Roll HarfBuzz to 1.3.0

third_party/harfbuzz-ng/src/hb-ot-shape-complex-indic.cc

 /*
  * Copyright © 2011,2012  Google, Inc.
  *
  *  This is part of HarfBuzz, a text shaping library.
  *
  * Permission is hereby granted, without written agreement and without
  * license or royalty fees, to use, copy, modify, and distribute this
  * software and its documentation for any purpose, provided that the
  * above copyright notice and the following two paragraphs appear in
  * all copies of this software.
@@ skipping 182 matching lines @@
   else if (unlikely (u == 0x1CEDu))
     cat = OT_A;
   /* The following take marks in standalone clusters, similar to Avagraha. */
   else if (unlikely (hb_in_ranges (u, 0xA8F2u, 0xA8F7u,
                                       0x1CE9u, 0x1CECu,
                                       0x1CEEu, 0x1CF1u)))
   {
     cat = OT_Symbol;
     ASSERT_STATIC ((int) INDIC_SYLLABIC_CATEGORY_AVAGRAHA == OT_Symbol);
   }
-  else if (unlikely (u == 0x17DDu)) /* https://github.com/roozbehp/unicode-data/issues/2 */
-  {
-    cat = OT_M;
-    pos = POS_ABOVE_C;
-  }
   else if (unlikely (u == 0x17C6u)) cat = OT_N; /* Khmer Bindu doesn't like to be repositioned. */
   else if (unlikely (hb_in_range (u, 0x2010u, 0x2011u)))
                                     cat = OT_PLACEHOLDER;
   else if (unlikely (u == 0x25CCu)) cat = OT_DOTTEDCIRCLE;
-  else if (unlikely (u == 0xA982u)) cat = OT_SM; /* Javanese repha. */
-  else if (unlikely (u == 0xA9BEu)) cat = OT_CM2; /* Javanese medial ya. */
-  else if (unlikely (u == 0xA9BDu)) { cat = OT_M; pos = POS_POST_C; } /* Javanese vocalic r. */
 
 
   /*
    * Re-assign position.
    */
 
   if ((FLAG_SAFE (cat) & CONSONANT_FLAGS))
   {
     pos = POS_BASE_C;
     if (is_ra (u))
@@ skipping 45 matching lines @@
 enum reph_mode_t {
   REPH_MODE_IMPLICIT,  /* Reph formed out of initial Ra,H sequence. */
   REPH_MODE_EXPLICIT,  /* Reph formed out of initial Ra,H,ZWJ sequence. */
   REPH_MODE_VIS_REPHA, /* Encoded Repha character, no reordering needed. */
   REPH_MODE_LOG_REPHA  /* Encoded Repha character, needs reordering. */
 };
 enum blwf_mode_t {
   BLWF_MODE_PRE_AND_POST, /* Below-forms feature applied to pre-base and post-base. */
   BLWF_MODE_POST_ONLY     /* Below-forms feature applied to post-base only. */
 };
-enum pref_len_t {
-  PREF_LEN_1 = 1,
-  PREF_LEN_2 = 2,
-  PREF_LEN_DONT_CARE = PREF_LEN_2
-};
 struct indic_config_t
 {
   hb_script_t     script;
   bool            has_old_spec;
   hb_codepoint_t  virama;
   base_position_t base_pos;
   reph_position_t reph_pos;
   reph_mode_t     reph_mode;
   blwf_mode_t     blwf_mode;
-  pref_len_t      pref_len;
 };
 
