Rewrite grapheme-formation in terms of new work

Also, don't attach ZWNJ to previous cluster.  Closer to Unicode graphemes.

src/hb-coretext.cc

@@ -626,7 +626,7 @@ _hb_coretext_shape (hb_shape_plan_t    *shape_plan,
   CGFloat y_mult = (CGFloat) font->y_scale / ct_font_size;
 
   /* Attach marks to their bases, to match the 'ot' shaper.
-   * Adapted from hb-ot-shape:hb_form_clusters().
+   * Adapted from a very old version of hb-ot-shape:hb_form_clusters().
    * Note that this only makes us be closer to the 'ot' shaper,
    * but by no means the same.  For example, if there's
    * B1 M1 B2 M2, and B1-B2 form a ligature, M2's cluster will

src/hb-ot-shape.cc

@@ -339,26 +339,12 @@ hb_form_clusters (hb_buffer_t *buffer)
   if (!(buffer->scratch_flags & HB_BUFFER_SCRATCH_FLAG_HAS_NON_ASCII))
     return;
 
-  /* Loop duplicated in hb_ensure_native_direction(), and in _hb-coretext.cc */
-  unsigned int base = 0;
-  unsigned int count = buffer->len;
-  hb_glyph_info_t *info = buffer->info;
-  for (unsigned int i = 1; i < count; i++)
-  {
-    if (likely (!HB_UNICODE_GENERAL_CATEGORY_IS_MARK (_hb_glyph_info_get_general_category (&info[i])) &&
-		!_hb_glyph_info_is_joiner (&info[i])))
-    {
-      if (buffer->cluster_level == HB_BUFFER_CLUSTER_LEVEL_MONOTONE_GRAPHEMES)
-	buffer->merge_clusters (base, i);
-      else
-	buffer->unsafe_to_break (base, i);
-      base = i;
-    }
-  }
   if (buffer->cluster_level == HB_BUFFER_CLUSTER_LEVEL_MONOTONE_GRAPHEMES)
-    buffer->merge_clusters (base, count);
+    foreach_grapheme (buffer, start, end)
+      buffer->merge_clusters (start, end);
   else
-    buffer->unsafe_to_break (base, count);
+    foreach_grapheme (buffer, start, end)
+      buffer->unsafe_to_break (start, end);
 }
 
 static void
@@ -376,25 +362,17 @@ hb_ensure_native_direction (hb_buffer_t *buffer)
       (HB_DIRECTION_IS_VERTICAL   (direction) &&
        direction != HB_DIRECTION_TTB))
   {
-    /* Same loop as hb_form_clusters().
-     * Since form_clusters() merged clusters already, we don't merge. */
-    unsigned int base = 0;
-    unsigned int count = buffer->len;
-    hb_glyph_info_t *info = buffer->info;
-    for (unsigned int i = 1; i < count; i++)
-    {
-      if (likely (!HB_UNICODE_GENERAL_CATEGORY_IS_MARK (_hb_glyph_info_get_general_category (&info[i]))))
-      {
-	if (buffer->cluster_level == HB_BUFFER_CLUSTER_LEVEL_MONOTONE_CHARACTERS)
-	  buffer->merge_clusters (base, i);
-	buffer->reverse_range (base, i);
 
-	base = i;
-      }
-    }
     if (buffer->cluster_level == HB_BUFFER_CLUSTER_LEVEL_MONOTONE_CHARACTERS)
-      buffer->merge_clusters (base, count);
-    buffer->reverse_range (base, count);
+      foreach_grapheme (buffer, start, end)
+      {
+	buffer->merge_clusters (start, end);
+	buffer->reverse_range (start, end);
+      }
+    else
+      foreach_grapheme (buffer, start, end)
+	/* form_clusters() merged clusters already, we don't merge. */
+	buffer->reverse_range (start, end);
 
     buffer->reverse ();
 

