<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2 Final//EN">
<html>
<head>
<!--
Copyright (c) 2000, 2001, Oracle and/or its affiliates. All rights reserved.
DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.

This code is free software; you can redistribute it and/or modify it
under the terms of the GNU General Public License version 2 only, as
published by the Free Software Foundation.  Oracle designates this
particular file as subject to the "Classpath" exception as provided
by Oracle in the LICENSE file that accompanied this code.

This code is distributed in the hope that it will be useful, but WITHOUT
ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
version 2 for more details (a copy is included in the LICENSE file that
accompanied this code).

You should have received a copy of the GNU General Public License version
2 along with this work; if not, write to the Free Software Foundation,
Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.

Please contact Sun Microsystems, Inc., 4150 Network Circle, Santa Clara,
CA 95054 USA or visit www.sun.com if you need additional information or
have any questions.
-->

<title>JPEG Metadata Format Specification and Usage Notes</title>
</head>

<body bgcolor="white">

<center><h1>
JPEG Metadata Format Specification and Usage Notes
</h1></center>

<p>
<a href=#metadata>JPEG Metadata</a><br>
<a href=#abbrev>Abbreviated Streams</a><br>
<a href=#tables>Sources of Tables</a><br>
<a href=#color>Colorspace Transformations and Conventional Markers</a><br>
<a href=#thumbs>Thumbnail Images</a><br>
<a href=#prog>Progressive Encoding</a><br>
<a href=#tree>Native Metadata Format Tree Structure and Editing</a><br>
<a href=#image>Image Metadata DTD</a><br>
<a href=#stream>Stream Metadata DTD</a>
<p>
<bold>NOTE</bold>: It is important to call <code>dispose()</code> 
on the JPEG reader and writer objects when they are no longer needed, as 
they consume significant native resources which are not adequately recovered 
by garbage collection.  Both reader and writer call <code>dispose()</code> 
in their finalizers, but those finalizers may not be called before the native 
code has exhausted native memory.

<p>

The JPEG writer does not support replacing pixels.

<p>
<h2>
<a name=metadata>JPEG Metadata</a>
</h2>
JPEG metadata consists of the data contained in marker segments in a JPEG
stream.  The image metadata object returned from a read describes the
contents of the marker segments between the <code>SOI</code> marker and 
the <code>EOI</code> marker for that image.  The image metadata object 
passed into a write determines the contents of the stream between the 
<code>SOI</code> marker and the <code>EOI</code> marker for that image, 
subject to the controls in any <code>ImageWriteParam</code>.

<p>
Stream metadata is used only for tables-only images found (or to be 
placed) at the beginning of a stream containing abbreviated images.  
Tables-only images are not treated as images and do not consume an 
image index.  The stream metadata object returned from a read describes the
contents of the marker segments between the <code>SOI</code> marker and 
the <code>EOI</code> marker for the single tables-only image at the 
beginning of the stream, if one is present.  If no tables-only image is 
present at the front of the stream, the <code>getStreamMetadata</code> 
method of <code>ImageReader</code> returns <code>null</code>.  If 
stream metadata is provided to the writer, a single tables-only image 
containing the tables from the stream metadata object will be written at 
the beginning of the stream.  If the stream metadata object contains no
tables, default tables will be written.  As the sole purpose of stream
metadata is for specifying tables-only images at the front of abbreviated 
streams, the stream metadata argument is useful only on the 
<code>ImageWriter.prepareWriteSequence</code> method.  It is ignored on all 
other methods.

<p> 
The <code>ImageWriter.getDefaultStreamMetadata</code> method returns an 
object containing the tables from the <code>ImageWriteParam</code> argument, 
if it is a <code>JPEGImageWriteParam</code> and contains tables.  Otherwise, 
the returned object will contain default tables.

<p>The <code>ImageWriter.getDefaultImageMetadata</code> method returns a 
metadata object containing <bold>no</bold> tables if the 
<code>ImageWriteParam</code> argument contains tables.  Otherwise the 
returned metadata object will contain default visually lossless tables.  
Of course, only a <code>JPEGImageWriteParam</code> may contain tables.

<p>
If <code>ignoreMetadata</code> is set to <code>true</code> when the input
is set on the reader, stream metadata will not be available, but image
metadata will.

<p>
<h2>
<a name=abbrev>Abbreviated Streams</a>
</h2>
Both the reader and the writer retain their tables from one operation to the 
next, thus permitting the use of abbreviated streams quite naturally, with a 
few minor restrictions:  
<ol>
  <li> Abbreviated streams may contain only one tables-only image, which must 
       come first in the stream.  Subsequent tables-only images will cause
       undefined behavior.</li>
  <li> Abbreviated streams must be read fully and in order.  No random access
       is allowed, in either direction.  The same image may be read multiple 
       times.  If a call is made with an image index that is not the same as
       or one greater than the most recent call (or 0 if no calls have been
       made), then an <code>IllegalArgumentException</code> is thrown.</li>
</ol>
These restrictions mean that streams may contain abbreviated images 
interspersed with images containing tables.  As required by JPEG, any tables
appearing in the stream override previous tables, regardless of the source
of the previous tables.  

<p>
Note that once a tables-only image has been read, it's contents is available
as stream metadata from the reader until either another tables-only image
is read from another stream or the reader is reset.  Changing the input does
not reset the stream metadata.  This is useful for reading the tables from
one file, then changing the input to read an abbreviated stream containing 
a sequence of images.  The tables will be used automatically, and will remain
available as "stream" metadata.

<p>
Abbreviated streams are written using the sequence methods of 
<code>ImageWriter</code>.  Stream metadata is used to write a tables-only 
image at the beginning of the stream, and the tables are set up for use, using
<code>ImageWriter.prepareWriteSequence</code>.  If no stream metadata is 
supplied to <code>ImageWriter.prepareWriteSequence</code>, then no 
tables-only image is written.  If stream metadata containing no tables is
supplied to <code>ImageWriter.prepareWriteSequence</code>, then a tables-only
image containing default visually lossless tables is written.

<h2>
<a name=tables>Sources of Tables</a>
</h2>
<p>
Images are written with tables if tables are present in their metadata objects 
or without them if no tables are present in their metadata objects.  If no
metadata object is present then the tables are written.  The tables used for 
compression are taken from one of the following sources, which are consulted 
in order: 
<ol>
  <li> If there is an <code>ImageWriteParam</code> and the compression mode is
       set to <code>EXPLICIT</code>, default tables constructed using the
       quality setting are used.  They are written only if the metadata 
       contains tables or if there is no metadata, but they replace the 
       tables in the metadata.</li>
  <li> If there is an <code>ImageWriteParam</code> and the compression mode is
       set to <code>DEFAULT</code>, default visually lossles tables are used.
       They are written only if the metadata contains tables or if
       there is no metadata, but they replace the tables in the
       metadata.</li>
  <li> Otherwise the compression mode on the <code>ImageWriteParam</code> must
       be MODE_COPY_FROM_<code>METADATA</code>, in which case the following
       are used:
    <ol>
      <li> the tables in the image metadata, if present</li>
      <li> the tables in the stream metadata, if present</li>
      <li> the tables in the <code>JPEGImageWriteParam</code>, if present</li>
      <li> default visually lossless tables</li>
    </ol> Tables are written only if they are taken from image metadata.
  </li>
</ol>

This ordering implements the design intention that tables should be included 
in <code>JPEGImageWriteParam</code>s only as a means of specifying tables 
when no other source is available, and this can occur only when writing to an
abbreviated stream without tables using known non-standard tables for 
compression.

