Proposed Specification 8 September 2011
A diff of changes from the previous draft is available at this link.
Please refer to the errata for this document, which may include some normative corrections.
Copyright © 2010, 2011 International Digital Publishing Forum™
All rights reserved. This work is protected under Title 17 of the United States Code. Reproduction and dissemination of this work with changes is prohibited except with the written permission of the International Digital Publishing Forum (IDPF).
EPUB is a registered trademark of the International Digital Publishing Forum.
Table of Contents
application/epub+zip
Media TypeThis section is informative
This specification, EPUB Open Container Format (OCF) 3.0, defines a file format and processing model for encapsulating the sets of related resources that comprise one or more EPUB® Publications into a single-file container.
This specification is one of a family of related specifications that compose EPUB 3, the third major revision of an interchange and delivery format for digital publications based on XML and Web Standards. It is meant to be read and understood in concert with the other specifications that make up EPUB 3:
The EPUB 3 Overview [EPUB3Overview], which provides an informative overview of EPUB and a roadmap to the rest of the EPUB 3 documents. The Overview should be read first.
EPUB Publications 3.0 [Publications30], which defines publication-level semantics and overarching conformance requirements for EPUB Publications.
EPUB Content Documents 3.0 [ContentDocs30], which defines profiles of XHTML, SVG and CSS for use in the context of EPUB Publications.
EPUB Media Overlays 3.0 [MediaOverlays30], which defines a format and a processing model for synchronization of text and audio.
OCF is the required container technology for EPUB Publications. OCF may play a role in the following workflows:
During the preparation steps in producing an electronic Publication, OCF may be used as the container format when exchanging in-progress Publications between different individuals and/or different organizations.
When providing an electronic Publication from publisher or conversion house to the distribution or sales channel, OCF is the recommended container format to be used as the transport format.
When delivering the final Publication to an EPUB Reading System or User, OCF is the required format for the container that holds all of the assets that make up the Publication.
The OCF specification defines the rules for structuring the file collection in the abstract: the "abstract container". It also defines the rules for the representation of this abstract container within a ZIP archive: the "physical container". The rules for ZIP physical containers build upon the ZIP technologies used by [ODF]. OCF also defines a standard method for obfuscating embedded fonts for those EPUB Publications that require this functionality.
This specification supersedes Open Container Format (OCF) 2.0.1 [OCF2]. Refer to [EPUB3Changes] for information on differences between this specification and its predecessor.
A logical document entity consisting of a set of interrelated resources and packaged in an EPUB Container, as defined by this specification and its sibling specifications.
A resource that contains content or instructions that contribute to the logic and rendering of the EPUB Publication. In the absence of this resource, the Publication might not render as intended by the Author. Examples of Publication Resources include the Package Document, EPUB Content Documents, EPUB Style Sheets, audio, video, images, embedded fonts and scripts.
With the exception of the Package Document itself, Publication Resources must be listed in the manifest [Publications30] and must be bundled in the EPUB container file unless specified otherwise in Publication Resource Locations [Publications30].
Examples of resources that are not Publication Resources include those identified by the Package Document link [Publications30] element and those identified in outbound hyperlinks that resolve outside the EPUB Container (e.g., referenced from an [HTML5] a
element href
attribute).
A Publication Resource that conforms to one of the EPUB Content Document definitions (XHTML or SVG).
An EPUB Content Document is a Core Media Type, and may therefore be included in the EPUB Publication without the provision of fallbacks [Publications30].
An EPUB Content Document conforming to the profile of [HTML5] defined in XHTML Content Documents [ContentDocs30].
XHTML Content Documents use the XHTML syntax of [HTML5].
An EPUB Content Document conforming to the constraints expressed in SVG Content Documents [ContentDocs30].
A set of Publication Resource types for which no fallback is required. Refer to Publication Resources [Publications30] for more information.
A Publication Resource carrying bibliographical and structural metadata about the EPUB Publication, as defined in Package Documents [Publications30].
The digital (or physical) embodiment of a work of intellectual content. Changes to the content such as significant revision, abridgement, translation, or the realization of the content in a different digital or physical form result in a new manifestation. There may be many individual but identical copies of a manifestation, termed 'instances' or 'items'. The ISBN is an example of a manifestation identifier, and is shared by all instances of that manifestation.
All instances of a manifestation need not be bit-for-bit identical, as minor corrections or revisions are not judged to create a new manifestation or work.
The Unique Identifier is the primary identifier for an EPUB Publication, as identified by the unique-identifier
attribute. The Unique Identifier may be shared by one or many Manifestations of the same work that conform to the EPUB standard and embody the same content, where the differences between the Manifestations are limited to those changes that take account of differences between EPUB Reading Systems (and which themselves may require changes in the ISBN).
