merge changes in emacs-23 branch

4 files changed, 164 insertions, 245 deletions

diff --git a/src/ChangeLog b/src/ChangeLog
index fc98bcc11e3..d1f789f5dcb 100644
--- a/src/ChangeLog
+++ b/src/ChangeLog
@@ -1,3 +1,33 @@
+2010-08-31  Kenichi Handa  <handa@m17n.org>
+
+        * dispextern.h (FACE_FOR_CHAR): Use an ASCII face for 8-bit
+        characters.
+
+        * term.c (encode_terminal_code): Fix the previous change.
+        (produce_glyphs): Don't set it->char_to_display here.  Don't
+        handle unibyte-display-via-language-environment here.
+        (produce_special_glyphs): Set temp_it.char_to_display before
+        calling produce_glyphs.
+
+        * xdisp.c (get_next_display_element): Set it->char_to_display
+        here.  Convert all 8-bit bytes from unibyte buffer/string to 8-bit
+        characters.
+        (get_overlay_arrow_glyph_row): Set it.char_to_display too before
+        calling PRODUCE_GLYPHS.
+        (append_space_for_newline): Save and store it->char_to_display.
+        Set it->char_to_display before calling PRODUCE_GLYPHS.
+        (extend_face_to_end_of_line): Set it->char_to_display before
+        calling PRODUCE_GLYPHS.
+        (get_glyph_face_and_encoding): Set the glyph code an 8-bit
+        character to its byte value.
+        (get_char_glyph_code): New function.
+        (produce_stretch_glyph): Set it2.char_to_display too before
+        calling x_produce_glyphs.
+        (x_produce_glyphs): Simplify by using the same code for ASCII and
+        non-ASCII characters.  Don't set it->char_to_display here.  Don't
+        handle unibyte-display-via-language-environment here. For a
+        charater of no glyph, use font->space_width instead of FONT_WIDTH.
+
 2010-08-31  Stefan Monnier  <monnier@iro.umontreal.ca>
 
         * keyboard.c (Fwindow_system): Fix compilation for USE_LISP_UNION_TYPE.
@@ -15,6 +45,11 @@
         * composite.c (composition_update_it): Fix computing of
         cmp_it->width.
 
+2010-08-29  Kenichi Handa  <handa@m17n.org>
+
+        * term.c (encode_terminal_code): Encode byte chars to the
+        correspnding bytes.
+
 2010-08-29  Jan Djärv  <jan.h.d@swipnet.se>
 
         * nsterm.m (ns_draw_window_cursor): Draw BAR_CURSOR correct for R2L.
diff --git a/src/dispextern.h b/src/dispextern.h
index c36db91ea02..5138958b6db 100644
--- a/src/dispextern.h
+++ b/src/dispextern.h
@@ -1718,7 +1718,7 @@ struct face_cache
    This macro is only meaningful for multibyte character CHAR.  */
 
 #define FACE_FOR_CHAR(F, FACE, CHAR, POS, OBJECT)       \
-  (ASCII_CHAR_P (CHAR)                                  \
+  ((ASCII_CHAR_P (CHAR) || CHAR_BYTE8_P (CHAR))         \
    ? (FACE)->ascii_face->id                             \
    : face_for_char ((F), (FACE), (CHAR), (POS), (OBJECT)))
 
@@ -2275,9 +2275,11 @@ struct it
      composition.  */
   struct composition_it cmp_it;
 
-  /* The character to display, possibly translated to multibyte
-     if unibyte_display_via_language_environment is set.  This
-     is set after produce_glyphs has been called.  */
+  /* The character to display, possibly translated to multibyte if
+     multibyte_p is zero or unibyte_display_via_language_environment
+     is set.  This is set after get_next_display_element has been
+     called.  If we are setting it->C directly before calling
+     PRODUCE_GLYPHS, this should be set beforehand too.  */
   int char_to_display;
 
