path: root/lib/orber/doc/src/ch_idl_to_erlang_mapping.xml

<?xml version="1.0" encoding="utf-8" ?>
<!DOCTYPE chapter SYSTEM "chapter.dtd">

<chapter>
  <header>
    <copyright>
      <year>1997</year><year>2016</year>
      <holder>Ericsson AB. All Rights Reserved.</holder>
    </copyright>
    <legalnotice>
      Licensed under the Apache License, Version 2.0 (the "License");
      you may not use this file except in compliance with the License.
      You may obtain a copy of the License at
 
          http://www.apache.org/licenses/LICENSE-2.0

      Unless required by applicable law or agreed to in writing, software
      distributed under the License is distributed on an "AS IS" BASIS,
      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
      See the License for the specific language governing permissions and
      limitations under the License.
    
    </legalnotice>

    <title>OMG IDL to Erlang Mapping</title>
    <prepared></prepared>
    <docno></docno>
    <date>1998-10-10</date>
    <rev></rev>
    <file>ch_idl_to_erlang_mapping.xml</file>
  </header>

  <section>
    <title>OMG IDL to Erlang Mapping - Overview</title>
    <p>The purpose of OMG IDL, <em>Interface Definition Language</em>, mapping
      is to act as translator between platforms and languages. An IDL 
      specification is supposed to describe data types, object types etc.</p>
    <p>CORBA is independent of the programming language used to construct
      clients or implementations. In order to use the ORB, it is
      necessary for programmers to know how to access ORB functionality 
      from their programming languages. It translates different IDL constructs
      to  a specific programming language. This chapter 
      describes the mapping of OMG IDL constructs to the Erlang programming
      language.</p>
  </section>

  <section>
    <title>OMG IDL Mapping Elements</title>
    <p>A complete language mapping will allow the programmer to have
      access to all ORB functionality in a way that is convenient for
      a specified programming language. 
      </p>
    <p>All mapping must define the following elements: 
      </p>
    <list type="bulleted">
      <item>All OMG IDL basic and constructed types</item>
      <item>References to constants defined in OMG IDL</item>
      <item>References to objects defined in OMG IDL</item>
      <item>Invocations of operations, including passing of
       parameters and receiving of results</item>
      <item>Exceptions, including what happens when an operation
       raises an exception and how the exception parameters are
       accessed</item>
      <item>Access to attributes</item>
      <item>Signatures for operations defined by the ORB, such as
       dynamic invocation interface, the object adapters etc.</item>
      <item>Scopes;
       OMG IDL has several levels of scopes, which are mapped to Erlang's 
       two scopes.</item>
    </list>
  </section>

  <section>
    <title>Getting Started</title>
    <p>To begin with, we should decide which type of objects (i.e. servers) we
      need and if two, or more, should export the same functionality. Let us
      assume that we want to create a system for DB (database) access for different
      kind of users. For example, anyone with a valid password may extract 
      data, but only a few may update the DB. Usually, an application
      is defined within a <c>module</c>, and all global datatypes are defined
      on the top-level. To begin with we create a module and the interfaces we
      need:</p>
    <code type="none">
// DB IDL
#ifndef _DB_IDL_
#define _DB_IDL_
// A module is simply a container
module DB {

  // An interface maps to a CORBA::Object.
  interface CommonUser {

  };

  // Inherit the Consumer interface
  interface Administrator : CommonUser {

  };

  interface Access {

  };

};
#endif    </code>
    <p>Since the <c>Administrator</c> should be able to do the same things as the
      <c>CommonUser</c>, the previous inherits from the latter. The <c>Access</c>
      interface will grant access to the DB.
      Now we are ready to define the functionality and data types we need. But, this
      requires that we know a little bit more about the OMG IDL.</p>
    <note>
      <p>The OMG defines a set of reserved case insensitive key-words, which may
        <em>NOT</em> be used as identifiers (e.g. module name). For more
        information, see
        <seealso marker="#key_words">Reserved Compiler Names and Keywords</seealso></p>
    </note>
  </section>

  <section>
    <title>Basic OMG IDL Types</title>
    <p>The OMG IDL mapping is strongly typed and, even if you have a good knowledge
      of CORBA types, it is essential to read carefully the following mapping to
      Erlang types.</p>
    <p>The mapping of basic types is straightforward. Note that the
      OMG IDL double type is mapped to an Erlang float which does not
      support the full double value range.</p>
    <table>
      <row>
        <cell align="left" valign="middle">OMG IDL type</cell>
        <cell align="left" valign="middle">Erlang type</cell>
        <cell align="left" valign="middle">Note</cell>
      </row>
      <row>
        <cell align="left" valign="middle">float</cell>
        <cell align="left" valign="middle">Erlang float</cell>
        <cell align="left" valign="middle"></cell>
      </row>
      <row>
        <cell align="left" valign="middle">double</cell>
        <cell align="left" valign="middle">Erlang float</cell>
        <cell align="left" valign="middle">value range not supported</cell>
      </row>
      <row>
        <cell align="left" valign="middle">short</cell>
        <cell align="left" valign="middle">Erlang integer</cell>
        <cell align="left" valign="middle">-2^15 .. 2^15-1</cell>
      </row>
      <row>
        <cell align="left" valign="middle">unsigned short</cell>
        <cell align="left" valign="middle">Erlang integer</cell>
        <cell align="left" valign="middle">0 .. 2^16-1</cell>
      </row>
      <row>
        <cell align="left" valign="middle">long</cell>
        <cell align="left" valign="middle">Erlang integer</cell>
        <cell align="left" valign="middle">-2^31 .. 2^31-1</cell>
      </row>
      <row>
        <cell align="left" valign="middle">unsigned long</cell>
        <cell align="left" valign="middle">Erlang integer</cell>
        <cell align="left" valign="middle">0 .. 2^32-1</cell>
      </row>
      <row>
        <cell align="left" valign="middle">long long</cell>
        <cell align="left" valign="middle">Erlang integer</cell>
        <cell align="left" valign="middle">-2^63 .. 2^63-1</cell>
      </row>
      <row>
        <cell align="left" valign="middle">unsigned long long</cell>
        <cell align="left" valign="middle">Erlang integer</cell>
        <cell align="left" valign="middle">0 .. 2^64-1</cell>
      </row>
      <row>
        <cell align="left" valign="middle">char</cell>
        <cell align="left" valign="middle">Erlang integer</cell>
        <cell align="left" valign="middle">ISO-8859-1</cell>
      </row>
      <row>
        <cell align="left" valign="middle">wchar</cell>
        <cell align="left" valign="middle">Erlang integer</cell>
        <cell align="left" valign="middle">UTF-16 (ISO-10646-1:1993)</cell>
      </row>
      <row>
        <cell align="left" valign="middle">boolean</cell>
        <cell align="left" valign="middle">Erlang atom</cell>
        <cell align="left" valign="middle">true/false</cell>
      </row>
      <row>
        <cell align="left" valign="middle">octet</cell>
        <cell align="left" valign="middle">Erlang integer</cell>
        <cell align="left" valign="middle"></cell>
      </row>
      <row>
        <cell align="left" valign="middle">any</cell>
        <cell align="left" valign="middle">Erlang record</cell>
        <cell align="left" valign="middle">#any{typecode, value}</cell>
      </row>
      <row>
        <cell align="left" valign="middle">long double</cell>
        <cell align="left" valign="middle">Not supported</cell>
        <cell align="left" valign="middle"></cell>
      </row>
      <row>
        <cell align="left" valign="middle">Object</cell>
        <cell align="left" valign="middle">Orber object reference</cell>
        <cell align="left" valign="middle">Internal Representation</cell>
      </row>
      <row>
        <cell align="left" valign="middle">void</cell>
        <cell align="left" valign="middle">Erlang atom</cell>
        <cell align="left" valign="middle">ok</cell>
      </row>
      <tcaption>OMG IDL basic types</tcaption>
    </table>
    <p>The <c>any</c> value is written as a record with the field typecode which
      contains the <term id="Type Code"><termdef>Type Code is a full definition of a type </termdef></term>representation,
      <seealso marker="#tk_values">see also the Type Code table</seealso>,
      and the value field itself.</p>
    <p>Functions with return type <c>void</c> will return the atom <c>ok</c>.</p>
  </section>

