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

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

<cref>
  <header>
    <copyright>
      <year>2001</year><year>2014</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>ei</title>
    <prepared>Jakob Cederlund</prepared>
    <responsible>Kent Boortz</responsible>
    <docno>1</docno>
    <approved>Kenneth Lundin</approved>
    <checked></checked>
    <date>2000-11-27</date>
    <rev>PA1</rev>
    <file>ei.sgml</file>
  </header>
  <lib>ei</lib>
  <libsummary>routines for handling the erlang binary term format</libsummary>
  <description>
    <p>The library <c><![CDATA[ei]]></c> contains macros and functions to encode
      and decode the erlang binary term format.</p>
    <p>With <c><![CDATA[ei]]></c>, you can convert atoms, lists, numbers and
      binaries to and from the binary format. This is useful when
      writing port programs and drivers. <c><![CDATA[ei]]></c> uses a given
      buffer, and no dynamic memory (with the exception of
      <c><![CDATA[ei_decode_fun()]]></c>), and is often quite fast.</p>
    <p>It also handles C-nodes, C-programs that talks erlang
      distribution with erlang nodes (or other C-nodes) using the
      erlang distribution format. The difference between <c><![CDATA[ei]]></c> and
      <c><![CDATA[erl_interface]]></c> is that <c><![CDATA[ei]]></c> uses the binary format
      directly when sending and receiving terms. It is also thread
      safe, and using threads, one process can handle multiple
      C-nodes. The <c><![CDATA[erl_interface]]></c> library is built on top of
      <c><![CDATA[ei]]></c>, but of legacy reasons, it doesn't allow for multiple
      C-nodes. In general, <c><![CDATA[ei]]></c> is the preferred way of doing
      C-nodes.</p>
    <p>The decode and encode functions use a buffer an index into the
      buffer, which points at the point where to encode and
      decode. The index is updated to point right after the term
      encoded/decoded. No checking is done whether the term fits in
      the buffer or not. If encoding goes outside the buffer, the
      program may crash.</p>
    <p>All functions takes two parameter, <c><![CDATA[buf]]></c> is a pointer to
      the buffer where the binary data is / will be, <c><![CDATA[index]]></c> is a
      pointer to an index into the buffer. This parameter will be
      incremented with the size of the term decoded / encoded. The
      data is thus at <c><![CDATA[buf[*index]]]></c> when an <c><![CDATA[ei]]></c> function is
      called.</p>
    <p>The encode functions all assumes that the <c><![CDATA[buf]]></c> and
      <c><![CDATA[index]]></c> parameters points to a buffer big enough for the
      data. To get the size of an encoded term, without encoding it,
      pass <c><![CDATA[NULL]]></c> instead of a buffer pointer. The <c><![CDATA[index]]></c>
      parameter will be incremented, but nothing will be encoded. This
      is the way in <c><![CDATA[ei]]></c> to "preflight" term encoding.</p>
    <p>There are also encode-functions that uses a dynamic buffer. It
      is often more convenient to use these to encode data. All encode
      functions comes in two versions: those starting with <c><![CDATA[ei_x]]></c>,
      uses a dynamic buffer.</p>
    <p>All functions return <c><![CDATA[0]]></c> if successful, and <c><![CDATA[-1]]></c> if
      not. (For instance, if a term is not of the expected type, or
      the data to decode is not a valid erlang term.)</p>
    <p>Some of the decode-functions needs a preallocated buffer. This
      buffer must be allocated big enough, and for non compound types
      the <c><![CDATA[ei_get_type()]]></c>
      function returns the size required (note that for strings an
      extra byte is needed for the 0 string terminator).</p>
  </description>
  <section>
    <title>DATA TYPES</title>

    <taglist>
      <tag><marker id="erlang_char_encoding"/>erlang_char_encoding</tag>
       <item>
	 <p/>
	 <code type="none">
typedef enum {
    ERLANG_ASCII = 1,
    ERLANG_LATIN1 = 2,
    ERLANG_UTF8 = 4
}erlang_char_encoding;
</code>
         <p>The character encodings used for atoms. <c>ERLANG_ASCII</c> represents 7-bit ASCII.
