path: root/shared-mime-info-spec.xml

<?xml version="1.0" standalone="no"?>
<!DOCTYPE article PUBLIC "-//OASIS//DTD DocBook XML V4.1.2//EN"
"/usr/share/sgml/docbook/dtd/xml/4.1.2/docbookx.dtd" [
  <!ENTITY updated "1 December 2005">
  <!ENTITY version "0.15">
]>
<article id="index">

<articleinfo>
	<authorgroup>
		<corpauthor>
			<ulink url="http://www.freedesktop.org">
			X Desktop Group
			</ulink>
		</corpauthor>
		<author>
			<firstname>Thomas</firstname>
			<surname>Leonard</surname>
			<affiliation>
				<address><email>tal197 at users.sf.net</email></address>
			</affiliation>
		</author>
	</authorgroup>

	<title>Shared MIME-info Database</title>
	<date>&updated;</date>
</articleinfo>

<sect1>
	<title>Introduction</title>
	<sect2>
		<title>Version</title>
		<para>
This is version &version; of the Shared MIME-info Database specification, last updated &updated;.</para>
	</sect2>
	<sect2>
		<title>What is this spec?</title>
		<para>
Many programs and desktops use the MIME system<citation>MIME</citation>
to represent the types of files. Frequently, it is necessary to work out the
correct MIME type for a file. This is generally done by examining the file's
name or contents, and looking up the correct MIME type in a database.
		</para>
		<para>
It is also useful to store information about each type, such as a textual
description of it, or a list of applications that can be used to view or edit
files of that type.
		</para>
		<para>
For interoperability, it is useful for different programs to use the same
database so that different programs agree on the type of a file and
information is not duplicated. It is also helpful for application authors to
only have to install new information in one place.
		</para>
		<para>
This specification attempts to unify the MIME database systems currently in
use by GNOME<citation>GNOME</citation>, KDE<citation>KDE</citation> and
ROX<citation>ROX</citation>, and provide room for future extensibility.
		</para>
		<para>
The MIME database does NOT store user preferences (such as a user's preferred
application for handling files of a particular type). It may be used to store
static information, such as that files of a certain type may be viewed with
a particular application.
		</para>
	</sect2>
	<sect2>
		<title>Language used in this specification</title>
		<para>
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD",
"SHOULD NOT", "RECOMMENDED",  "MAY", and "OPTIONAL" in this document are to be
interpreted as described in RFC 2119<citation>RFC-2119</citation>.
		</para>
	</sect2>
</sect1>

<sect1>
	<title>Unified system</title>
	<para>
In discussions about the previous systems used by GNOME, KDE and ROX (see the
"History and related systems" document), it was clear that the differences
between the databases were simply a result of them being separate, and not due
to any fundamental disagreements between developers. Everyone is keen to see
them merged.
	</para>
	<para>
This specification proposes:

