[cmap] Implement MacRoman encoding

For other Mac encodings just map the ASCII range.

Fixes https://github.com/harfbuzz/harfbuzz/issues/4540

src/hb-algs.hh

@@ -1053,6 +1053,18 @@ _hb_cmp_method (const void *pkey, const void *pval, Ts... ds)
   return val.cmp (key, ds...);
 }
 
+template <typename K, typename V>
+static int
+_hb_cmp_operator (const void *pkey, const void *pval)
+{
+  const K& key = * (const K*) pkey;
+  const V& val = * (const V*) pval;
+
+  if (key < val) return -1;
+  if (key > val) return  1;
+  return 0;
+}
+
 template <typename V, typename K, typename ...Ts>
 static inline bool
 hb_bsearch_impl (unsigned *pos, /* Out */

src/hb-ot-cmap-table.hh

@@ -41,6 +41,30 @@
 
 namespace OT {
 
+static inline uint8_t unicode_to_macroman (hb_codepoint_t u)
+{
+  uint16_t mapping[] = {
+    0x00C4, 0x00C5, 0x00C7, 0x00C9, 0x00D1, 0x00D6, 0x00DC, 0x00E1,
+    0x00E0, 0x00E2, 0x00E4, 0x00E3, 0x00E5, 0x00E7, 0x00E9, 0x00E8,
+    0x00EA, 0x00EB, 0x00ED, 0x00EC, 0x00EE, 0x00EF, 0x00F1, 0x00F3,
+    0x00F2, 0x00F4, 0x00F6, 0x00F5, 0x00FA, 0x00F9, 0x00FB, 0x00FC,
+    0x2020, 0x00B0, 0x00A2, 0x00A3, 0x00A7, 0x2022, 0x00B6, 0x00DF,
+    0x00AE, 0x00A9, 0x2122, 0x00B4, 0x00A8, 0x2260, 0x00C6, 0x00D8,
+    0x221E, 0x00B1, 0x2264, 0x2265, 0x00A5, 0x00B5, 0x2202, 0x2211,
+    0x220F, 0x03C0, 0x222B, 0x00AA, 0x00BA, 0x03A9, 0x00E6, 0x00F8,
+    0x00BF, 0x00A1, 0x00AC, 0x221A, 0x0192, 0x2248, 0x2206, 0x00AB,
+    0x00BB, 0x2026, 0x00A0, 0x00C0, 0x00C3, 0x00D5, 0x0152, 0x0153,
+    0x2013, 0x2014, 0x201C, 0x201D, 0x2018, 0x2019, 0x00F7, 0x25CA,
+    0x00FF, 0x0178, 0x2044, 0x20AC, 0x2039, 0x203A, 0xFB01, 0xFB02,
+    0x2021, 0x00B7, 0x201A, 0x201E, 0x2030, 0x00C2, 0x00CA, 0x00C1,
+    0x00CB, 0x00C8, 0x00CD, 0x00CE, 0x00CF, 0x00CC, 0x00D3, 0x00D4,
+    0xF8FF, 0x00D2, 0x00DA, 0x00DB, 0x00D9, 0x0131, 0x02C6, 0x02DC,
+    0x00AF, 0x02D8, 0x02D9, 0x02DA, 0x00B8, 0x02DD, 0x02DB, 0x02C7
+  };
+  uint16_t *c = hb_bsearch (u, mapping, ARRAY_LENGTH (mapping), sizeof (mapping[0]),
+			    _hb_cmp_operator<uint16_t, uint16_t>);
+  return c ? (c - mapping) + 0x7F : 0;
+}
 
 struct CmapSubtableFormat0
 {
@@ -1465,8 +1489,11 @@ struct EncodingRecord
     int ret;
     ret = platformID.cmp (other.platformID);
     if (ret) return ret;
-    ret = encodingID.cmp (other.encodingID);
-    if (ret) return ret;
+    if (other.encodingID != 0xFFFF)
+    {
+      ret = encodingID.cmp (other.encodingID);
+      if (ret) return ret;
+    }
     return 0;
   }
 