	 Latin1 and UTF8 are different extensions of 7-bit ASCII. All 7-bit ASCII characters
	 are valid Latin1 and UTF8 characters. ASCII and Latin1 both represent each character
	 by one byte. A UTF8 character can consist of one to four bytes. Note that these
	 constants are bit-flags and can be combined with bitwise-or.</p>
      </item>
    </taglist>
  </section>
  <funcs>
    <func>
      <name><ret>void</ret><nametext>ei_set_compat_rel(release_number)</nametext></name>
      <fsummary>Set the ei library in compatibility mode</fsummary>
      <type>
        <v>unsigned release_number;</v>
      </type>
      <desc>
        <marker id="ei_set_compat_rel"></marker>
        <p>By default, the <c><![CDATA[ei]]></c> library is only guaranteed
          to be compatible with other Erlang/OTP components from the same
          release as the <c><![CDATA[ei]]></c> library itself. For example, <c><![CDATA[ei]]></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[ei_set_compat_rel(release_number)]]></c> sets the
          <c><![CDATA[ei]]></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
            and must be called before any other functions in the <c><![CDATA[ei]]></c>
            library is called.</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>ei_encode_version(char *buf, int *index)</nametext></name>
      <name><ret>int</ret><nametext>ei_x_encode_version(ei_x_buff* x)</nametext></name>
      <fsummary>Encode version</fsummary>
      <desc>
        <p>Encodes a version magic number for the binary format. Must
          be the first token in a binary term.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_encode_long(char *buf, int *index, long p)</nametext></name>
      <name><ret>int</ret><nametext>ei_x_encode_long(ei_x_buff* x, long p)</nametext></name>
      <fsummary>Encode integer</fsummary>
      <desc>
        <p>Encodes a long integer in the binary format.
          Note that if the code is 64 bits the function ei_encode_long() is
          exactly the same as ei_encode_longlong().</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_encode_ulong(char *buf, int *index, unsigned long p)</nametext></name>
      <name><ret>int</ret><nametext>ei_x_encode_ulong(ei_x_buff* x, unsigned long p)</nametext></name>
      <fsummary>Encode unsigned integer</fsummary>
      <desc>
        <p>Encodes an unsigned long integer in the binary format.
          Note that if the code is 64 bits the function ei_encode_ulong() is
          exactly the same as ei_encode_ulonglong().</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_encode_longlong(char *buf, int *index, long long p)</nametext></name>
      <name><ret>int</ret><nametext>ei_x_encode_longlong(ei_x_buff* x, long long p)</nametext></name>
      <fsummary>Encode integer</fsummary>
      <desc>
        <p>Encodes a GCC <c><![CDATA[long long]]></c> or Visual C++ <c><![CDATA[__int64]]></c> (64 bit)
          integer in the binary format. Note that this function is missing
          in the VxWorks port.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_encode_ulonglong(char *buf, int *index, unsigned long long p)</nametext></name>
      <name><ret>int</ret><nametext>ei_x_encode_ulonglong(ei_x_buff* x, unsigned long long p)</nametext></name>
      <fsummary>Encode unsigned integer</fsummary>
      <desc>
        <p>Encodes a GCC <c><![CDATA[unsigned long long]]></c> or Visual C++ <c><![CDATA[unsigned __int64]]></c> (64 bit) integer in the binary format.  Note that
          this function is missing in the VxWorks port.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_encode_bignum(char *buf, int *index, mpz_t obj)</nametext></name>
      <name><ret>int</ret><nametext>ei_x_encode_bignum(ei_x_buff *x, mpz_t obj)</nametext></name>
      <fsummary>Encode an arbitrary precision integer</fsummary>
      <desc>
        <p>Encodes a GMP <c><![CDATA[mpz_t]]></c> integer to binary format.
          To use this function the ei library needs to be configured and compiled
          to use the GMP library. </p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_encode_double(char *buf, int *index, double p)</nametext></name>
      <name><ret>int</ret><nametext>ei_x_encode_double(ei_x_buff* x, double p)</nametext></name>
      <fsummary>Encode a double float</fsummary>
      <desc>
        <p>Encodes a double-precision (64 bit) floating point number in
          the binary format.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_encode_boolean(char *buf, int *index, int p)</nametext></name>
      <name><ret>int</ret><nametext>ei_x_encode_boolean(ei_x_buff* x, int p)</nametext></name>
      <fsummary>Encode a boolean</fsummary>
      <desc>
        <p>Encodes a boolean value, as the atom <c><![CDATA[true]]></c> if p is not
          zero or <c><![CDATA[false]]></c> if p is zero.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_encode_char(char *buf, int *index, char p)</nametext></name>
      <name><ret>int</ret><nametext>ei_x_encode_char(ei_x_buff* x, char p)</nametext></name>
      <fsummary>Encode an 8-bit integer between 0-255</fsummary>
      <desc>
        <p>Encodes a char (8-bit) as an integer between 0-255 in the binary format.