The Unique Identifier is less granular than the ISBN. However, significant revision, abridgement, etc. of the content requires a new Unique Identifier.
A CSS Style Sheet conforming to the CSS profile defined in EPUB Style Sheets [ContentDocs30].
The region of an EPUB Reading System in which the content of an EPUB Publication is rendered visually to a User.
The ZIP-based packaging and distribution format for EPUB Publications defined in OCF ZIP Container.
A software application that processes EPUB Containers according to this specification.
The person(s) or organization responsible for the creation of an EPUB Publication, which is not necessarily the creator of the content and resources it contains.
An individual that consumes an EPUB Publication using an EPUB Reading System.
A system that processes EPUB Publications for presentation to a User in a manner conformant with this specification and its sibling specifications.
The keywords "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in [RFC2119].
All sections of this specification are normative except where identified by the informative status label "This section is informative". The application of informative status to sections and appendices applies to all child content and subsections they may contain.
All examples in this specification are informative.
› An OCF Abstract Container must meet the conformance constraints defined in OCF Abstract Container.
› An OCF ZIP Container (also referred to as an EPUB Container) must meet the conformance constraints defined in OCF ZIP Container.
An EPUB Reading System must meet all of the following criteria:
› It must process the OCF ZIP Container in conformance with all Reading System conformance constraints expressed in OCF ZIP Container.
› If it has a Viewport, it must support deobfuscation of fonts as defined in Font Obfuscation.
This section is informative
An OCF Abstract Container defines a file system model for the contents of the container. The file system model uses a single common root directory for all of the contents of the container. All (non-remote) resources for embedded Publications are located within the directory tree headed by the container’s root directory, although no specific file system structure is mandated for this. The file system model also includes a mandatory directory named META-INF
that is a direct child of the container's root directory and is used to store the following special files:
container.xml
[required]
Identifies the file that is the point of entry for each embedded Publication.
signatures.xml
[optional]
Contains digital signatures for various assets.
encryption.xml
[optional]
Contains information about the encryption of Publication resources. (This file is required if font obfuscation is used.)
metadata.xml
[optional]
Used to store metadata about the container.
rights.xml
[optional]
Used to store information about digital rights.
manifest.xml
[allowed]
A manifest of container contents as allowed by Open Document Format [ODF].
Complete conformance requirements for the various files in META-INF
are found in META-INF.
The virtual file system for the OCF Abstract Container must have a single common root directory for all of the contents of the container.
The OCF Abstract Container must include a directory named META-INF
that is a direct child of the container's root directory. Requirements for the contents of this directory are described in META-INF.
The file name mimetype
in the root directory is reserved for use by OCF ZIP Containers, as explained in OCF ZIP Container.
All other files within the OCF Abstract Container may be in any location descendant from the container's root directory except within the META-INF
directory.
It is recommended that the contents of each of the individual Publications be stored within its own dedicated directory under the container's root.
Files within the OCF Abstract Container must reference each other via Relative IRI References ([RFC3987] and [RFC3986]). For example, if a file named chapter1.html
references an image file named image1.jpg
that is located in the same directory, then chapter1.html
might contain the following as part of its content:
<img src="image1.jpg" alt="…" />
For Relative IRI References, the Base IRI [RFC3986] is determined by the relevant language specifications for the given file formats. For example, the CSS specification defines how relative IRI references work in the context of CSS style sheets and property declarations. Note that some language specifications reference RFCs that preceded RFC3987, in which case the earlier RFC applies for content in that particular language.
Unlike most language specifications, the Base IRIs for all files within the META-INF
directory use the root directory for the Abstract Container as the default Base IRI. For example, if META-INF/container.xml
has the following content:
<?xml version="1.0"?> <container version="1.0" xmlns="urn:oasis:names:tc:opendocument:xmlns:container"> <rootfiles> <rootfile full-path="OEBPS/Great Expectations.opf" media-type="application/oebps-package+xml" /> </rootfiles> </container>
then the path OEBPS/Great Expectations.opf
is relative to the root directory for the OCF Abstract Container and not relative to the META-INF
directory.
The term File Name
represents the name of any type of file, either a directory or an ordinary file within a directory within an OCF Abstract Container.
For a given directory within the OCF Abstract Container, the Path Name
is a string holding all directory File Names in the full path concatenated together with a /
(U+002F
) character separating the directory File Names. For a given file within the Abstract Container, the Path Name is the string holding all directory File Names concatenated together with a /
character separating the directory File Names, followed by a /
character and then the File Name of the file.
