<?xml version="1.0" encoding="latin1" ?>
<!DOCTYPE cref SYSTEM "cref.dtd">
<cref>
<header>
<copyright>
<year>1996</year><year>2009</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_marshal</title>
<prepared>Torbjörn Törnkvist</prepared>
<responsible>Torbjörn Törnkvist</responsible>
<docno></docno>
<approved>Bjarne Däcker</approved>
<checked>Torbjörn Törnkvist</checked>
<date>980703</date>
<rev>A</rev>
<file>erl_marshal.sgml</file>
</header>
<lib>erl_marshal</lib>
<libsummary>Encoding and Decoding of Erlang terms</libsummary>
<description>
<p>This module contains functions for encoding Erlang terms into
a sequence of bytes, and for decoding Erlang terms from a
sequence of bytes.</p>
</description>
<funcs>
<func>
<name><ret>int</ret><nametext>erl_compare_ext(bufp1, bufp2)</nametext></name>
<fsummary>Compares encoded byte sequences</fsummary>
<type>
<v>unsigned char *bufp1,*bufp2;</v>
</type>
<desc>
<p>This function compares two encoded terms.
</p>
<p><c><![CDATA[bufp1]]></c> is a buffer containing an encoded Erlang
term term1.
</p>
<p><c><![CDATA[bufp2]]></c> is a buffer containing an encoded Erlang
term term2.
</p>
<p>The function returns 0 if the terms are equal, -1 if term1
is less than term2, or 1 if term2 is less than term1.
</p>
</desc>
</func>
<func>
<name><ret>ETERM *</ret><nametext>erl_decode(bufp)</nametext></name>
<name><ret>ETERM *</ret><nametext>erl_decode_buf(bufpp)</nametext></name>
<fsummary>Converts a term from Erlang external format</fsummary>
<type>
<v>unsigned char *bufp;</v>
<v>unsigned char **bufpp;</v>
</type>
<desc>
<p><c><![CDATA[erl_decode()]]></c> and <c><![CDATA[erl_decode_buf()]]></c> decode
the contents of a buffer and return the corresponding
Erlang term. <c><![CDATA[erl_decode_buf()]]></c> provides a simple
mechanism for dealing with several encoded terms stored
consecutively in the buffer.</p>
<p><c><![CDATA[bufp]]></c> is a pointer to a buffer containing one or
more encoded Erlang terms.
</p>
<p><c><![CDATA[bufpp]]></c> is the address of a buffer pointer. The buffer
contains one or more consecutively encoded Erlang terms.
Following a successful call to <c><![CDATA[erl_decode_buf()]]></c>,
<c><![CDATA[bufpp]]></c> will be updated so that it points to the next
encoded term.
</p>
<p><c><![CDATA[erl_decode()]]></c> returns an Erlang term
corresponding to the contents of <c><![CDATA[bufp]]></c> on success, or
NULL on failure. <c><![CDATA[erl_decode_buf()]]></c> returns an Erlang
term corresponding to the first of the consecutive terms in
<c><![CDATA[bufpp]]></c> and moves <c><![CDATA[bufpp]]></c> forward to point to the
next term in the buffer. On failure, each of the functions
returns NULL.
</p>
</desc>
</func>
<func>
<name><ret>int</ret><nametext>erl_encode(term, bufp)</nametext></name>
<name><ret>int</ret><nametext>erl_encode_buf(term, bufpp)</nametext></name>
<fsummary>Converts a term into Erlang external format</fsummary>
<type>
<v>ETERM *term;</v>
<v>unsigned char *bufp;</v>
<v>unsigned char **bufpp;</v>
</type>
<desc>
<p><c><![CDATA[erl_encode()]]></c> and <c><![CDATA[erl_encode_buf()]]></c> encode
Erlang terms into external format for storage or transmission.
<c><![CDATA[erl_encode_buf()]]></c> provides a simple mechanism for
encoding several terms consecutively in the same
buffer.
</p>
<p><c>term</c> is an Erlang term to be encoded.
</p>
<p><c>bufp</c> is a pointer to a buffer containing one or
more encoded Erlang terms.
</p>
<p><c>bufpp</c> is a pointer to a pointer to a buffer
containing one or more consecutively encoded Erlang terms.
Following a successful call to <c><![CDATA[erl_encode_buf()]]></c>,
<c>bufpp</c> will be updated so that it points to the
position for the next encoded term.
</p>
<p>
These functions returns the number of bytes written to buffer
if successful, otherwise returns 0.
</p>
<p>Note that no bounds checking is done on the buffer. It is
the caller's responsibility to make sure that the buffer is
large enough to hold the encoded terms. You can either use a
static buffer that is large enough to hold the terms you
expect to need in your program, or use <c><![CDATA[erl_term_len()]]></c>
to determine the exact requirements for a given term.