          Note that for historical reasons the integer argument is of
          type <c><![CDATA[char]]></c>. Your C code should consider the
          given argument to be of type <c><![CDATA[unsigned char]]></c> even if
          the C compilers and system may define <c><![CDATA[char]]></c> to be
          signed.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_encode_string(char *buf, int *index, const char *p)</nametext></name>
      <name><ret>int</ret><nametext>ei_encode_string_len(char *buf, int *index, const char *p, int len)</nametext></name>
      <name><ret>int</ret><nametext>ei_x_encode_string(ei_x_buff* x, const char *p)</nametext></name>
      <name><ret>int</ret><nametext>ei_x_encode_string_len(ei_x_buff* x, const char* s, int len)</nametext></name>
      <fsummary>Encode a string</fsummary>
      <desc>
        <p>Encodes a string in the binary format. (A string in erlang
          is a list, but is encoded as a character array in the binary
          format.) The string should be zero-terminated, except for
          the <c><![CDATA[ei_x_encode_string_len()]]></c> function.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_encode_atom(char *buf, int *index, const char *p)</nametext></name>
      <name><ret>int</ret><nametext>ei_encode_atom_len(char *buf, int *index, const char *p, int len)</nametext></name>
      <name><ret>int</ret><nametext>ei_x_encode_atom(ei_x_buff* x, const char *p)</nametext></name>
      <name><ret>int</ret><nametext>ei_x_encode_atom_len(ei_x_buff* x, const char *p, int len)</nametext></name>
      <fsummary>Encode an atom</fsummary>
      <desc>
        <p>Encodes an atom in the binary format. The <c><![CDATA[p]]></c> parameter
          is the name of the atom in latin1 encoding. Only upto <c>MAXATOMLEN-1</c> bytes
          are encoded. The name should be zero-terminated, except for
          the <c><![CDATA[ei_x_encode_atom_len()]]></c> function.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_encode_atom_as(char *buf, int *index, const char *p, erlang_char_encoding from_enc, erlang_char_encoding to_enc)</nametext></name>
      <name><ret>int</ret><nametext>ei_encode_atom_len_as(char *buf, int *index, const char *p, int len, erlang_char_encoding from_enc, erlang_char_encoding to_enc)</nametext></name>
      <name><ret>int</ret><nametext>ei_x_encode_atom_as(ei_x_buff* x, const char *p, erlang_char_encoding from_enc, erlang_char_encoding to_enc)</nametext></name>
      <name><ret>int</ret><nametext>ei_x_encode_atom_len_as(ei_x_buff* x, const char *p, int len, erlang_char_encoding from_enc, erlang_char_encoding to_enc)</nametext></name>
      <fsummary>Encode an atom</fsummary>
      <desc>
        <p>Encodes an atom in the binary format with character encoding
	<c><seealso marker="#erlang_char_encoding">to_enc</seealso></c> (latin1 or utf8).
	The <c>p</c> parameter is the name of the atom with character encoding
	<c><seealso marker="#erlang_char_encoding">from_enc</seealso></c>  (ascii, latin1 or utf8).
	The name must either be zero-terminated or a function variant with a <c>len</c>
	parameter must be used. If <c>to_enc</c> is set to the bitwise-or'd combination
	<c>(ERLANG_LATIN1|ERLANG_UTF8)</c>, utf8 encoding is only used if the atom string
	can not be represented in latin1 encoding.</p>
	<p>The encoding will fail if <c>p</c> is not a valid string in encoding <c>from_enc</c>,
	if the string is too long or if it can not be represented with character encoding <c>to_enc</c>.</p>
	<p>These functions were introduced in R16 release of Erlang/OTP as part of a first step
	to support UTF8 atoms. Atoms encoded with <c>ERLANG_UTF8</c>
	can not be decoded by earlier releases than R16.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_encode_binary(char *buf, int *index, const void *p, long len)</nametext></name>
      <name><ret>int</ret><nametext>ei_x_encode_binary(ei_x_buff* x, const void *p, long len)</nametext></name>
      <fsummary>Encode a binary</fsummary>
      <desc>
        <p>Encodes a binary in the binary format. The data is at
          <c><![CDATA[p]]></c>, of <c><![CDATA[len]]></c> bytes length.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_encode_pid(char *buf, int *index, const erlang_pid *p)</nametext></name>
      <name><ret>int</ret><nametext>ei_x_encode_pid(ei_x_buff* x, const erlang_pid *p)</nametext></name>
      <fsummary>Encode a pid</fsummary>
      <desc>
        <p>Encodes an erlang process identifier, pid, in the binary
          format. The <c><![CDATA[p]]></c> parameter points to an
          <c><![CDATA[erlang_pid]]></c> structure (which should have been obtained
          earlier with <c><![CDATA[ei_decode_pid()]]></c>).</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_encode_fun(char *buf, int *index, const erlang_fun *p)</nametext></name>