		<itemizedlist>
			<listitem><para>
A standard way for applications to install new MIME related information.
			</para></listitem>
			<listitem><para>
A standard way of getting the MIME type for a file.
			</para></listitem>
			<listitem><para>
A standard way of getting information about a MIME type.
			</para></listitem>
			<listitem><para>
Standard locations for all the files, and methods of resolving conflicts.
			</para></listitem>
		</itemizedlist>
Further, the existing databases have been merged into a single package
<citation>SharedMIME</citation>.
	</para>
	<sect2 id="s2_layout">
		<title>Directory layout</title>
		<para>
There are two important requirements for the way the MIME database is stored:
			<itemizedlist>
				<listitem><para>
Applications must be able to extend the database in any way when they are installed,
to add both new rules for determining type, and new information about specific types.
				</para></listitem>
				<listitem><para>
It must be possible to install applications in /usr, /usr/local and the user's home directory
(in the normal Unix way) and have the MIME information used.
				</para></listitem>
			</itemizedlist>
		</para>
		<para>
This specification uses the XDG Base Directory Specification<citation>BaseDir</citation> to
define the prefixes below which the database is stored.
In the rest of this document, paths shown with the prefix
<filename>&lt;MIME&gt;</filename> indicate the files should be loaded from
the <filename>mime</filename> subdirectory of every directory in
<envar>XDG_DATA_HOME</envar>:<envar>XDG_DATA_DIRS</envar>.
		</para>
		<para>
For example, when using the default paths, <quote>Load all the
<filename>&lt;MIME&gt;/text/html.xml</filename> files</quote> means to
load <filename>/usr/share/mime/text/html.xml</filename>,
<filename>/usr/local/share/mime/text/html.xml</filename>, and
<filename>~/.local/share/mime/text/html.xml</filename> (if they exist).
		</para>
		<para>
Each application that wishes to contribute to the MIME database will install a
single XML file, named after the application, into one of the three
<filename>&lt;MIME&gt;/packages/</filename> directories (depending on where the user requested
the application be installed). After installing, uninstalling or modifying this
file, the application MUST run the <command>update-mime-database</command> command,
which is provided by the freedesktop.org shared database<citation>SharedMIME</citation>.
		</para>
		<para>
<command>update-mime-database</command> is passed the <filename>mime</filename>
directory containing the <filename>packages</filename> subdirectory which was
modified as its only argument. It scans all the XML files in the <filename>packages</filename>
subdirectory, combines the information in them, and creates a number of output files.
		</para>
		<para>
Where the information from these files is conflicting, information from directories
lower in the list takes precedence.
Any file named <filename>Override.xml</filename> takes precedence over all other files in
the same <filename>packages</filename> directory. This can be used by
tools which let the user edit the database to ensure that the user's
changes take effect.
		</para>
		<para>
The files created by <command>update-mime-database</command> are:
			<itemizedlist>
				<listitem><para>
<filename>&lt;MIME&gt;/globs</filename> (contains a mapping from names to MIME types)
				</para></listitem>
				<listitem><para>
<filename>&lt;MIME&gt;/magic</filename> (contains a mapping from file contents to MIME types)
				</para></listitem>
				<listitem><para>
<filename>&lt;MIME&gt;/subclasses</filename> (contains a mapping from MIME types to types they inherit from)
				</para></listitem>
				<listitem><para>
<filename>&lt;MIME&gt;/aliases</filename> (contains a mapping from aliases to MIME types)
				</para></listitem>
				<listitem><para>
<filename>&lt;MIME&gt;/XMLnamespaces</filename> (contains a mapping from XML
(namespaceURI, localName) pairs to MIME types)
				</para></listitem>
 				<listitem><para>
<filename>&lt;MIME&gt;/MEDIA/SUBTYPE.xml</filename> (one file for each MIME
type, giving details about the type)
				</para></listitem>
				<listitem><para>
<filename>&lt;MIME&gt;/mime.cache</filename> (contains the same information as the <filename>globs</filename>,
<filename>magic</filename>, <filename>subclasses</filename>, <filename>aliases</filename> and 
<filename>XMLnamespaces</filename> files, in a binary, mmappable format)
				</para></listitem>
			</itemizedlist>
The format of these generated files and the source files in <filename>packages</filename>
are explained in the following sections. This step serves several purposes. First, it allows
applications to quickly get the data they need without parsing all the source XML files (the
base package alone is over 700K). Second, it allows the database to be used for other
purposes (such as creating the <filename>/etc/mime.types</filename> file if
desired). Third, it allows validation to be performed on the input data,
and removes the need for other applications to carefully check the input for
errors themselves.
		</para>
	</sect2>
	<sect2>
		<title>The source XML files</title>
		<para>
Each application provides only a single XML source file, which is installed in the
<filename>packages</filename> directory as described above. This file is an XML file
whose document element is named <userinput>mime-info</userinput> and whose namespace URI
is <ulink url="http://www.freedesktop.org/standards/shared-mime-info"/>. All elements
described in this specification MUST have this namespace too.
		</para><para>
The document element may contain zero or more <userinput>mime-type</userinput> child nodes,
in any order, each describing a single MIME type. Each element has a <userinput>type</userinput>
attribute giving the MIME type that it describes.
		</para><para>
Each <userinput>mime-type</userinput> node may contain any combination of the following elements,
and in any order:
			<itemizedlist>
				<listitem><para>
<userinput>glob</userinput> elements have a <userinput>pattern</userinput> attribute. Any file
whose name matches this pattern will be given this MIME type (subject to conflicting rules in
other files, of course).
		</para>
		<para>
KDE's glob system replaces GNOME's and ROX's ext/regex fields, since it
is trivial to detect a pattern in the form '*.ext' and store it in an
extension hash table internally. The full power of regular expressions was
not being used by either desktop, and glob patterns are more suitable for
filename matching anyway.
				</para></listitem>
				<listitem><para>
<userinput>magic</userinput> elements contain a list of
<userinput>match</userinput> elements, any of which may match, and an optional
<userinput>priority</userinput> attribute for all of the contained rules. Low
numbers should be used for more generic types (such as 'gzip compressed data')
and higher values for specific subtypes (such as a word processor format that
happens to use gzip to compress the file). The default priority value is 50, and
the maximum is 100.
				</para><para>
Each <userinput>match</userinput> element has a number of attributes:

<informaltable>
	<tgroup cols="3">
	<thead><row><entry>Attribute</entry><entry>Required?</entry><entry>Value</entry></row></thead>
	<tbody>

	<row><entry>type</entry><entry>Yes</entry><entry>
<userinput>string</userinput>, <userinput>host16</userinput>,
<userinput>host32</userinput>, <userinput>big16</userinput>,
<userinput>big32</userinput>, <userinput>little16</userinput>,
<userinput>little32</userinput> or <userinput>byte</userinput>.
	</entry></row>

	<row><entry>offset</entry><entry>Yes</entry><entry>The byte offset(s)
	in the file to check. This may be a single number or a range in the
	form `start:end', indicating that all offsets in the range should be
	checked. The range is inclusive.</entry></row>

	<row><entry>value</entry><entry>Yes</entry><entry>
	The value to compare the file contents with, in the format indicated by the type
	attribute.
	</entry></row>

	<row><entry>mask</entry><entry>No</entry><entry>
	The number to AND the value in the file with before comparing it to
	`value'. Masks for numerical types can be any number, while masks for strings
	must be in base 16, and start with 0x.
	</entry></row>

	</tbody></tgroup>
</informaltable>

Each element corresponds to one line of
<citerefentry><refentrytitle>file</refentrytitle>
<manvolnum>1</manvolnum></citerefentry>'s <filename>magic.mime</filename> file.
They can be nested in the same way to provide the equivalent of continuation
lines. That is, <![CDATA[<a><b/><c/></a>]]> means 'a and (b or c)'.
				</para></listitem>
				<listitem><para>
<userinput>alias</userinput> elements indicate that the type is also sometimes
known by another name, given by the <userinput>type</userinput> attribute. For
example, <userinput>audio/midi</userinput> has an alias of
<userinput>audio/x-midi</userinput>. Note that there should not be a
<userinput>mime-type</userinput> element defining each alias; a single
element defines the canonical name for the type and lists all its aliases.
				</para></listitem>
				<listitem><para>
<userinput>sub-class-of</userinput> elements indicate that any data of this
type is also some other type, given by the <userinput>type</userinput>
attribute. See <xref linkend="subclassing"/>.
				</para></listitem>
				<listitem><para>
<userinput>comment</userinput> elements give a human-readable textual description of the MIME
type, usually composed of an acronym of the file name extension and a short description, like
"ODS spreadsheet".
There may be many of these elements with different <userinput>xml:lang</userinput> attributes
to provide the text in multiple languages.
				</para></listitem>
				<listitem><para>
<userinput>acronym</userinput> elements give experienced users a terse idea of the document contents.
for example "ODS", "GEDCOM", "JPEG" and "XML".
There may be many of these elements with different <userinput>xml:lang</userinput> attributes
to provide the text in multiple languages, although these should only be used if absolutely neccessary.
				</para></listitem>
				<listitem><para>
<userinput>expanded-acronym</userinput> elements are the expanded versions of the acronym elements,
for example "OpenDocument Spreadsheet", "GEnealogical Data COMmunication", and "eXtensible Markup Language".
The purpose of these elements is to provide users a way to look up information on various MIME types or
file formats in third-party resources.
There may be many of these elements with different <userinput>xml:lang</userinput> attributes
to provide the text in multiple languages, although these should only be used if absolutely neccessary.
				</para></listitem>
				<listitem><para>
<userinput>root-XML</userinput> elements have <userinput>namespaceURI</userinput> 
and <userinput>localName</userinput> attributes. If a file is identified as being an XML file,
these rules allow a more specific MIME type to be chosen based on the namespace and localname
of the document element.
				</para><para>
If <userinput>localName</userinput> is present but empty then the document element may have
any name, but the namespace must still match.
				</para></listitem>
			</itemizedlist>
Applications may also define their own elements, provided they are namespaced to prevent collisions.
Unknown elements are copied directly to the output XML files like <userinput>comment</userinput>
elements. A typical use for this would be to indicate the default handler
application for a particular desktop
("Galeon is the GNOME default text/html browser"). Note that this doesn't
indicate the user's preferred application, only the (fixed) default.
		</para>
		<para>
Here is an example source file, named <filename>diff.xml</filename>:
		<programlisting><![CDATA[
<?xml version="1.0"?>
<mime-info xmlns='http://www.freedesktop.org/standards/shared-mime-info'>
  <mime-type type="text/x-diff">
    <comment>Differences between files</comment>
    <comment xml:lang="af">verskille tussen lêers</comment>
    ...
    <magic priority="50">
      <match type="string" offset="0" value="diff\t"/>
      <match type="string" offset="0" value="***\t"/>
      <match type="string" offset="0" value="Common subdirectories: "/>
    </magic>
    <glob pattern="*.diff"/>
    <glob pattern="*.patch"/>
  </mime-type>
</mime-info>
]]></programlisting>
		</para><para>
In practice, common types such as text/x-diff are provided by the freedesktop.org shared
database. Also, only new information needs to be provided, since this information will be merged
with other information about the same type.
		</para>
	</sect2>
	<sect2>
		<title>The MEDIA/SUBTYPE.xml files</title>
		<para>
These files have a <userinput>mime-type</userinput> element as the root node. The format is
as described above. They are created by merging all the <userinput>mime-type</userinput>
elements from the source files and creating one output file per MIME type. Each file may contain
information from multiple source files. The <userinput>magic</userinput>,
<userinput>glob</userinput> and <userinput>root-XML</userinput> elements will
have been removed.
		</para>
		<para>
The example source file given above would (on its own) create an output file called
<filename>&lt;MIME&gt;/text/x-diff.xml</filename> containing the following:
			<programlisting><![CDATA[
<?xml version="1.0" encoding="utf-8"?>
<mime-type xmlns="http://www.freedesktop.org/standards/shared-mime-info" type="text/x-diff">
<!--Created automatically by update-mime-database. DO NOT EDIT!-->
  <comment>Differences between files</comment>
  <comment xml:lang="af">verskille tussen lêers</comment>
  ...
</mime-type>