  <section>
    <title>Template OMG IDL Types and Complex Declarators</title>
    <p>Constructed types all have native mappings as shown in the table
      below.</p>
    <table>
      <row>
        <cell align="left" valign="middle"><em>Type</em></cell>
        <cell align="left" valign="middle"><em>IDL code</em></cell>
        <cell align="left" valign="middle"><em>Maps to</em></cell>
        <cell align="left" valign="middle"><em>Erlang code</em></cell>
      </row>
      <row>
        <cell align="left" valign="middle"><em>string</em></cell>
        <cell align="left" valign="middle">typedef string S;        <br></br>
void op(in S a);</cell>
        <cell align="left" valign="middle">Erlang string</cell>
        <cell align="left" valign="middle">ok = op(Obj, "Hello World"),</cell>
      </row>
      <row>
        <cell align="left" valign="middle"><em>wstring</em></cell>
        <cell align="left" valign="middle">typedef wstring S;        <br></br>
void op(in S a);</cell>
        <cell align="left" valign="middle">Erlang list of Integers</cell>
        <cell align="left" valign="middle">ok = op(Obj, "Hello World"),</cell>
      </row>
      <row>
        <cell align="left" valign="middle"><em>sequence</em></cell>
        <cell align="left" valign="middle">typedef sequence &lt;long, 3&gt; S;        <br></br>
void op(in S a);</cell>
        <cell align="left" valign="middle">Erlang list</cell>
        <cell align="left" valign="middle">ok = op(Obj, [1, 2, 3]),</cell>
      </row>
      <row>
        <cell align="left" valign="middle"><em>array</em></cell>
        <cell align="left" valign="middle">typedef string S[2];        <br></br>
void op(in S a);</cell>
        <cell align="left" valign="middle">Erlang tuple</cell>
        <cell align="left" valign="middle">ok = op(Obj, {"one", "two"}),</cell>
      </row>
      <row>
        <cell align="left" valign="middle"><em>fixed</em></cell>
        <cell align="left" valign="middle">typedef fixed&lt;3,2> myFixed;        <br></br>
void op(in myFixed a);</cell>
        <cell align="left" valign="middle">Erlang tuple</cell>
        <cell align="left" valign="middle">MF = fixed:create(3, 2, 314),        <br></br>
ok = op(Obj, MF),</cell>
      </row>
      <tcaption>OMG IDL Template and Complex Declarators</tcaption>
    </table>

    <section>
      <title>String/WString Data Types</title>
      <p>A <c>string</c> consists of all possible 8-bit quantities except null.
        Most ORB:s uses, including Orber, the character set Latin-1 (ISO-8859-1).
        The <c>wstring</c> type is represented as a list of integers, where
        each integer represents a wide character. In this case Orber uses, as
        most other ORB:s, the UTF-16 (ISO-10646-1:1993) character set.</p>
      <p>When defining a a string or wstring they can be of limited length or
        null terminated:</p>
      <code type="none"><![CDATA[
// Null terminated
typedef string myString;
typedef wstring myWString;
// Maximum length 10 
typedef string<10> myString10;
typedef wstring<10> myWString10;
      ]]></code>
      <p>If we want to define a char/string or wchar/wstring constant, we can
        use octal (\OOO - one, two or three octal digits), 
        hexadecimal (\xHH - one or two hexadecimal digits) and unicode (\uHHHH -
        one, two, three or four hexadecimal digits.) representation as well.
        For example:</p>
      <code type="none">
const string  SwedensBestSoccerTeam = "\101" "\x49" "\u004B";
const wstring SwedensBestHockeyTeam = L"\101\x49\u004B";
const char  aChar  = '\u004B';  
const wchar aWchar = L'\u004C';
      </code>
      <p>Naturally, we can use <c>"Erlang"</c>, <c>L"Rocks"</c>, <c>'A'</c>
        and <c>L'A'</c> as well.</p>
    </section>

    <section>
      <title>Sequence Data Type</title>
      <p>A sequence can be defined to be of a maximum length or unbounded, and may 
        contain Basic and Template types and scoped names:</p>
      <code type="none"><![CDATA[
typedef sequence <short, 1> aShortSequence;
typedef sequence <long> aLongSequence;
typedef sequence <aLongSequence> anEvenLongerSequence;
      ]]></code>
    </section>

    <section>
      <title>Array Data Type</title>
      <p>Arrays are multidimensional, fixed-size arrays. The indices is language
        mapping specific, which is why one should not pass them as arguments
        to another ORB.</p>
      <code type="none">
typedef long myMatrix[2][3];
      </code>
    </section>

    <section>
      <title>Fixed Data Type</title>
      <p>A Fixed Point literal consists of an integer part (decimal digits), 
        decimal point and a fraction part (decimal digits),
        followed by a <c>D</c> or <c>d</c>. Either the integer part or the
        fraction part may be missing; the decimal point may be missing, 
        but not d/D. The integer part must be a positive integer less than 32.
        The Fraction part must be a positive integer less than or equal to
        the Integer part.</p>
      <code type="none">
const fixed myFixed1 = 3.14D;
const fixed myFixed2 = .14D;
const fixed myFixed3 = 0.14D;
const fixed myFixed4 = 3.D;
const fixed myFixed5 = 3D;
      </code>
      <p>It is also possible to use unary (+-) and binary (+-*/) operators:</p>
      <code type="none">
const fixed myFixed6 = 3D + 0.14D;
const fixed myFixed7 = -3.14D;
      </code>
      <p>The Fixed Point examples above are, so called, <em>anonymous</em> 
        definitions. In later CORBA specifications these have been deprecated
        as function parameters or return values. Hence, we strongly recommend that
        you do not use them. Instead, you should use:</p>
      <code type="none"><![CDATA[
typedef fixed<5,3> myFixed53;
const myFixed53 myFixed53constant = 03.140d;
typedef fixed<3,2> myFixed32;
const myFixed32 myFixed32constant = 3.14d;

myFixed53 foo(in myFixed32 MF); // OK
void bar(in fixed<5,3> MF); // Illegal
      ]]></code>
    </section>
    <p>For more information, see <seealso marker="fixed">Fixed</seealso> in 
      Orber's Reference Manual.</p>
    <p>Now we continue to work on our IDL specification. To begin with, we want
      to limit the size of the logon parameters (Id and password). Since the
      <c>UserID</c> and <c>Password</c> parameters, only will be used when
      invoking operations on the <c>Access</c> interface, we may choose to define
      them within the scope that interface. To keep it simple our DB will contain
      employee information. Hence, as the DB key we choose an integer
      (<c>EmployeeNo</c>).</p>
    <code type="none"><![CDATA[
// DB IDL
#ifndef _DB_IDL_
#define _DB_IDL_
module DB {

  typedef unsigned long EmployeeNo;

  interface CommonUser {

     any lookup(in EmployeeNo ENo);