 static const indic_config_t indic_configs[] =
 {
   /* Default.  Should be first. */
-  {HB_SCRIPT_INVALID,   false,      0,BASE_POS_LAST, REPH_POS_BEFORE_POST,REPH_MODE_IMPLICIT, BLWF_MODE_PRE_AND_POST, PREF_LEN_1},
-  {HB_SCRIPT_DEVANAGARI,true, 0x094Du,BASE_POS_LAST, REPH_POS_BEFORE_POST,REPH_MODE_IMPLICIT, BLWF_MODE_PRE_AND_POST, PREF_LEN_DONT_CARE},
-  {HB_SCRIPT_BENGALI,   true, 0x09CDu,BASE_POS_LAST, REPH_POS_AFTER_SUB,  REPH_MODE_IMPLICIT, BLWF_MODE_PRE_AND_POST, PREF_LEN_DONT_CARE},
-  {HB_SCRIPT_GURMUKHI,  true, 0x0A4Du,BASE_POS_LAST, REPH_POS_BEFORE_SUB, REPH_MODE_IMPLICIT, BLWF_MODE_PRE_AND_POST, PREF_LEN_DONT_CARE},
-  {HB_SCRIPT_GUJARATI,  true, 0x0ACDu,BASE_POS_LAST, REPH_POS_BEFORE_POST,REPH_MODE_IMPLICIT, BLWF_MODE_PRE_AND_POST, PREF_LEN_DONT_CARE},
-  {HB_SCRIPT_ORIYA,     true, 0x0B4Du,BASE_POS_LAST, REPH_POS_AFTER_MAIN, REPH_MODE_IMPLICIT, BLWF_MODE_PRE_AND_POST, PREF_LEN_DONT_CARE},
-  {HB_SCRIPT_TAMIL,     true, 0x0BCDu,BASE_POS_LAST, REPH_POS_AFTER_POST, REPH_MODE_IMPLICIT, BLWF_MODE_PRE_AND_POST, PREF_LEN_2},
-  {HB_SCRIPT_TELUGU,    true, 0x0C4Du,BASE_POS_LAST, REPH_POS_AFTER_POST, REPH_MODE_EXPLICIT, BLWF_MODE_POST_ONLY,    PREF_LEN_2},
-  {HB_SCRIPT_KANNADA,   true, 0x0CCDu,BASE_POS_LAST, REPH_POS_AFTER_POST, REPH_MODE_IMPLICIT, BLWF_MODE_POST_ONLY,    PREF_LEN_2},
-  {HB_SCRIPT_MALAYALAM, true, 0x0D4Du,BASE_POS_LAST, REPH_POS_AFTER_MAIN, REPH_MODE_LOG_REPHA,BLWF_MODE_PRE_AND_POST, PREF_LEN_2},
+  {HB_SCRIPT_INVALID,   false,      0,BASE_POS_LAST, REPH_POS_BEFORE_POST,REPH_MODE_IMPLICIT, BLWF_MODE_PRE_AND_POST},
+  {HB_SCRIPT_DEVANAGARI,true, 0x094Du,BASE_POS_LAST, REPH_POS_BEFORE_POST,REPH_MODE_IMPLICIT, BLWF_MODE_PRE_AND_POST},
+  {HB_SCRIPT_BENGALI,   true, 0x09CDu,BASE_POS_LAST, REPH_POS_AFTER_SUB,  REPH_MODE_IMPLICIT, BLWF_MODE_PRE_AND_POST},
+  {HB_SCRIPT_GURMUKHI,  true, 0x0A4Du,BASE_POS_LAST, REPH_POS_BEFORE_SUB, REPH_MODE_IMPLICIT, BLWF_MODE_PRE_AND_POST},
+  {HB_SCRIPT_GUJARATI,  true, 0x0ACDu,BASE_POS_LAST, REPH_POS_BEFORE_POST,REPH_MODE_IMPLICIT, BLWF_MODE_PRE_AND_POST},
+  {HB_SCRIPT_ORIYA,     true, 0x0B4Du,BASE_POS_LAST, REPH_POS_AFTER_MAIN, REPH_MODE_IMPLICIT, BLWF_MODE_PRE_AND_POST},
+  {HB_SCRIPT_TAMIL,     true, 0x0BCDu,BASE_POS_LAST, REPH_POS_AFTER_POST, REPH_MODE_IMPLICIT, BLWF_MODE_PRE_AND_POST},
+  {HB_SCRIPT_TELUGU,    true, 0x0C4Du,BASE_POS_LAST, REPH_POS_AFTER_POST, REPH_MODE_EXPLICIT, BLWF_MODE_POST_ONLY},
+  {HB_SCRIPT_KANNADA,   true, 0x0CCDu,BASE_POS_LAST, REPH_POS_AFTER_POST, REPH_MODE_IMPLICIT, BLWF_MODE_POST_ONLY},
+  {HB_SCRIPT_MALAYALAM, true, 0x0D4Du,BASE_POS_LAST, REPH_POS_AFTER_MAIN, REPH_MODE_LOG_REPHA,BLWF_MODE_PRE_AND_POST},
   {HB_SCRIPT_SINHALA,   false,0x0DCAu,BASE_POS_LAST_SINHALA,
-                                                     REPH_POS_AFTER_MAIN, REPH_MODE_EXPLICIT, BLWF_MODE_PRE_AND_POST, PREF_LEN_DONT_CARE},
-  {HB_SCRIPT_KHMER,     false,0x17D2u,BASE_POS_FIRST,REPH_POS_DONT_CARE,  REPH_MODE_VIS_REPHA,BLWF_MODE_PRE_AND_POST, PREF_LEN_2},
-  {HB_SCRIPT_JAVANESE,  false,0xA9C0u,BASE_POS_FIRST,REPH_POS_DONT_CARE,  REPH_MODE_VIS_REPHA,BLWF_MODE_PRE_AND_POST, PREF_LEN_1},
+                                                     REPH_POS_AFTER_MAIN, REPH_MODE_EXPLICIT, BLWF_MODE_PRE_AND_POST},
+  {HB_SCRIPT_KHMER,     false,0x17D2u,BASE_POS_FIRST,REPH_POS_DONT_CARE,  REPH_MODE_VIS_REPHA,BLWF_MODE_PRE_AND_POST},
 };
 
 
 