<p>
When reading, tables in a <code>JPEGImageReadParam</code> are consulted only 
if tables have not been set by any previous read.  Tables set from a 
<code>JPEGImageReadParam</code> are overridden by any tables present in the 
stream being read.

<p>
Note that if no image metadata object is specified for a particular image, a 
default object is used, which includes default tables.

<p>
<h2>
<a name=color>Colorspace Transformations and Conventional Markers</a>
</h2>
Colorspace transformations are controlled by the destination type for
both reading and writing of images.  When <code>Raster</code>s are
read, no colorspace transformation is performed, and any destination type 
is ignored.  A warning is sent to any listeners if a destination type is 
specified in this case.  When <code>Raster</code>s are written, any 
destination type is used to interpret the bands.  This might result in a
JFIF or Adobe header being written, or different component ids being written 
to the frame and scan headers.  If values present in a metadata object do not 
match the destination type, the destination type is used and a warning is sent 
to any listeners.

<p>

When reading, the contents of the stream are interpreted by the usual
JPEG conventions, as follows:

<ul>
  <li> If a JFIF <code>APP0</code> marker segment is present, the colorspace
       is known to be either grayscale or YCbCr.  If an <code>APP2</code>
       marker segment containing an embedded ICC profile is also present, then
       the YCbCr is converted to RGB according to the formulas given in the
       JFIF spec, and the ICC profile is assumed to refer to the resulting RGB
       space.
  <p>
  <li> If an Adobe <code>APP14</code> marker segment is present, the 
       colorspace is determined by consulting the <code>transform</code> flag.
       The <code>transform</code> flag takes one of three values:
    <ul>
      <li> 2 - The image is encoded as YCCK (implicitly converted from
           CMYK on encoding).

      <li> 1 - The image is encoded as YCbCr (implicitly converted from RGB
           on encoding).

      <li> 0 - Unknown.  3-channel images are assumed to be RGB, 4-channel 
           images are assumed to be CMYK.
    </ul>
   <p>
  <li> If neither marker segment is present, the following procedure is
       followed: Single-channel images are assumed to be grayscale, and
       2-channel images are assumed to be grayscale with an alpha channel.
       For 3- and 4-channel images, the component ids are consulted.  If these
       values are 1-3 for a 3-channel image, then the image is assumed to be
       YCbCr. If these values are 1-4 for a 4-channel image, then the image 
       is assumed to be YCbCrA.  If these values are > 4, they are checked 
       against the ASCII codes for 'R', 'G', 'B', 'A', 'C', 'c'.  These can 
       encode the following colorspaces:
       <p>
       <br>RGB
       <br>RGBA
       <br>YCC (as 'Y','C','c'), assumed to be PhotoYCC
       <br>YCCA (as 'Y','C','c','A'), assumed to be PhotoYCCA
       <p>
       Otherwise, 3-channel subsampled images are assumed to be YCbCr, 
       3-channel non-subsampled images are assumed to be RGB, 4-channel 
       subsampled images are assumed to be YCCK, and 4-channel, non-subsampled
       images are assumed to be CMYK.

  <p>
  <li> All other images are declared uninterpretable and an exception is
       thrown if an attempt is made to read one as a 
       <code>BufferedImage</code>.  Such an image may be read only as a 
       <code>Raster</code>.  If an image is interpretable but there is no Java
       <code>ColorSpace</code> available corresponding to the encoded 
       colorspace (<italic>e.g.</italic> YCbCr), then 
       <code>ImageReader.getRawImageType</code> will return <code>null</code>.
</ul>

Once an encoded colorspace is determined, then the target colorspace is 
determined as follows:

<ul>
  <li> If a destination type is not set, then the following default
       transformations take place after upsampling: YCbCr (and YCbCrA) images 
       are converted to RGB (and RGBA) using the conversion provided by the
       underlying IJG library and either the built-in sRGB 
       <code>ColorSpace</code> or a custom RGB <code>ColorSpace</code> object
       based on an embedded ICC profile is used to create the output
       <code>ColorModel</code>.  PhotoYCC and PhotoYCCA images are not
       converted.  CMYK and YCCK images are currently not supported.</li>

  <li> If a destination image or type is set, it is used as follows:
       If the IJG library provides an appropriate conversion, it is used.
       Otherwise the default library conversion is followed by a colorspace 
       conversion in Java.</li>

  <li> Bands are selected AFTER any library colorspace conversion.  If a 
       subset of either source or destination bands is used, then the default 
       library conversions are used with no further conversion in Java,
       regardless of any destination type.</li>

  <li> An exception is thrown if an attempt is made to read an image in an 
       unsupported jpeg colorspace as a <code>BufferedImage</code> 
       (<italic>e.g.</italic> CMYK).  Such images may be read as
       <code>Raster</code>s.  If an image colorspace is unsupported or
       uninterpretable, then <code>ImageReader.getImageTypes</code> will
       return an empty <code>Iterator</code>.   If a subset of the raw bands
       are required, a <code>Raster</code> must be obtained first and the
       bands obtained from that. </li>
</ul>

<p>
For writing, the color transformation to apply is determined as
follows:

<p>
If a subset of the source bands is to be written, no color conversion is 
performed.  Any destination, if set, must match the number of bands that will 
be written, and serves as an interpretation of the selected bands, rather than 
a conversion request.  This behavior is identical to that for 
<code>Raster</code>s.  If all the bands are to be written and an image 
(as opposed to a <code>Raster</code>) is being written, any destination type 
is ignored and a warning is sent to any listeners.

<p>
If a destination type is used and any aspect of the metadata object, if there 
is one, is not compatible with that type, the destination type is used, the 
metadata written is modified from that provided, and a warning is sent to 
listeners.  This includes the <code>app0JFIF</code> and 
<code>app14Adobe</code> nodes.  The component ids in the <code>sof</code> and 
<code>sos</code> nodes are not modified, however, as unless a 
<code>app0JFIF</code> node is present, any values may be used.
<p>

When a full image is written, a destination colorspace will be 
chosen based on the image contents and the metadata settings, according to 
the following algorithm:

<p>

If no metadata object is specified, then the following defaults apply:

<ul>
  <li> Grayscale images are written with a JFIF <code>APP0</code> marker
       segment.  Grayscale images with alpha are written with no special
       marker.  As required by JFIF, the component ids in the frame and
       scan header is set to 1.