]]></programlisting>
		</para>
	</sect2>
	<sect2>
		<title>The glob files</title>
		<para>
This is a simple list of lines containing a MIME type and pattern, separated by a colon.
For example:
			<programlisting><![CDATA[
# This file was automatically generated by the
# update-mime-database command. DO NOT EDIT!
...
text/x-diff:*.diff
text/x-diff:*.patch
...
]]></programlisting>
		</para>
		<para>
Applications MUST first try a case-sensitive match, then try again with the
filename converted to lower-case if that fails.
This is so that <filename>main.C</filename> will be seen as a C++ file,
but <filename>IMAGE.GIF</filename> will still use the *.gif pattern.
		</para>
		<para>
If several patterns match then the longest pattern SHOULD be used. In
particular, files with multiple extensions (such as
<filename>Data.tar.gz</filename>) MUST match the longest sequence of extensions
(eg '*.tar.gz' in preference to '*.gz'). Literal patterns (eg, 'Makefile') must
be matched before all others. It is suggested that patterns beginning with `*.'
and containing no other special characters (`*?[') should be placed in a hash
table for efficient lookup, since this covers the majority of the patterns. Thus,
patterns of this form should be matched before other wildcarded patterns.
		</para>
		<para>
If a matching pattern is provided by two or more MIME types, applications
SHOULD not rely on one of them. They are instead supposed to use magic data
(see below) to detect the actual MIME type. This is for instance required to
deal with container formats like Ogg or AVI, that map various video and/or
audio-encoded data to one extension.
		</para>
		<para>
There may be several rules mapping to the same type. They should all be merged.
If the same pattern is defined twice, then they MUST be ordered by the
directory the rule came from, as described above.
		</para>
		<para>
Lines beginning with `#' are comments and should be ignored. Everything from
the `:' character to the newline is part of the pattern; spaces should not be
stripped. The file is in the UTF-8 encoding. The format of the glob pattern
is as for fnmatch(3). The format does not allow a pattern to contain a literal
newline character, but this is not expected to be a problem.
		</para>
		<para>
Common types (such as MS Word Documents) will be provided in the X Desktop
Group's package, which MUST be required by all applications using this
specification. Since each application will then only be providing information
about its own types, conflicts should be rare.
		</para>
	</sect2>
	<sect2>
		<title>The magic files</title>
		<para>
The magic data is stored in a binary format for ease of parsing. The old magic database
had complex escaping rules; these are now handled by <command>update-mime-database</command>.
		</para><para>
The file starts with the magic string "MIME-Magic\0\n".
There is no version number in the file. Incompatible changes will be handled by
creating both the current `magic' file and a newer `magic2' in the new format.
Where possible, compatible changes only will be made.
All numbers are big-endian, so need to be byte-swapped on little-endian machines.
		</para><para>
The rest of the file is made up of a sequence of small sections.
Each section is introduced by giving the priority and type in brackets, followed by
a newline character. Higher priority entries come first. Example:
<screen>[50:text/x-diff]\n</screen>
Each line in the section takes the form:
<screen>[ indent ] ">" start-offset "=" value
[ "&amp;" mask ] [ "~" word-size ] [ "+" range-length ] "\n"</screen>
<informaltable>
	<tgroup cols="3">
	<thead><row><entry>Part</entry><entry>Example</entry><entry>Meaning</entry></row></thead>
	<tbody>