  };

  interface Administrator : CommonUser {

     void delete(in EmployeeNo ENo);

  };

  interface Access {

      typedef string<10> UserID;
      typedef string<10> Password;

      CommonUser logon(in UserID ID, in Password PW);

  };

};
#endif    ]]></code>
    <p>But what should, for example, the <c>lookup</c> operation return? One option
      is to use the <c>any</c> data type. But, depending on what kind of data it
      encapsulates, this datatype can be rather expensive to use. We might find a 
      solution to our problems among the <c>Constructed</c> IDL types.</p>
  </section>

  <section>
    <title>Constructed OMG IDL Types</title>
    <p>Constructed types all have native mappings as shown in the table
      below.</p>
    <table>
      <row>
        <cell align="left" valign="middle"><em>Type</em></cell>
        <cell align="left" valign="middle"><em>IDL code</em></cell>
        <cell align="left" valign="middle"><em>Maps to</em></cell>
        <cell align="left" valign="middle"><em>Erlang code</em></cell>
      </row>
      <row>
        <cell align="left" valign="middle"><em>struct</em></cell>
        <cell align="left" valign="middle">struct myStruct {        <br></br>
long a;         <br></br>
short b;        <br></br>
};        <br></br>
void op(in myStruct a);</cell>
        <cell align="left" valign="middle">Erlang record</cell>
        <cell align="left" valign="middle">ok = op(Obj, #'myStruct'{a=300, b=127}),</cell>
      </row>
      <row>
        <cell align="left" valign="middle"><em>union</em></cell>
        <cell align="left" valign="middle">union myUnion switch(long) {        <br></br>
case 1: long a;        <br></br>
};        <br></br>
void op(in myUnion a);</cell>
        <cell align="left" valign="middle">Erlang record</cell>
        <cell align="left" valign="middle">ok = op(Obj, #'myUnion'{label=1, value=66}),</cell>
      </row>
      <row>
        <cell align="left" valign="middle"><em>enum</em></cell>
        <cell align="left" valign="middle">enum myEnum {one, two};        <br></br>
void op(in myEnum a);</cell>
        <cell align="left" valign="middle">Erlang atom</cell>
        <cell align="left" valign="middle">ok = op(Obj, one),</cell>
      </row>
      <tcaption>OMG IDL constructed types</tcaption>
    </table>

    <section>
      <title>Struct Data Type</title>
      <p>A <c>struct</c> may have Basic, Template, Scoped Names and Constructed 
        types as members. By using forward declaration we can define a recursive struct:</p>
      <code type="none"><![CDATA[
struct myStruct; // Forward declaration
typedef sequence<myStruct> myStructSeq;
struct myStruct {
    myStructSeq chain;
};

// Deprecated definition (anonymous) not supported by IC
struct myStruct {
    sequence<myStruct> chain;
};
      ]]></code>
    </section>

    <section>
      <title>Enum Data Type</title>
      <p>The maximum number of identifiers which may defined in an enumeration
        is 2&sup3;&sup2;. The order in which the identifiers are named in the
        specification of an enumeration defines the relative order of the
        identifiers.</p>
    </section>

    <section>
      <title>Union Data Type</title>
      <p>A <c>union</c> may consist of:</p>
      <list type="bulleted">
        <item>Identifier</item>
        <item>Switch - may be an integer, char, boolean, enum or scoped name.</item>
        <item>Body - with or without a <c>default</c> case; may appear at 
         most once.</item>
      </list>
      <p>A case label must match the defined type of the discriminator, and may only
        contain a default case if the values given in the non-default labels do
        not cover the entire range of the union's discriminant type. For example:</p>
      <code type="none">
// Illegal default; all cases covered by 
// non-default cases.
union BooleanUnion switch(boolean) {
  case TRUE:  long TrueValue;
  case FALSE: long FalseValue;
  default: long DefaultValue; 
};
// OK
union BooleanUnion2 switch(boolean) {
  case TRUE:  long TrueValue;
  default: long DefaultValue; 
};
      </code>
      <p>It is not necessary to list all possible values of the union discriminator
        in the body. Hence, the value of a union is the value of the discriminator
        and, in given order, one of the following:</p>
      <list type="ordered">
        <item>If the discriminator match a label, explicitly listed in a 
         case statement, the value must be of the same type.</item>
        <item>If the union contains a default label, the value must match the
         type of the default label.</item>
        <item>No value. Orber then inserts the Erlang atom <c>undefined</c>
         in the value field when receiving a union from an external 
         ORB.</item>
      </list>
      <p>The above can be summed up to:</p>
      <code type="none">
// If the discriminator equals 1 or 2 the value 
// is a long. Otherwise, the atom undefined.
union LongUnion switch(long) {
  case 1:
  case 2:  long TrueValue;
};
// If the discriminator equals 1 or 2 the value 
// is a long. Otherwise, a boolean.
union LongUnion2 switch(long) {
  case 1:
  case 2:  long TrueValue;
  default: boolean DefaultValue; 
};
      </code>
      <p>In the same way as structs, unions can be recursive if forward
         declaration is used (anonymous types is deprecated and not supported):</p>
      <code type="none"><![CDATA[
// Forward declaration
union myUnion;
typedef sequence<myUnion>myUnionSeq;
union myUnion switch (long) {
    case 1 : myUnionSeq chain;
    default: boolean DefaultValue;
};
      ]]></code>

    <note>
      <p>Recursive types (union and struct) require Light IFR. I.e. the
         IC option {light_ifr, true} is used and that Orber is configured in such a way that
         Light IFR is activated. Recursive TypeCode is currently not supported, which is
         why these cannot be encapsulated in an any data type.</p>
    </note>

    </section>
    <warning>
      <p>Every field in, for example, a struct must be initiated. Otherwise
        it will be set to the atom <c>undefined</c>, which Orber cannot
        encode when communicating via IIOP. In the example above, invoking
        the operation with #'myStruct'{a=300} will fail (equal to 
        #'myStruct'{a=300, b=undefined})</p>
    </warning>
    <p>Now we can continue to work on our IDL specification. To begin with, we should
      determine the return value of the <c>lookup</c> operation. Since the <c>any</c>
      type can be rather expensive we can use a <c>struct</c> or a <c>union</c> instead.
      If we intend to return the same information about a employee every time we can
      use a struct. Let us assume that the DB contains the name, address, employee 
      number and department.</p>
    <code type="none"><![CDATA[
// DB IDL
#ifndef _DB_IDL_
#define _DB_IDL_
module DB {

  typedef unsigned long EmployeeNo;

  enum Department {Department1, Department2};

  struct employee {
        EmployeeNo No;
        string Name; 
        string Address; 
        Department Dpt;
  };

  typedef employee EmployeeData;

  interface CommonUser {

     EmployeeData lookup(in EmployeeNo ENo);