  <li> RGB images are converted to YCbCr, subsampled in the chrominance
       channels by half both vertically and horizontally, and written with a
       JFIF <code>APP0</code> marker segment.  If the <code>ColorSpace</code>
       of the image is based on an <code>ICCProfile</code> (it is an instance
       of <code>ICC_ColorSpace</code>, but is not one of the standard built-in 
       <code>ColorSpaces</code>), then that profile is embedded in an 
       <code>APP2</code> marker segment.  As required by JFIF, the 
       component ids in the frame and scan headers are set to 1, 2, and 3.


  <li> RGBA images are converted to YCbCrA, subsampled in the
       chrominance channels by half both vertically and horizontally, and
       written without any special marker segments.  The component ids 
       in the frame and scan headers are set to 1, 2, 3, and 4.

  <li> PhotoYCC and YCCAimages are subsampled by half in the chrominance
       channels both vertically and horizontally and written with an
       Adobe <code>APP14</code> marker segment and 'Y','C', and 'c' (and
       'A' if an alpha channel is present) as component ids in the frame
        and scan headers.
</ul>

Default metadata objects for these image types will reflect these settings.

<p>

If a metadata object is specified, then the number of channels in the
frame and scan headers must always match the number of bands to be 
written, or an exception is thrown.  <code>app0JFIF</code> and 
<code>app14Adobe</code> nodes may appear in the same metadata object only 
if the <code>app14Adobe</code> node indicates YCbCr, and the component ids 
are JFIF compatible (0-2).  The various image types are processed in the 
following ways:

<br>

(All multi-channel images are subsampled according to the sampling factors 
in the frame header node of the metadata object, regardless of color space.)

<ul>
  <li> Grayscale Images:
    <ul>
      <li> If an <code>app0JFIF</code> node is present in the metadata object,
           a JFIF <code>APP0</code> marker segment is written.
      <li> If an <code>app14Adobe</code> node is present in the metadata 
           object, it is checked for validity (<code>transform</code> must be
           <code>UNKNOWN</code>) and written.
      <li> If neither node is present in the metadata object, no special 
           marker segment is written.
    </ul>

  <li> Grayscale Images with an Alpha Channel:
    <ul>
      <li> If an <code>app0JFIF</code> node is present in the metadata object,
           it is ignored and a warning is sent to listeners, as JFIF does not
           support 2-channel images.
      <li> If an <code>app14Adobe</code> node is present in the metadata
           object, it is checked for validity (<code>transform</code> must be
           <code>UNKNOWN</code>) and written.  If <code>transform</code> is
           not <code>UNKNOWN</code>, a warning is sent to listeners and the
           correct transform is written.
      <li> If neither node is present in the metadata object, no special 
           marker segment is written.
    </ul>

  <li> RGB Images:
    <ul>
      <li> If an <code>app0JFIF</code> node is present in the metadata object,
           the image is converted to YCbCr and written with a JFIF 
           <code>APP0</code> marker segment.  If the <code>ColorSpace</code>
           of the image is based on a non-standard ICC Profile, then that
           profile is embedded in an <code>APP2</code> marker segment.  If the
           <code>ColorSpace</code> is not based on a non-standard ICC Profile,
           but an <code>app2ICC</code> node appears in the metadata, then an
           <code>APP2</code> marker segment is written with the appropriate
           standard profile.  Note that the profile must specify an RGB color
           space, as the file must be JFIF compliant.

      <li> If an <code>app14Adobe</code> node is present in the metadata
           object, the image is converted according to the color transform
           setting and written with an Adobe <code>APP14</code> marker
           segment.  Component ids are written just as they appear in the
           frame and scan headers.  The color transform must be either YCbCr
           or <code>UNKNOWN</code>.  If it is <code>UNKNOWN</code>, the image
           is not color converted.

      <li> If neither node is present, the component ids in the frame
           header are consulted.  If these indicate a colorspace as described
           above, then the image is converted to that colorspace if possible.  
           If the component ids do not indicate a colorspace, then the
           sampling factors are consulted.  If the image is to be subsampled,
           it is converted to YCbCr first.  If the image is not to be
           subsampled, then no conversion is applied.  No special marker
           segmentss are written.
    </ul>

  <li> RGBA images:
    <ul>
      <li> If an <code>app0JFIF</code> node is present in the metadata object,
           it is ignored and a warning is sent to listeners, as JFIF does not
           support 4-channel images.

      <li> If an <code>app14Adobe</code> node is present in the metadata
           object, the image is written with an Adobe <code>APP14</code> marker
           segment.  No colorspace conversion is performed.  Component ids
           are written just as they appear in the frame and scan  headers.
           The color transform must be <code>UNKNOWN</code>.  If it is 
           not, a warning is sent to listeners.

      <li> If no <code>app14Adobe</code> node is present, the component ids in
           the frame header are consulted.  If these indicate a colorspace as
           described above, then the image is converted to that colorspace if
           possible.  If the component ids do not indicate a colorspace, then
           the sampling factors are consulted.  If the image is to be
           subsampled, it is converted to YCbCrA.  If the image is not to be
           subsampled, then no conversion is applied.  No special marker
           segments are written.
    </ul>

  <li> PhotoYCC Images:
    <ul>
      <li> If an <code>app0JFIF</code> node is present in the metadata object,
           the image is converted to sRGB, and then to YCbCr during encoding,
           and a JFIF <code>APP0</code> marker segment is written.

      <li> If an <code>app14Adobe</code> node is present in the metadata
           object, no conversion is applied, and an Adobe <code>APP14</code>
           marker segment is written.  The color transform must be YCC.  If it
           is not, a warning is sent to listeners.

      <li> If neither node is present in the metadata object, no conversion
           is applied, and no special marker segment is written.
    </ul>

  <li> PhotoYCCA Images:
    <ul>
      <li> If an <code>app0JFIF</code> node is present in the metadata object,
           it is ignored and a warning is sent to listeners, as JFIF does not
           support 4-channel images.

      <li> If an <code>app14Adobe</code> node is present in the metadata
           object, no conversion is applied, and an Adobe <code>APP14</code>
           marker segment is written.  The color transform must be
           <code>UNKNOWN</code>.  If it is not, a warning is sent to
           listeners.

