object-api/openstack-object-storage-dev/section_object-api-archive-...

115 lines
5.3 KiB
XML

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE section [
<!-- Useful for describing APIs -->
<!ENTITY PUT '<command xmlns="http://docbook.org/ns/docbook">PUT</command>'>
]>
<section xmlns="http://docbook.org/ns/docbook"
xmlns:xi="http://www.w3.org/2001/XInclude"
xmlns:xlink="http://www.w3.org/1999/xlink" version="5.0"
xml:id="archive-auto-extract">
<title>Auto-extract archive files</title>
<para>To discover whether your Object Storage system supports
this feature, see <xref linkend="discoverability"
/>. Alternatively, check with your service provider.</para>
<para>Use the auto-extract archive feature to upload a tar(1)
archive file.</para>
<para>The Object Storage system extracts files from the archive
file and creates an object.</para>
<section xml:id="archive-auto-extract-put">
<title>Auto-extract archive PUT request</title>
<para>To upload an archive file, you make a &PUT; request. Add
the
<parameter>extract-archive=<replaceable>format</replaceable></parameter>
query parameter to indicate that you are uploading a
tar(1) archive file instead of normal content.</para>
<para>Valid values for the <replaceable>format</replaceable>
variable are <literal>tar</literal>,
<literal>tar.gz</literal>, or
<literal>tar.bz2</literal>.</para>
<para>The path you specify in the &PUT; request is a prefix
for the resulting object names.</para>
<para>For example, if the first object in the tar(1) archive
is <filename>/home/file1.txt</filename> and you specify
the
<filename>/v1/12345678912345/mybackup/castor/</filename>
path, the operation creates the
<filename>castor/home/file1.txt</filename> object in
the <literal>mybackup</literal> container in the
<literal>12345678912345</literal> account.</para>
<para>In the &PUT; request, you can specify the path
for:</para>
<itemizedlist>
<listitem>
<para>An account</para>
</listitem>
<listitem>
<para>Optionally, a specific container</para>
</listitem>
<listitem>
<para>Optionally, a specific object prefix</para>
</listitem>
</itemizedlist>
</section>
<section xml:id="archive-auto-extract-create">
<title>Create an archive for auto-extract</title>
<para>You must use the tar(1) utility to create the tar(1)
archive file.</para>
<para>You can upload regular files but you cannot upload other
items, such as empty directories, symbolic links, and so
on.</para>
<para>You must UTF-8-encode the member names.</para>
<para>The archive auto-extract feature supports these
formats:</para>
<itemizedlist>
<listitem>
<para>The POSIX.1-1988 Ustar format.</para>
</listitem>
<listitem>
<para>The GNU tar format. Includes the long name, long
link, and sparse extensions.</para>
</listitem>
<listitem>
<para>The POSIX.1-2001 pax format.</para>
<para>Use gzip(1) or bzip2(1) to compress the
archive.</para>
<para>Use the <parameter>extract-archive</parameter>
query parameter to specify the format. Valid
values for this parameter are <literal>tar</literal>,
<literal>tar.gz</literal>, or
<literal>tar.bz2</literal>.</para>
</listitem>
</itemizedlist>
</section>
<section xml:id="archive-auto-extract-response">
<title>Auto-extract archive response: JSON</title>
<para>When Object Storage processes the request, it performs
multiple sub-operations. Even if all sub-operations fail,
the operation returns a <returnvalue>201</returnvalue>
<literal>Created</literal> status. You must examine the
response body to determine which members failed to result
in an object creation.</para>
<para>You can set the <literal>Accept</literal> request header
to one of these values, which defines the response format:</para>
<itemizedlist>
<listitem>
<para><literal>text/plain</literal>. Formats response
as plain text. If you omit the
<literal>Accept</literal> header,
<literal>text/plain</literal> is the
default.</para>
</listitem>
<listitem>
<para><literal>application/json</literal>. Formats
response as JSON.</para>
</listitem>
<listitem>
<para><literal>application/xml</literal> or
<literal>text/xml</literal>. Formats response
as XML.</para>
</listitem>
</itemizedlist>
<para>For more information, see <xref
linkend="archive-auto-extract-response"/>.</para>
</section>
</section>