The File Name restrictions described below are designed to allow Path Names and File Names to be used without modification on most commonly used operating systems. This specification does not specify how an OCF Processor that is unable to represent OCF File and Path Names would compensate for this incompatibility.
In the context of an OCF Abstract Container, File and Path Names must meet all of the following criteria:
› File Names must not exceed 255 bytes.
› The Path Name for any directory or file within the Abstract Container must not exceed 65535 bytes.
› File Names must not use the following [Unicode] characters, as these characters might not be supported consistently across commonly-used operating systems:
SOLIDUS: /
(U+002F
)
QUOTATION MARK: "
(U+0022
)
ASTERISK: *
(U+002A
)
FULL STOP as the last character: .
(U+002E
)
COLON: :
(U+003A
)
LESS-THAN SIGN: <
(U+003C
)
GREATER-THAN SIGN: >
(U+003E
)
QUESTION MARK: ?
(U+003F
)
REVERSE SOLIDUS: \
(U+005C
)
DEL (U+007F
)
C0 range (U+0000 … U+001F
)
C1 range (U+0080 … U+009F
)
Private Use Area (U+E000 … U+F8FF
)
Non characters in Arabic Presentation Forms-A (U+FDDO … U+FDEF
)
Specials (U+FFF0 … U+FFFF
)
Tags and Variation Selectors Supplement (U+E0000 … U+E0FFF
)
Supplementary Private Use Area-A (U+F0000 … U+FFFFF
)
Supplementary Private Use Area-B (U+100000 … U+10FFFF
)
› File Names are case sensitive.
› All File Names within the same directory must be unique following case normalization as described in section 3.13 of [Unicode].
› All File Names within the same directory should be unique following NFC or NFD normalization [TR15].
Some commercial ZIP tools do not support the full Unicode range and may support only the ASCII range for File Names. Content creators who want to use ZIP tools that have these restrictions may find it is best to restrict their File Names to the ASCII range. If the names of files cannot be preserved during the unzipping process, it will be necessary to compensate for any name translation which took place when the files are referenced by URI from within the content.
All OCF Abstract Containers must include a directory called META-INF
. This directory contains the files specified below. Files other than the ones defined below may be included in the META-INF
directory; OCF Processors must not fail when encountering such files.
All OCF Containers must include a file called container.xml
within the META-INF
directory. The container.xml
file must identify the media type of, and paths to, the root files for the EPUB Publications included within the container.
The container.xml
file must not be encrypted.
The schema for container.xml
files is available in Schema for container.xml
; container.xml
files must be valid according to this schema after removing all elements and attributes from other namespaces (including all attributes and contents of such elements).
The rootfiles
element must contain one or more rootfile
elements, each of which must uniquely reference a Package Document representing a single Publication. The Publications stored in an OCF should be different renditions of the same Manifestation.
An OCF Processor must consider the first rootfile
element within the rootfiles
element to represent the default rendition for the contained Publication. Reading Systems are not required to use any rendition except the default one.
The following example shows a sample container.xml
for an EPUB Publication with the root file OEBPS/My Crazy Life.opf
(the Package Document):
<?xml version="1.0"?> <container version="1.0" xmlns="urn:oasis:names:tc:opendocument:xmlns:container"> <rootfiles> <rootfile full-path="OEBPS/My Crazy Life.opf" media-type="application/oebps-package+xml" /> </rootfiles> </container>
The following example shows SVG and XHTML renditions of The Sandman bundled in the same container:
<?xml version="1.0"?> <container version="1.0" xmlns="urn:oasis:names:tc:opendocument:xmlns:container"> <rootfiles> <rootfile full-path="SVG/Sandman.opf" media-type="application/oebps-package+xml" /> <rootfile full-path="XHTML/Sandman.opf" media-type="application/oebps-package+xml" /> </rootfiles> </container>
The manifest
element contained within the Package Document specifies the one and only manifest used for EPUB processing. Ancillary manifest information contained in the ZIP archive or in the optional manifest.xml
file must not be used for EPUB processing purposes. Any extra files in the ZIP archive must not be used in the processing of the EPUB Publication (i.e., files within the ZIP archive that are not listed within the Package Document's manifest
element, such as META-INF
files or alternate derived renditions of the Publication).
The value of the full-path
attribute must contain a path component
(as defined by RFC3986) which must take the form of a path-rootless
only (also defined by RFC 3986). The path components are relative to the root of the container in which they are used.
OCF Processors must ignore foreign elements and attributes within a container.xml
file.