  };

  interface Administrator : CommonUser {

     void delete(in EmployeeNo ENo);

  };

  interface Access {

      typedef string<10> UserID;
      typedef string<10> Password;

      // Since Administrator inherits from CommonUser
      // the returned Object can be of either type.
      CommonUser logon(in UserID ID, in Password PW);

  };

};
#endif    ]]></code>
    <p>We can also define exceptions (i.e. not system exception) thrown by
      each interface. Since exceptions are thoroughly described in the chapter
      <seealso marker="ch_exceptions">System and User Defined Exceptions</seealso>,
      we choose not to. Hence, we are now ready to compile our IDL-file by 
      invoking:</p>
    <pre>
$ <input>erlc DB.idl</input>
    </pre>
    <p>or:</p>
    <pre>
$ <input>erl</input>
Erlang (BEAM) emulator version 5.1.1 [threads:0]

Eshell V5.1.1  (abort with ^G)
1> <input>ic:gen('DB').</input>
ok
2> <input>halt().</input>
    </pre>
    <p>The next step is to implement our servers. But, to be able to do that,
      we need to know how we can access data type definitions. For example,
      since a struct is mapped to an Erlang record we must include an hrl-file
      in our callback module.</p>
  </section>

  <section>
    <title>Scoped Names and Generated Files</title>

    <section>
      <title>Scoped Names</title>
      <p>Within a scope all identifiers must be unique. The following kinds of
        definitions form scopes in the OMG IDL:</p>
      <list type="bulleted">
        <item><em>module</em></item>
        <item><em>interface</em></item>
        <item><em>operation</em></item>
        <item><em>valuetype</em></item>
        <item><em>struct</em></item>
        <item><em>union</em></item>
        <item><em>exception</em></item>
      </list>
      <p>For example, since enumerants do not form a scope, the following IDL code
        is not valid:</p>
      <code type="none">
module MyModule {
     // 'two' is not unique
     enum MyEnum {one, two};
     enum MyOtherEnum {two, three};
};
      </code>
      <p>But, since Erlang only has two levels of scope, <em>module</em> and 
        <em>function</em>, the OMG IDL scope is mapped as follows:</p>
      <list type="bulleted">
        <item><em>Function Scope</em> - used for constants, operations and attributes.</item>
        <item><em>Erlang Module Scope</em> - the Erlang module scope
         handles the remaining OMG IDL scopes.</item>
      </list>
      <p>An Erlang module, corresponding to an IDL global name, is derived by 
        converting occurrences of "::" to underscore, and eliminating
        the leading "::". Hence, accessing <c>MyEnum</c> from another module, one 
        use <c>MyModule::MyEnum</c></p>
      <p>For example, an operation <c>foo</c> defined in interface <c>I</c>, which
        is defined in module <c>M</c>, would be written in IDL as <c>M::I::foo</c>
        and as <c>'M_I':foo</c> in Erlang - <c>foo</c> is the function
        name and <c>'M_I'</c> is the name of the Erlang module. Applying this
        knowledge to a stripped version of the DB.idl gives:</p>
      <code type="none"><![CDATA[
// DB IDL
#ifndef _DB_IDL_
#define _DB_IDL_
// ++ topmost scope ++ 
// IC generates oe_XX.erl and oe_XX.hrl.
// XX is equal to the name of the IDL-file.
// Tips: create one IDL-file for each top module
// and give the file the same name (DB.idl).
// The oe_XX.erl module is used to register data
// in the IFR.
module DB {

  // ++ Module scope ++
  // To access 'EmployeeNo' from another scope, use:
  // DB::EmployeeNo, DB::Access etc.
  typedef unsigned long EmployeeNo;

  enum Department {Department1, Department2};

  // Definitions of this struct is contained in:
  // DB.hrl
  // Access functions exported by:
  // DB_employee.erl
  struct employee {
     ... CUT ...
  };

  typedef employee EmployeeData;

  ... CUT ...

  // If this interface should inherit an interface
  // in another module (e.g. OtherModule) use:
  // interface Access : OtherModule::OtherInterface
  interface Access {

      // ++ interface scope ++
      // Types within this scope is accessible via:
      // DB::Access::UserID
      // The Stub/Skeleton for this interface is
      // placed in the module:
      // DB_Access.erl
      typedef string<10> UserID;
      typedef string<10> Password;

      // Since Administrator inherits from CommonUser
      // the returned Object can be of either type.
      // This operation is exported from:
      // DB_Access.erl
      CommonUser logon(in UserID ID, in Password PW);

  };

};
#endif      ]]></code>
      <p>Using underscores in IDL names can lead to ambiguities
        due to the name mapping described above. It is advisable to 
        avoid the use of underscores in identifiers. For example, the following
        definition would generate two structures named <c>x_y_z</c>.</p>
      <code type="none">
module x {

    struct y_z {
    ...
    };

    interface y {

    struct z {
        ...
    };
    };
};
      </code>
    </section>

    <section>
      <title>Generated Files</title>
      <p>Several files can be generated for each scope.</p>
      <list type="bulleted">
        <item>An Erlang source code file (<c>.erl</c>) is generated
         for top level scope as well as the Erlang header file.</item>
        <item>An Erlang header file (<c>.hrl</c>) will be generated for 
         each scope. The header file will contain record definitions 
         for all <c>struct</c>, <c>union</c> and <c>exception</c> 
         types in that scope.</item>
        <item>Modules that contain at least one constant definition,
         will produce Erlang source code files (<c>.erl</c>).
         That Erlang file will contain constant functions for 
         that scope.
         Modules that contain no constant definitions are considered
         empty and no code will be produced for them, but only for
         their included modules/interfaces.</item>
        <item>Interfaces will produce Erlang source code files (<c>.erl</c>),
         this code will contain all operation stub code and implementation
         functions.</item>
        <item>In addition to the scope-related files, an Erlang source file will
         be generated for each definition of the types <c>struct</c>,
        <c>union</c> and <c>exception</c> (these are the types that
         will be represented in Erlang as records).
         This file will contain special access functions for that record.</item>
        <item>The top level scope will produce two files, one header file
         (<c>.hrl</c>) and one Erlang source file (<c>.erl</c>).
         These files are named as the IDL file, prefixed with <c>oe_</c>.</item>
      </list>
      <p>After compiling DB.idl, the following files have been generated:</p>
      <list type="bulleted">
        <item><c>oe_DB.hrl</c> and <c>oe_DB.erl</c> for the top scope level.</item>
        <item><c>DB.hrl</c> for the module <c>DB</c>.</item>
        <item><c>DB_Access.hrl</c> and <c>DB_Access.erl</c> for the interface 
        <c>DB_Access</c>.</item>
        <item><c>DB_CommonUser.hrl</c> and <c>DB_CommonUser.erl</c> for the interface 
        <c>DB_CommonUser</c>.</item>
        <item><c>DB_Administrator.hrl</c> and <c>DB_Administrator.erl</c> for the interface 
        <c>DB_Administrator</c>.</item>
        <item><c>DB_employee.erl</c> for the structure <c>employee</c> in module 
        <c>DB</c>.</item>
      </list>
      <p>Since the <c>employee</c> struct is defined in the top level scope,
        the Erlang record definition is found in <c>DB.hrl</c>. IC also generates
        stubs/skeletons (e.g. <c>DB_CommonUser.erl</c>) and access functions for
        some datatypes (e.g. <c>DB_employee.erl</c>). How the stubs/skeletons are
        used is thoroughly described in
        <seealso marker="ch_stubs">Stubs/Skeletons</seealso> and
        <seealso marker="Module_Interface">Module_Interface</seealso>.</p>
    </section>
  </section>