      <li> If neither node is present in the metadata object, no conversion
           is applied, and no special marker segment is written.
    </ul>
</ul>

<p>
<h2>
<a name=thumbs>Thumbnail Images</a>
</h2>
Thumbnails are supported by the use of JFIF and JFIF extension marker segments.
Thumbnails provided on the write methods determine the thumbnails that will be 
included.  <code>app0JFIF</code> and <code>app0JFXX</code> nodes present in 
the metadata do not contain any thumbnail pixel data.  However, the kinds of 
thumbnails written depend on the contents of the metadata object, as follows.  
Any thumbnail which is to be written as an indexed or RGB image and which is 
larger than 255 by 255 will be clipped, not scaled, to 255 by 255.  Thumbnails 
written as JPEG images may be any size.  A warning is sent to any listeners 
whenever a thumbnail is clipped.
<ul>
  <li> If there is a single thumbnail, it is processed as follows:
    <ul>
      <li> If the thumbnail image is an RGB palette image, it is processed as
           follows:
        <ul>
          <li> If no <code>app0JFXX</code> node is present in the metadata, or
               the first <code>app0JFXX</code> node present in the metadata
               contains a <code>JFIFthumbPalette</code> element, a
               palette thumbnail is written in a JFXX <code>APP0</code> marker
               segment.
          <li> If the first <code>app0JFXX</code> node present in the metadata
               contains another thumbnail form (RGB or JPEG), the palette
               image is expanded to RGB and the indicated thumbnail form is
               written.  
        </ul>

      <li> If the thumbnail image is an RGB image, it is processed as follows:
        <ul>
          <li> If no <code>app0JFXX</code> node is present in the metadata,
               the thumbnail is written as part of the JFIF <code>APP0</code>
               marker segment. 
          <li> If the first <code>app0JFXX</code> node present in the metadata
               contains a <code>JFIFthumbRGB</code> element, an
               RGB thumbnail is written in a JFXX <code>APP0</code> marker
               segment.
          <li> If the first <code>app0JFXX</code> node present in the metadata
               contains a <code>JFIFthumbJPEG</code> element, a
               JPEG thumbnail is written in a JFXX <code>APP0</code> marker
               segment.
          <li> If the first <code>app0JFXX</code> node present in the metadata
               contains a <code>JFIFthumbPalette</code> element, an
               RGB thumbnail is written in a JFXX <code>APP0</code> marker
               segment and a warning is sent to any listeners.
        </ul>

      <li> If the thumbnail image is a grayscale image, it is processed as
           follows:
        <ul>
          <li> If no <code>app0JFXX</code> node is present in the metadata,
               the thumbnail is expanded to RGB and written as part of the
               JFIF <code>APP0</code> marker segment. 
          <li> If the first <code>app0JFXX</code> node present in the metadata
               contains a <code>JFIFthumbRGB</code> element, the thumbnail is
               expanded to RGB and written in a separate <code>JFXX</code> RGB
               marker segment.
          <li> If the first <code>app0JFXX</code> node present in the metadata
               contains a <code>JFIFthumbJPEG</code> element, a
               JPEG thumbnail is written in a JFXX <code>APP0</code> marker
               segment.
          <li> If the first <code>app0JFXX</code> node present in the metadata
               contains a <code>JFIFthumbPalette</code> element, a
               JPEG thumbnail is written in a JFXX <code>APP0</code> marker
               segment and a warning is sent to any listeners.
        </ul>

      <li> Any other thumbnail image types are ignored and a warning is sent
           to any listeners.
    </ul>

  <li> If there are multiple thumbnails, each one is processed as above, except
       that no thumbnail is placed in the JFIF <code>APP0</code> segment, and
       the <code>app0JFXX</code> node consulted for each thumbnail is the
       <code>app0JFXX</code> node from the metadata that occurs in the same
       sequence as the thumbnail.  <italic>I.e.</italic> the first
       <code>app0JFXX</code> node applies to the first thumbnail, the second
       node to the second thumbnail, and so on.  If there are fewer
       <code>app0JFXX</code> nodes in the metadata than thumbnails, then 
       those thumbnails are considered to have no matching
       <code>app0JFXX</code> node.  An RGB thumbnail with no matching
       <code>app0JFXX</code> node is written in a JFXX <code>APP0</code> marker
       segment.  A grayscale thumbnail with no matching
       <code>app0JFXX</code> node is written as a JPEG image to a JFXX 
       <code>APP0</code> marker segment.
</ul>
<p>

Note that as the only mechanism for storing thumbnails is via the
JFIF or JFIF extension marker segments, only grayscale or RGB images 
may have thumbnails.  If thumbnails are present when writing any other type 
of image, the thumbnails are ignored and a warning is sent to any warning
listeners.

<p>
<h2>
<a name=prog>Progressive Encoding</a>
</h2>

Progressive encoding must be enabled on the <code>ImageWriteParam</code>
passed in to a write operation, or the image will be written sequentially,
regardless of the scan headers included in the metadata object.  If 
progressive encoding is enabled and set to copy from metadata, then
the sequence of scan headers from the metadata is used to write the 
image.  If progressive encoding is enabled and set to use a default, 
then the scans in the metadata are ignored and a default set of scans 
is used.  Progressive encoding always forces optimized Huffman tables to
be used.  Any Huffman tables present in the metadata will be ignored,
and a warning will be sent to any warning listeners.

If Huffman-table optimization is requested on the <code>ImageWriteParam</code>,
all Huffman tables in the metadata or in the <code>ImageWriteParam</code> 
itself are ignored, and a warning will be sent to any warning listeners if 
any such tables are present.

<p>
<h2>
<a name=tree>Native Metadata Format Tree Structure and Editing</a>
</h2>

The DTDs below describe just the trees of metadata objects actually returned
by the <code>IIOMetadata</code> object.  They do not include nodes 
corresponding to <code>SOI</code>, <code>EOI</code>, or <code>RST</code> 
markers, as these parsing delimiters do not carry any meaningful metadata.  
<p>

The first node is always a <code>JPEGvariety</code> node.  In the
<code>javax_imageio_jpeg_image_1.0</code> version of the JPEG metadata
format, this node may have one child, an <code>app0JFIF</code> node,
indicating that the JPEG stream contains a JFIF marker segment and related
data, or no children, indicating that the stream contains no JFIF marker.
In future versions of the JPEG metadata format, other varieties of JPEG
metadata may be supported (e.g. Exif) by defining other types of nodes
which may appear as a child of the <code>JPEGvariety</code> node.
<p>

(Note that an application wishing to interpret Exif metadata given
a metadata tree structure in the <code>javax_imageio_jpeg_image_1.0</code>
format must check for an <code>unknown</code> marker segment with a tag
indicating an <code>APP1</code> marker and containing data identifying it
as an Exif marker segment.  Then it may use application-specific code to
interpret the data in the marker segment.  If such an application were
to encounter a metadata tree formatted according to a future version of
the JPEG metadata format, the Exif marker segment might not be
<code>unknown</code> in that format - it might be structured as a
child node of the <code>JPEGvariety</code> node.  Thus, it is important
for an application to specify which version to use by passing the string
identifying the version to the method/constructor used to obtain an
<code>IIOMetadata</code> object.)

