path: root/lib/erl_interface/doc/src/erl_eterm.xml

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

<cref>
  <header>
    <copyright>
      <year>1996</year><year>2013</year>
      <holder>Ericsson AB. All Rights Reserved.</holder>
    </copyright>
    <legalnotice>
      The contents of this file are subject to the Erlang Public License,
      Version 1.1, (the "License"); you may not use this file except in
      compliance with the License. You should have received a copy of the
      Erlang Public License along with this software. If not, it can be
      retrieved online at http://www.erlang.org/.
    
      Software distributed under the License is distributed on an "AS IS"
      basis, WITHOUT WARRANTY OF ANY KIND, either express or implied. See
      the License for the specific language governing rights and limitations
      under the License.
    
    </legalnotice>

    <title>erl_eterm</title>
    <prepared>Torbj&ouml;rn T&ouml;rnkvist</prepared>
    <responsible>Torbj&ouml;rn T&ouml;rnkvist</responsible>
    <docno></docno>
    <approved>Bjarne D&auml;cker</approved>
    <checked>Torbj&ouml;rn T&ouml;rnkvist</checked>
    <date>980703</date>
    <rev>A</rev>
    <file>erl_eterm.sgml</file>
  </header>
  <lib>erl_eterm</lib>
  <libsummary>Functions for Erlang Term Construction</libsummary>
  <description>
    <p>This module contains functions for creating and manipulating
      Erlang terms. </p>
    <p>An Erlang term is represented by a C structure of type
      <c><![CDATA[ETERM]]></c>. Applications should not reference any fields in this
      structure directly, because it may be changed in future releases
      to provide faster and more compact term storage. Instead,
      applications should us the macros and functions provided. </p>
    <p>The following macros each take a single ETERM pointer as an
      argument. They return a non-zero value if the test is true, and 0
      otherwise:</p>
    <taglist>
      <tag><c><![CDATA[ERL_IS_INTEGER(t)]]></c></tag>
      <item>True if <c><![CDATA[t]]></c> is an integer.</item>
      <tag><c><![CDATA[ERL_IS_UNSIGNED_INTEGER(t)]]></c></tag>
      <item>True if <c><![CDATA[t]]></c> is an integer.</item>
      <tag><c><![CDATA[ERL_IS_FLOAT(t)]]></c></tag>
      <item>True if <c><![CDATA[t]]></c> is a floating point number.</item>
      <tag><c><![CDATA[ERL_IS_ATOM(t)]]></c></tag>
      <item>True if <c><![CDATA[t]]></c> is an atom.</item>
      <tag><c><![CDATA[ERL_IS_PID(t)]]></c></tag>
      <item>True if <c><![CDATA[t]]></c> is a Pid (process identifier).</item>
      <tag><c><![CDATA[ERL_IS_PORT(t)]]></c></tag>
      <item>True if <c><![CDATA[t]]></c> is a port.</item>
      <tag><c><![CDATA[ERL_IS_REF(t)]]></c></tag>
      <item>True if <c><![CDATA[t]]></c> is a reference.</item>
      <tag><c><![CDATA[ERL_IS_TUPLE(t)]]></c></tag>
      <item>True if <c><![CDATA[t]]></c> is a tuple.</item>
      <tag><c><![CDATA[ERL_IS_BINARY(t)]]></c></tag>
      <item>True if <c><![CDATA[t]]></c> is a binary.</item>
      <tag><c><![CDATA[ERL_IS_LIST(t)]]></c></tag>
      <item>True if <c><![CDATA[t]]></c> is a list with zero or more elements.</item>
      <tag><c><![CDATA[ERL_IS_EMPTY_LIST(t)]]></c></tag>
      <item>True if <c><![CDATA[t]]></c> is an empty list.</item>
      <tag><c><![CDATA[ERL_IS_CONS(t)]]></c></tag>
      <item>True if <c><![CDATA[t]]></c> is a list with at least one element.</item>
    </taglist>
    <p>The following macros can be used for retrieving parts of Erlang
      terms. None of these do any type checking; results are undefined
      if you pass an ETERM* containing the wrong type. For example,
      passing a tuple to ERL_ATOM_PTR() will likely result in garbage.