	<row><entry>indent</entry><entry>1</entry><entry>The nesting
	depth of the rule, corresponding to the number of '>' characters in the traditional file format.</entry></row>
	<row><entry>">" start-offset</entry><entry>&gt;4</entry><entry>The offset into the
	file to look for a match.</entry></row>
	<row><entry>"=" value</entry><entry>=\0x0\0x2\0x55\0x40</entry><entry>
	Two bytes giving the (big-endian) length of the value, followed by the value itself.
	</entry></row>
	<row><entry>"&amp;" mask</entry><entry>&amp;\0xff\0xf0</entry><entry>
	The mask, which (if present) is exactly the same length as the value.
	</entry></row>
	<row><entry>"~" word-size</entry><entry>~2</entry><entry>On little-endian machines, the
	size of each group to byte-swap.</entry></row>
	<row><entry>"+" range-length</entry><entry>+8</entry><entry>The length of the region
	in the file to check.
	</entry></row>
	</tbody>
	</tgroup>
</informaltable>
		</para><para>
Note that the value, value length and mask are all binary, whereas everything
else is textual. Each of the elements begins with a single character to
identify it, except for the indent level.
		</para><para>
The word size is used for byte-swapping. Little-endian systems should reverse
the order of groups of bytes in the value and mask if this is greater than one.
This only affects `host' matches (`big32' entries still have a word size of 1,
for example, because no swapping is necessary, whereas `host32' has a word size
of 4).
		</para><para>
The indent, range-length, word-size and mask components are optional. If
missing, indent defaults to 0, range-length to 1, the word-size to 1, and the
mask to all 'one' bits.
		</para><para>
Indent corresponds to the nesting depth of the rule. Top-level rules have an
indent of zero. The parent of an entry is the preceding entry with an indent
one less than the entry.
		</para><para>
If an unknown character is found where a newline is expected then the whole
line should be ignored (there will be no binary data after the new
character, so the next line starts after the next "\n" character). This is for
future extensions.
		</para><para>
The text/x-diff above example would (on its own) create this magic file:
			<programlisting><![CDATA[
00000000  4d 49 4d 45 2d 4d 61 67  69 63 00 0a 5b 35 30 3a  |MIME-Magic..[50:|
00000010  74 65 78 74 2f 78 2d 64  69 66 66 5d 0a 3e 30 3d  |text/x-diff].>0=|
00000020  00 05 64 69 66 66 09 0a  3e 30 3d 00 04 2a 2a 2a  |..diff..>0=..***|
00000030  09 0a 3e 30 3d 00 17 43  6f 6d 6d 6f 6e 20 73 75  |..>0=..Common su|
00000040  62 64 69 72 65 63 74 6f  72 69 65 73 3a 20 0a     |bdirectories: .|
]]></programlisting>
		</para>
	</sect2>
	<sect2>
		<title>The XMLnamespaces files</title>
		<para>
Each <filename>XMLnamespaces</filename> file is a list of lines in the form:
<screen>namespaceURI " " localName " " MIME-Type "\n"</screen>
For example:
<screen>
http://www.w3.org/1999/xhtml html application/xhtml+xml
</screen>
The lines are sorted (using strcmp) and there are no lines with the same namespaceURI and
localName in one file. If the localName was empty then there will be two spaces following
the namespaceURI.
		</para>
	</sect2>
	<sect2>
		<title>The mime.cache files</title>
		<para>
The <filename>mime.cache</filename> files contain the same information as the 
<filename>globs</filename>, <filename>magic</filename>, <filename>subclasses</filename>, 
<filename>aliases</filename> and <filename>XMLnamespaces</filename> files, in a binary, 
mmappable format:
</para>
<programlisting>
Header:
2			CARD16		MAJOR_VERSION	1	
2			CARD16		MINOR_VERSION	0	
4			CARD32		ALIAS_LIST_OFFSET
4			CARD32		PARENT_LIST_OFFSET
4			CARD32		LITERAL_LIST_OFFSET
4			CARD32		SUFFIX_LIST_OFFSET
4			CARD32		GLOB_LIST_OFFSET
4			CARD32		MAGIC_LIST_OFFSET
4			CARD32		NAMESPACE_LIST_OFFSET