An optional encryption.xml
file within the META-INF
directory at the root level of the container file system holds all encryption information on the contents of the container. This file is an XML document whose root element is encryption
. The encryption
element contains child elements of type EncryptedKey
and EncryptedData
as defined by [XML ENC Core]. Each EncryptedData
element describes how one or more files within the container are encrypted. Consequently, if any resource within the container is encrypted, encryption.xml
must be present to indicate that the resource is encrypted and provide information on how it is encrypted.
An EncryptedKey
element describes each encryption key used in the container, while an EncryptedData
element describes each encrypted file. Each EncryptedData
element refers to an EncryptedKey
element, as described in XML Encryption.
The schema for encryption.xml
files is available in Schema for encryption.xml
; encryption.xml
files must be valid according to this schema.
OCF encrypts individual files independently, trading off some security for improved performance, allowing the container contents to be incrementally decrypted. Encryption in this way exposes the directory structure and file naming of the whole package.
OCF uses XML Encryption [XML ENC Core] to provide a framework for encryption, allowing a variety of algorithms to be used. XML Encryption specifies a process for encrypting arbitrary data and representing the result in XML. Even though an OCF Abstract Container may contain non-XML data, XML Encryption can be used to encrypt all data in an OCF Abstract Container. OCF encryption supports only the encryption of entire files within the container, not parts of files. The encryption.xml
file, if present, must not be encrypted.
Encrypted data replaces unencrypted data in an OCF Abstract Container. For example, if an image named photo.jpeg
is encrypted, the contents of the photo.jpeg
resource should be replaced by its encrypted contents. When stored in a ZIP container, streams of data must be compressed before they are encrypted and Deflate compression must be used. Within the ZIP directory, encrypted files should be stored rather than Deflate-compressed.
Some situations require obfuscating the storage of embedded fonts referenced by an EPUB Publication to tie them to the "parent" Publication and make them more difficult to extract for unrestricted use. In these cases, encryption.xml
should be used to provide requisite font decoding information according to Font Obfuscation.
The following files must never be encrypted, regardless of whether default or specific encryption is requested:
mimetype |
META-INF/container.xml |
META-INF/encryption.xml |
META-INF/manifest.xml |
META-INF/metadata.xml |
META-INF/rights.xml |
META-INF/signatures.xml |
EPUB rootfile (the Package Document) |
Signed resources may subsequently be encrypted using the Decryption Transform for XML Signature [XML SIG Decrypt]. This feature enables an application such as an OCF agent to distinguish data that was encrypted before signing from data that was encrypted after signing. Only data that was encrypted after signing must be decrypted before computing the digest used to validate the signature.
In the following example, adapted from Section 2.2.1 of [XML ENC Core] the resource image.jpeg
is encrypted using a symmetric key algorithm (AES) and the symmetric key is further encrypted using an asymmetric key algorithm (RSA) with a key of John Smith.
<encryption xmlns ="urn:oasis:names:tc:opendocument:xmlns:container" xmlns:enc="http://www.w3.org/2001/04/xmlenc#" xmlns:ds="http://www.w3.org/2000/09/xmldsig#"> <enc:EncryptedKey Id="EK"> <enc:EncryptionMethod Algorithm="http://www.w3.org/2001/04/xmlenc#rsa-1_5"/> <ds:KeyInfo> <ds:KeyName>John Smith</ds:KeyName> </ds:KeyInfo> <enc:CipherData> <enc:CipherValue>xyzabc</enc:CipherValue> </enc:CipherData> </enc:EncryptedKey> <enc:EncryptedData Id="ED1"> <enc:EncryptionMethod Algorithm="http://www.w3.org/2001/04/xmlenc#kw-aes128"/> <ds:KeyInfo> <ds:RetrievalMethod URI="#EK" Type="http://www.w3.org/2001/04/xmlenc#EncryptedKey"/> </ds:KeyInfo> <enc:CipherData> <enc:CipherReference URI="image.jpeg"/> </enc:CipherData> </enc:EncryptedData> </encryption>
An optional file with the reserved name manifest.xml
may be included within the META-INF
directory at the root level of the container file system.
The manifest.xml
file, if present, must not be encrypted.
An optional file with the reserved name metadata.xml
may be included within the META-INF
directory at the root level of the container file system. This file, if present, must be used for container-level metadata. This version of the OCF specification does not specify any container-level metadata.
If the META-INF/metadata.xml
file is present, its contents should be only namespace-qualified elements [XMLNS] to avoid collision with future versions of OCF that may specify a particular format for this file.
The metadata.xml
file, if present, must not be encrypted.