      </p>
    <taglist>
      <tag><c><![CDATA[char *ERL_ATOM_PTR(t)]]></c></tag>
      <tag><c><![CDATA[char *ERL_ATOM_PTR_UTF8(t)]]></c></tag>
      <item>A string representing atom <c><![CDATA[t]]></c>.
      </item>
      <tag><c><![CDATA[int ERL_ATOM_SIZE(t)]]></c></tag>
      <tag><c><![CDATA[int ERL_ATOM_SIZE_UTF8(t)]]></c></tag>
      <item>The length (in bytes) of atom t.</item>
      <tag><c><![CDATA[void *ERL_BIN_PTR(t)]]></c></tag>
      <item>A pointer to the contents of <c><![CDATA[t]]></c></item>
      <tag><c><![CDATA[int ERL_BIN_SIZE(t)]]></c></tag>
      <item>The length (in bytes) of binary object <c><![CDATA[t]]></c>.</item>
      <tag><c><![CDATA[int ERL_INT_VALUE(t)]]></c></tag>
      <item>The integer of <c><![CDATA[t]]></c>.</item>
      <tag><c><![CDATA[unsigned int ERL_INT_UVALUE(t)]]></c></tag>
      <item>The unsigned integer value of <c><![CDATA[t]]></c>.</item>
      <tag><c><![CDATA[double ERL_FLOAT_VALUE(t)]]></c></tag>
      <item>The floating point value of <c><![CDATA[t]]></c>.</item>
      <tag><c><![CDATA[ETERM *ERL_PID_NODE(t)]]></c></tag>
      <tag><c><![CDATA[ETERM *ERL_PID_NODE_UTF8(t)]]></c></tag>
      <item>The Node in pid <c><![CDATA[t]]></c>.</item>
      <tag><c><![CDATA[int ERL_PID_NUMBER(t)]]></c></tag>
      <item>The sequence number in pid <c><![CDATA[t]]></c>.</item>
      <tag><c><![CDATA[int ERL_PID_SERIAL(t)]]></c></tag>
      <item>The serial number in pid <c><![CDATA[t]]></c>.</item>
      <tag><c><![CDATA[int ERL_PID_CREATION(t)]]></c></tag>
      <item>The creation number in pid <c><![CDATA[t]]></c>.</item>
      <tag><c><![CDATA[int ERL_PORT_NUMBER(t)]]></c></tag>
      <item>The sequence number in port <c><![CDATA[t]]></c>.</item>
      <tag><c><![CDATA[int ERL_PORT_CREATION(t)]]></c></tag>
      <item>The creation number in port <c><![CDATA[t]]></c>.</item>
      <tag><c><![CDATA[ETERM *ERL_PORT_NODE(t)]]></c></tag>
      <tag><c><![CDATA[ETERM *ERL_PORT_NODE_UTF8(t)]]></c></tag>
      <item>The node in port <c><![CDATA[t]]></c>.</item>
      <tag><c><![CDATA[int ERL_REF_NUMBER(t)]]></c></tag>
      <item>The first part of the reference number in ref <c><![CDATA[t]]></c>. Use
       only for compatibility.</item>
      <tag><c><![CDATA[int ERL_REF_NUMBERS(t)]]></c></tag>
      <item>Pointer to the array of reference numbers in ref <c><![CDATA[t]]></c>.</item>
      <tag><c><![CDATA[int ERL_REF_LEN(t)]]></c></tag>
      <item>The number of used reference numbers in ref <c><![CDATA[t]]></c>.</item>
      <tag><c><![CDATA[int ERL_REF_CREATION(t)]]></c></tag>
      <item>The creation number in ref <c><![CDATA[t]]></c>.</item>
      <tag><c><![CDATA[int ERL_TUPLE_SIZE(t)]]></c></tag>
      <item>The number of elements in tuple <c><![CDATA[t]]></c>.</item>
      <tag><c><![CDATA[ETERM *ERL_CONS_HEAD(t)]]></c></tag>
      <item>The head element of list <c><![CDATA[t]]></c>.</item>
      <tag><c><![CDATA[ETERM *ERL_CONS_TAIL(t)]]></c></tag>
      <item>A List representing the tail elements of list <c><![CDATA[t]]></c>.</item>
    </taglist>
  </description>
  <funcs>
    <func>
      <name><ret>ETERM *</ret><nametext>erl_cons(head, tail)</nametext></name>
      <fsummary>Prepends a term to the head of a list.</fsummary>
      <type>
        <v>ETERM *head;</v>
        <v>ETERM *tail;</v>
      </type>
      <desc>
        <p>This function concatenates two Erlang terms, prepending
          <c><![CDATA[head]]></c> onto <c><![CDATA[tail]]></c> and thereby creating a <c><![CDATA[cons]]></c> cell.