  <section>
    <title>Typecode, Identity and Name Access Functions</title>
    <p>As mentioned in a previous section, <c>struct</c>, <c>union</c> and 
      <c>exception</c> types yield record definitions and access code 
      for that record. 
      For <c>struct</c>, <c>union</c>, <c>exception</c>, <c>array</c> and 
      <c>sequence</c> types, a special file is generated that holds access
      functions for <c>TypeCode</c>, <c>Identity</c> and <c>Name</c>.
      These functions are put in the file corresponding to the scope where 
      they are defined. For example, the module <c>DB_employee.erl</c>, 
      representing the <c>employee</c> struct, exports the following functions:</p>
    <list type="bulleted">
      <item>tc/0 - returns the type code for the struct.</item>
      <item>id/0 - returns the IFR identity of the struct. In this case
       the returned value is <c>"IDL:DB/employee:1.0"</c>, but
       if the struct was defined in the scope of <c>CommonUser</c>,
       the result would be <c>"IDL:DB/CommonUser/employee:1.0"</c>.
       However, the user usually do not need to know the Id, just
       which Erlang module contains the correct Id.</item>
      <item>name/0 - returns the scoped name of the struct. The <c>employee</c>
       struct name is <c>"DB_employee"</c>.</item>
    </list>
    <p><term id="Type Codes"><termdef>Type codes give a complete description  of the type including all its components and structure.</termdef></term>are, for example, used in <seealso marker="any">Any</seealso> values.
      Hence, we can encapsulate the <c>employee</c> struct in an <c>any</c>
      type by:</p>
    <code type="none">
%% Erlang code
....
AnEmployee = #'DB_employee'{'No'      = 1,
                            'Name'    = "Adam Ivan Kendall",
                            'Address' = "Rasunda, Solna",
                            'Dpt'     = 'Department1'},
EmployeeTC = 'DB_employee':tc(),
EmployeeAny = any:create(EmployeeTC, AnEmployee),
....
    </code>
    <p>For more information, see the
      <seealso marker="#tk_values">Type Code listing</seealso>.</p>
  </section>

  <section>
    <title>References to Constants</title>
    <p>Constants are generated as Erlang functions, and are accessed by a
      single function call. The functions are put in the file
      corresponding to the scope where they are defined. There is no
      need for an object to be started to access a constant.</p>
    <p>Example:</p>
    <code type="none">
// m.idl
module m {
    const float pi = 3.14;

    interface i {
    const float pi = 3.1415;
    };
};
    </code>
    <p>Since the two constants are defined in different scopes, the IDL code
      above is valid, but not necessarily a good approach. After compiling
      <c>m.idl</c>, the constant definitions can be extracted by invoking:</p>
    <pre>
$ <input>erlc m.idl</input>
$ <input>erlc m.erl</input>
$ <input>erl</input>
Erlang (BEAM) emulator version 5.1.1 [threads:0]

Eshell V5.1.1  (abort with ^G)
1> <input>m:pi().</input>
3.14
2> <input>m_i:pi().</input>
3.1415
3> <input>halt().</input>
    </pre>
  </section>

  <section>
    <title>References to Objects Defined in OMG IDL</title>
    <p>Objects are accessed by object references. An object reference
      is an opaque Erlang term created and maintained by the ORB.</p>
    <p>Objects are implemented by providing implementations for all
      operations and attributes of the Object, <seealso marker="#op_impl">see operation implementation</seealso>.</p>
  </section>

  <section>
    <title>Exceptions</title>
    <p>Exceptions are handled as Erlang catch and throws. Exceptions
      are translated to messages over an IIOP bridge but converted
      back to a throw on the receiving side. Object implementations
      that invoke operations on other objects must be aware of the
      possibility of a non-local return. This includes invocation of
      ORB and IFR services. See also the
      <seealso marker="ch_exceptions">Exceptions</seealso> section.</p>
    <p>Exception parameters are mapped as an Erlang record and accessed
      as such.</p>
    <p>An object implementation that raises an exception will use the
      <c>corba:raise/1</c> function, passing the exception record as
      parameter.</p>
  </section>

  <section>
    <title>Access to Attributes</title>
    <p>Attributes are accessed through their access functions. An
      attribute implicitly defines the <c>_get</c> and <c>_set</c>
      operations. These operations are handled in the same way as
      normal operations. The <c>_get</c> operation is defined as a <c>readonly</c> 
      attribute.</p>
    <code type="none">
readonly attribute long RAttribute;
attribute long RWAttribute;
    </code>
    <p>The <c>RAttribute</c> requires that you implement, in your call-back module,
      <c>_get_RAttribute</c>. For the <c>RWAttribute</c> it is necessary to implement
      <c>_get_RWAttribute</c> and <c>_set_RWAttribute</c>.</p>
  </section>

  <section>
    <title>Invocations of Operations</title>
    <marker id="op_impl"></marker>
    <p>A standard Erlang <c>gen_server</c> behavior is used for
      object implementation. The <c>gen_server</c> state is then
      used as the object internal state. Implementation of the object 
      function is achieved by implementing its methods and attribute operations.
      These functions will usually have the internal state as their first parameter, 
      followed by any <c>in</c> and <c>inout</c> parameters. </p>
    <p>Do not confuse the
      object internal state with its object reference. The object internal state is
      an Erlang term which has a format defined by the user.</p>
    <note>
      <p>It is not always the case that the internal state will be the first parameter, as stubs can use their own object reference as the first parameter (see the IC documentation).</p>
    </note>
    <p>A function call will invoke an operation. The first
      parameter of the function should be the object reference and then
      all <c>in</c> and <c>inout</c> parameters follow in the same
      order as specified in the IDL specification. The result will be a return value
      unless the function has <c>inout</c> or <c>out</c> parameters specified; 
      in which case, a tuple of the return value, followed by the parameters will
      be returned.</p>
    <p>Example:</p>
    <code type="none">
// IDL
module m {
  interface i {
      readonly attribute long RAttribute;
      attribute long RWAttribute;
      long foo(in short a);
      long bar(in char c, inout string s, out long count);
      void baz(out long Id);
  };
};
    </code>
    <p>Is used in Erlang as :</p>
    <code type="none">
%% Erlang code
....
Obj = ...    %% get object reference
RAttr  = m_i:'_get_RAttribute'(Obj),
RWAttr = m_i:'_get_RWAttribute'(Obj),
ok = m_i:'_set_RWAttribute'(Obj, Long),
R1 = m_i:foo(Obj, 55),
{R2, S, Count} = m_i:bar(Obj, $a, "hello"),
....
    </code>
    <p>Note how the <c>inout</c> parameter is passed <em>and</em>
      returned. There is no way to use a single occurrence of a
      variable for this in Erlang. Also note, that <c>ok</c>, Orber's 
      representation of the IDL-type <c>void</c>, must be returned by
      <c>baz</c> and <c>'_set_RWAttribute'</c>. 
      These operations can be implemented in the call-back module as:</p>
    <code type="none">
'_set_RWAttribute'(State, Long) ->
    {reply, ok, State}.