   /* If what == IT_IMAGE, the id of the image to display.  */
diff --git a/src/term.c b/src/term.c
index d1279498060..f090cdd2792 100644
--- a/src/term.c
+++ b/src/term.c
@@ -689,7 +689,8 @@ encode_terminal_code (struct glyph *src, int src_len, struct coding_system *codi
                                                   encode_terminal_src_size);
                   buf = encode_terminal_src + nbytes;
                 }
-              if (char_charset (c, charset_list, NULL))
+              if (CHAR_BYTE8_P (c)
+                  || char_charset (c, charset_list, NULL))
                 {
                   /* Store the multibyte form of C at BUF.  */
                   buf += CHAR_STRING (c, buf);
@@ -1614,18 +1615,15 @@ produce_glyphs (struct it *it)
       goto done;
     }
 
-  /* Maybe translate single-byte characters to multibyte.  */
-  it->char_to_display = it->c;
-
-  if (it->c >= 040 && it->c < 0177)
+  if (it->char_to_display >= 040 && it->char_to_display < 0177)
     {
       it->pixel_width = it->nglyphs = 1;
       if (it->glyph_row)
         append_glyph (it);
     }
-  else if (it->c == '\n')
+  else if (it->char_to_display == '\n')
     it->pixel_width = it->nglyphs = 0;
-  else if (it->c == '\t')
+  else if (it->char_to_display == '\t')
     {
       int absolute_x = (it->current_x
                         + it->continuation_lines_width);
@@ -1656,32 +1654,19 @@ produce_glyphs (struct it *it)
       it->pixel_width = nspaces;
       it->nglyphs = nspaces;
     }
-  else if (CHAR_BYTE8_P (it->c))
+  else if (CHAR_BYTE8_P (it->char_to_display))
     {
-      if (unibyte_display_via_language_environment
-          && (it->c >= 0240))
-        {
-          it->char_to_display = BYTE8_TO_CHAR (it->c);
-          it->pixel_width = CHAR_WIDTH (it->char_to_display);
-          it->nglyphs = it->pixel_width;
-          if (it->glyph_row)
-            append_glyph (it);
-        }
-      else
-        {
-          /* Coming here means that it->c is from display table, thus
-             we must send the raw 8-bit byte as is to the terminal.
-             Although there's no way to know how many columns it
-             occupies on a screen, it is a good assumption that a
-             single byte code has 1-column width.  */
-          it->pixel_width = it->nglyphs = 1;
-          if (it->glyph_row)
-            append_glyph (it);
-        }
+      /* Coming here means that we must send the raw 8-bit byte as is
+         to the terminal.  Although there's no way to know how many
+         columns it occupies on a screen, it is a good assumption that
+         a single byte code has 1-column width.  */
+      it->pixel_width = it->nglyphs = 1;
+      if (it->glyph_row)
+        append_glyph (it);
     }
   else
     {
-      it->pixel_width = CHAR_WIDTH (it->c);
+      it->pixel_width = CHAR_WIDTH (it->char_to_display);
       it->nglyphs = it->pixel_width;
 
       if (it->glyph_row)
@@ -1917,7 +1902,7 @@ produce_special_glyphs (struct it *it, enum display_element_type what)
   else
     abort ();
 
-  temp_it.c = GLYPH_CHAR (glyph);
+  temp_it.c = temp_it.char_to_display = GLYPH_CHAR (glyph);
   temp_it.face_id = GLYPH_FACE (glyph);
   temp_it.len = CHAR_BYTES (temp_it.c);
 
diff --git a/src/xdisp.c b/src/xdisp.c
index 07078a199a3..2ec271cdb6b 100644
--- a/src/xdisp.c
+++ b/src/xdisp.c
@@ -5762,10 +5762,23 @@ get_next_display_element (struct it *it)
           struct charset *unibyte = CHARSET_FROM_ID (charset_unibyte);
           enum { char_is_other = 0, char_is_nbsp, char_is_soft_hyphen }
           nbsp_or_shy = char_is_other;
-          int decoded = it->c;
+          int c = it->c;        /* This is the character to display.  */
+
+          if (! it->multibyte_p && ! ASCII_CHAR_P (c))
+            {
+              xassert (SINGLE_BYTE_CHAR_P (c));
+              if (unibyte_display_via_language_environment)
+                {
+                  c = DECODE_CHAR (unibyte, c);
+                  if (c < 0)
+                    c = BYTE8_TO_CHAR (it->c);
+                }
+              else
+                c = BYTE8_TO_CHAR (it->c);
+            }
 