          To make a proper list, <c><![CDATA[tail]]></c> should always be a
          list or an empty list. Note that NULL is not a valid list.</p>
        <p><c><![CDATA[head]]></c> is the new term to be added.</p>
        <p><c><![CDATA[tail]]></c> is the existing list to which <c><![CDATA[head]]></c> will
          be concatenated.</p>
        <p>The function returns a new list.</p>
        <p><c><![CDATA[ERL_CONS_HEAD(list)]]></c> and <c><![CDATA[ERL_CONS_TAIL(list)]]></c>
          can be used to retrieve the head and tail components
          from the list. <c><![CDATA[erl_hd(list)]]></c> and <c><![CDATA[erl_tl(list)]]></c> will do
          the same thing, but check that the argument really is a list.</p>
        <p>For example:</p>
        <code type="none"><![CDATA[
ETERM *list,*anAtom,*anInt;
anAtom = erl_mk_atom("madonna");
anInt  = erl_mk_int(21);
list   = erl_mk_empty_list();
list   = erl_cons(anAtom, list);
list   = erl_cons(anInt, list);
 ... /* do some work */
erl_free_compound(list);
        ]]></code>
      </desc>
    </func>
    <func>
      <name><ret>ETERM *</ret><nametext>erl_copy_term(term)</nametext></name>
      <fsummary>Creates a copy of an Erlang term</fsummary>
      <type>
        <v>ETERM *term;</v>
      </type>
      <desc>
        <p>This function creates and returns a copy of the Erlang term 
          <c><![CDATA[term]]></c>.</p>
      </desc>
    </func>
    <func>
      <name><ret>ETERM *</ret><nametext>erl_element(position, tuple)</nametext></name>
      <fsummary>Extracts an element from an Erlang tuple</fsummary>
      <type>
        <v>int position;</v>
        <v>ETERM *tuple;</v>
      </type>
      <desc>
        <p>This function extracts a specified element from an Erlang
          tuple. </p>
        <p><c><![CDATA[position]]></c> specifies which element to retrieve from
          <c><![CDATA[tuple]]></c>. The elements are numbered starting from 1.</p>
        <p><c><![CDATA[tuple]]></c> is an Erlang term containing at least
          <c><![CDATA[position]]></c> elements.</p>
        <p>The function returns a new Erlang term corresponding to the
          requested element, or NULL if <c><![CDATA[position]]></c> was greater than
          the arity of <c><![CDATA[tuple]]></c>.</p>
      </desc>
    </func>
    <func>
      <name><ret>void</ret><nametext>erl_init(NULL, 0)</nametext></name>
      <fsummary>Initialization routine</fsummary>
      <type>
        <v>void *NULL;</v>
        <v>int 0;</v>
      </type>
      <desc>
        <marker id="erl_init"></marker>
        <p>This function must be called before any of the others in 
          the <c><![CDATA[erl_interface]]></c> library in order to initialize the
          library functions. The arguments must be specified as 
          <c><![CDATA[erl_init(NULL,0)]]></c>.</p>
      </desc>
    </func>
    <func>
      <name><ret>ETERM *</ret><nametext>erl_hd(list)</nametext></name>
      <fsummary>Extracts the first element from a list</fsummary>
      <type>
        <v>ETERM *list;</v>
      </type>
      <desc>
        <p>Extracts the first element from a list.</p>
        <p><c><![CDATA[list]]></c> is an Erlang term containing a list.</p>
        <p>The function returns an Erlang term corresponding to the
          head element in the list, or a NULL pointer if <c><![CDATA[list]]></c> was
          not a list.</p>
      </desc>
    </func>
    <func>
      <name><ret>ETERM *</ret><nametext>erl_iolist_to_binary(term)</nametext></name>
      <fsummary>Converts an IO list to a binary</fsummary>
      <type>
        <v>ETERM *list;</v>
      </type>
      <desc>
        <p>This function converts an IO list to a binary term.</p>
        <p><c><![CDATA[list]]></c> is an Erlang term containing a list.</p>
        <p>This function an Erlang binary term, or NULL if <c><![CDATA[list]]></c>
          was not an IO list. </p>
        <p>Informally, an IO list is a deep list of characters and
          binaries which can be sent to an Erlang port. In BNF, an IO
          list is formally defined as follows: </p>
        <code type="none"><![CDATA[
iolist ::= []
        |   Binary
        |   [iohead | iolist]
        ;
iohead ::= Binary
        |   Byte (integer in the range [0..255])
        |   iolist
        ;
        ]]></code>
      </desc>
