You must supply names (identifiers) to satisfy the syntax of SQL
    statements that require user-supplied names. In statement syntax
    diagrams, user-supplied names are shown in lowercase type.

    User-supplied names must:

    o  Be no more than 31 octets (8-bit characters).

    o  Conform to one of the following rules:

       -  If the identifier character set is MCS, the name must
          contain only alphanumeric characters and begin with an
          uppercase or lowercase letter. Alphanumeric characters are
          uppercase or lowercase letters (A, a), including letters
          with diacritical marks (À),  digits, dollar signs ($), and
          underscores (_).

          Uppercase and lowercase letters are treated equally.

          Although dollar signs are valid characters in names, to
          avoid conflicts it is recommended that you do not use them.

          You cannot begin a user-supplied name with a numeric
          character.

       -  If the identifier character set is other than DEC_MCS, it
          can contain only a valid sequence of characters as defined
          by the standard for that character set.

       -  The name can be a delimited identifier. A delimited
          identifier is a user-supplied name enclosed in double
          quotation marks (").  It can start with and contain
          alphanumeric characters, special characters, control
          characters, and spaces. The quotation mark (")  character
          can be included in a delimited identifier by typing two
          quotation marks together (see the following example).
          Trailing spaces are not significant. See the Oracle Rdb
          SQL Reference Manual for a list of special characters.
          The alphabetic characters can be uppercase or lowercase.
          The following example shows many variations of delimited
          identifiers:

          SQL> SET QUOTING RULES 'SQL99
          SQL> CREATE TABLE "This is a Test"
          cont> (""""               CHAR(5),
          cont>  "_NAME"            CHAR(20),
          cont>  "  City"           CHAR(20),
          cont>  "1st_date"         DATE,
          cont>  "A ""B and C"""    CHAR(10),
          cont>  "$_Amount"         INT,
          cont>  """Test"           CHAR(5)
          cont> );
          SQL> SHOW TABLE (COLUMN) "This is a Test";
          Information for table This is a Test

          Columns for table This is a Test:
          Column Name                     Data Type        Domain
          -----------                     ---------        ------
          "                               CHAR(5)
          _NAME                           CHAR(20)
            City                          CHAR(20)
          1st_date                        DATE VMS
          A "B and C"                     CHAR(10)
          $_Amount                        INTEGER
          "Test                           CHAR(5)

          If you use a delimited identifier, SQL maintains the
          case of the identifier. That is, if you enclose the
          identifier Employee_ID in quotation marks ("Employee_ID"),
          SQL preserves the uppercase and lowercase letters. The
          delimited identifiers "Employee_ID", "EMPLOYEE_ID", and
          "employee_id" are distinct from each other.

       You must enable ANSI/ISO SQL quoting when using delimited
       identifiers.

    If you want to use a keyword as a user-supplied name, you must
    set the quoting rules or dialect to SQL99 and use the delimited
    identifier.

    SQL uses the identifier character set as the character set for
    database object names. However, because SQL interprets the names
    of some database objects as file names or path names, you must
    use only ASCII alphanumeric characters for the names of the
    following database objects:

    o  Database file name

    o  Database path name

    o  Snapshot files

    o  Storage areas

    o  Journal files

    o  Alias names