           if (it->dp
-              && (dv = DISP_CHAR_VECTOR (it->dp, it->c),
+              && (dv = DISP_CHAR_VECTOR (it->dp, c),
                   VECTORP (dv)))
             {
               struct Lisp_Vector *v = XVECTOR (dv);
@@ -5791,21 +5804,10 @@ get_next_display_element (struct it *it)
               goto get_next;
             }
 
-          if (unibyte_display_via_language_environment
-              && !ASCII_CHAR_P (it->c))
-            decoded = DECODE_CHAR (unibyte, it->c);
-
-          if (it->c >= 0x80 && ! NILP (Vnobreak_char_display))
-            {
-              if (it->multibyte_p)
-                nbsp_or_shy = (it->c == 0xA0   ? char_is_nbsp
-                               : it->c == 0xAD ? char_is_soft_hyphen
-                               :                 char_is_other);
-              else if (unibyte_display_via_language_environment)
-                nbsp_or_shy = (decoded == 0xA0   ? char_is_nbsp
-                               : decoded == 0xAD ? char_is_soft_hyphen
-                               :                   char_is_other);
-            }
+          if (! ASCII_CHAR_P (c) && ! NILP (Vnobreak_char_display))
+            nbsp_or_shy = (c == 0xA0   ? char_is_nbsp
+                           : c == 0xAD ? char_is_soft_hyphen
+                           :             char_is_other);
 
           /* Translate control characters into `\003' or `^C' form.
              Control characters coming from a display table entry are
@@ -5813,27 +5815,23 @@ get_next_display_element (struct it *it)
              the translation.  This could easily be changed but I
              don't believe that it is worth doing.
 
-             If it->multibyte_p is nonzero, non-printable non-ASCII
-             characters are also translated to octal form.
+             NBSP and SOFT-HYPEN are property translated too.
 
-             If it->multibyte_p is zero, eight-bit characters that
-             don't have corresponding multibyte char code are also
+             Non-printable characters and raw-byte characters are also
              translated to octal form.  */
-          if ((it->c < ' '
+          if (((c < ' ' || c == 127) /* ASCII control chars */
                ? (it->area != TEXT_AREA
                   /* In mode line, treat \n, \t like other crl chars.  */
-                  || (it->c != '\t'
+                  || (c != '\t'
                       && it->glyph_row
                       && (it->glyph_row->mode_line_p || it->avoid_cursor_p))
-                  || (it->c != '\n' && it->c != '\t'))
+                  || (c != '\n' && c != '\t'))
                : (nbsp_or_shy
-                  || (it->multibyte_p
-                      ? ! CHAR_PRINTABLE_P (it->c)
-                      : (! unibyte_display_via_language_environment
-                         ? it->c >= 0x80
-                         : (decoded >= 0x80 && decoded < 0xA0))))))
+                  || CHAR_BYTE8_P (c)
+                  || ! CHAR_PRINTABLE_P (c))))
             {
-              /* IT->c is a control character which must be displayed
+              /* C is a control character, NBSP, SOFT-HYPEN, raw-byte,
+                 or a non-printable character which must be displayed
                  either as '\003' or as `^C' where the '\\' and '^'
                  can be defined in the display table.  Fill
                  IT->ctl_chars with glyphs for what we have to
@@ -5845,7 +5843,7 @@ get_next_display_element (struct it *it)
 
               /* Handle control characters with ^.  */
 
-              if (it->c < 128 && it->ctl_arrow_p)
+              if (ASCII_CHAR_P (c) && it->ctl_arrow_p)
                 {
                   int g;
 
@@ -5878,7 +5876,7 @@ get_next_display_element (struct it *it)
                     }
 
