */
                term_seen_key_event(term);
-               for (i = 0; i < n; i += 2) {
-                   if (ldisc)
+               /* don't divide SURROGATE PAIR */
+               if (ldisc) {
+                    for (i = 0; i < n; i += 2) {
+                       WCHAR hs = *(unsigned short *)(buff+i);
+                       if (IS_HIGH_SURROGATE(hs) && i+2 < n) {
+                           WCHAR ls = *(unsigned short *)(buff+i+2);
+                           if (IS_LOW_SURROGATE(ls)) {
+                               luni_send(ldisc, (unsigned short *)(buff+i), 2, 1);
+                               i += 2;
+                               continue;
+                           }
+                       }
                        luni_send(ldisc, (unsigned short *)(buff+i), 1, 1);
+                    }
                }
                free(buff);
            }
@@ -3309,6 +3326,7 @@ void do_text_internal(Context ctx, int x, int y, wchar_t *text, int len,
     static int *lpDx = NULL;
     static int lpDx_len = 0;
     int *lpDx_maybe;
+    int len2; /* for SURROGATE PAIR */
 
     lattr &= LATTR_MODE;
 
@@ -3379,7 +3397,8 @@ void do_text_internal(Context ctx, int x, int y, wchar_t *text, int len,
     }
 
     /* Anything left as an original character set is unprintable. */
-    if (DIRECT_CHAR(text[0])) {
+    if (DIRECT_CHAR(text[0]) &&
+        (len < 2 || !IS_SURROGATE_PAIR(text[0], text[1]))) {
        int i;
        for (i = 0; i < len; i++)
            text[i] = 0xFFFD;
@@ -3432,6 +3451,24 @@ void do_text_internal(Context ctx, int x, int y, wchar_t *text, int len,
     line_box.top = y;
     line_box.right = x + char_width * len;
     line_box.bottom = y + font_height;
+    /* adjust line_box.right for SURROGATE PAIR & VARIATION SELECTOR */
+    {
+       int i;
+       int rc_width = 0;
+       for (i = 0; i < len ; i++) {
+           if (i+1 < len && IS_HIGH_VARSEL(text[i], text[i+1])) {
+               i++;
+           } else if (i+1 < len && IS_SURROGATE_PAIR(text[i], text[i+1])) {
+               rc_width += char_width;
+               i++;
+           } else if (IS_LOW_VARSEL(text[i])) {
+               /* do nothing */
+            } else {
+               rc_width += char_width;
+            }
+       }
+       line_box.right = line_box.left + rc_width;
+    }
 
     /* Only want the left half of double width lines */
     if (line_box.right > font_width*term->cols+offset_width)
@@ -3462,8 +3499,19 @@ void do_text_internal(Context ctx, int x, int y, wchar_t *text, int len,
 
     opaque = TRUE;                     /* start by erasing the rectangle */
     for (remaining = len; remaining > 0;
-         text += len, remaining -= len, x += char_width * len) {
+         text += len, remaining -= len, x += char_width * len2) {
         len = (maxlen < remaining ? maxlen : remaining);
+        /* don't divide SURROGATE PAIR and VARIATION SELECTOR */
+        len2 = len;
+        if (maxlen == 1) {
+            if (remaining >= 1 && IS_SURROGATE_PAIR(text[0], text[1]))
+                len++;
+            if (remaining-len >= 1 && IS_LOW_VARSEL(text[len]))
+                len++;
+            else if (remaining-len >= 2 &&
+                     IS_HIGH_VARSEL(text[len], text[len+1]))
+                len += 2;
+        }
 
         if (len > lpDx_len) {
             if (len > lpDx_len) {
@@ -3473,8 +3521,24 @@ void do_text_internal(Context ctx, int x, int y, wchar_t *text, int len,
         }
         {
             int i;
-            for (i = 0; i < len; i++)
+            /* only last char has dx width in SURROGATE PAIR and
+             * VARIATION sequence */
+            for (i = 0; i < len; i++) {
                 lpDx[i] = char_width;
+                if (i+1 < len && IS_HIGH_VARSEL(text[i], text[i+1])) {
+                    if (i > 0) lpDx[i-1] = 0;
+                    lpDx[i] = 0;
+                    i++;
+                    lpDx[i] = char_width;
+                } else if (i+1 < len && IS_SURROGATE_PAIR(text[i],text[i+1])) {
+                    lpDx[i] = 0;
+                    i++;
+                    lpDx[i] = char_width;
+                } else if (IS_LOW_VARSEL(text[i])) {
+                    if (i > 0) lpDx[i-1] = 0;
+                    lpDx[i] = char_width;
+                }
+            }
         }
 
         /* We're using a private area for direct to font. (512 chars.) */
@@ -3623,9 +3687,35 @@ void do_text(Context ctx, int x, int y, wchar_t *text, int len,
 {
     if (attr & TATTR_COMBINING) {
        unsigned long a = 0;
-       attr &= ~TATTR_COMBINING;
+       int len0 = 1;
+        /* don't divide SURROGATE PAIR and VARIATION SELECTOR */
+       if (len >= 2 && IS_SURROGATE_PAIR(text[0], text[1]))
+           len0 = 2;
+       if (len-len0 >= 1 && IS_LOW_VARSEL(text[len0])) {
+           attr &= ~TATTR_COMBINING;
+           do_text_internal(ctx, x, y, text, len0+1, attr, lattr);
+           text += len0+1;
+           len -= len0+1;
+           a = TATTR_COMBINING;
+       } else if (len-len0 >= 2 && IS_HIGH_VARSEL(text[len0], text[len0+1])) {
+           attr &= ~TATTR_COMBINING;
+           do_text_internal(ctx, x, y, text, len0+2, attr, lattr);
+           text += len0+2;
+           len -= len0+2;
+           a = TATTR_COMBINING;
+       } else {
+            attr &= ~TATTR_COMBINING;
+        }
+
        while (len--) {
-           do_text_internal(ctx, x, y, text, 1, attr | a, lattr);
+           if (len >= 1 && IS_SURROGATE_PAIR(text[0], text[1])) {
+               do_text_internal(ctx, x, y, text, 2, attr | a, lattr);
+               len--;
+               text++;
+           } else {
+                do_text_internal(ctx, x, y, text, 1, attr | a, lattr);
+            }
+
            text++;
            a = TATTR_COMBINING;
        }