InnoDB support for hlindexes and mhnsw

* mhnsw:
  * use primary key, innodb loves and (and the index cannot have dupes anyway)
    * MyISAM is ok with that, performance-wise
  * must be ha_rnd_init(0) because we aren't going to scan
    * MyISAM resets the position on ha_rnd_init(0) so query it before
    * oh, and use the correct handler, just in case
  * HA_ERR_RECORD_IS_THE_SAME is no error
* innodb:
  * return ref_length on create
  * don't assume table->pos_in_table_list is set
  * ok, assume away, but only for system versioned tables
* set alter_info on create (InnoDB needs to check for FKs)
* pair external_lock/external_unlock correctly

mysql-test/main/vector.combinations

+[innodb]
+innodb
+default-storage-engine=innodb
+
+[myisam]
+default-storage-engine=myisam

mysql-test/main/vector.test

+replace_result InnoDB MyISAM;
 error ER_NO_INDEX_ON_TEMPORARY;
 create temporary table t1 (id int auto_increment primary key, v blob not null, vector index (v));
 
@@ -7,6 +8,7 @@ create table t1 (id int auto_increment primary key,
   v blob not null, vector index (v));
 
 create table t1 (id int auto_increment primary key, v blob not null, vector index (v));
+replace_result InnoDB MyISAM;
 show create table t1;
 show keys from t1;
 query_vertical select * from information_schema.statistics where table_name='t1';

sql/handler.cc

@@ -6413,9 +6413,12 @@ int ha_create_table(THD *thd, const char *path, const char *db,
     DBUG_ASSERT(share.key_info[share.keys].algorithm == HA_KEY_ALG_VECTOR);
     TABLE_SHARE index_share;
     char file_name[FN_REFLEN+1];
+    Alter_info index_ainfo;
     HA_CREATE_INFO index_cinfo;
     char *path_end= strmov(file_name, path);
 
+    index_cinfo.alter_info= &index_ainfo;
+
     if ((error= share.path.length > sizeof(file_name) - HLINDEX_BUF_LEN))
       goto err;
 

sql/sql_base.cc

@@ -9893,8 +9893,6 @@ int TABLE::open_hlindexes_for_write()
   for (uint i= s->keys; i < s->total_keys; i++)
   {
     KEY *key= s->key_info + i;
-    if (hlindex)
-      hlindex->in_use= 0;
     for (uint j=0; j < key->usable_key_parts; j++)
       if (bitmap_is_set(write_set, key->key_part[j].fieldnr - 1))
       {
@@ -9908,8 +9906,11 @@ int TABLE::open_hlindexes_for_write()
 
 int TABLE::reset_hlindexes()
 {
-  if (hlindex)
+  if (hlindex && hlindex->in_use)
+  {
     hlindex->file->ha_external_unlock(in_use);
+    hlindex->in_use= 0;
+  }
   return 0;
 }
 

sql/vector_mhnsw.cc

@@ -283,8 +283,11 @@ static int write_neighbors(MHNSW_Context *ctx, size_t layer,
     err= graph->file->ha_index_read_map(graph->record[1], key,
                                             HA_WHOLE_KEY, HA_READ_KEY_EXACT);
     if (!err)
+    {
       err= graph->file->ha_update_row(graph->record[1], graph->record[0]);
-
+      if (err == HA_ERR_RECORD_IS_THE_SAME)
+        err= 0;
+    }
   }
   my_safe_afree(neighbor_array_bytes, total_size);
   return err;
@@ -426,7 +429,9 @@ int mhnsw_insert(TABLE *table, KEY *keyinfo)
 
   const double NORMALIZATION_FACTOR= 1 / std::log(thd->variables.hnsw_max_connection_per_layer);
 
-  if (int err= h->ha_rnd_init(1))
+  table->file->position(table->record[0]);
+
+  if (int err= h->ha_rnd_init(0))
     return err;
 
   SCOPE_EXIT([h](){ h->ha_rnd_end(); });
@@ -436,15 +441,13 @@ int mhnsw_insert(TABLE *table, KEY *keyinfo)
 
   SCOPE_EXIT([graph](){ graph->file->ha_index_end(); });
 
-  h->position(table->record[0]);
-
   if (int err= graph->file->ha_index_last(graph->record[0]))
   {
     if (err != HA_ERR_END_OF_FILE)
       return err;
 
     // First insert!
-    FVectorNode target(&ctx, h->ref);
+    FVectorNode target(&ctx, table->file->ref);
     ctx.target= ⌖
     return write_neighbors(&ctx, 0, target);
   }
@@ -465,7 +468,7 @@ int mhnsw_insert(TABLE *table, KEY *keyinfo)
   if (ctx.vec_len * sizeof(float) != res->length())
     return bad_value_on_insert(vec_field);
 
-  FVectorNode target(&ctx, h->ref, res->ptr());
+  FVectorNode target(&ctx, table->file->ref, res->ptr());
   ctx.target= ⌖
 
   double new_num= my_rnd(&thd->rand);
@@ -612,7 +615,7 @@ const LEX_CSTRING mhnsw_hlindex_table_def(THD *thd, uint ref_length)
                      "  layer int not null,              "
                      "  src varbinary(%u) not null,      "
                      "  neighbors varbinary(%u) not null,"
-                     "  index (layer, src))              ";
+                     "  primary key (layer, src))        ";
   size_t len= sizeof(templ) + 32;
   char *s= thd->alloc(len);
   len= my_snprintf(s, len, templ, ref_length, 2 * ref_length *

storage/innobase/handler/ha_innodb.cc

@@ -8322,13 +8322,15 @@ calc_row_difference(
 
 	ut_a(buf <= (byte*) original_upd_buff + buff_len);
 
-	const TABLE_LIST *tl= table->pos_in_table_list;
-	const uint8 op_map= tl->trg_event_map | tl->slave_fk_event_map;
-	/* Used to avoid reading history in FK check on DELETE (see MDEV-16210). */
-	prebuilt->upd_node->is_delete =
-		(op_map & trg2bit(TRG_EVENT_DELETE)
-		 && table->versioned(VERS_TIMESTAMP))
-		? VERSIONED_DELETE : NO_DELETE;
+	if (const TABLE_LIST *tl= table->pos_in_table_list)
+	{
+		const uint8 op_map= tl->trg_event_map | tl->slave_fk_event_map;
+		/* Used to avoid reading history in FK check on DELETE (see MDEV-16210). */
+		prebuilt->upd_node->is_delete =
+			(op_map & trg2bit(TRG_EVENT_DELETE)
+			&& table->versioned(VERS_TIMESTAMP))
+			? VERSIONED_DELETE : NO_DELETE;
+	}
 
 	if (prebuilt->versioned_write) {
 		/* Guaranteed by CREATE TABLE, but anyway we make sure we
@@ -13201,6 +13203,11 @@ ha_innobase::create(const char *name, TABLE *form, HA_CREATE_INFO *create_info,
   else if (!error && m_prebuilt)
     m_prebuilt->table= info.table();
 
+  if (form->s->primary_key >= MAX_KEY)
+    ref_length = DATA_ROW_ID_LEN;
+  else
+    ref_length = form->key_info[form->s->primary_key].key_length;
+
   DBUG_RETURN(error);
 }