Hi, Rucha.

I decided i dislike the overall design of the struct json_schema.
Seems better to do it with C++ classes using virtual methods.
So instead of structures we'd have classes as
   json_schema_const
   json_schema_enum
   json_scnema_string
   json_schema_number
  etc.
  they all inherited from the
  class json_schema_item
 {
     virtual parse()
     virtual check()
}
Parsing the schema we can behave like this:
json_parse_schema()
   checks all the keys on the current level
   looking for 'const' 'enum' and 'typ[e' keywords only.
   If it gets 'const' or 'enum' it creates the
    json_schema_const/enum respectively
  For the "type" it creates the appropriate json_schema_xxx and
  callst ->parse() with the same JSON object. So the parse would go over it once
  again looking for the type-specific keywords.
That way we don't need the hash_list as the destructor of the json_schema_xxx class
would free it's resources.

Minor comments.
You compare keys like strncmp((const char*)curr_key, "const", key_len)
all over the code. But it is not correct.
That way 'cons' or 'con' or 'co' or even 'c' - they all will be considered equal to the 'const'.
Just add the 'key_cmp()' function that would compare key lengths then do the strncmp.

> double val= je->s.cs->strntod((char *) je->value, je->value_len, &end;
You do this even if the value_type is not number. Not very safe. Better to avoid doing
that here at all.

That queriy just crashes the server. Didn't yet found why
SET @schema_array= '{
                       "typ": "arra",
                       "prefixItem": [
                                       { "typ": "objec"},
                                       { "typ": "objec",
                                         "propertie": {
                                                         "number1": {"typ":"numbe"},
                                                         "string1": {"typ":"strin"},
                                                         "obj1" :   {"typ":"objec",
                                                                   "propertie": {
                                                                                    "key1" : {"type":"string"}
                                                                                 }
                                                                   }
                                                        },
                                          "required":["obj1"]
                                       }
                                      ],
                        "items":false
                    }';

SELECT JSON_SCHEMA_VALID(@schema_array, '[{"key1":"value1"}, {"number1":1, "string1":"str1", "obj1":{"key1":"val1"}}]');


Feel free to discuss the design details!

Best regards.
HF