'_get_RWAttribute'(State) ->
    {reply, Long, State}.

'_get_RAttribute'(State) ->
    {reply, Long, State}.

foo(State, AShort) ->
    {reply, ALong, State}.
 
bar(State, AShort, AString) ->
    {reply, {ALong, "MyString", ALong}, State}.
 
baz(State) ->
    {reply, {ok, AId}, State}.
    </code>
    <p>The operations may require more arguments (depends on IC options used). For
      more information, see <seealso marker="ch_stubs">Stubs/Skeletons</seealso>
      and <seealso marker="Module_Interface">Module_Interface</seealso>.</p>
    <warning>
      <p>A function can also be defined to be <c>oneway</c>, i.e. 
        asynchronous. But, since the behavior of a oneway operation is not
        defined in the OMG specifications (i.e. the behavior can differ depending on
        which other ORB Orber is communicating with), one should avoid using it.</p>
    </warning>
  </section>

  <section>
    <title>Implementing the DB Application</title>
    <p>Now we are ready to implement the call-back modules. There are three modules
      we must create:</p>
    <list type="bulleted">
      <item>DB_Access_impl.erl</item>
      <item>DB_CommonUser_impl.erl</item>
      <item>DB_Administrator_impl.erl</item>
    </list>
    <p>An easy way to accomplish that, is to use the IC backend <c>erl_template</c>,
      which will generate a complete call-back module. One should also add
      the same compile options, for example <c>this</c> or <c>from</c>,
      used when generating the stub/skeleton modules:</p>
    <code type="none">
$> erlc +"{be,erl_template}" DB.idl
    </code>
    <p>We begin with implementing the <c>DB_Access_impl.erl</c> module, which,
      if we used <c>erl_template</c>, will look like the following. All we need
      to do is to add the logic to the <c>logon</c> operation.</p>
    <code type="none"><![CDATA[
%%----------------------------------------------------------------------
%% <LICENSE>
%% 
%%     $Id$
%%
%%----------------------------------------------------------------------
%% Module       : DB_Access_impl.erl
%% 
%% Source       : /home/user/example/DB.idl
%% 
%% Description  : 
%% 
%% Creation date: 2005-05-20
%%
%%----------------------------------------------------------------------
-module('DB_Access_impl').

-export([logon/3]).

%%----------------------------------------------------------------------
%% Internal Exports
%%----------------------------------------------------------------------
-export([init/1,
         terminate/2,
         code_change/3,
         handle_info/2]).

%%----------------------------------------------------------------------
%% Include Files
%%----------------------------------------------------------------------


%%----------------------------------------------------------------------
%% Macros
%%----------------------------------------------------------------------


%%----------------------------------------------------------------------
%% Records
%%----------------------------------------------------------------------
-record(state, {}).

%%======================================================================
%% API Functions
%%======================================================================
%%----------------------------------------------------------------------
%% Function   : logon/3
%% Arguments  : State - term()
%%              ID = String()
%%              PW = String()
%% Returns    : ReturnValue = OE_Reply
%%              OE_Reply = Object_Ref()
%% Raises     : 
%% Description: 
%%----------------------------------------------------------------------
logon(State, ID, PW) ->
    %% Check if the ID/PW is valid and what 
    %% type of user it is (Common or Administrator).
    OE_Reply
            = case check_user(ID, PW) of
             {ok, administrator} ->
                'DB_Administrator':oe_create();
             {ok, common} ->
                'DB_CommonUser':oe_create();
             error ->
                %% Here we should throw an exception              
                corba:raise(....)
        end,
    {reply, OE_Reply, State}.

%%======================================================================
%% Internal Functions
%%======================================================================
%%----------------------------------------------------------------------
%% Function   : init/1
%% Arguments  : Env = term()
%% Returns    : {ok, State}          |
%%              {ok, State, Timeout} |
%%              ignore               |
%%              {stop, Reason}
%% Raises     : -
%% Description: Initiates the server
%%----------------------------------------------------------------------
init(_Env) ->
    {ok, #state{}}.


%%----------------------------------------------------------------------
%% Function   : terminate/2
%% Arguments  : Reason = normal | shutdown | term()
%%              State = term()
%% Returns    : ok
%% Raises     : -
%% Description: Invoked when the object is terminating.
%%----------------------------------------------------------------------
terminate(_Reason, _State) ->
    ok.


%%----------------------------------------------------------------------
%% Function   : code_change/3
%% Arguments  : OldVsn = undefined | term()
%%              State = NewState = term()
%%              Extra = term()
%% Returns    : {ok, NewState}
%% Raises     : -
%% Description: Invoked when the object should update its internal state
%%              due to code replacement.
%%----------------------------------------------------------------------
code_change(_OldVsn, State, _Extra) ->
    {ok, State}.


%%----------------------------------------------------------------------
%% Function   : handle_info/2
%% Arguments  : Info = normal | shutdown | term()
%%              State = NewState = term()
%% Returns    : {noreply, NewState}          |
%%              {noreply, NewState, Timeout} |
%%              {stop, Reason, NewState}
%% Raises     : -
%% Description: Invoked when, for example, the server traps exits.
%%----------------------------------------------------------------------
handle_info(_Info, State) ->
    {noreply, State}.
    ]]></code>
    <p>Since <c>DB_Administrator</c> inherits from <c>DB_CommonUser</c>,
      we must implement <c>delete</c> in the <c>DB_Administrator_impl.erl</c>
      module, and <c>lookup</c> in <c>DB_Administrator_impl.erl</c><em>and</em><c>DB_CommonUser_impl.erl</c>. But wait, is that really necessary? Actually,
      it is not. We simple use the IC compile option <em>impl</em>:</p>
    <pre>
$ <input>erlc +'{{impl, "DB::CommonUser"}, "DBUser_impl"}'\
 +'{{impl, "DB::Administrator"}, "DBUser_impl"}' DB.idl</input>
$ <input>erlc *.erl</input>
    </pre>
    <p>Instead of creating, and not the least, maintaining two call-back modules,
      we only have to deal with <c>DBUser_impl.erl</c>. If we generated the
      templates, we simply rename <c>DB_Administrator_impl.erl</c> to
      <c>DBUser_impl.erl</c>. See also the
      <seealso marker="ch_exceptions">Exceptions</seealso> chapter.
      In the following example, only the implementation of the API functions
      are shown:</p>
    <code type="none">
%%======================================================================
%% API Functions
%%======================================================================
%%----------------------------------------------------------------------
%% Function   : delete/2
%% Arguments  : State - term()
%%              ENo = unsigned_Long()
%% Returns    : ReturnValue = ok
%% Raises     : 
%% Description: 
%%----------------------------------------------------------------------
delete(State, ENo) ->
        %% How we access the DB, for example mnesia, is not shown here.
        case delete_employee(No) of
            ok ->
                {reply, ok, State};
            error ->
                %% Here we should throw an exception if
                %% there is no match.
                corba:raise(....)
        end.