    </func>
    <func>
      <name><ret>char *</ret><nametext>erl_iolist_to_string(list)</nametext></name>
      <fsummary>Converts an IO list to a zero terminated string</fsummary>
      <type>
        <v>ETERM *list;</v>
      </type>
      <desc>
        <p>This function converts an IO list to a '\0' terminated C
          string. </p>
        <p><c><![CDATA[list]]></c> is an Erlang term containing an IO list. The IO
          list must not contain the integer 0, since C strings may not
          contain this value except as a terminating marker.</p>
        <p>This function returns a pointer to a dynamically allocated
          buffer containing a string. If <c><![CDATA[list]]></c> is not an IO list,
          or if <c><![CDATA[list]]></c> contains the integer 0, NULL is returned. It
          is the caller's responsibility free the allocated buffer
          with <c><![CDATA[erl_free()]]></c>. </p>
        <p>Refer to <c><![CDATA[erl_iolist_to_binary()]]></c> for the definition of an
          IO list. </p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>erl_iolist_length(list)</nametext></name>
      <fsummary>Return the length of an IO list</fsummary>
      <type>
        <v>ETERM *list;</v>
      </type>
      <desc>
        <p>Returns the length of an IO list.
          </p>
        <p><c><![CDATA[list]]></c> is an Erlang term containing an IO list. </p>
        <p>The function returns the length of <c><![CDATA[list]]></c>, or -1 if
          <c><![CDATA[list]]></c> is not an IO list.</p>
        <p>Refer to <c><![CDATA[erl_iolist_to_binary()]]></c> for the definition of
          an IO list. </p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>erl_length(list)</nametext></name>
      <fsummary>Determines the length of a list</fsummary>
      <type>
        <v>ETERM *list;</v>
      </type>
      <desc>
        <p>Determines the length of a proper list.</p>
        <p><c><![CDATA[list]]></c> is an Erlang term containing proper list. In a
          proper list, all tails except the last point to another list
          cell, and the last tail points to an empty list.</p>
        <p>Returns -1 if <c><![CDATA[list]]></c> is not a proper list.</p>
      </desc>
    </func>
    <func>
      <name><ret>ETERM *</ret><nametext>erl_mk_atom(string)</nametext></name>
      <fsummary>Creates an atom</fsummary>
      <type>
        <v>const char *string;</v>
      </type>
      <desc>
        <p>Creates an atom.</p>
        <p><c><![CDATA[string]]></c> is the sequence of characters that will be
          used to create the atom.</p>
        <p>Returns an Erlang term containing an atom. Note that it is
          the callers responsibility to make sure that <c><![CDATA[string]]></c>
          contains a valid name for an atom.</p>
        <p><c><![CDATA[ERL_ATOM_PTR(atom)]]></c> and <c><![CDATA[ERL_ATOM_PTR_UTF8(atom)]]></c>
	can be used to retrieve the atom name (as a null terminated string). <c><![CDATA[ERL_ATOM_SIZE(atom)]]></c>
	and <c><![CDATA[ERL_ATOM_SIZE_UTF8(atom)]]></c> returns the length of the atom name.</p>
	<note><p>Note that the UTF8 variants were introduced in Erlang/OTP releases R16
	and the string returned by <c>ERL_ATOM_PTR(atom)</c> was not null terminated on older releases.</p>
	</note>
      </desc>
    </func>
    <func>
      <name><ret>ETERM *</ret><nametext>erl_mk_binary(bptr, size)</nametext></name>
      <fsummary>Creates a binary object</fsummary>
      <type>
        <v>char *bptr;</v>
        <v>int size;</v>
      </type>
      <desc>
        <p>This function produces an Erlang binary object from a
          buffer containing a sequence of bytes.</p>
        <p><c><![CDATA[bptr]]></c> is a pointer to a buffer containing data to be converted.</p>
        <p><c><![CDATA[size]]></c> indicates the length of <c><![CDATA[bptr]]></c>.</p>
        <p>The function returns an Erlang binary object.</p>
        <p><c><![CDATA[ERL_BIN_PTR(bin)]]></c> retrieves a pointer to
          the binary data. <c><![CDATA[ERL_BIN_SIZE(bin)]]></c> retrieves the 
          size. </p>
      </desc>
    </func>
    <func>
      <name><ret>ETERM *</ret><nametext>erl_mk_empty_list()</nametext></name>
      <fsummary>Creates an empty Erlang list</fsummary>
      <desc>
        <p>This function creates and returns an empty Erlang list.