<p>

On reading, <code>JFXX</code> and <code>app2ICC</code> nodes occur as 
children of an <code>app0JFIF</code> node.
This is true regardless of where the JFXX <code>APP0</code> and 
<code>APP2</code> marker segments actually occur in the stream.  The ordering 
of nodes within the <code>markerSequence</code> node corresponds to the 
ordering of marker segments found in the JPEG stream.  
<p>
On writing, any <code>JFXX</code> and <code>app2ICC</code> nodes must 
occur as children of an <code>app0JFIF</code> node, itself a child of a 
<code>JPEGvariety</code> node, which must always be the first node.
(If the stream is not to be JFIF compliant, no <code>app0JFIF</code> node
should be provided, and the <code>JPEGvariety</code> node should have no
children.)  Any 
JFIF <code>APP0</code>, JFXX <code>APP0</code>, and <code>APP2</code> marker 
segments are written first, followed by all Adobe <code>APP14</code>, 
<code>APPn</code>, <code>COM</code> and unknown segments in the 
order in which their corresponding nodes appear in the 
<code>markerSequence</code> node, followed by <code>DQT</code> (and 
<code>DHT</code> for non-progressive writes) marker segments, followed by the 
<code>SOF</code> and <code>SOS</code> marker segments.  For progressive writes
using metadata to control progression, the <code>SOS</code> segments are used 
in the order in which their corresponding nodes occur in the 
<code>markerSequence</code> node.
<p>

The <code>reset</code>, <code>mergeTree</code> and <code>setFromTree</code> 
operations have the following semantics for the JPEG plug-in metadata object:

<p> <code>reset</code> - A call to <code>reset</code> will restore the 
metadata object to the same state it had immediately after creation, whether 
this came about from reading a stream or by obtaining a default object from 
the <code>ImageWriter</code>.  This is true regardless of how many times the 
metadata object has been modified since creation.

<p> <code>mergeTree</code> - Native Format
<br> The <code>mergeTree</code> operation accepts valid trees conforming to 
the DTD below, and merges the nodes using the following ordering rules.  In 
all cases, only data present in the new node is changed in a corresponding 
existing node, if any.  This means that nodes cannot be removed using 
<code>mergeTree</code>.  To remove nodes, use <code>setFromTree</code>.  The 
tree must consist of <code>IIOMetadataNode</code>s.
<ul>
  <li> <code>app0JFIF</code>
        <ul>
          <li> If an <code>app0JFIF</code> node already exists, the contents 
               of the new one modify the existing one. 
          <li> If there is no such node, a new one is created and inserted in
               the appropriate position.
        </ul>    
  <li> <code>dqt</code>
        <ul>
          <li> If there already exist <code>dqt</code> nodes in the sequence,
               then each table in the node replaces the first table, in any
               <code>dqt</code> node, with the same table id.  
          <li> If none of the existing <code>dqt</code> nodes contain a table
               with the same id, then the table is added to the last existing
               <code>dqt</code> node.
          <li> If there are no <code>dqt</code> nodes, then a new one is
               created and added as follows:
            <ul>
              <li> If there are <code>dht</code> nodes, the new
                   <code>dqt</code> node is inserted before the first one.  
              <li> If there are no <code>dht</code> nodes, the new
                   <code>dqt</code> node is inserted  before an
                   <code>sof</code> node, if there is one. 
              <li> If there is no <code>sof</code> node, the new
                   <code>dqt</code> node is inserted before the first
                   <code>sos</code> node, if there is one.
              <li> If there is no <code>sos</code> node, the new
                   <code>dqt</code> node is added to the end of the sequence.
            </ul>
        </ul>
  <li> <code>dht</code>
        <ul>
          <li> If there already exist <code>dht</code> nodes in the sequence,
               then each table in the node replaces the first table, in any
               <code>dht</code> node, with the same table class and table id.
          <li> If none of the existing <code>dht</code> nodes contain a table
               with the same class and id, then the table is added to the last
               existing <code>dht</code> node.
          <li> If there are no <code>dht</code> nodes, then a new one is
               created and added as follows:
            <ul>
              <li> If there are <code>dqt</code> nodes, the new
                   <code>dht</code> node is inserted immediately following the
                   last <code>dqt</code> node.
              <li> If there are no <code>dqt</code> nodes, the new
                   <code>dht</code> node is inserted before an
                   <code>sof</code> node, if there is one. 
              <li> If there is no <code>sof</code> node, the new
                   <code>dht</code> node is inserted before the first
                   <code>sos</code> node, if there is one.
              <li> If there is no <code>sos</code> node, the new
                   <code>dht</code> node is added to the end of the sequence.
            </ul>
        </ul>
  <li> <code>dri</code>
        <ul>
          <li> If there already exists a <code>dri</code> node, the restart 
               interval value is updated.
          <li> If there is no <code>dri</code> node, then a new one is created
               and added as follows:
            <ul>
              <li> If there is an <code>sof</code> node, the new
                   <code>dri</code> node is inserted before it.
              <li> If there is no <code>sof</code> node, the new
                   <code>dri</code> node is inserted before the first
                   <code>sos</code> node, if there is one.
              <li> If there is no <code>sos</code> node, the new
                   <code>dri</code> node is added to the end of the sequence.
            </ul>
        </ul>
  <li> <code>com</code>
        <br> A new <code>com</code> node is created and inserted as follows:
        <ul>
          <li> If there already exist <code>com</code> nodes, the new one is
               inserted after the last one.
          <li> If there are no <code>com</code> nodes, the new
               <code>com</code> node is inserted after the
               <code>app14Adobe</code> node, if there is one.
          <li> If there is no <code>app14Adobe</code> node, the new
               <code>com</code> node is inserted at the beginning of the
               sequence.
        </ul>
  <li> <code>app14Adobe</code>
        <ul>
          <li> If there already exists an <code>app14Adobe</code> node, then 
               its attributes are updated from the node.
          <li> If there is no <code>app14Adobe</code> node, then a new one is 
               created and  added as follows:
            <ul> 
              <li> The new <code>app14Adobe</code> node is inserted after the
                   last <code>unknown</code> node, if there are any.
              <li> If there are no <code>unknown</code> nodes, the new
                   <code>app14Adobe</code> node is inserted at the beginning
                   of the sequence.
            </ul>
        </ul>
  <li> <code>unknown</code>
       <br> A new <code>unknown</code> node is created and added to the
            sequence as follows:
        <ul>
          <li> If there already exist <code>unknown</code> marker nodes, the
               new one is inserted after the last one.
          <li> If there are no <code>unknown</code> nodes, the new
               <code>unknown</code> node is inserted before the
               <code>app14Adobe</code> node, if there is one.
          <li> If there is no <code>app14Adobe</code> node, the new
               <code>unknown</code> node is inserted at the beginning of the
               sequence.
        </ul>
  <li> <code>sof</code>
        <ul>
          <li> If there already exists an <code>sof</code> node in the
               sequence, then its values are updated from the node.
          <li> If there is no <code>sof</code> node, then a new one is created
               and added as follows:
            <ul>
              <li> If there are any <code>sos</code> nodes, the new
                   <code>sof</code> node is inserted before the first one.
              <li> If there is no <code>sos</code> node, the new
                   <code>sof</code> node is added to the end of the sequence.
            </ul>
        </ul>
  <li> <code>sos</code>
        <ul>
          <li> If there already exists a single <code>sos</code> node, then
               the values are updated from the node.
          <li> If there are more than one existing <code>sos</code> nodes,
               then an <code>IIOInvalidTreeException</code> is thrown, as
               <code>sos</code> nodes cannot be merged into a set of
               progressive scans.
          <li> If there are no <code>sos</code> nodes, a new one is created
               and added to the end of the sequence.
        </ul>
</ul>