      <name><ret>int</ret><nametext>ei_x_encode_fun(ei_x_buff* x, const erlang_fun* fun)</nametext></name>
      <fsummary>Encode a fun</fsummary>
      <desc>
        <p>Encodes a fun in the binary format. The <c><![CDATA[p]]></c> parameter
          points to an <c><![CDATA[erlang_fun]]></c> structure. The
          <c><![CDATA[erlang_fun]]></c> is not freed automatically, the
          <c><![CDATA[free_fun]]></c> should be called if the fun is not needed
          after encoding.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_encode_port(char *buf, int *index, const erlang_port *p)</nametext></name>
      <name><ret>int</ret><nametext>ei_x_encode_port(ei_x_buff* x, const erlang_port *p)</nametext></name>
      <fsummary>Encodes a port</fsummary>
      <desc>
        <p>Encodes an erlang port in the binary format. The <c><![CDATA[p]]></c>
          parameter points to a <c><![CDATA[erlang_port]]></c> structure (which
          should have been obtained earlier with
          <c><![CDATA[ei_decode_port()]]></c>.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_encode_ref(char *buf, int *index, const erlang_ref *p)</nametext></name>
      <name><ret>int</ret><nametext>ei_x_encode_ref(ei_x_buff* x, const erlang_ref *p)</nametext></name>
      <fsummary>Encodes a ref</fsummary>
      <desc>
        <p>Encodes an erlang reference in the binary format. The
          <c><![CDATA[p]]></c> parameter points to a <c><![CDATA[erlang_ref]]></c> structure
          (which should have been obtained earlier with
          <c><![CDATA[ei_decode_ref()]]></c>.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_encode_term(char *buf, int *index, void *t)</nametext></name>
      <name><ret>int</ret><nametext>ei_x_encode_term(ei_x_buff* x, void *t)</nametext></name>
      <fsummary>Encode an <c><![CDATA[erl_interface]]></c>term</fsummary>
      <desc>
        <p>This function encodes an <c><![CDATA[ETERM]]></c>, as obtained from
          <c><![CDATA[erl_interface]]></c>. The <c><![CDATA[t]]></c> parameter is actually an
          <c><![CDATA[ETERM]]></c> pointer. This function doesn't free the
          <c><![CDATA[ETERM]]></c>.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_encode_trace(char *buf, int *index, const erlang_trace *p)</nametext></name>
      <name><ret>int</ret><nametext>ei_x_encode_trace(ei_x_buff* x, const erlang_trace *p)</nametext></name>
      <fsummary>Encode a trace token</fsummary>
      <desc>
        <p>This function encodes an erlang trace token in the binary
          format. The <c><![CDATA[p]]></c> parameter points to a
          <c><![CDATA[erlang_trace]]></c> structure (which should have been
          obtained earlier with <c><![CDATA[ei_decode_trace()]]></c>.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_encode_tuple_header(char *buf, int *index, int arity)</nametext></name>
      <name><ret>int</ret><nametext>ei_x_encode_tuple_header(ei_x_buff* x, int arity)</nametext></name>
      <fsummary>Encode a tuple</fsummary>
      <desc>
        <p>This function encodes a tuple header, with a specified
          arity. The next <c><![CDATA[arity]]></c> terms encoded will be the
          elements of the tuple. Tuples and lists are encoded
          recursively, so that a tuple may contain another tuple or
          list.</p>
        <p>E.g. to encode the tuple <c><![CDATA[{a, {b, {}}}]]></c>:</p>
        <pre>
ei_encode_tuple_header(buf, &amp;i, 2);
ei_encode_atom(buf, &amp;i, "a");
ei_encode_tuple_header(buf, &amp;i, 2);
ei_encode_atom(buf, &amp;i, "b");
ei_encode_tuple_header(buf, &amp;i, 0);
        </pre>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_encode_list_header(char *buf, int *index, int arity)</nametext></name>
      <name><ret>int</ret><nametext>ei_x_encode_list_header(ei_x_buff* x, int arity)</nametext></name>
      <fsummary>Encode a list</fsummary>
      <desc>
        <p>This function encodes a list header, with a specified
          arity. The next <c><![CDATA[arity+1]]></c> terms are the elements
          (actually its <c><![CDATA[arity]]></c> cons cells) and the tail of the
          list. Lists and tuples are encoded recursively, so that a
          list may contain another list or tuple.</p>
        <p>E.g. to encode the list <c><![CDATA[[c, d, [e | f]]]]></c>:</p>
        <pre>
ei_encode_list_header(buf, &amp;i, 3);
ei_encode_atom(buf, &amp;i, "c");
ei_encode_atom(buf, &amp;i, "d");
ei_encode_list_header(buf, &amp;i, 1);
ei_encode_atom(buf, &amp;i, "e");
ei_encode_atom(buf, &amp;i, "f");
ei_encode_empty_list(buf, &amp;i);
        </pre>
        <note>
          <p>It may seem that there is no way to create a list without
            knowing the number of elements in advance. But indeed
            there is a way. Note that the list <c><![CDATA[[a, b, c]]]></c> can be