AliasList:
4			CARD32		N_ALIASES
8*N_ALIASES		AliasListEntry

AliasListEntry:
4			CARD32		ALIAS_OFFSET
4			CARD32		MIME_TYPE_OFFSET

ParentList:
4			CARD32		N_ENTRIES 
8*N_ENTRIES		ParentListEntry

ParentListEntry:
4			CARD32		MIME_TYPE_OFFSET
4			CARD32		PARENTS_OFFSET

Parents:
4			CARD32		N_PARENTS
4*N_PARENTS		CARD32		MIME_TYPE_OFFSET

LiteralList:
4			CARD32		N_LITERALS
8*N_LITERALS		LiteralEntry	

LiteralEntry:
4			CARD32		LITERAL_OFFSET
4			CARD32		MIME_TYPE_OFFSET

GlobList:
4			CARD32		N_GLOBS
8*N_GLOBS		GlobEntry	

GlobEntry:
4			CARD32		GLOB_OFFSET
4			CARD32		MIME_TYPE_OFFSET

SuffixTree:
4			CARD32		N_ROOTS
4	 		CARD32		FIRST_ROOT_OFFSET

SuffixTreeNode:
4			CARD32		CHARACTER
4			CARD32		MIME_TYPE_OFFSET
4			CARD32		N_CHILDREN			
4			CARD32		FIRST_CHILD_OFFSET

MagicList:
4			CARD32		N_MATCHES
4			CARD32		MAX_EXTENT
4			CARD32		FIRST_MATCH_OFFSET

Match:
4			CARD32		PRIORITY
4			CARD32		MIME_TYPE_OFFSET
4			CARD32		N_MATCHLETS
4			CARD32		FIRST_MATCHLET_OFFSET

Matchlet:
4			CARD32		RANGE_START
4			CARD32		RANGE_LENGTH
4			CARD32		WORD_SIZE
4			CARD32		VALUE_LENGTH
4			CARD32		VALUE
4			CARD32		MASK
4			CARD32		N_CHILDREN
4			CARD32		FIRST_CHILD_OFFSET