%%----------------------------------------------------------------------
%% Function   : lookup/2
%% Arguments  : State - term()
%%              ENo = unsigned_Long()
%% Returns    : ReturnValue = OE_Reply
%%              OE_Reply = #'DB_employee'{No,Name,Address,Dpt}
%%              No = unsigned_Long()
%%              Name = String()
%%              Address = String()
%%              Dpt = Department
%%              Department = 'Department1' | 'Department2' 
%% Raises     : 
%% Description: 
%%----------------------------------------------------------------------
lookup(State, ENo) ->
        %% How we access the DB, for example mnesia, is not shown here.
        case lookup_employee(ENo) of
            %% We assume that we receive a 'DB_employee' struct
            {ok, Employee} ->
                OE_Reply = Employee,
                {reply, OE_Reply, State};
            error ->
                %% Here we should throw an exception if
                %% there is no match.
                corba:raise(....)
        end.
    </code>
    <p>After you have compiled both call-back modules, and implemented the missing
      functionality (e.g. lookup_employee/1), we can test our application:</p>
    <code type="none">
%% Erlang code
....
%% Create an Access object
Acc = 'DB_Access':oe_create(),

%% Login is Common user and Administrator
Adm = 'DB_Access':logon(A, "admin", "pw"),
Com = 'DB_Access':logon(A, "comm", "pw"),

%% Lookup existing employee
Employee = 'DB_Administrator':lookup(Adm, 1),
Employee = 'DB_CommonUser':lookup(Adm, 1),

%% If we try the same using the DB_CommonUser interface 
%% it result in an exit since that operation is not exported.
{'EXIT', _} = (catch 'DB_CommonUser':delete(Adm, 1)),

%% Try to delete the employee via the CommonUser Object
{'EXCEPTION', _} = (catch 'DB_Administrator':delete(Com, 1)),

%% Invoke delete operation on the Administrator object
ok = 'DB_Administrator':delete(Adm, 1),
....
    </code>
  </section>

  <section>
    <title>Reserved Compiler Names and Keywords</title>
    <marker id="key_words"></marker>
    <p>The use of some names is  strongly discouraged due to 
      ambiguities. However, the use of some names is prohibited 
      when using the Erlang mapping , as they are strictly reserved for IC.</p>
    <p>IC reserves all identifiers starting with <c>OE_</c> and <c>oe_</c> 
      for internal use.</p>
    <p>Note also, that an identifier in IDL can  contain alphabetic,
      digits and underscore characters, but the first character 
      <em>must</em> be alphabetic.
      </p>
    <p>The OMG defines a set of reserved words, shown below, for use as keywords. 
      These may <em>not</em> be used as, for example, identifiers. The keywords
      which are not in bold face was introduced in the OMG CORBA-3.0 
      specification.</p>
    <table>
      <row>
        <cell align="left" valign="middle"><em>abstract</em></cell>
        <cell align="left" valign="middle"><em>exception</em></cell>
        <cell align="left" valign="middle"><em>inout</em></cell>
        <cell align="left" valign="middle">provides</cell>
        <cell align="left" valign="middle"><em>truncatable</em></cell>
      </row>
      <row>
        <cell align="left" valign="middle"><em>any</em></cell>
        <cell align="left" valign="middle">emits</cell>
        <cell align="left" valign="middle"><em>interface</em></cell>
        <cell align="left" valign="middle"><em>public</em></cell>
        <cell align="left" valign="middle"><em>typedef</em></cell>
      </row>
      <row>
        <cell align="left" valign="middle"><em>attribute</em></cell>
        <cell align="left" valign="middle"><em>enum</em></cell>
        <cell align="left" valign="middle"><em>local</em></cell>
        <cell align="left" valign="middle">publishes</cell>
        <cell align="left" valign="middle">typeid</cell>
      </row>
      <row>
        <cell align="left" valign="middle"><em>boolean</em></cell>
        <cell align="left" valign="middle">eventtype</cell>
        <cell align="left" valign="middle"><em>long</em></cell>
        <cell align="left" valign="middle"><em>raises</em></cell>
        <cell align="left" valign="middle">typeprefix</cell>
      </row>
      <row>
        <cell align="left" valign="middle"><em>case</em></cell>
        <cell align="left" valign="middle"><em>factory</em></cell>
        <cell align="left" valign="middle"><em>module</em></cell>
        <cell align="left" valign="middle"><em>readonly</em></cell>
        <cell align="left" valign="middle"><em>unsigned</em></cell>
      </row>
      <row>
        <cell align="left" valign="middle"><em>char</em></cell>
        <cell align="left" valign="middle"><em>FALSE</em></cell>
        <cell align="left" valign="middle">multiple</cell>
        <cell align="left" valign="middle">setraises</cell>
        <cell align="left" valign="middle"><em>union</em></cell>
      </row>
      <row>
        <cell align="left" valign="middle">component</cell>
        <cell align="left" valign="middle">finder</cell>
        <cell align="left" valign="middle"><em>native</em></cell>
        <cell align="left" valign="middle"><em>sequence</em></cell>
        <cell align="left" valign="middle">uses</cell>
      </row>
      <row>
        <cell align="left" valign="middle"><em>const</em></cell>
        <cell align="left" valign="middle"><em>fixed</em></cell>
        <cell align="left" valign="middle"><em>Object</em></cell>
        <cell align="left" valign="middle"><em>short</em></cell>
        <cell align="left" valign="middle"><em>ValueBase</em></cell>
      </row>
      <row>
        <cell align="left" valign="middle">consumes</cell>
        <cell align="left" valign="middle"><em>float</em></cell>
        <cell align="left" valign="middle"><em>octet</em></cell>
        <cell align="left" valign="middle"><em>string</em></cell>
        <cell align="left" valign="middle"><em>valuetype</em></cell>
      </row>
      <row>
        <cell align="left" valign="middle"><em>context</em></cell>
        <cell align="left" valign="middle">getraises</cell>
        <cell align="left" valign="middle"><em>oneway</em></cell>
        <cell align="left" valign="middle"><em>struct</em></cell>
        <cell align="left" valign="middle"><em>void</em></cell>
      </row>
      <row>
        <cell align="left" valign="middle"><em>custom</em></cell>
        <cell align="left" valign="middle">home</cell>
        <cell align="left" valign="middle"><em>out</em></cell>
        <cell align="left" valign="middle"><em>supports</em></cell>
        <cell align="left" valign="middle"><em>wchar</em></cell>
      </row>
      <row>
        <cell align="left" valign="middle"><em>default</em></cell>
        <cell align="left" valign="middle">import</cell>
        <cell align="left" valign="middle">primarykey</cell>
        <cell align="left" valign="middle"><em>switch</em></cell>
        <cell align="left" valign="middle"><em>wstring</em></cell>
      </row>
      <row>
        <cell align="left" valign="middle"><em>double</em></cell>
        <cell align="left" valign="middle"><em>in</em></cell>
        <cell align="left" valign="middle"><em>private</em></cell>
        <cell align="left" valign="middle"><em>TRUE</em></cell>
        <cell align="left" valign="middle"></cell>
      </row>
      <tcaption>OMG IDL keywords</tcaption>
    </table>
    <p>The keywords listed above must be written exactly as shown. Any usage
      of identifiers that collide with a keyword is illegal. For example,
      <em>long</em> is a valid keyword; <em>Long</em> and <em>LONG</em> are
      illegal as keywords and identifiers. But, since the OMG must be able
      to expand the IDL grammar, it is possible to use <em>Escaped  Identifiers</em>. For example, it is not unlikely that <c>native</c>
      have been used in IDL-specifications as identifiers. One option is to
      change all occurrences to <c>myNative</c>. Usually, it is necessary
      to change programming language code that depends upon that IDL as well.
      Since Escaped Identifiers just disable type checking (i.e. if it is a reserved
      word or not) and leaves everything else unchanged, it is only necessary to
      update the IDL-specification. To escape an identifier, simply prefix it
      with <em>_</em>. The following IDL-code is illegal:</p>
    <code type="none">
typedef string native;
interface i {
   void foo(in native Arg);
   };
};
    </code>
    <p>With Escaped Identifiers the code will look like:</p>
    <code type="none">
typedef string _native;
interface i {
   void foo(in _native Arg);
   };
};
    </code>
  </section>