          Note that NULL is not used to represent an empty list;
          Use this function instead.</p>
      </desc>
    </func>
    <func>
      <name><ret>ETERM *</ret><nametext>erl_mk_estring(string, len)</nametext></name>
      <fsummary>Creates an Erlang string</fsummary>
      <type>
        <v>char *string;</v>
        <v>int len;</v>
      </type>
      <desc>
        <p>This function creates a list from a sequence of bytes.</p>
        <p><c><![CDATA[string]]></c> is a buffer containing a sequence of
          bytes. The buffer does not need to be zero-terminated.</p>
        <p><c><![CDATA[len]]></c> is the length of <c><![CDATA[string]]></c>.</p>
        <p>The function returns an Erlang list object corresponding to
          the character sequence in <c><![CDATA[string]]></c>.</p>
      </desc>
    </func>
    <func>
      <name><ret>ETERM *</ret><nametext>erl_mk_float(f)</nametext></name>
      <fsummary>Creates an Erlang float</fsummary>
      <type>
        <v>double f;</v>
      </type>
      <desc>
        <p>Creates an Erlang float.</p>
        <p><c><![CDATA[f]]></c> is a value to be converted to an Erlang float.</p>
        <p></p>
        <p>The function returns an Erlang float object with the value
            specified in <c><![CDATA[f]]></c> or <c><![CDATA[NULL]]></c> if
            <c><![CDATA[f]]></c> is not finite.
        </p>
        <p><c><![CDATA[ERL_FLOAT_VALUE(t)]]></c> can be used to retrieve the
            value from an Erlang float.</p>
      </desc>
    </func>
    <func>
      <name><ret>ETERM *</ret><nametext>erl_mk_int(n)</nametext></name>
      <fsummary>Creates an Erlang integer</fsummary>
      <type>
        <v>int n;</v>
      </type>
      <desc>
        <p>Creates an Erlang integer.</p>
        <p><c><![CDATA[n]]></c> is a value to be converted to an Erlang integer.</p>
        <p></p>
        <p>The function returns an Erlang integer object with the
          value specified in <c><![CDATA[n]]></c>.</p>
        <p><c><![CDATA[ERL_INT_VALUE(t)]]></c> can be used to retrieve the value 
          value from an Erlang integer.</p>
      </desc>
    </func>
    <func>
      <name><ret>ETERM *</ret><nametext>erl_mk_list(array, arrsize)</nametext></name>
      <fsummary>Creates a list from an array</fsummary>
      <type>
        <v>ETERM **array;</v>
        <v>int arrsize;</v>
      </type>
      <desc>
        <p>Creates an Erlang list from an array of Erlang terms, such
          that each element in the list corresponds to one element in
          the array. </p>
        <p><c><![CDATA[array]]></c> is an array of Erlang terms.</p>
        <p><c><![CDATA[arrsize]]></c> is the number of elements in <c><![CDATA[array]]></c>.</p>
        <p>The function creates an Erlang list object, whose length
          <c><![CDATA[arrsize]]></c> and whose elements are taken from the terms in
          <c><![CDATA[array]]></c>.</p>
      </desc>
    </func>
    <func>
      <name><ret>ETERM *</ret><nametext>erl_mk_pid(node, number, serial, creation)</nametext></name>
      <fsummary>Creates a process identifier</fsummary>
      <type>
        <v>const char *node;</v>
        <v>unsigned int number;</v>
        <v>unsigned int serial;</v>
        <v>unsigned int creation;</v>
      </type>
      <desc>
        <p>This function creates an Erlang process identifier. The
          resulting pid can be used by Erlang processes wishing to
          communicate with the C node.</p>
        <p><c><![CDATA[node]]></c> is the name of the C node.</p>
        <p><c><![CDATA[number]]></c>, <c><![CDATA[serial]]></c> and <c><![CDATA[creation]]></c> are
