path: root/lib/stdlib/doc/src/unicode.xml

<?xml version="1.0" encoding="latin1" ?>
<!DOCTYPE erlref SYSTEM "erlref.dtd">

<erlref>
  <header>
    <copyright>
      <year>1996</year>
      <year>2011</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.

  The Initial Developer of the Original Code is Ericsson AB.
    </legalnotice>

    <title>unicode</title>
    <prepared></prepared>
    <docno></docno>
    <date></date>
    <rev></rev>
  </header>
  <module>unicode</module>
  <modulesummary>Functions for converting Unicode characters</modulesummary>
  <description>
  <p>This module contains functions for converting between different character representations. Basically it converts between iso-latin-1 characters and Unicode ditto, but it can also convert between different Unicode encodings (like UTF-8, UTF-16 and UTF-32).</p>
  <p>The default Unicode encoding in Erlang is in binaries UTF-8, which is also the format in which built in functions and libraries in OTP expect to find binary Unicode data. In lists, Unicode data is encoded as integers, each integer representing one character and encoded simply as the Unicode codepoint for the character.</p> 
  <p>Other Unicode encodings than integers representing codepoints or UTF-8 in binaries are referred to as &quot;external encodings&quot;. The iso-latin-1 encoding is in binaries and lists referred to as latin1-encoding.</p>
  <p>It is recommended to only use external encodings for communication with external entities where this is required. When working inside the Erlang/OTP environment, it is recommended to keep binaries in UTF-8 when representing Unicode characters. Latin1 encoding is supported both for backward compatibility and for communication with external entities not supporting Unicode character sets.</p>
  </description>

  <datatypes>
    <datatype>
      <name name="encoding"/>
    </datatype>
    <datatype>
      <name name="endian"/>
    </datatype>
    <datatype>
      <name name="unicode_binary"/>
      <desc>
        <p>A binary() with characters encoded in the UTF-8 coding standard.</p>
      </desc>
    </datatype>
    <datatype>
      <name name="unicode_char"/>
      <desc>
        <p>An integer() representing a valid unicode codepoint.</p>
      </desc>
    </datatype>
    <datatype>
      <name name="chardata"/>
    </datatype>
    <datatype>
      <name name="charlist"/>
      <desc>
        <p>A unicode_binary is allowed as the tail of the list.</p>
      </desc>
    </datatype>
    <datatype>
      <name name="external_unicode_binary"/>
      <desc>
        <p>A <c>binary()</c> with characters coded in a user specified Unicode
           encoding other than UTF-8 (UTF-16 or UTF-32).</p>
      </desc>
    </datatype>
    <datatype>
      <name name="external_chardata"/>
    </datatype>
    <datatype>
      <name name="external_charlist"/>
      <desc>
        <p>An <c>external_unicode_binary()</c> is allowed as the tail
           of the list.</p>
      </desc>
    </datatype>
    <datatype>
      <name name="latin1_binary"/>
      <desc><p>A <c>binary()</c> with characters coded in iso-latin-1.</p>
      </desc>
    </datatype>
    <datatype>
      <name name="latin1_char"/>
      <desc><p>An <c>integer()</c> representing valid latin1
         character (0-255).</p>
      </desc>
    </datatype>
    <datatype>
      <name name="latin1_chardata"/>
    </datatype>
    <datatype>
      <name name="latin1_charlist"/>
      <desc><p>A <c>latin1_binary()</c> is allowed as the tail of
        the list.</p>
      </desc>
    </datatype>
  </datatypes>

  <funcs>
    <func>
      <name name="bom_to_encoding" arity="1"/>
      <fsummary>Identify UTF byte order marks in a binary.</fsummary>
      <type name="endian"/>
      <type_desc variable="Bin">A binary() of byte_size 4 or more.</type_desc>
      <desc>

      <p>Check for a UTF byte order mark (BOM) in the beginning of a
      binary.  If the supplied binary <c><anno>Bin</anno></c> begins with a valid
      byte order mark for either UTF-8, UTF-16 or UTF-32, the function
      returns the encoding identified along with the length of the BOM
      in bytes.</p>