            written as <c><![CDATA[[a | [b | [c]]]]]></c>.  Using this, a list can
            be written as conses.</p>
        </note>
        <p>To encode a list, without knowing the arity in advance:</p>
        <pre>
while (something()) {
    ei_x_encode_list_header(&amp;x, 1);
    ei_x_encode_ulong(&amp;x, i); /* just an example */
}
ei_x_encode_empty_list(&amp;x);
        </pre>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_encode_empty_list(char* buf, int* index)</nametext></name>
      <name><ret>int</ret><nametext>ei_x_encode_empty_list(ei_x_buff* x)</nametext></name>
      <fsummary>Encode an empty list (<c><![CDATA[nil]]></c>)</fsummary>
      <desc>
        <p>This function encodes an empty list. It's often used at the
          tail of a list.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_encode_map_header(char *buf, int *index, int arity)</nametext></name>
      <name><ret>int</ret><nametext>ei_x_encode_map_header(ei_x_buff* x, int arity)</nametext></name>
      <fsummary>Encode a map</fsummary>
      <desc>
        <p>This function encodes a map header, with a specified arity. The next
	   <c>arity</c> terms encoded will be the keys of the map, and the next
	   <c>arity</c> terms after that will be the corresponding values in
	   same order.</p>
        <p>E.g. to encode the map <c>#{a => "Apple", b => "Banana"}</c>:</p>
        <pre>
ei_x_encode_map_header(&amp;x, 2);
ei_x_encode_atom(&amp;x, "a");
ei_x_encode_atom(&amp;x, "b");
ei_x_encode_string(&amp;x, "Apple");
ei_x_encode_string(&amp;x, "Banana");
        </pre>
	<p>A correctly encoded map can not have duplicate keys, but no check
	for duplicate keys is done by this function.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_get_type(const char *buf, const int *index, int *type, int *size)</nametext></name>
      <fsummary>Fetch the type and size of an encoded term</fsummary>
      <desc>
        <p>This function returns the type in <c><![CDATA[type]]></c> and size in
          <c><![CDATA[size]]></c> of the encoded term.
          For strings and atoms, size
          is the number of characters <em>not</em> including the
          terminating 0. For binaries, <c><![CDATA[size]]></c> is the number of
          bytes. For lists and tuples, <c><![CDATA[size]]></c> is the arity of the
          object. For other types, <c><![CDATA[size]]></c> is 0. In all cases,
          <c><![CDATA[index]]></c> is left unchanged.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_decode_version(const char *buf, int *index, int *version)</nametext></name>
      <fsummary>Encode an empty list (<c><![CDATA[nil]]></c>)</fsummary>
      <desc>
        <p>This function decodes the version magic number for the
          erlang binary term format. It must be the first token in a
          binary term.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_decode_long(const char *buf, int *index, long *p)</nametext></name>
      <fsummary>Decode integer</fsummary>
      <desc>
        <p>This function decodes a long integer from the binary format.
          Note that if the code is 64 bits the function ei_decode_long() is
          exactly the same as ei_decode_longlong().</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_decode_ulong(const char *buf, int *index, unsigned long *p)</nametext></name>
      <fsummary>Decode unsigned integer</fsummary>
      <desc>
        <p>This function decodes an unsigned long integer from
          the binary format.
          Note that if the code is 64 bits the function ei_decode_ulong() is
          exactly the same as ei_decode_ulonglong().</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_decode_longlong(const char *buf, int *index, long long *p)</nametext></name>
      <fsummary>Decode integer</fsummary>
      <desc>
        <p>This function decodes a GCC <c><![CDATA[long long]]></c> or Visual C++ <c><![CDATA[__int64]]></c>
          (64 bit) integer from the binary format.  Note that this
          function is missing in the VxWorks port.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_decode_ulonglong(const char *buf, int *index, unsigned long long *p)</nametext></name>
      <fsummary>Decode unsigned integer</fsummary>
      <desc>
        <p>This function decodes a GCC <c><![CDATA[unsigned long long]]></c> or Visual C++
          <c><![CDATA[unsigned __int64]]></c> (64 bit) integer from the binary format.
          Note that this function is missing in the VxWorks port.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_decode_bignum(const char *buf, int *index, mpz_t obj)</nametext></name>
      <fsummary>Decode a GMP arbitrary precision integer</fsummary>
      <desc>
        <p>This function decodes an integer in the binary format to a GMP <c><![CDATA[mpz_t]]></c> integer.