          arbitrary numbers. Note though, that these are limited in
          precision, so only the low 15, 3 and 2 bits of these numbers
          are actually used.</p>
        <p>The function returns an Erlang pid object.</p>
        <p><c><![CDATA[ERL_PID_NODE(pid)]]></c>, <c><![CDATA[ERL_PID_NUMBER(pid)]]></c>,
          <c><![CDATA[ERL_PID_SERIAL(pid)]]></c> and <c><![CDATA[ERL_PID_CREATION(pid)]]></c>
          can be used to retrieve the four values used to create the pid.</p>
      </desc>
    </func>
    <func>
      <name><ret>ETERM *</ret><nametext>erl_mk_port(node, number, creation)</nametext></name>
      <fsummary>Creates a port identifier</fsummary>
      <type>
        <v>const char *node;</v>
        <v>unsigned int number;</v>
        <v>unsigned int creation;</v>
      </type>
      <desc>
        <p>This function creates an Erlang port identifier. </p>
        <p><c><![CDATA[node]]></c> is the name of the C node.</p>
        <p><c><![CDATA[number]]></c> and <c><![CDATA[creation]]></c> are arbitrary numbers.
          Note though, that these are limited in
          precision, so only the low 18 and 2 bits of these numbers
          are actually used.</p>
        <p>The function returns an Erlang port object.</p>
        <p><c><![CDATA[ERL_PORT_NODE(port)]]></c>, <c><![CDATA[ERL_PORT_NUMBER(port)]]></c>
          and <c><![CDATA[ERL_PORT_CREATION]]></c> can be used to retrieve the three
          values used to create the port. </p>
      </desc>
    </func>
    <func>
      <name><ret>ETERM *</ret><nametext>erl_mk_ref(node, number, creation)</nametext></name>
      <fsummary>Creates an old Erlang reference</fsummary>
      <type>
        <v>const char *node;</v>
        <v>unsigned int number;</v>
        <v>unsigned int creation;</v>
      </type>
      <desc>
        <p>This function creates an old Erlang reference, with
          only 18 bits - use <c><![CDATA[erl_mk_long_ref]]></c> instead.</p>
        <p><c><![CDATA[node]]></c> is the name of the C node.</p>
        <p><c><![CDATA[number]]></c> should be chosen uniquely for each reference
          created for a given C node.</p>
        <p><c><![CDATA[creation]]></c> is an arbitrary number.</p>
        <p>Note  that <c><![CDATA[number]]></c> and <c><![CDATA[creation]]></c> are limited in
          precision, so only the low 18 and 2 bits of these numbers
          are actually used.
          </p>
        <p>The function returns an Erlang reference object.</p>
        <p><c><![CDATA[ERL_REF_NODE(ref)]]></c>, <c><![CDATA[ERL_REF_NUMBER(ref)]]></c>, and
          <c><![CDATA[ERL_REF_CREATION(ref)]]></c> to retrieve the three values used
          to create the reference. </p>
      </desc>
    </func>
    <func>
      <name><ret>ETERM *</ret><nametext>erl_mk_long_ref(node, n1, n2, n3, creation)</nametext></name>
      <fsummary>Creates an Erlang reference</fsummary>
      <type>
        <v>const char *node;</v>
        <v>unsigned int n1, n2, n3;</v>
        <v>unsigned int creation;</v>
      </type>
      <desc>
        <p>This function creates an Erlang reference, with 82 bits.</p>
        <p><c><![CDATA[node]]></c> is the name of the C node.</p>
        <p><c><![CDATA[n1]]></c>, <c><![CDATA[n2]]></c> and <c><![CDATA[n3]]></c> can be seen as one big number
          <c><![CDATA[n1*2^64+n2*2^32+n3]]></c> which should be chosen uniquely for
          each reference
          created for a given C node.</p>
        <p><c><![CDATA[creation]]></c> is an arbitrary number.</p>
        <p>Note that <c><![CDATA[n3]]></c> and <c><![CDATA[creation]]></c> are limited in
          precision, so only the low 18 and 2 bits of these numbers
          are actually used.