      <p>If no BOM is found, the function returns <c>{latin1,0}</c></p>
      </desc>
    </func>
    <func>
      <name name="characters_to_list" arity="1"/>
      <fsummary>Convert a collection of characters to list of Unicode characters</fsummary>
      <desc>
      <p>Same as characters_to_list(<anno>Data</anno>,unicode).</p>
      </desc>
    </func>
    <func>
      <name>characters_to_list(Data, InEncoding) -> Result</name>
      <fsummary>Convert a collection of characters to list of Unicode characters</fsummary>
      <type>
        <v>Data = <seealso marker="#type-latin1_chardata">latin1_chardata()</seealso>
                | <seealso marker="#type-chardata">chardata()</seealso>
                | <seealso marker="#type-external_chardata">external_chardata()</seealso></v>
        <v>Result = list() | {error, list(), RestData} | {incomplete, list(), binary()}</v>
        <v>RestData = <seealso marker="#type-latin1_chardata">latin1_chardata()</seealso>
                    | <seealso marker="#type-chardata">chardata()</seealso>
                    | <seealso marker="#type-external_chardata">external_chardata()</seealso></v>
        <v>InEncoding = <seealso marker="#type-encoding">encoding()</seealso></v>
      </type>
      <desc>

      <p>This function converts a possibly deep list of integers and
      binaries into a list of integers representing unicode
      characters. The binaries in the input may have characters
      encoded as latin1 (0 - 255, one character per byte), in which
      case the <c>InEncoding</c> parameter should be given as
      <c>latin1</c>, or have characters encoded as one of the
      UTF-encodings, which is given as the <c>InEncoding</c>
      parameter. Only when the <c>InEncoding</c> is one of the UTF
      encodings, integers in the list are allowed to be grater than
      255.</p>
      
      <p>If <c>InEncoding</c> is <c>latin1</c>, the <c>Data</c> parameter
      corresponds to the <c>iodata()</c> type, but for <c>unicode</c>,
      the <c>Data</c> parameter can contain integers greater than 255
      (unicode characters beyond the iso-latin-1 range), which would
      make it invalid as <c>iodata()</c>.</p>

      <p>The purpose of the function is mainly to be able to convert
      combinations of unicode characters into a pure unicode
      string in list representation for further processing. For
      writing the data to an external entity, the reverse function
      <seealso
      marker="#characters_to_binary/3">characters_to_binary/3</seealso>
      comes in handy.</p>
 
       <p>The option <c>unicode</c> is an alias for <c>utf8</c>, as this is the
      preferred encoding for Unicode characters in
      binaries. <c>utf16</c> is an alias for <c>{utf16,big}</c> and
      <c>utf32</c> is an alias for <c>{utf32,big}</c>. The <c>big</c>
      and <c>little</c> atoms denote big or little endian
      encoding.</p> 

      <p>If for some reason, the data cannot be converted, either
      because of illegal unicode/latin1 characters in the list, or
      because of invalid UTF encoding in any binaries, an error
      tuple is returned. The error tuple contains the tag
      <c>error</c>, a list representing the characters that could be
      converted before the error occurred and a representation of the
      characters including and after the offending integer/bytes. The
      last part is mostly for debugging as it still constitutes a
      possibly deep and/or mixed list, not necessarily of the same
      depth as the original data. The error occurs when traversing the
      list and whatever's left to decode is simply returned as is.</p>

      <p>However, if the input <c>Data</c> is a pure binary, the third
      part of the error tuple is guaranteed to be a binary as
      well.</p>

      <p>Errors occur for the following reasons:</p>
      <list type="bulleted">

           <item>Integers out of range - If <c>InEncoding</c> is
	   <c>latin1</c>, an error occurs whenever an integer greater
	   than 255 is found in the lists. If <c>InEncoding</c> is
	   of a Unicode type, an error occurs whenever an integer
	   <list type="bulleted">
	     <item>greater than <c>16#10FFFF</c>
	     (the maximum unicode character),</item>
	     <item>in the range <c>16#D800</c> to <c>16#DFFF</c>
	       (invalid range reserved for UTF-16 surrogate pairs)</item>
	   </list>
	   is found.
	   </item>

	   <item>UTF encoding incorrect - If <c>InEncoding</c> is
	   one of the UTF types, the bytes in any binaries have to be valid
	   in that encoding. Errors can occur for various
	   reasons, including &quot;pure&quot; decoding errors 
	   (like the upper
	   bits of the bytes being wrong), the bytes are decoded to a
	   too large number, the bytes are decoded to a code-point in the 
	   invalid unicode
	   range or encoding is &quot;overlong&quot;, meaning that a
	   number should have been encoded in fewer bytes. The
	   case of a truncated UTF is handled specially, see the
	   paragraph about incomplete binaries below. If
	   <c>InEncoding</c> is <c>latin1</c>, binaries are always valid 
	   as long as they contain whole bytes,
	   as each byte falls into the valid iso-latin-1 range.</item>

      </list>	   

      <p>A special type of error is when no actual invalid integers or
      bytes are found, but a trailing <c>binary()</c> consists of too
      few bytes to decode the last character. This error might occur
      if bytes are read from a file in chunks or binaries in other
      ways are split on non UTF character boundaries. In this case an
      <c>incomplete</c> tuple is returned instead of the <c>error</c>
      tuple. It consists of the same parts as the <c>error</c> tuple, but
      the tag is <c>incomplete</c> instead of <c>error</c> and the
      last element is always guaranteed to be a binary consisting of
      the first part of a (so far) valid UTF character.</p>