<p> <code>mergeTree</code> - Standard Format
<br>
The <code>mergeTree</code> operation, when given a tree in the standard 
format, will modify the native tree in the following ways:
<ul>
   <li> <code>Chroma</code> - The <code>ColorSpaceType</code> subnode of a
        <code>Chroma</code> node may change the target colorspace of the
        compressed image.  The selection of a new colorspace can cause a number
        of changes, in keeping with the algorithms described above: 
        <code>app0JFIF</code> and <code>app14Adobe</code> nodes may be added
        or removed, subsampling may be added or removed, component ids may
        be changed, and <code>sof</code> and <code>sos</code> nodes will be
        updated accordingly.  If necessary, additional quantization and
        huffman tables are added.  In the case of quantization tables, the
        default will be scaled to match the quality level of any existing
        tables.  No tables are added to metadata that does not already contain
        tables.  If the existing metadata specifies progressive encoding, then
        the number of channels must not change.  Any <code>Transparency</code>
        node is also taken into account, as an explicit value of
        <code>none</code> for the <code>Alpha</code> subnode can cause the
        removal of an alpha channel, and anything other than <code>none</code>
        can cause the addition of an alpha channel.
   <li> <code>Dimension</code> - A <code>PixelAspectRatio</code> specification
        can cause the contents of an <code>app0JFIF</code> node to change, if
        there is one present, or the addition of an <code>app0JFIF</code> node
        containing appropriate values, if there can be one.  An appropriate
        pair of integers is computed from the floating-point ratio for
        inclusion in the node.
   <li> <code>Text</code> - Each uncompressed text item is converted to a
        <code>com</code> node and inserted according to the rules above for
        merging <code>com</code> nodes.
</ul>

<p> <code>setFromTree</code> - Native Format
<br>
The <code>setFromTree</code> operation, when given a tree in the native 
format described below, will simply replace the existing tree in its entirety 
with the new one.  The tree must consist of <code>IIOMetadataNode</code>s.

<p> <code>setFromTree</code> - Standard Format
<br>
The <code>setFromTree</code> operation, when given a tree in the standard 
format, performs a <code>reset</code> followed by a merge of the new tree.

<h2>
<a name=image>Image Metadata DTD</a>
</h2>

<pre>
&lt;!DOCTYPE "javax_imageio_jpeg_image_1.0" [

  &lt;!ELEMENT "javax_imageio_jpeg_image_1.0" (JPEGvariety, markerSequence)&gt;

    &lt;!ELEMENT "JPEGvariety" (app0JFIF)&gt;
      &lt;!-- A node grouping all marker segments specific to the variety of
              stream being read/written (e.g. JFIF) - may be empty --&gt; 

      &lt;!ELEMENT "app0JFIF" (JFXX?, app2ICC?)&gt;
        &lt;!ATTLIST "app0JFIF" "majorVersion" #CDATA "1"&gt;
          &lt;!-- The major JFIF version number --&gt; 
          &lt;!-- Data type: Integer --&gt;
          &lt;!-- Min value: 0 (inclusive) --&gt;
          &lt;!-- Max value: 255 (inclusive) --&gt;
        &lt;!ATTLIST "app0JFIF" "minorVersion" #CDATA "2"&gt;
          &lt;!-- The minor JFIF version number --&gt; 
          &lt;!-- Data type: Integer --&gt;
          &lt;!-- Min value: 0 (inclusive) --&gt;
          &lt;!-- Max value: 255 (inclusive) --&gt;
        &lt;!ATTLIST "app0JFIF" "resUnits" ("0" | "1" | "2") "0"&gt;
          &lt;!-- The resolution units for Xdensisty and Ydensity (0 = no 
               units, just aspect ratio; 1 = dots/inch; 2 = dots/cm) --&gt; 
        &lt;!ATTLIST "app0JFIF" "Xdensity" #CDATA "1"&gt;
          &lt;!-- The horizontal density or aspect ratio numerator --&gt; 
          &lt;!-- Data type: Integer --&gt;
          &lt;!-- Min value: 1 (inclusive) --&gt;
          &lt;!-- Max value: 65535 (inclusive) --&gt;
        &lt;!ATTLIST "app0JFIF" "Ydensity" #CDATA "1"&gt;
          &lt;!-- The vertical density or aspect ratio denominator --&gt; 
          &lt;!-- Data type: Integer --&gt;
          &lt;!-- Min value: 1 (inclusive) --&gt;
          &lt;!-- Max value: 65535 (inclusive) --&gt;
        &lt;!ATTLIST "app0JFIF" "thumbWidth" #CDATA "0"&gt;
          &lt;!-- The width of the thumbnail, or 0 if there isn't one --&gt; 
          &lt;!-- Data type: Integer --&gt;
          &lt;!-- Min value: 0 (inclusive) --&gt;
          &lt;!-- Max value: 255 (inclusive) --&gt;
        &lt;!ATTLIST "app0JFIF" "thumbHeight" #CDATA "0"&gt;
          &lt;!-- The height of the thumbnail, or 0 if there isn't one --&gt; 
          &lt;!-- Data type: Integer --&gt;
          &lt;!-- Min value: 0 (inclusive) --&gt;
          &lt;!-- Max value: 255 (inclusive) --&gt;

        &lt;!ELEMENT "JFXX" (app0JFXX)*&gt;
          &lt;!-- Min children: 1 --&gt;

        &lt;!ELEMENT "app0JFXX" (JFIFthumbJPEG | JFIFthumbPalette | 
          JFIFthumbRGB)&gt;
          &lt;!-- A JFIF extension marker segment --&gt; 
          &lt;!ATTLIST "app0JFXX" "extensionCode" ("16" | "17" | "19")
             #IMPLIED&gt;
            &lt;!-- The JFXX extension code identifying thumbnail type: (16 = 
                 JPEG, 17 = indexed, 19 = RGB --&gt; 