                   XSETINT (it->ctl_chars[0], g);
-                  XSETINT (it->ctl_chars[1], it->c ^ 0100);
+                  XSETINT (it->ctl_chars[1], c ^ 0100);
                   ctl_len = 2;
                   goto display_control;
                 }
@@ -5893,7 +5891,7 @@ get_next_display_element (struct it *it)
                   face_id = merge_faces (it->f, Qnobreak_space, 0,
                                          it->face_id);
 
-                  it->c = ' ';
+                  c = ' ';
                   XSETINT (it->ctl_chars[0], ' ');
                   ctl_len = 1;
                   goto display_control;
@@ -5939,7 +5937,6 @@ get_next_display_element (struct it *it)
               if (EQ (Vnobreak_char_display, Qt)
                   && nbsp_or_shy == char_is_soft_hyphen)
                 {
-                  it->c = '-';
                   XSETINT (it->ctl_chars[0], '-');
                   ctl_len = 1;
                   goto display_control;
@@ -5951,55 +5948,25 @@ get_next_display_element (struct it *it)
               if (nbsp_or_shy)
                 {
                   XSETINT (it->ctl_chars[0], escape_glyph);
-                  it->c = (nbsp_or_shy == char_is_nbsp ? ' ' : '-');
-                  XSETINT (it->ctl_chars[1], it->c);
+                  c = (nbsp_or_shy == char_is_nbsp ? ' ' : '-');
+                  XSETINT (it->ctl_chars[1], c);
                   ctl_len = 2;
                   goto display_control;
                 }
 
               {
-                unsigned char str[MAX_MULTIBYTE_LENGTH];
-                int len;
-                int i;
+                char str[10];
+                int len, i;
 
-                /* Set IT->ctl_chars[0] to the glyph for `\\'.  */
-                if (CHAR_BYTE8_P (it->c))
-                  {
-                    str[0] = CHAR_TO_BYTE8 (it->c);
-                    len = 1;
-                  }
-                else if (it->c < 256)
-                  {
-                    str[0] = it->c;
-                    len = 1;
-                  }
-                else
-                  {
-                    /* It's an invalid character, which shouldn't
-                       happen actually, but due to bugs it may
-                       happen.  Let's print the char as is, there's
-                       not much meaningful we can do with it.  */
-                    str[0] = it->c;
-                    str[1] = it->c >> 8;
-                    str[2] = it->c >> 16;
-                    str[3] = it->c >> 24;
-                    len = 4;
-                  }
+                if (CHAR_BYTE8_P (c))
+                  /* Display \200 instead of \17777600.  */
+                  c = CHAR_TO_BYTE8 (c);
+                len = sprintf (str, "%03o", c);
 
+                XSETINT (it->ctl_chars[0], escape_glyph);
                 for (i = 0; i < len; i++)
-                  {
-                    int g;
-                    XSETINT (it->ctl_chars[i * 4], escape_glyph);
-                    /* Insert three more glyphs into IT->ctl_chars for
-                       the octal display of the character.  */
-                    g = ((str[i] >> 6) & 7) + '0';
-                    XSETINT (it->ctl_chars[i * 4 + 1], g);
-                    g = ((str[i] >> 3) & 7) + '0';
-                    XSETINT (it->ctl_chars[i * 4 + 2], g);
-                    g = (str[i] & 7) + '0';
-                    XSETINT (it->ctl_chars[i * 4 + 3], g);
-                  }
-                ctl_len = len * 4;
+                  XSETINT (it->ctl_chars[i + 1], str[i]);
+                ctl_len = len + 1;
               }
 
             display_control:
@@ -6014,6 +5981,11 @@ get_next_display_element (struct it *it)
               it->ellipsis_p = 0;
               goto get_next;
             }
+          it->char_to_display = c;
+        }
+      else if (success_p)
+        {
+          it->char_to_display = it->c;
         }
     }
 
@@ -6040,7 +6012,8 @@ get_next_display_element (struct it *it)
                      : STRINGP (it->string) ? IT_STRING_CHARPOS (*it)
                      : IT_CHARPOS (*it));
 