          </p>
        <p>The function returns an Erlang reference object.</p>
        <p><c><![CDATA[ERL_REF_NODE(ref)]]></c>, <c><![CDATA[ERL_REF_NUMBERS(ref)]]></c>,
          <c><![CDATA[ERL_REF_LEN(ref)]]></c> and
          <c><![CDATA[ERL_REF_CREATION(ref)]]></c> to retrieve the values used
          to create the reference. </p>
      </desc>
    </func>
    <func>
      <name><ret>ETERM *</ret><nametext>erl_mk_string(string)</nametext></name>
      <fsummary>Creates a string</fsummary>
      <type>
        <v>char *string;</v>
      </type>
      <desc>
        <p>This function creates a list from a zero terminated string.</p>
        <p><c><![CDATA[string]]></c> is the zero-terminated sequence of characters
          (i.e. a C string) from which the list will be created.</p>
        <p>The function returns an Erlang list.</p>
      </desc>
    </func>
    <func>
      <name><ret>ETERM *</ret><nametext>erl_mk_tuple(array, arrsize)</nametext></name>
      <fsummary>Creates an Erlang tuple from an array</fsummary>
      <type>
        <v>ETERM **array;</v>
        <v>int arrsize;</v>
      </type>
      <desc>
        <p>Creates an Erlang tuple from an array of Erlang terms.</p>
        <p><c><![CDATA[array]]></c> is an array of Erlang terms.</p>
        <p><c><![CDATA[arrsize]]></c> is the number of elements in <c><![CDATA[array]]></c>.</p>
        <p>The function creates an Erlang tuple, whose arity is
          <c><![CDATA[size]]></c> and whose elements are taken from the terms in
          <c><![CDATA[array]]></c>.</p>
        <p>To retrieve the size of a tuple, either use the
          <c><![CDATA[erl_size]]></c> function (which checks the type of the checked
          term and works for a binary as well as for a tuple), or the
          <c><![CDATA[ERL_TUPLE_SIZE(tuple)]]></c> returns the arity of a tuple.
          <c><![CDATA[erl_size()]]></c> will do the same thing, but it checks that
          the argument really is a tuple.
          <c><![CDATA[erl_element(index,tuple)]]></c> returns the element
          corresponding to a given position in the tuple. </p>
      </desc>
    </func>
    <func>
      <name><ret>ETERM *</ret><nametext>erl_mk_uint(n)</nametext></name>
      <fsummary>Creates an unsigned integer</fsummary>
      <type>
        <v>unsigned int n;</v>
      </type>
      <desc>
        <p>Creates an Erlang unsigned integer.</p>
        <p><c><![CDATA[n]]></c> is a value to be converted to an Erlang
          unsigned integer.</p>
        <p></p>
        <p>The function returns an Erlang unsigned integer object with
          the value specified in <c><![CDATA[n]]></c>.</p>
        <p><c><![CDATA[ERL_INT_UVALUE(t)]]></c> can be used to retrieve the
          value from an Erlang unsigned integer.</p>
      </desc>
    </func>
    <func>
      <name><ret>ETERM *</ret><nametext>erl_mk_var(name)</nametext></name>
      <fsummary>Creates an Erlang variable</fsummary>
      <type>
        <v>char *name;</v>
      </type>
      <desc>
        <p>This function creates an unbound Erlang variable. The
          variable can later be bound through pattern matching or assignment.</p>
        <p><c><![CDATA[name]]></c> specifies a name for the variable.</p>
        <p>The function returns an Erlang variable object with the
          name <c><![CDATA[name]]></c>. </p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>erl_print_term(stream, term)</nametext></name>
      <fsummary>Prints an Erlang term</fsummary>
      <type>
        <v>FILE *stream;</v>
        <v>ETERM *term;</v>
      </type>
      <desc>
        <p>This function prints the specified Erlang term to the given
          output stream.</p>
        <p><c><![CDATA[stream]]></c> indicates where the function should send its
          output.</p>
        <p><c><![CDATA[term]]></c> is the Erlang term to print.</p>
        <p>The function returns the number of characters written, or a
          negative value if there was an error.</p>
      </desc>
    </func>
    <func>