</p>
<p>The following can help you estimate the buffer
requirements for a term. Note that this information is
implementation specific, and may change in future versions.
If you are unsure, use <c><![CDATA[erl_term_len()]]></c>.
</p>
<p>Erlang terms are encoded with a 1 byte tag that
identifies the type of object, a 2- or 4-byte length field,
and then the data itself. Specifically:
</p>
<taglist>
<tag><c><![CDATA[Tuples]]></c></tag>
<item>need 5 bytes, plus the space for each element.</item>
<tag><c><![CDATA[Lists]]></c></tag>
<item>need 5 bytes, plus the space for each element, and 1
additional byte for the empty list at the end.</item>
<tag><c><![CDATA[Strings and atoms]]></c></tag>
<item>need 3 bytes, plus 1 byte for each character (the
terminating 0 is not encoded). Really long strings (more
than 64k characters) are encoded as lists. Atoms cannot
contain more than 256 characters.</item>
<tag><c><![CDATA[Integers]]></c></tag>
<item>need 5 bytes.</item>
<tag><c><![CDATA[Characters]]></c></tag>
<item>(integers < 256) need 2 bytes.</item>
<tag><c><![CDATA[Floating point numbers]]></c></tag>
<item>need 32 bytes.</item>
<tag><c><![CDATA[Pids]]></c></tag>
<item>need 10 bytes, plus the space for the node name, which
is an atom.</item>
<tag><c><![CDATA[Ports and Refs]]></c></tag>
<item>need 6 bytes, plus the space for the node name, which
is an atom.</item>
</taglist>
<p>The total space required will be the result calculated
from the information above, plus 1 additional byte for a
version identifier.
</p>
</desc>
</func>
<func>
<name><ret>int</ret><nametext>erl_ext_size(bufp)</nametext></name>
<fsummary>Counts elements in encoded term</fsummary>
<type>
<v>unsigned char *bufp;</v>
</type>
<desc>
<p>This function returns the number of elements in an
encoded term.</p>
</desc>
</func>
<func>
<name><ret>unsigned char</ret><nametext>erl_ext_type(bufp)</nametext></name>
<fsummary>Determines type of an encoded byte sequence</fsummary>
<type>
<v>unsigned char *bufp;</v>
</type>
<desc>
<p>This function identifies and returns the type of Erlang term encoded
in a buffer. It will skip a trailing <em>magic</em> identifier.
Returns <c><![CDATA[0]]></c> if the type can't be determined or one of</p>
<list type="bulleted">
<item>
<p>ERL_INTEGER</p>
</item>
<item>
<p>ERL_ATOM</p>
</item>
<item>
<p>ERL_PID <c><![CDATA[/* Erlang process identifier */]]></c></p>
</item>
<item>
<p>ERL_PORT</p>
</item>
<item>
<p>ERL_REF <c><![CDATA[/* Erlang reference */]]></c></p>
</item>
<item>
<p>ERL_EMPTY_LIST</p>
</item>
<item>
<p>ERL_LIST</p>
</item>
<item>
<p>ERL_TUPLE</p>
</item>
<item>
<p>ERL_FLOAT</p>
</item>
<item>
<p>ERL_BINARY</p>
</item>
<item>
<p>ERL_FUNCTION</p>
</item>
</list>
</desc>
</func>
<func>
<name><ret>unsigned char *</ret><nametext>erl_peek_ext(bufp, pos)</nametext></name>
<fsummary>Steps over encoded term</fsummary>
<type>
<v>unsigned char *bufp;</v>
<v>int pos;</v>
</type>
<desc>
<p>This function is used for stepping over one or more
encoded terms in a buffer, in order to directly access a
later term.
</p>
<p><c><![CDATA[bufp]]></c> is a pointer to a buffer containing one or
more encoded Erlang terms.
</p>
<p><c><![CDATA[pos]]></c> indicates how many terms to step over in the
buffer.
</p>
<p>The function returns a pointer to a sub-term that can be
used in a subsequent call to <c><![CDATA[erl_decode()]]></c> in order to retrieve
the term at that position. If there is no term, or <c><![CDATA[pos]]></c>
would exceed the size of the terms in the buffer, NULL is returned.
</p>
</desc>
</func>
<func>
<name><ret>int</ret><nametext>erl_term_len(t)</nametext></name>
<fsummary>Determines encoded size of term</fsummary>
<type>
<v>ETERM *t;</v>
</type>
<desc>
<p>This function determines the buffer space that would be
needed by <c><![CDATA[t]]></c> if it were encoded into Erlang external
format by <c><![CDATA[erl_encode()]]></c>.
</p>
<p>The size in bytes is returned.
</p>
</desc>
</func>
</funcs>
</cref>