-          it->face_id = FACE_FOR_CHAR (it->f, face, it->c, pos, it->string);
+          it->face_id = FACE_FOR_CHAR (it->f, face, it->char_to_display, pos,
+                                       it->string);
         }
     }
 #endif
@@ -16481,15 +16454,19 @@ get_overlay_arrow_glyph_row (struct window *w, Lisp_Object overlay_arrow_string)
 
       /* Get the next character.  */
       if (multibyte_p)
-        it.c = string_char_and_length (p, &it.len);
+        it.c = it.char_to_display = string_char_and_length (p, &it.len);
       else
-        it.c = *p, it.len = 1;
+        {
+          it.c = it.char_to_display = *p, it.len = 1;
+          if (! ASCII_CHAR_P (it.c))
+            it.char_to_display = BYTE8_TO_CHAR (it.c);
+        }
       p += it.len;
 
       /* Get its face.  */
       ilisp = make_number (p - arrow_string);
       face = Fget_text_property (ilisp, Qface, overlay_arrow_string);
-      it.face_id = compute_char_face (f, it.c, face);
+      it.face_id = compute_char_face (f, it.char_to_display, face);
 
       /* Compute its width, get its glyphs.  */
       n_glyphs_before = it.glyph_row->used[TEXT_AREA];
@@ -16721,6 +16698,7 @@ append_space_for_newline (struct it *it, int default_face_p)
              append_space_for_newline has been called.  */
           enum display_element_type saved_what = it->what;
           int saved_c = it->c, saved_len = it->len;
+          int saved_char_to_display = it->char_to_display;
           int saved_x = it->current_x;
           int saved_face_id = it->face_id;
           struct text_pos saved_pos;
@@ -16733,7 +16711,7 @@ append_space_for_newline (struct it *it, int default_face_p)
           it->what = IT_CHARACTER;
           memset (&it->position, 0, sizeof it->position);
           it->object = make_number (0);
-          it->c = ' ';
+          it->c = it->char_to_display = ' ';
           it->len = 1;
 
           if (default_face_p)
@@ -16754,6 +16732,7 @@ append_space_for_newline (struct it *it, int default_face_p)
           it->face_id = saved_face_id;
           it->len = saved_len;
           it->c = saved_c;
+          it->char_to_display = saved_char_to_display;
           return 1;
         }
     }
@@ -16886,7 +16865,7 @@ extend_face_to_end_of_line (struct it *it)
       it->what = IT_CHARACTER;
       memset (&it->position, 0, sizeof it->position);
       it->object = make_number (0);
-      it->c = ' ';
+      it->c = it->char_to_display = ' ';
       it->len = 1;
       /* The last row's blank glyphs should get the default face, to
          avoid painting the rest of the window with the region face,
@@ -20490,7 +20469,12 @@ get_glyph_face_and_encoding (struct frame *f, struct glyph *glyph,
 
   if (face->font)
     {
-      unsigned code = face->font->driver->encode_char (face->font, glyph->u.ch);
+      unsigned code;
+
+      if (CHAR_BYTE8_P (glyph->u.ch))
+        code = CHAR_TO_BYTE8 (glyph->u.ch);
+      else
+        code = face->font->driver->encode_char (face->font, glyph->u.ch);
 
       if (code != FONT_INVALID_CODE)
         STORE_XCHAR2B (char2b, (code >> 8), (code & 0xFF));
@@ -20505,6 +20489,26 @@ get_glyph_face_and_encoding (struct frame *f, struct glyph *glyph,
 }
 
 
+/* Get glyph code of character C in FONT in the two-byte form CHAR2B.
+   Retunr 1 if FONT has a glyph for C, otherwise return 0.  */
+
+static INLINE int
+get_char_glyph_code (int c, struct font *font, XChar2b *char2b)
+{
+  unsigned code;
+
+  if (CHAR_BYTE8_P (c))
+    code = CHAR_TO_BYTE8 (c);
+  else
+    code = font->driver->encode_char (font, c);
+
+  if (code == FONT_INVALID_CODE)
+    return 0;
+  STORE_XCHAR2B (char2b, (code >> 8), (code & 0xFF));
+  return 1;
+}
+
+
 /* Fill glyph string S with composition components specified by S->cmp.
 