          To use this function the ei library needs to be configured and compiled
          to use the GMP library. </p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_decode_double(const char *buf, int *index, double *p)</nametext></name>
      <fsummary>Decode a double</fsummary>
      <desc>
        <p>This function decodes an double-precision (64 bit) floating
          point number from the binary format.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_decode_boolean(const char *buf, int *index, int *p)</nametext></name>
      <fsummary>Decode a boolean</fsummary>
      <desc>
        <p>This function decodes a boolean value from the binary
          format. A boolean is actually an atom, <c><![CDATA[true]]></c> decodes 1
          and <c><![CDATA[false]]></c> decodes 0.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_decode_char(const char *buf, int *index, char *p)</nametext></name>
      <fsummary>Decode an 8-bit integer between 0-255</fsummary>
      <desc>
        <p>This function decodes a char (8-bit) integer between 0-255
          from the binary format.
          Note that for historical reasons the returned integer is of
          type <c><![CDATA[char]]></c>. Your C code should consider the
          returned value to be of type <c><![CDATA[unsigned char]]></c> even if
          the C compilers and system may define <c><![CDATA[char]]></c> to be
          signed.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_decode_string(const char *buf, int *index, char *p)</nametext></name>
      <fsummary>Decode a string</fsummary>
      <desc>
        <p>This function decodes a string from the binary format. A
          string in erlang is a list of integers between 0 and
          255. Note that since the string is just a list, sometimes
          lists are encoded as strings by <c><![CDATA[term_to_binary/1]]></c>,
          even if it was not intended.</p>
        <p>The string is copied to <c><![CDATA[p]]></c>, and enough space must be
          allocated. The returned string is null terminated so you
          need to add an extra byte to the memory requirement.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_decode_atom(const char *buf, int *index, char *p)</nametext></name>
      <fsummary>Decode an atom</fsummary>
      <desc>
        <p>This function decodes an atom from the binary format.  The
	null terminated name of the atom is placed at <c><![CDATA[p]]></c>. There can be at most
          <c><![CDATA[MAXATOMLEN]]></c> bytes placed in the buffer.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_decode_atom_as(const char *buf, int *index, char *p, int plen, erlang_char_encoding want, erlang_char_encoding* was, erlang_char_encoding* result)</nametext></name>
      <fsummary>Decode an atom</fsummary>
      <desc>
        <p>This function decodes an atom from the binary format. The
	null terminated name of the atom is placed in buffer at <c>p</c> of length
	<c>plen</c> bytes.</p>
	<p>The wanted string encoding is specified by <c><seealso marker="#erlang_char_encoding">
	want</seealso></c>. The original encoding used in the
	binary format (latin1 or utf8) can be obtained from <c>*was</c>. The actual encoding of the resulting string
	(7-bit ascii, latin1 or utf8) can be obtained from <c>*result</c>. Both <c>was</c> and <c>result</c> can be <c>NULL</c>.
	
	<c>*result</c> may differ from <c>want</c> if <c>want</c> is a bitwise-or'd combination like
	<c>ERLANG_LATIN1|ERLANG_UTF8</c> or if <c>*result</c> turn out to be pure 7-bit ascii
	(compatible with both latin1 and utf8).</p>
	<p>This function fails if the atom is too long for the buffer
	or if it can not be represented with encoding <c>want</c>.</p>
	<p>This function was introduced in R16 release of Erlang/OTP as part of a first step
	to support UTF8 atoms.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_decode_binary(const char *buf, int *index, void *p, long *len)</nametext></name>
      <fsummary>Decode a binary</fsummary>
      <desc>
        <p>This function decodes a binary from the binary format. The
          <c><![CDATA[len]]></c> parameter is set to the actual size of the
          binary. Note that <c><![CDATA[ei_decode_binary()]]></c> assumes that there
          are enough room for the binary. The size required can be
          fetched by <c><![CDATA[ei_get_type()]]></c>.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_decode_fun(const char *buf, int *index, erlang_fun *p)</nametext></name>
      <name><ret>void</ret><nametext>free_fun(erlang_fun* f)</nametext></name>
      <fsummary>Decode a fun</fsummary>
      <desc>
        <p>This function decodes a fun from the binary format. The
          <c><![CDATA[p]]></c> parameter should be NULL or point to an
          <c><![CDATA[erlang_fun]]></c> structure. This is the only decode
          function that allocates memory; when the <c><![CDATA[erlang_fun]]></c>
          is no longer needed, it should be freed with
          <c><![CDATA[free_fun]]></c>. (This has to do with the arbitrary size of