 /*
  * Indic shaper.
  */
 
 struct feature_list_t {
   hb_tag_t tag;
@@ skipping 257 matching lines @@
    * 930,94D in 'blwf', not the expected 94D,930 (with new-spec
    * table).  As such, we simply match both sequences.  Seems
    * to work. */
   hb_codepoint_t glyphs[3] = {virama, consonant, virama};
   if (indic_plan->blwf.would_substitute (glyphs  , 2, face) ||
       indic_plan->blwf.would_substitute (glyphs+1, 2, face))
     return POS_BELOW_C;
   if (indic_plan->pstf.would_substitute (glyphs  , 2, face) ||
       indic_plan->pstf.would_substitute (glyphs+1, 2, face))
     return POS_POST_C;
-  unsigned int pref_len = indic_plan->config->pref_len;
-  if ((pref_len == PREF_LEN_2 &&
-       (indic_plan->pref.would_substitute (glyphs  , 2, face) ||
-        indic_plan->pref.would_substitute (glyphs+1, 2, face)))
-   || (pref_len == PREF_LEN_1 &&
-       indic_plan->pref.would_substitute (glyphs+1, 1, face)))
+  if (indic_plan->pref.would_substitute (glyphs  , 2, face) ||
+      indic_plan->pref.would_substitute (glyphs+1, 2, face))
     return POS_POST_C;
   return POS_BASE_C;
 }
 
 
 enum syllable_type_t {
   consonant_syllable,
   vowel_syllable,
   standalone_cluster,
   symbol_cluster,
@@ skipping 485 matching lines @@
       if (info[i  ].indic_category() == OT_Ra &&
           info[i+1].indic_category() == OT_H  &&
           (i + 2 == base ||
            info[i+2].indic_category() != OT_ZWJ))
       {
         info[i  ].mask |= indic_plan->mask_array[BLWF];
         info[i+1].mask |= indic_plan->mask_array[BLWF];
       }
   }
 
-  unsigned int pref_len = indic_plan->config->pref_len;
+  unsigned int pref_len = 2;
   if (indic_plan->mask_array[PREF] && base + pref_len < end)
   {
-    assert (1 <= pref_len && pref_len <= 2);
     /* Find a Halant,Ra sequence and mark it for pre-base reordering processing. */
     for (unsigned int i = base + 1; i + pref_len - 1 < end; i++) {
       hb_codepoint_t glyphs[2];
       for (unsigned int j = 0; j < pref_len; j++)
         glyphs[j] = info[i + j].codepoint;
       if (indic_plan->pref.would_substitute (glyphs, pref_len, face))
       {
         for (unsigned int j = 0; j < pref_len; j++)
           info[i++].mask |= indic_plan->mask_array[PREF];
 
@@ skipping 189 matching lines @@
    * cluster.
    */
 
   bool try_pref = !!indic_plan->mask_array[PREF];
 
   /* Find base again */
   unsigned int base;
   for (base = start; base < end; base++)
     if (info[base].indic_position() >= POS_BASE_C)
     {
-      if (try_pref && base + 1 < end && indic_plan->config->pref_len == 2)
+      if (try_pref && base + 1 < end)
       {
         for (unsigned int i = base + 1; i < end; i++)
           if ((info[i].mask & indic_plan->mask_array[PREF]) != 0)
           {
             if (!(_hb_glyph_info_substituted (&info[i]) &&
                   _hb_glyph_info_ligated_and_didnt_multiply (&info[i])))
             {
               /* Ok, this was a 'pref' candidate but didn't form any.
                * Base is around here... */
               base = i;
@@ skipping 261 matching lines @@
 
 
   /*   o Reorder pre-base reordering consonants:
    *
    *     If a pre-base reordering consonant is found, reorder it according to
    *     the following rules:
    */
 
   if (try_pref && base + 1 < end) /* Otherwise there can't be any pre-base reordering Ra. */
   {
-    unsigned int pref_len = indic_plan->config->pref_len;
     for (unsigned int i = base + 1; i < end; i++)
       if ((info[i].mask & indic_plan->mask_array[PREF]) != 0)
       {
         /*       1. Only reorder a glyph produced by substitution during application
          *          of the <pref> feature. (Note that a font may shape a Ra consonant with
          *          the feature generally but block it in certain contexts.)
          */
         /* Note: We just check that something got substituted.  We don't check that
          * the <pref> feature actually did it...
          *
-         * If pref len is longer than one, then only reorder if it ligated.  If
-         * pref len is one, only reorder if it didn't ligate with other things. */
-        if (_hb_glyph_info_substituted (&info[i]) &&
-            ((pref_len == 1) ^ _hb_glyph_info_ligated_and_didnt_multiply (&info[i])))
+         * Reorder pref only if it ligated. */
+        if (_hb_glyph_info_ligated_and_didnt_multiply (&info[i]))
         {
           /*
            *       2. Try to find a target position the same way as for pre-base matra.
            *          If it is found, reorder pre-base consonant glyph.
            *
            *       3. If position is not found, reorder immediately before main
            *          consonant.
            */
 
           unsigned int new_pos = base;
@@ skipping 216 matching lines @@
   data_destroy_indic,
   NULL, /* preprocess_text */
   NULL, /* postprocess_glyphs */
   HB_OT_SHAPE_NORMALIZATION_MODE_COMPOSED_DIACRITICS_NO_SHORT_CIRCUIT,
   decompose_indic,
   compose_indic,
   setup_masks_indic,
   HB_OT_SHAPE_ZERO_WIDTH_MARKS_NONE,
   false, /* fallback_position */
 };