          &lt;!ELEMENT "JFIFthumbJPEG" (markerSequence?)&gt;
            &lt;!-- A JFIF thumbnail in JPEG format (no JFIF segments 
                 permitted) --&gt; 

          &lt;!ELEMENT "JFIFthumbPalette" EMPTY&gt;
            &lt;!-- A JFIF thumbnail as an RGB indexed image --&gt; 
            &lt;!ATTLIST "JFIFthumbPalette" "thumbWidth" #CDATA #IMPLIED&gt;
              &lt;!-- The width of the thumbnail --&gt; 
              &lt;!-- Data type: Integer --&gt;
              &lt;!-- Min value: 0 (inclusive) --&gt;
              &lt;!-- Max value: 255 (inclusive) --&gt;
            &lt;!ATTLIST "JFIFthumbPalette" "thumbHeight" #CDATA #IMPLIED&gt;
              &lt;!-- The height of the thumbnail --&gt; 
              &lt;!-- Data type: Integer --&gt;
              &lt;!-- Min value: 0 (inclusive) --&gt;
              &lt;!-- Max value: 255 (inclusive) --&gt;

          &lt;!ELEMENT "JFIFthumbRGB" EMPTY&gt;
            &lt;!-- A JFIF thumbnail as an RGB image --&gt; 
            &lt;!ATTLIST "JFIFthumbRGB" "thumbWidth" #CDATA #IMPLIED&gt;
              &lt;!-- The width of the thumbnail --&gt; 
              &lt;!-- Data type: Integer --&gt;
              &lt;!-- Min value: 0 (inclusive) --&gt;
              &lt;!-- Max value: 255 (inclusive) --&gt;
            &lt;!ATTLIST "JFIFthumbRGB" "thumbHeight" #CDATA #IMPLIED&gt;
              &lt;!-- The height of the thumbnail --&gt; 
              &lt;!-- Data type: Integer --&gt;
              &lt;!-- Min value: 0 (inclusive) --&gt;
              &lt;!-- Max value: 255 (inclusive) --&gt;

        &lt;!ELEMENT "app2ICC" EMPTY&gt;
          &lt;!-- An ICC profile APP2 marker segment --&gt; 
          &lt;!-- Optional User object: java.awt.color.ICC_Profile --&gt;

    &lt;!ELEMENT "markerSequence" (dqt | dht | dri | com | unknown | 
      app14Adobe | sof | sos)*&gt;
      &lt;!-- A node grouping all non-jfif marker segments --&gt; 

      &lt;!ELEMENT "dqt" (dqtable)*&gt;
        &lt;!-- A Define Quantization Table(s) marker segment --&gt; 
        &lt;!-- Min children: 1 --&gt;
        &lt;!-- Max children: 4 --&gt;

        &lt;!ELEMENT "dqtable" EMPTY&gt;
          &lt;!-- A single quantization table --&gt; 
          &lt;!-- User object: javax.imageio.plugins.jpeg.JPEGQTable --&gt;
          &lt;!ATTLIST "dqtable" "elementPrecision" #CDATA "0"&gt;
            &lt;!-- The number of bits in each table element (0 = 8, 1 = 16) 
                 --&gt; 
            &lt;!-- Data type: Integer --&gt;
          &lt;!ATTLIST "dqtable" "qtableId" ("0" | "1" | "2" | "3") #REQUIRED&gt;

      &lt;!ELEMENT "dht" (dhtable)*&gt;
        &lt;!-- A Define Huffman Table(s) marker segment --&gt; 
        &lt;!-- Min children: 1 --&gt;
        &lt;!-- Max children: 4 --&gt;

        &lt;!ELEMENT "dhtable" EMPTY&gt;
          &lt;!-- A single Huffman table --&gt; 
          &lt;!-- User object: javax.imageio.plugins.jpeg.JPEGHuffmanTable --&gt;
          &lt;!ATTLIST "dhtable" "class" ("0" | "1") #REQUIRED&gt;
            &lt;!-- Indicates whether this is a DC (0) or an AC (1) table --&gt; 
          &lt;!ATTLIST "dhtable" "htableId" ("0" | "1" | "2" | "3") #REQUIRED&gt;
            &lt;!-- The table id --&gt; 

      &lt;!ELEMENT "dri" EMPTY&gt;
        &lt;!-- A Define Restart Interval marker segment --&gt; 
        &lt;!ATTLIST "dri" "interval" #CDATA #REQUIRED&gt;
          &lt;!-- The restart interval in MCUs --&gt; 
          &lt;!-- Data type: Integer --&gt;
          &lt;!-- Min value: 0 (inclusive) --&gt;
          &lt;!-- Max value: 65535 (inclusive) --&gt;

      &lt;!ELEMENT "com" EMPTY&gt;
        &lt;!-- A Comment marker segment. The user object contains the actual 
             bytes. --&gt; 
        &lt;!-- User object: array of [B --&gt;
        &lt;!-- Min length: 1 --&gt;
        &lt;!-- Max length: 65533 --&gt;
        &lt;!ATTLIST "com" "comment" #CDATA #IMPLIED&gt;
          &lt;!-- The comment as a string (used only if user object is null) 
               --&gt; 
          &lt;!-- Data type: String --&gt;

      &lt;!ELEMENT "unknown" EMPTY&gt;
        &lt;!-- An unrecognized marker segment. The user object contains the 
             data not including length. --&gt; 
        &lt;!-- User object: array of [B --&gt;
        &lt;!-- Min length: 1 --&gt;
        &lt;!-- Max length: 65533 --&gt;
        &lt;!ATTLIST "unknown" "MarkerTag" #CDATA #REQUIRED&gt;
          &lt;!-- The tag identifying this marker segment --&gt; 
          &lt;!-- Data type: Integer --&gt;
          &lt;!-- Min value: 0 (inclusive) --&gt;
          &lt;!-- Max value: 255 (inclusive) --&gt;

      &lt;!ELEMENT "app14Adobe" EMPTY&gt;
        &lt;!-- An Adobe APP14 marker segment --&gt; 
        &lt;!ATTLIST "app14Adobe" "version" #CDATA "100"&gt;
          &lt;!-- The version of Adobe APP14 marker segment --&gt; 
          &lt;!-- Data type: Integer --&gt;
          &lt;!-- Min value: 100 (inclusive) --&gt;
          &lt;!-- Max value: 255 (inclusive) --&gt;
        &lt;!ATTLIST "app14Adobe" "flags0" #CDATA "0"&gt;
          &lt;!-- The flags0 variable of an APP14 marker segment --&gt; 
          &lt;!-- Data type: Integer --&gt;
          &lt;!-- Min value: 0 (inclusive) --&gt;
          &lt;!-- Max value: 65535 (inclusive) --&gt;
        &lt;!ATTLIST "app14Adobe" "flags1" #CDATA "0"&gt;
          &lt;!-- The flags1 variable of an APP14 marker segment --&gt; 
          &lt;!-- Data type: Integer --&gt;
          &lt;!-- Min value: 0 (inclusive) --&gt;
          &lt;!-- Max value: 65535 (inclusive) --&gt;
        &lt;!ATTLIST "app14Adobe" "transform" ("0" | "1" | "2") #REQUIRED&gt;
          &lt;!-- The color transform applied to the image (0 = Unknown, 1 = 
               YCbCr, 2 = YCCK) --&gt; 