    BASE_FACE is the base face of the composition.
@@ -21909,10 +21913,14 @@ produce_stretch_glyph (struct it *it)
         {
           int maxlen = ((IT_BYTEPOS (*it) >= GPT ? ZV : GPT)
                         - IT_BYTEPOS (*it));
-          it2.c = STRING_CHAR_AND_LENGTH (p, it2.len);
+          it2.c = it2.char_to_display = STRING_CHAR_AND_LENGTH (p, it2.len);
         }
       else
-        it2.c = *p, it2.len = 1;
+        {
+          it2.c = it2.char_to_display = *p, it2.len = 1;
+          if (! ASCII_CHAR_P (it2.c))
+            it2.char_to_display = BYTE8_TO_CHAR (it2.c);
+        }
 
       it2.glyph_row = NULL;
       it2.what = IT_CHARACTER;
@@ -22082,49 +22090,12 @@ x_produce_glyphs (struct it *it)
   if (it->what == IT_CHARACTER)
     {
       XChar2b char2b;
-      struct font *font;
       struct face *face = FACE_FROM_ID (it->f, it->face_id);
-      struct font_metrics *pcm;
-      int font_not_found_p;
+      struct font *font = face->font;
+      int font_not_found_p = font == NULL;
+      struct font_metrics *pcm = NULL;
       int boff;                 /* baseline offset */
-      /* We may change it->multibyte_p upon unibyte<->multibyte
-         conversion.  So, save the current value now and restore it
-         later.
-
-         Note: It seems that we don't have to record multibyte_p in
-         struct glyph because the character code itself tells whether
-         or not the character is multibyte.  Thus, in the future, we
-         must consider eliminating the field `multibyte_p' in the
-         struct glyph.  */
-      int saved_multibyte_p = it->multibyte_p;
-
-      /* Maybe translate single-byte characters to multibyte, or the
-         other way.  */
-      it->char_to_display = it->c;
-      if (!ASCII_BYTE_P (it->c)
-          && ! it->multibyte_p)
-        {
-          if (SINGLE_BYTE_CHAR_P (it->c)
-              && unibyte_display_via_language_environment)
-            {
-              struct charset *unibyte = CHARSET_FROM_ID (charset_unibyte);
-
-              /* get_next_display_element assures that this decoding
-                 never fails.  */
-              it->char_to_display = DECODE_CHAR (unibyte, it->c);
-              it->multibyte_p = 1;
-              it->face_id = FACE_FOR_CHAR (it->f, face, it->char_to_display,
-                                           -1, Qnil);
-              face = FACE_FROM_ID (it->f, it->face_id);
-            }
-        }
-
-      /* Get font to use.  Encode IT->char_to_display.  */
-      get_char_face_and_encoding (it->f, it->char_to_display, it->face_id,
-                                  &char2b, it->multibyte_p, 0);
-      font = face->font;
 
-      font_not_found_p = font == NULL;
       if (font_not_found_p)
         {
           /* When no suitable font found, display an empty box based
@@ -22144,16 +22115,12 @@ x_produce_glyphs (struct it *it)
             boff = VCENTER_BASELINE_OFFSET (font, it->f) - boff;
         }
 
-      if (it->char_to_display >= ' '
-          && (!it->multibyte_p || it->char_to_display < 128))
+      if (it->char_to_display != '\n' && it->char_to_display != '\t')
         {
-          /* Either unibyte or ASCII.  */
           int stretched_p;
 
           it->nglyphs = 1;
 
-          pcm = get_per_char_metric (it->f, font, &char2b);
-
           if (it->override_ascent >= 0)
             {
               it->ascent = it->override_ascent;
@@ -22166,6 +22133,15 @@ x_produce_glyphs (struct it *it)
               it->descent = FONT_DESCENT (font) - boff;
             }
 