test/shaping/data/in-house/tests/indic-joiner-candrabindu.tests

 ../fonts/5028afb650b1bb718ed2131e872fbcce57828fff.ttf::U+0B13,U+200D,U+0B01:[omorya=0+1450]
-../fonts/5028afb650b1bb718ed2131e872fbcce57828fff.ttf::U+0B13,U+200C,U+0B01:[oorya=0+1309|space=0+0|candrabinduorya=0+0]
+../fonts/5028afb650b1bb718ed2131e872fbcce57828fff.ttf::U+0B13,U+200C,U+0B01:[oorya=0+1309|space=1+0|candrabinduorya=1+0]

test/shaping/data/in-house/tests/indic-joiners.tests

-../fonts/f443753e8ffe8e8aae606cfba158e00334b6efb1.ttf::U+179A,U+1784,U+17D2,U+179F,U+200C,U+17CA,U+17B8,U+0020:[uni179a=0+775|uni1784=1+1550|uni179f.sub=1+775|space=1+0|uni17ca=1+0|uni17b8=1@0,300+0|space=7+600]
+../fonts/f443753e8ffe8e8aae606cfba158e00334b6efb1.ttf::U+179A,U+1784,U+17D2,U+179F,U+200C,U+17CA,U+17B8,U+0020:[uni179a=0+775|uni1784=1+1550|uni179f.sub=1+775|space=4+0|uni17ca=4+0|uni17b8=4@0,300+0|space=7+600]
 ../fonts/f443753e8ffe8e8aae606cfba158e00334b6efb1.ttf::U+179A,U+1784,U+17D2,U+179F,U+17CA,U+17B8:[uni179a=0+775|uni1784=1+1550|uni179f.sub=1+775|uni17bb=1@-75,-700+0|uni17b8=1+0]
-../fonts/8116e5d8fedfbec74e45dc350d2416d810bed8c4.ttf:--font-funcs=ft:U+091F,U+094D,U+200C,U+092F,U+093F:[uni091F=0+876|uni094D=0@4,0+0|space=0+0|uni093F.750=3+397|uni092F=3+924]
+../fonts/8116e5d8fedfbec74e45dc350d2416d810bed8c4.ttf:--font-funcs=ft:U+091F,U+094D,U+200C,U+092F,U+093F:[uni091F=0+876|uni094D=0@4,0+0|space=2+0|uni093F.750=3+397|uni092F=3+924]
 ../fonts/8116e5d8fedfbec74e45dc350d2416d810bed8c4.ttf:--font-funcs=ft:U+091F,U+094D,U+200D,U+092F,U+093F:[uni093F=0+398|uni091F=0+876|uni094D=0@4,0+0|space=0+0|uni092F=0+924]
-../fonts/8116e5d8fedfbec74e45dc350d2416d810bed8c4.ttf:--font-funcs=ft:U+091F,U+094D,U+200D,U+091F,U+094D,U+200C,U+091F,U+094D,U+200D,U+092F,U+093F:[uni091F=0+876|uni094D=0@4,0+0|space=0+0|uni091F=3+876|uni094D=3@4,0+0|space=3+0|uni093F=6+398|uni091F=6+876|uni094D=6@4,0+0|space=6+0|uni092F=6+924]
+../fonts/8116e5d8fedfbec74e45dc350d2416d810bed8c4.ttf:--font-funcs=ft:U+091F,U+094D,U+200D,U+091F,U+094D,U+200C,U+091F,U+094D,U+200D,U+092F,U+093F:[uni091F=0+876|uni094D=0@4,0+0|space=0+0|uni091F=3+876|uni094D=3@4,0+0|space=5+0|uni093F=6+398|uni091F=6+876|uni094D=6@4,0+0|space=6+0|uni092F=6+924]
 ../fonts/8116e5d8fedfbec74e45dc350d2416d810bed8c4.ttf:--font-funcs=ft:U+091F,U+094D,U+200D,U+091F,U+094D,U+200D,U+091F,U+094D,U+200D,U+092F,U+093F:[uni093F=0+398|uni091F=0+876|uni094D=0@4,0+0|space=0+0|uni091F=0+876|uni094D=0@4,0+0|space=0+0|uni091F=0+876|uni094D=0@4,0+0|space=0+0|uni092F=0+924]