NamespaceList:
4			CARD32		N_NAMESPACES
12*N_NAMESPACES		NamespaceEntry	

NamespaceEntry:
4			CARD32		NAMESPACE_URI_OFFSET
4			CARD32		LOCAL_NAME_OFFSET
4			CARD32		MIME_TYPE_OFFSET
</programlisting>
<para>
Lists in the file are sorted, to enable binary searching. The list of 
aliases is sorted by alias, the list of literal globs is sorted by the 
literal. The SuffixTreeNode siblings are sorted by character. 
The list of namespaces is sorted by namespace uri.
</para>
<para>
Identical globs are stored in the suffix tree by appending suffix
tree nodes with '\0' as character.
</para>
<para>
All offsets are in bytes from the beginning of the file.
</para>
<para>
Strings are zero-terminated.
</para>
<para>
All numbers are in network (big-endian) order. This is necessary because the data will be stored in 
arch-independent directories like <filename>/usr/share/mime</filename> or even in user's home directories.
</para>
<para>
Cache files have to be written atomically - write to a temporary name, then move over the old file - so 
that clients that have the old cache file open and mmap'ed won't get corrupt data.
</para>
	</sect2>
	<sect2>
		<title>Storing the MIME type using Extended Attributes</title>
		<para>
An implementation MAY also get a file's MIME type from the
<userinput>user.mime_type</userinput> extended attribute. <!-- The attr(5) man
page documents this name --> The type given here should normally be used in
preference to any guessed type, since the user is able to set it explicitly.
Applications MAY choose to set the type when saving files. Since many
applications and filesystems do not support extended attributes,
implementations MUST NOT rely on this method being available.
		</para>
	</sect2>
	<sect2 id="subclassing">
		<title>Subclassing</title>
		<para>
A type is a subclass of another type if any instance of the first type is
also an instance of the second. For example, all image/svg files are also
text/xml, text/plain and application/octet-stream files. Subclassing is about
the format, rather than the catagory of the data (for example, there is no
'generic spreadsheet' class that all spreadsheets inherit from).
		</para>
		<para>
Some subclass rules are implicit:
			<itemizedlist>
<listitem><para>All text/* types are subclasses of text/plain.</para></listitem>
<listitem><para>All streamable types (ie, everything except the inode/* types)
are subclasses of application/octet-stream.</para></listitem>
			</itemizedlist>
In addition to these rules, explicit subclass information may be given using
the <userinput>sub-class-of</userinput> element.
		</para>
		<para>
Note that some file formats are also compressed files (application/x-jar files
are also application/zip files). However, this is different to a case such as a
compressed postscript file, which is not a valid postscript file itself (so
application/x-gzpostscript does not inherit from application/postscript,
because an application that can handle the latter may not cope with the
former).
		</para>
		<para>
Some types may or may not be instances of other types. For example, a
spreadsheet file may be compressed or not. It is a valid spreadsheet file
either way, but only inherits from application/x-gzip in one case. This
information cannot be represented statically; instead an application
interested in this information should run all of the magic rules, and
use the list of types returned as the subclasses.
		</para>
	</sect2>
	<sect2>
		<title>Recommended checking order</title>
		<para>
Because different applications have different requirements, they may choose to
use the various methods provided by this specification in any order. However, the
RECOMMENDED order to perform the checks is:
			<itemizedlist>
				<listitem><para>
If a MIME type is provided explicitly (eg, by a ContentType HTTP header, a MIME
email attachment, an extended attribute or some other means) then that should
be used instead of guessing.
				</para></listitem>
				<listitem><para>
If no explicit type is present, magic rules with a priority of 80 or more
should be tried next. These rules have a very low false-positive rate.
				</para></listitem>
				<listitem><para>
If there is still no match, the glob rules should be applied to the name to
get the type.
				</para></listitem>
				<listitem><para>
If no glob rules match, the remaining magic rules should be tried next.
				</para></listitem>
				<listitem><para>
If nothing matches, the default type of application/octet-stream should be used
for binary data, or text/plain for textual data. Checking the first 32
bytes of the file for ASCII control characters is a good way to guess
whether a file is binary or text, but note that files with high-bit-set
characters should still be treated as text since these can appear in UTF-8
text, unlike control characters.
				</para></listitem>
			</itemizedlist>
		</para>
		<para>
There are several reasons for checking most of the glob patterns before the magic.
Some applications don't check the magic at all, and this makes it more likely
that both will get the same type. Users can easily understand why calling their
text file <filename>README.mp3</filename> makes the system think it's an MP3,
whereas they have trouble understanding why their computer thinks
<filename>README.txt</filename> is a PostScript file. If the system guesses wrongly,
the user can often rename the file to fix the problem.
		</para>
	</sect2>
	<sect2>
		<title>Non-regular files</title>
		<para>
Sometimes it is useful to assign MIME types to other objects in the filesystem,
such as directories, sockets and device files. This could be useful when looking up
an icon for a type, or for providing a textual description of one of these objects.
The media type 'inode' is provided for this purpose, with the following types corresponding
to the standard types of object found in a Unix filesystem:
		</para>
		<simplelist>
			<member>inode/blockdevice</member>
			<member>inode/chardevice</member>
			<member>inode/directory</member>
			<member>inode/fifo</member>
			<member>inode/mount-point</member>
			<member>inode/socket</member>
			<member>inode/symlink</member>
		</simplelist>
		<para>
An inode/mount-point is a subclass of inode/directory. It can be useful when adding extra
actions for these directories, such as 'mount' or 'eject'. Mounted directories can be
detected by comparing the 'st_dev' of a directory with that of its parent. If
they differ, they are from different devices and the directory is a mount
point.
		</para>
	</sect2>
	<sect2>
		<title>Security implications</title>
		<para>
The system described in this document is intended to allow different programs
to see the same file as having the same type. This is to help interoperability.
The type determined in this way is only a guess, and an application MUST NOT
trust a file based simply on its MIME type. For example, a downloader should
not pass a file directly to a launcher application without confirmation simply
because the type looks `harmless' (eg, text/plain).
		</para>
		<para>
Do not rely on two applications getting the same type for the same file, even
if they both use this system. The spec allows some leeway in implementation,
and in any case the programs may be following different versions of the spec.
		</para>
	</sect2>
	<sect2>
		<title>User modification</title>
		<para>
The MIME database is NOT intended to store user preferences. Users should never
edit the database. If they wish to make corrections or provide MIME entries for
software that doesn't provide these itself, they should do so by means of the
Override.xml mentioned in <xref linkend="s2_layout"/>. Information such as
"text/html files need to be opened with Mozilla" should NOT go in the database.
		</para>
	</sect2>
</sect1>