          the environment for a fun.)</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_decode_pid(const char *buf, int *index, erlang_pid *p)</nametext></name>
      <fsummary>Decode a <c><![CDATA[pid]]></c></fsummary>
      <desc>
        <p>Decodes a pid, process identifier, from the binary format.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_decode_port(const char *buf, int *index, erlang_port *p)</nametext></name>
      <fsummary>Decode a port</fsummary>
      <desc>
        <p>This function decodes a port identifier from the binary
          format.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_decode_ref(const char *buf, int *index, erlang_ref *p)</nametext></name>
      <fsummary>Decode a reference</fsummary>
      <desc>
        <p>This function decodes a reference from the binary format.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_decode_trace(const char *buf, int *index, erlang_trace *p)</nametext></name>
      <fsummary>Decode a trace token</fsummary>
      <desc>
        <p>Decodes an erlang trace token from the binary format.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_decode_tuple_header(const char *buf, int *index, int *arity)</nametext></name>
      <fsummary>Decode a tuple</fsummary>
      <desc>
        <p>This function decodes a tuple header, the number of elements
          is returned in <c><![CDATA[arity]]></c>. The tuple elements follows in order in
          the buffer.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_decode_list_header(const char *buf, int *index, int *arity)</nametext></name>
      <fsummary>Decode a list</fsummary>
      <desc>
        <p>This function decodes a list header from the binary
          format. The number of elements is returned in
          <c><![CDATA[arity]]></c>. The <c><![CDATA[arity+1]]></c> elements follows (the last
          one is the tail of the list, normally an empty list.) If
          <c><![CDATA[arity]]></c> is <c><![CDATA[0]]></c>, it's an empty list.</p>
        <p>Note that lists are encoded as strings, if they consist
          entirely of integers in the range 0..255. This function will
          not decode such strings, use <c><![CDATA[ei_decode_string()]]></c>
          instead.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_decode_map_header(const char *buf, int *index, int *arity)</nametext></name>
      <fsummary>Decode a map</fsummary>
      <desc>
        <p>This function decodes a map header from the binary
          format. The number of key-value pairs is returned in
          <c>arity</c>. Keys and values follows, first all keys and then all values,
	  which makes a total of <c>arity*2</c> terms.
	  Keys and values are paired according to their order, the first key
	  with the first value and so on. If <c>arity</c> is zero, it's an empty map.
	  A correctly encoded map does not have duplicate keys.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_decode_ei_term(const char* buf, int* index, ei_term* term)</nametext></name>
      <fsummary>Decode a term, without prior knowledge of type</fsummary>
      <desc>
        <p>This function decodes any term, or at least tries to. If the
          term pointed at by <c><![CDATA[*index]]></c> in <c><![CDATA[buf]]></c> fits in the
          <c><![CDATA[term]]></c> union, it is decoded, and the appropriate field
          in <c><![CDATA[term->value]]></c> is set, and <c><![CDATA[*index]]></c> is
          incremented by the term size.</p>
        <p>The function returns 1 on successful decoding, -1 on error,
          and 0 if the term seems alright, but does not fit in the
          <c><![CDATA[term]]></c> structure. If it returns 1, the <c><![CDATA[index]]></c>
          will be incremented, and the <c><![CDATA[term]]></c> contains the
          decoded term.</p>
        <p>The <c><![CDATA[term]]></c> structure will contain the arity for a tuple
          or list, size for a binary, string or atom. It will contains
          a term if it's any of the following: integer, float, atom,
          pid, port or ref.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_decode_term(const char *buf, int *index, void *t)</nametext></name>
      <fsummary>Decode a <c><![CDATA[ETERM]]></c></fsummary>
      <desc>
        <p>This function decodes a term from the binary format. The
          term is return in <c><![CDATA[t]]></c> as a <c><![CDATA[ETERM*]]></c>, so <c><![CDATA[t]]></c>
          is actually an <c><![CDATA[ETERM**]]></c> (see
          <c><![CDATA[erl_interface(3)]]></c>. The term should later be
          deallocated.</p>
        <p>Note that this function is located in the erl_interface
          library.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_print_term(FILE* fp, const char* buf, int* index)</nametext></name>
      <name><ret>int</ret><nametext>ei_s_print_term(char** s, const char* buf, int* index)</nametext></name>
      <fsummary>Print a term in clear text</fsummary>
      <desc>
        <p>This function prints a term, in clear text, to the file
          given by <c><![CDATA[fp]]></c>, or the buffer pointed to by <c><![CDATA[s]]></c>. It