      <p>If one UTF characters is split over two consecutive
      binaries in the <c>Data</c>, the conversion succeeds. This means
      that a character can be decoded from a range of binaries as long
      as the whole range is given as input without errors
      occurring. Example:</p>

<code>
     decode_data(Data) ->
         case unicode:characters_to_list(Data,unicode) of
             {incomplete,Encoded, Rest} ->
	           More = get_some_more_data(),
		   Encoded ++ decode_data([Rest, More]);
	     {error,Encoded,Rest} ->
	           handle_error(Encoded,Rest);
             List ->
	           List
         end.
</code>
      <p>Bit-strings that are not whole bytes are however not allowed,
      so a UTF character has to be split along 8-bit boundaries to
      ever be decoded.</p>

      <p>If any parameters are of the wrong type, the list structure
      is invalid (a number as tail) or the binaries does not contain
      whole bytes (bit-strings), a <c>badarg</c> exception is
      thrown.</p>
 
      </desc>
    </func>
    <func>
     <name name="characters_to_binary" arity="1"/>
      <fsummary>Convert a collection of characters to an UTF-8 binary</fsummary>
      <desc>
      <p>Same as characters_to_binary(Data, unicode, unicode).</p>
      </desc>
    </func>
    <func>    
     <name>characters_to_binary(Data,InEncoding) -> Result</name>
      <fsummary>Convert a collection of characters to an UTF-8 binary</fsummary>

      <type>
        <v>Data = <seealso marker="#type-latin1_chardata">latin1_chardata()</seealso>
                | <seealso marker="#type-chardata">chardata()</seealso>
                | <seealso marker="#type-external_chardata">external_chardata()</seealso></v>
        <v>Result = binary() | {error, binary(), RestData} | {incomplete, binary(), binary()}</v>
        <v>RestData = <seealso marker="#type-latin1_chardata">latin1_chardata()</seealso>
                    | <seealso marker="#type-chardata">chardata()</seealso>
                    | <seealso marker="#type-external_chardata">external_chardata()</seealso></v>
        <v>InEncoding = <seealso marker="#type-encoding">encoding()</seealso></v>
      </type>
      <desc>
      <p>Same as characters_to_binary(Data, InEncoding, unicode).</p>
      </desc>
    </func>    
    <func>
      <name name="characters_to_binary" arity="3"/>
      <fsummary>Convert a collection of characters to an UTF-8 binary</fsummary>
      <desc>

      <p>This function behaves as <seealso
      marker="#characters_to_list/2">
      characters_to_list/2</seealso>, but produces an binary
      instead of a unicode list. The
      <c><anno>InEncoding</anno></c> defines how input is to be interpreted if
      binaries are present in the <c>Data</c>, while
      <c><anno>OutEncoding</anno></c> defines in what format output is to be
      generated.</p>

      <p>The option <c>unicode</c> is an alias for <c>utf8</c>, as this is the
      preferred encoding for Unicode characters in
      binaries. <c>utf16</c> is an alias for <c>{utf16,big}</c> and
      <c>utf32</c> is an alias for <c>{utf32,big}</c>. The <c>big</c>
      and <c>little</c> atoms denote big or little endian
      encoding.</p> 

      <p>Errors and exceptions occur as in <seealso
      marker="#characters_to_list/2">
      characters_to_list/2</seealso>, but the second element
      in the <c>error</c> or
      <c>incomplete</c> tuple will be a <c>binary()</c> and not a
      <c>list()</c>.</p>

      </desc>
    </func>
    <func>
      <name name="encoding_to_bom" arity="1"/>
      <fsummary>Create a binary UTF byte order mark from encoding.</fsummary>
      <type_desc variable="Bin">A binary() of byte_size 4 or more.</type_desc>
      <desc>

      <p>Create an UTF byte order mark (BOM) as a binary from the
      supplied <c><anno>InEncoding</anno></c>. The BOM is, if supported at all,
      expected to be placed first in UTF encoded files or
      messages.</p>

      <p>The function returns <c>&lt;&lt;&gt;&gt;</c> for the
      <c>latin1</c> encoding, there is no BOM for ISO-latin-1.</p>

      <p>It can be noted that the BOM for UTF-8 is seldom used, and it
      is really not a <em>byte order</em> mark. There are obviously no
      byte order issues with UTF-8, so the BOM is only there to
      differentiate UTF-8 encoding from other UTF formats.</p>

      </desc>
    </func>
  </funcs>
</erlref>