      <name><ret>void</ret><nametext>erl_set_compat_rel(release_number)</nametext></name>
      <fsummary>Set the erl_interface library in compatibility mode</fsummary>
      <type>
        <v>unsigned release_number;</v>
      </type>
      <desc>
        <marker id="erl_set_compat_rel"></marker>
        <p>By default, the <c><![CDATA[erl_interface]]></c> library is only guaranteed
          to be compatible with other Erlang/OTP components from the same
          release as the <c><![CDATA[erl_interface]]></c> library itself. For example,
          <c><![CDATA[erl_interface]]></c> from the OTP R10 release is not compatible
          with an Erlang emulator from the OTP R9 release by default.</p>
        <p>A call to <c><![CDATA[erl_set_compat_rel(release_number)]]></c> sets the
          <c><![CDATA[erl_interface]]></c> library in compatibility mode of release
          <c><![CDATA[release_number]]></c>. Valid range of <c><![CDATA[release_number]]></c>
          is [7, current release]. This makes it possible to
          communicate with Erlang/OTP components from earlier releases.</p>
        <note>
          <p>If this function is called, it may only be called once
            directly after the call to the
            <seealso marker="#erl_init">erl_init()</seealso> function.</p>
        </note>
        <warning>
          <p>You may run into trouble if this feature is used
            carelessly. Always make sure that all communicating
            components are either from the same Erlang/OTP release, or
            from release X and release Y where all components
            from release Y are in compatibility mode of release X.</p>
        </warning>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>erl_size(term)</nametext></name>
      <fsummary>Return the arity of a tuple or binary</fsummary>
      <type>
        <v>ETERM *term;</v>
      </type>
      <desc>
        <p>Returns the arity of an Erlang tuple, or the
          number of bytes in an Erlang binary object. </p>
        <p><c><![CDATA[term]]></c> is an Erlang tuple or an Erlang binary object.</p>
        <p>The function returns the size of <c><![CDATA[term]]></c> as described
          above, or -1 if <c><![CDATA[term]]></c> is not one of the two supported
          types. </p>
      </desc>
    </func>
    <func>
      <name><ret>ETERM *</ret><nametext>erl_tl(list)</nametext></name>
      <fsummary>Extracts the tail from a list</fsummary>
      <type>
        <v>ETERM *list;</v>
      </type>
      <desc>
        <p>Extracts the tail from a list.</p>
        <p><c><![CDATA[list]]></c> is an Erlang term containing a list.</p>
        <p>The function returns an Erlang list corresponding to the
          original list minus the first element, or NULL pointer if
          <c><![CDATA[list]]></c> was not a list.</p>
      </desc>
    </func>
    <func>
      <name><ret>ETERM *</ret><nametext>erl_var_content(term, name)</nametext></name>
      <fsummary>Extracts the content of a variable</fsummary>
      <type>
        <v>ETERM *term;</v>
        <v>char *name;</v>
      </type>
      <desc>
        <p>This function returns the contents of the specified
          variable in an Erlang term. 
          </p>
        <p><c><![CDATA[term]]></c> is an Erlang term. In order for this function
          to succeed, <c><![CDATA[term]]></c> must be an Erlang variable with the
          specified name, or it must be an Erlang list or tuple
          containing a variable with the specified name. Other Erlang
          types cannot contain variables.</p>
        <p><c><![CDATA[name]]></c> is the name of an Erlang variable.</p>
        <p>Returns the Erlang object corresponding to the value of
          <c><![CDATA[name]]></c> in <c><![CDATA[term]]></c>. If no variable with the name
          <c><![CDATA[name]]></c> was found in <c><![CDATA[term]]></c>, or if <c><![CDATA[term]]></c> is
          not a valid Erlang term, NULL is returned.</p>
      </desc>
    </func>
  </funcs>
</cref>