          tries to resemble the term printing in the erlang shell.</p>
        <p>In <c><![CDATA[ei_s_print_term()]]></c>, the parameter <c><![CDATA[s]]></c> should
          point to a dynamically (malloc) allocated string of
          <c><![CDATA[BUFSIZ]]></c> bytes or a NULL pointer. The string may be
          reallocated (and <c><![CDATA[*s]]></c> may be updated) by this function
          if the result is more than <c><![CDATA[BUFSIZ]]></c> characters. The
          string returned is zero-terminated.</p>
        <p>The return value is the number of characters written to the
          file or string, or -1 if <c><![CDATA[buf[index]]]></c> doesn't contain a
          valid term. Unfortunately, I/O errors on <c><![CDATA[fp]]></c> is not
          checked.</p>
        <p>The argument <c><![CDATA[index]]></c> is updated, i.e. this function can
          be viewed as en decode function that decodes a term into a
          human readable format.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_x_format(ei_x_buff* x, const char* fmt, ...)</nametext></name>
      <name><ret>int</ret><nametext>ei_x_format_wo_ver(ei_x_buff* x, const char *fmt, ... )</nametext></name>
      <fsummary>Format a term from a format string and parameters.</fsummary>
      <desc>
        <p>Format a term, given as a string, to a buffer. This
          functions works like a sprintf for erlang terms. The
          <c><![CDATA[fmt]]></c> contains a format string, with arguments like
          <c><![CDATA[~d]]></c>, to insert terms from variables. The following
          formats are supported (with the C types given):</p>
        <p></p>
        <pre>
~a - an atom, char*
~c - a character, char
~s - a string, char*
~i - an integer, int
~l - a long integer, long int
~u - a unsigned long integer, unsigned long int
~f - a float, float
~d - a double float, double float
~p - an Erlang PID, erlang_pid*
        </pre>
        <p>For instance, to encode a tuple with some stuff:</p>
        <pre>
ei_x_format("{~a,~i,~d}", "numbers", 12, 3.14159)
encodes the tuple {numbers,12,3.14159}
        </pre>
        <p>The <c><![CDATA[ei_x_format_wo_ver()]]></c> formats into a buffer, without
          the initial version byte.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_x_new(ei_x_buff* x)</nametext></name>
      <name><ret>int</ret><nametext>ei_x_new_with_version(ei_x_buff* x)</nametext></name>
      <fsummary>Allocate a new buffer</fsummary>
      <desc>
        <p>This function allocates a new <c><![CDATA[ei_x_buff]]></c> buffer. The
          fields of the structure pointed to by <c><![CDATA[x]]></c> parameter is
          filled in, and a default buffer is allocated. The
          <c><![CDATA[ei_x_new_with_version()]]></c> also puts an initial version
          byte, that is used in the binary format. (So that
          <c><![CDATA[ei_x_encode_version()]]></c> won't be needed.)</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_x_free(ei_x_buff* x)</nametext></name>
      <fsummary>Frees a buffer</fsummary>
      <desc>
        <p>This function frees an <c><![CDATA[ei_x_buff]]></c> buffer. The memory
          used by the buffer is returned to the OS.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_x_append(ei_x_buff* x, const ei_x_buff* x2)</nametext></name>
      <name><ret>int</ret><nametext>ei_x_append_buf(ei_x_buff* x, const char* buf, int len)</nametext></name>
      <fsummary>Appends a buffer at the end</fsummary>
      <desc>
        <p>These functions appends data at the end of the buffer <c><![CDATA[x]]></c>.</p>
      </desc>
    </func>
    <func>
      <name><ret>int</ret><nametext>ei_skip_term(const char* buf, int* index)</nametext></name>
      <fsummary>skip a term</fsummary>
      <desc>
        <p>This function skips a term in the given buffer, it
          recursively skips elements of lists and tuples, so that a
          full term is skipped. This is a way to get the size of an
          erlang term.</p>
        <p><c><![CDATA[buf]]></c> is the buffer.</p>
        <p><c><![CDATA[index]]></c> is updated to point right after the term in the
          buffer.</p>
        <note>
          <p>This can be useful when you want to hold arbitrary
            terms: just skip them and copy the binary term data to some
            buffer.</p>
        </note>
        <p>The function returns <c><![CDATA[0]]></c> on success and <c><![CDATA[-1]]></c> on
          failure.</p>
      </desc>
    </func>
  </funcs>

  <section>
    <title>Debug Information</title>
    <p>Some tips on what to check when the emulator doesn't seem to
      receive the terms that you send.</p>
    <list type="bulleted">
      <item>be careful with the version header, use
      <c><![CDATA[ei_x_new_with_version()]]></c> when appropriate</item>
      <item>turn on distribution tracing on the erlang node</item>
      <item>check the result codes from ei_decode_-calls</item>
    </list>
  </section>

  <section>
    <title>See Also</title>
    <p>erl_interface(3)</p>
  </section>
</cref>