      &lt;!ELEMENT "sof" (componentSpec)*&gt;
        &lt;!-- A Start Of Frame marker segment --&gt; 
        &lt;!-- Min children: 1 --&gt;
        &lt;!-- Max children: 4 --&gt;
        &lt;!ATTLIST "sof" "process" ("0" | "1" | "2") #IMPLIED&gt;
          &lt;!-- The JPEG process (0 = Baseline sequential, 1 = Extended 
               sequential, 2 = Progressive) --&gt; 
        &lt;!ATTLIST "sof" "samplePrecision" #CDATA "8"&gt;
          &lt;!-- The number of bits per sample --&gt; 
          &lt;!-- Data type: Integer --&gt;
        &lt;!ATTLIST "sof" "numLines" #CDATA #IMPLIED&gt;
          &lt;!-- The number of lines in the image --&gt; 
          &lt;!-- Data type: Integer --&gt;
          &lt;!-- Min value: 0 (inclusive) --&gt;
          &lt;!-- Max value: 65535 (inclusive) --&gt;
        &lt;!ATTLIST "sof" "samplesPerLine" #CDATA #IMPLIED&gt;
          &lt;!-- The number of samples per line --&gt; 
          &lt;!-- Data type: Integer --&gt;
          &lt;!-- Min value: 0 (inclusive) --&gt;
          &lt;!-- Max value: 65535 (inclusive) --&gt;
        &lt;!ATTLIST "sof" "numFrameComponents" ("1" | "2" | "3" | "4")
           #IMPLIED&gt;
          &lt;!-- The number of components in the image --&gt; 

        &lt;!ELEMENT "componentSpec" EMPTY&gt;
          &lt;!-- A component specification for a frame --&gt; 
          &lt;!ATTLIST "componentSpec" "componentId" #CDATA #REQUIRED&gt;
            &lt;!-- The id for this component --&gt; 
            &lt;!-- Data type: Integer --&gt;
            &lt;!-- Min value: 0 (inclusive) --&gt;
            &lt;!-- Max value: 255 (inclusive) --&gt;
          &lt;!ATTLIST "componentSpec" "HsamplingFactor" #CDATA #REQUIRED&gt;
            &lt;!-- The horizontal sampling factor for this component --&gt; 
            &lt;!-- Data type: Integer --&gt;
            &lt;!-- Min value: 1 (inclusive) --&gt;
            &lt;!-- Max value: 255 (inclusive) --&gt;
          &lt;!ATTLIST "componentSpec" "VsamplingFactor" #CDATA #REQUIRED&gt;
            &lt;!-- The vertical sampling factor for this component --&gt; 
            &lt;!-- Data type: Integer --&gt;
            &lt;!-- Min value: 1 (inclusive) --&gt;
            &lt;!-- Max value: 255 (inclusive) --&gt;
          &lt;!ATTLIST "componentSpec" "QtableSelector" ("0" | "1" | "2" | 
            "3") #REQUIRED&gt;
            &lt;!-- The quantization table to use for this component --&gt; 

      &lt;!ELEMENT "sos" (scanComponentSpec)*&gt;
        &lt;!-- A Start Of Scan marker segment --&gt; 
        &lt;!-- Min children: 1 --&gt;
        &lt;!-- Max children: 4 --&gt;
        &lt;!ATTLIST "sos" "numScanComponents" ("1" | "2" | "3" | "4")
           #REQUIRED&gt;
          &lt;!-- The number of components in the scan --&gt; 
        &lt;!ATTLIST "sos" "startSpectralSelection" #CDATA "0"&gt;
          &lt;!-- The first spectral band included in this scan --&gt; 
          &lt;!-- Data type: Integer --&gt;
          &lt;!-- Min value: 0 (inclusive) --&gt;
          &lt;!-- Max value: 63 (inclusive) --&gt;
        &lt;!ATTLIST "sos" "endSpectralSelection" #CDATA "63"&gt;
          &lt;!-- The last spectral band included in this scan --&gt; 
          &lt;!-- Data type: Integer --&gt;
          &lt;!-- Min value: 0 (inclusive) --&gt;
          &lt;!-- Max value: 63 (inclusive) --&gt;
        &lt;!ATTLIST "sos" "approxHigh" #CDATA "0"&gt;
          &lt;!-- The highest bit position included in this scan --&gt; 
          &lt;!-- Data type: Integer --&gt;
          &lt;!-- Min value: 0 (inclusive) --&gt;
          &lt;!-- Max value: 15 (inclusive) --&gt;
        &lt;!ATTLIST "sos" "approxLow" #CDATA "0"&gt;
          &lt;!-- The lowest bit position included in this scan --&gt; 
          &lt;!-- Data type: Integer --&gt;
          &lt;!-- Min value: 0 (inclusive) --&gt;
          &lt;!-- Max value: 15 (inclusive) --&gt;

        &lt;!ELEMENT "scanComponentSpec" EMPTY&gt;
          &lt;!-- A component specification for a scan --&gt; 
          &lt;!ATTLIST "scanComponentSpec" "componentSelector" #CDATA
             #REQUIRED&gt;
            &lt;!-- The id of this component --&gt; 
            &lt;!-- Data type: Integer --&gt;
            &lt;!-- Min value: 0 (inclusive) --&gt;
            &lt;!-- Max value: 255 (inclusive) --&gt;
          &lt;!ATTLIST "scanComponentSpec" "dcHuffTable" ("0" | "1" | "2" | 
            "3") #REQUIRED&gt;
            &lt;!-- The huffman table to use for encoding DC coefficients --&gt; 
          &lt;!ATTLIST "scanComponentSpec" "acHuffTable" ("0" | "1" | "2" | 
            "3") #REQUIRED&gt;
            &lt;!-- The huffman table to use for encoding AC coefficients --&gt; 
]&gt;
</pre>

<h2>
<a name=stream>Stream Metadata DTD</a>
</h2>

<pre>
&lt;!DOCTYPE "javax_imageio_jpeg_stream_1.0" [
  &lt;!ELEMENT "javax_imageio_jpeg_stream_1.0" (dqt |
                      dht | 
                      dri | 
                      com | 
                      unknown)*&gt;
   
  &lt;!-- All elements are as defined above for image metadata --&gt;
]&gt;
</pre>

</body>
</html>