+          if (! font_not_found_p
+              && get_char_glyph_code (it->char_to_display, font, &char2b))
+            {
+              pcm = get_per_char_metric (it->f, font, &char2b);
+              if (pcm->width == 0
+                  && pcm->rbearing == 0 && pcm->lbearing == 0)
+                pcm = NULL;
+            }
+
           if (pcm)
             {
               it->phys_ascent = pcm->ascent + boff;
@@ -22177,7 +22153,7 @@ x_produce_glyphs (struct it *it)
               it->glyph_not_available_p = 1;
               it->phys_ascent = it->ascent;
               it->phys_descent = it->descent;
-              it->pixel_width = FONT_WIDTH (font);
+              it->pixel_width = font->space_width;
             }
 
           if (it->constrain_row_ascent_descent_p)
@@ -22351,7 +22327,7 @@ x_produce_glyphs (struct it *it)
                 }
             }
         }
-      else if (it->char_to_display == '\t')
+      else                    /* i.e. (it->char_to_display == '\t') */
         {
           if (font->space_width > 0)
             {
@@ -22382,85 +22358,6 @@ x_produce_glyphs (struct it *it)
               it->nglyphs = 1;
             }
         }
-      else
-        {
-          /* A multi-byte character.  Assume that the display width of the
-             character is the width of the character multiplied by the
-             width of the font.  */
-
-          /* If we found a font, this font should give us the right
-             metrics.  If we didn't find a font, use the frame's
-             default font and calculate the width of the character by
-             multiplying the width of font by the width of the
-             character.  */
-
-          pcm = get_per_char_metric (it->f, font, &char2b);
-
-          if (font_not_found_p || !pcm)
-            {
-              int char_width = CHAR_WIDTH (it->char_to_display);
-
-              if (char_width == 0)
-                /* This is a non spacing character.  But, as we are
-                   going to display an empty box, the box must occupy
-                   at least one column.  */
-                char_width = 1;
-              it->glyph_not_available_p = 1;
-              it->pixel_width = font->space_width * char_width;
-              it->phys_ascent = FONT_BASE (font) + boff;
-              it->phys_descent = FONT_DESCENT (font) - boff;
-            }
-          else
-            {
-              it->pixel_width = pcm->width;
-              it->phys_ascent = pcm->ascent + boff;
-              it->phys_descent = pcm->descent - boff;
-              if (it->glyph_row
-                  && (pcm->lbearing < 0
-                      || pcm->rbearing > pcm->width))
-                it->glyph_row->contains_overlapping_glyphs_p = 1;
-            }
-          it->nglyphs = 1;
-          it->ascent = FONT_BASE (font) + boff;
-          it->descent = FONT_DESCENT (font) - boff;
-          if (face->box != FACE_NO_BOX)
-            {
-              int thick = face->box_line_width;
-
-              if (thick > 0)
-                {
-                  it->ascent += thick;
-                  it->descent += thick;
-                }
-              else
-                thick = - thick;
-
-              if (it->start_of_box_run_p)
-                it->pixel_width += thick;
-              if (it->end_of_box_run_p)
-                it->pixel_width += thick;
-            }
-
-          /* If face has an overline, add the height of the overline
-             (1 pixel) and a 1 pixel margin to the character height.  */
-          if (face->overline_p)
-            it->ascent += overline_margin;
-
-          take_vertical_position_into_account (it);
-
-          if (it->ascent < 0)
-            it->ascent = 0;
-          if (it->descent < 0)
-            it->descent = 0;
-
-          if (it->glyph_row)
-            append_glyph (it);
-          if (it->pixel_width == 0)
-            /* We assure that all visible glyphs have at least 1-pixel
-               width.  */
-            it->pixel_width = 1;
-        }
-      it->multibyte_p = saved_multibyte_p;
     }
   else if (it->what == IT_COMPOSITION && it->cmp_it.ch < 0)
     {
@@ -22556,7 +22453,7 @@ x_produce_glyphs (struct it *it)
             }
           else
             {
-              width = FONT_WIDTH (font);
+              width = font->space_width;
               ascent = FONT_BASE (font);
               descent = FONT_DESCENT (font);
               lbearing = 0;