@@ -1814,9 +1841,13 @@ struct cmap
                                          c->plan));
   }
 
-  const CmapSubtable *find_best_subtable (bool *symbol = nullptr) const
+  const CmapSubtable *find_best_subtable (bool *symbol = nullptr,
+					  bool *mac = nullptr,
+					  bool *macroman = nullptr) const
   {
     if (symbol) *symbol = false;
+    if (mac) *mac = false;
+    if (macroman) *macroman = false;
 
     const CmapSubtable *subtable;
 
@@ -1841,6 +1872,20 @@ struct cmap
     if ((subtable = this->find_subtable (0, 1))) return subtable;
     if ((subtable = this->find_subtable (0, 0))) return subtable;
 
+    /* MacRoman subtable. */
+    if ((subtable = this->find_subtable (1, 0)))
+    {
+      if (mac) *mac = true;
+      if (macroman) *macroman = true;
+      return subtable;
+    }
+    /* Any other Mac subtable; we just map ASCII for these. */
+    if ((subtable = this->find_subtable (1, 0xFFFF)))
+    {
+      if (mac) *mac = true;
+      return subtable;
+    }
+
     /* Meh. */
     return &Null (CmapSubtable);
   }
@@ -1852,8 +1897,8 @@ struct cmap
     accelerator_t (hb_face_t *face)
     {
       this->table = hb_sanitize_context_t ().reference_table<cmap> (face);
-      bool symbol;
-      this->subtable = table->find_best_subtable (&symbol);
+      bool symbol, mac, macroman;
+      this->subtable = table->find_best_subtable (&symbol, &mac, &macroman);
       this->subtable_uvs = &Null (CmapSubtableFormat14);
       {
 	const CmapSubtable *st = table->find_subtable (0, 5);
@@ -1862,6 +1907,7 @@ struct cmap
       }
 
       this->get_glyph_data = subtable;
+#ifndef HB_NO_CMAP_LEGACY_SUBTABLES
       if (unlikely (symbol))
       {
 	switch ((unsigned) face->table.OS2->get_font_page ()) {
@@ -1881,7 +1927,16 @@ struct cmap
 	  break;
 	}
       }
+      else if (unlikely (macroman))
+      {
+	this->get_glyph_funcZ = get_glyph_from_macroman<CmapSubtable>;
+      }
+      else if (unlikely (mac))
+      {
+	this->get_glyph_funcZ = get_glyph_from_ascii<CmapSubtable>;
+      }
       else
+#endif
       {
 	switch (subtable->u.format) {
 	/* Accelerate format 4 and format 12. */
@@ -2006,6 +2061,28 @@ struct cmap
       return false;
     }
 
+    template <typename Type>
+    HB_INTERNAL static bool get_glyph_from_ascii (const void *obj,
+						  hb_codepoint_t codepoint,
+						  hb_codepoint_t *glyph)
+    {
+      const Type *typed_obj = (const Type *) obj;
+      return codepoint < 0x80 && typed_obj->get_glyph (codepoint, glyph);
+    }
+
+    template <typename Type>
+    HB_INTERNAL static bool get_glyph_from_macroman (const void *obj,
+						     hb_codepoint_t codepoint,
+						     hb_codepoint_t *glyph)
+    {
+      if (get_glyph_from_ascii<Type> (obj, codepoint, glyph))
+	return true;
+
+      const Type *typed_obj = (const Type *) obj;
+      unsigned c = unicode_to_macroman (codepoint);
+      return c && typed_obj->get_glyph (c, glyph);
+    }
+
     private:
     hb_nonnull_ptr_t<const CmapSubtable> subtable;
     hb_nonnull_ptr_t<const CmapSubtableFormat14> subtable_uvs;