  <section>
    <title>Type Code Representation</title>
    <marker id="tk_values"></marker>
    <p>Type Codes are used in <c>any</c> values. To avoid mistakes, you should
      use access functions exported by the Data Types modules
      (e.g. struct, union etc) or the <seealso marker="orber_tc">orber_tc</seealso>
      module.</p>
    <table>
      <row>
        <cell align="left" valign="middle"><em>Type Code</em></cell>
        <cell align="left" valign="middle"><em>Example</em></cell>
      </row>
      <row>
        <cell align="left" valign="middle">tk_null</cell>
        <cell align="left" valign="middle"></cell>
      </row>
      <row>
        <cell align="left" valign="middle">tk_void</cell>
        <cell align="left" valign="middle"></cell>
      </row>
      <row>
        <cell align="left" valign="middle">tk_short</cell>
        <cell align="left" valign="middle"></cell>
      </row>
      <row>
        <cell align="left" valign="middle">tk_long</cell>
        <cell align="left" valign="middle"></cell>
      </row>
      <row>
        <cell align="left" valign="middle">tk_longlong</cell>
        <cell align="left" valign="middle"></cell>
      </row>
      <row>
        <cell align="left" valign="middle">tk_ushort</cell>
        <cell align="left" valign="middle"></cell>
      </row>
      <row>
        <cell align="left" valign="middle">tk_ulong</cell>
        <cell align="left" valign="middle"></cell>
      </row>
      <row>
        <cell align="left" valign="middle">tk_ulonglong</cell>
        <cell align="left" valign="middle"></cell>
      </row>
      <row>
        <cell align="left" valign="middle">tk_float</cell>
        <cell align="left" valign="middle"></cell>
      </row>
      <row>
        <cell align="left" valign="middle">tk_double</cell>
        <cell align="left" valign="middle"></cell>
      </row>
      <row>
        <cell align="left" valign="middle">tk_boolean</cell>
        <cell align="left" valign="middle"></cell>
      </row>
      <row>
        <cell align="left" valign="middle">tk_char</cell>
        <cell align="left" valign="middle"></cell>
      </row>
      <row>
        <cell align="left" valign="middle">tk_wchar</cell>
        <cell align="left" valign="middle"></cell>
      </row>
      <row>
        <cell align="left" valign="middle">tk_octet</cell>
        <cell align="left" valign="middle"></cell>
      </row>
      <row>
        <cell align="left" valign="middle">tk_any</cell>
        <cell align="left" valign="middle"></cell>
      </row>
      <row>
        <cell align="left" valign="middle">tk_TypeCode</cell>
        <cell align="left" valign="middle"></cell>
      </row>
      <row>
        <cell align="left" valign="middle">tk_Principal</cell>
        <cell align="left" valign="middle"></cell>
      </row>
      <row>
        <cell align="left" valign="middle">{tk_objref, IFRId, Name}</cell>
        <cell align="left" valign="middle">{tk_objref, "IDL:M1\I1:1.0", "I1"}</cell>
      </row>
      <row>
        <cell align="left" valign="middle">{tk_struct, IFRId, Name, [{ElemName, ElemTC}]}</cell>
        <cell align="left" valign="middle">{tk_struct, "IDL:M1\S1:1.0", "S1", [{"a", tk_long}, {"b", tk_char}]}</cell>
      </row>
      <row>
        <cell align="left" valign="middle">{tk_union, IFRId, Name, DiscrTC, DefaultNr, [{Label, ElemName, ElemTC}]}         <br></br>
Note: DefaultNr tells which of tuples in the case list that is default, or -1 if no default</cell>
        <cell align="left" valign="middle">{tk_union, "IDL:U1:1.0", "U1", tk_long, 1, [{1, "a", tk_long}, {default, "b", tk_char}]}</cell>
      </row>
      <row>
        <cell align="left" valign="middle">{tk_enum, IFRId, Name, [ElemName]}</cell>
        <cell align="left" valign="middle">{tk_enum, "IDL:E1:1.0", "E1", ["a1", "a2"]}</cell>
      </row>
      <row>
        <cell align="left" valign="middle">{tk_string, Length}</cell>
        <cell align="left" valign="middle">{tk_string, 5}</cell>
      </row>
      <row>
        <cell align="left" valign="middle">{tk_wstring, Length}</cell>
        <cell align="left" valign="middle">{tk_wstring, 7}</cell>
      </row>
      <row>
        <cell align="left" valign="middle">{tk_fixed, Digits, Scale}</cell>
        <cell align="left" valign="middle">{tk_fixed, 3, 2}</cell>
      </row>
      <row>
        <cell align="left" valign="middle">{tk_sequence, ElemTC, Length}</cell>
        <cell align="left" valign="middle">{tk_sequence, tk_long, 4}</cell>
      </row>
      <row>
        <cell align="left" valign="middle">{tk_array, ElemTC, Length}</cell>
        <cell align="left" valign="middle">{tk_array, tk_char, 9}</cell>
      </row>
      <row>
        <cell align="left" valign="middle">{tk_alias, IFRId, Name, TC}</cell>
        <cell align="left" valign="middle">{tk_alias, "IDL:T1:1.0", "T1", tk_short}</cell>
      </row>
      <row>
        <cell align="left" valign="middle">{tk_except, IFRId, Name, [{ElemName, ElemTC}]}</cell>
        <cell align="left" valign="middle">{tk_except, "IDL:Exc1:1.0", "Exc1", [{"a", tk_long}, {"b", {tk_string, 0}}]}</cell>
      </row>
      <tcaption>Type Code tuples</tcaption>
    </table>
  </section>
</chapter>