<sect1>
	<title>Contributors</title>
	<simplelist>
		<member>
			Thomas Leonard <email>tal197 at users.sf.net</email>
		</member>
		<member>
			David Faure <email>david at mandrakesoft.com</email>
		</member>
		<member>
			Alex Larsson <email>alexl at redhat.com</email>
		</member>
		<member>
			Seth Nickell <email>snickell at stanford.edu</email>
		</member>
		<member>
			Keith Packard <email>keithp at keithp.com</email>
		</member>
		<member>
			Filip Van Raemdonck <email>mechanix at debian.org</email>
		</member>
		<member>
			Christos Zoulas <email>christos at zoulas.com</email>
		</member>
		<member>
			Matthias Clasen <email>mclasen at redhat.com</email>
		</member>
	</simplelist>
</sect1>

<bibliography>
	<title>References</title>

	<bibliomixed>
		<abbrev>GNOME</abbrev><citetitle>The GNOME desktop,
		<ulink url="http://www.gnome.org"/></citetitle>
	</bibliomixed>
	<bibliomixed>
		<abbrev>KDE</abbrev><citetitle>The KDE desktop,
		<ulink url="http://www.kde.org"/></citetitle>
	</bibliomixed>
	<bibliomixed>
		<abbrev>ROX</abbrev><citetitle>The ROX desktop,
		<ulink url="http://rox.sourceforge.net"/></citetitle>
	</bibliomixed>
	<bibliomixed>
		<abbrev>DesktopEntries</abbrev><citetitle>Desktop Entry Specification,
		<ulink url="http://www.freedesktop.org/standards/desktop-entry-spec.html"/>
		</citetitle>
	</bibliomixed>
	<bibliomixed>
		<abbrev>SharedMIME</abbrev><citetitle>Shared MIME-info Database,
		<ulink url="http://www.freedesktop.org/standards/shared-mime-info.html"/>
		</citetitle>
	</bibliomixed>
	<bibliomixed>
		<abbrev>RFC-2119</abbrev>
		<citetitle>Key words for use in RFCs to Indicate Requirement Levels,
		<ulink url="http://www.ietf.org/rfc/rfc2119.txt?number=2119"/>
		</citetitle>
	</bibliomixed>
	<bibliomixed>
		<abbrev>BaseDir</abbrev>
		<citetitle>XDG Base Directory Specification
		<ulink url="http://www.freedesktop.org/standards/basedir/draft/basedir-spec/basedir-spec.html"/>
		</citetitle>
	</bibliomixed>
	<bibliomixed>
		<abbrev>ACAP</abbrev>
		<citetitle>ACAP Media Type Dataset Class
		<ulink url="ftp://ftp.ietf.org/internet-drafts/draft-ietf-acap-mediatype-01.txt"/>
		</citetitle>
	</bibliomixed>
</bibliography>

</article>