An optional file with the reserved name rights.xml
may be included within the META-INF
directory at the root level of the container file system. This file is reserved for digital rights management (DRM) information for trusted exchange of Publications among rights holders, intermediaries, and users. This version of the OCF specification does not specify a required format for DRM information, but a future version may specify a particular format for DRM information.
If the META-INF/rights.xml
file is present, its contents should be only namespace-qualified elements [XMLNS] to avoid collision with future versions of OCF that may specify a particular format for this file.
The rights.xml
file must not be encrypted.
When the rights.xml
file is not present, the OCF container provides no information indicating any part of the container is rights governed.
An optional signatures.xml
within the META-INF
directory at the root level of the container file system holds digital signatures of the container and its contents. This file is an XML document whose root element is signatures
. The signatures
element contains child elements of type Signature
as defined by [XML DSIG Core]. Signatures can be applied to the Publication and any alternate renditions as a whole or to parts of the Publication and renditions. XML Signature can specify the signing of any kind of data, not just XML.
The signatures.xml
file must not be encrypted.
When the signatures.xml
file is not present, the OCF container provides no information indicating any part of the container is digitally signed at the container level. It is possible that digital signing exists within any optional alternate contained renditions, however.
The schema for signatures.xml
files is available in Schema for signatures.xml
; signatures.xml
files must be valid according to this schema.
When an OCF agent creates a signature of data in a container, it should add the new signature as the last child Signature
element of the signatures
element in the signatures.xml
file.
Each Signature
in the signatures.xml
file identifies by IRI the data to which the signature applies, using the XML Signature Manifest
element and its Reference
sub-elements. Individual contained files may be signed separately or together. Separately signing each file creates a digest value for the resource that can be validated independently. This approach may make a Signature element larger. If files are signed together, the set of signed files can be listed in a single XML Signature Manifest
element and referenced by one or more Signature
elements.
Any or all files in the container can be signed in their entirety with the exception of the signatures.xml
file since that file will contain the computed signature information. Whether and how the signatures.xml
file should be signed depends on the objective of the signer.
If the signer wants to allow signatures to be added or removed from the container without invalidating the signer’s signature, the signatures.xml
file should not be signed.
If the signer wants any addition or removal of a signature to invalidate the signer’s signature, the Enveloped Signature transform (defined in Section 6.6.4 of [XML DSIG Core]) can be used to sign the entire preexisting signature file excluding the Signature
being created. This transform would sign all previous signatures, and it would become invalid if a subsequent signature was added to the package.
If the signer wants the removal of an existing signature to invalidate the signer’s signature but also wants to allow the addition of signatures, an XPath transform can be used to sign just the existing signatures. (This is only a suggestion. The particular XPath transform is not a part of the OCF specification.)
XML-Signature does not associate any semantics with a signature; an agent may include semantic information, for example, by adding information to the Signature element that describes the signature. XML Signature describes how additional information can be added to a signature (for example, by using the SignatureProperties
element).
The following XML expression shows the content of an example signatures.xml
file, and is based on the examples found in Section 2 of [XML DSIG Core]. It contains one signature, and the signature applies to two resources, OEBFPS/book.html
and OEBFPS/images/cover.jpeg
, in the container.
<signatures xmlns="urn:oasis:names:tc:opendocument:xmlns:container"> <Signature Id="sig" xmlns="http://www.w3.org/2000/09/xmldsig#"> <SignedInfo> <CanonicalizationMethod Algorithm="http://www.w3.org/TR/2001/REC-xml-c14n-20010315"/> <SignatureMethod Algorithm="http://www.w3.org/2000/09/xmldsig#dsa-sha1"/> <Reference URI="#Manifest1"> <DigestMethod Algorithm="http://www.w3.org/2000/09/xmldsig#sha1"/> <DigestValue>j6lwx3rvEPO0vKtMup4NbeVu8nk=</DigestValue> </Reference> </SignedInfo> <SignatureValue>…</SignatureValue> <KeyInfo> <KeyValue> <DSAKeyValue> <P>…</P><Q>…</Q><G>…</G><Y>…</Y> </DSAKeyValue> </KeyValue> </KeyInfo> <Object> <Manifest Id="Manifest1"> <Reference URI="OEBFPS/book.xml"> <Transforms> <Transform Algorithm="http://www.w3.org/TR/2001/REC-xml-c14n-20010315"/> </Transforms> <DigestMethod Algorithm="http://www.w3.org/2000/09/xmldsig#sha1"/> <DigestValue></DigestValue> </Reference> <Reference URI="OEBFPS/images/cover.jpeg"> <Transforms> <Transform Algorithm="http://www.w3.org/TR/2001/REC-xml-c14n-20010315"/> </Transforms> <DigestMethod Algorithm="http://www.w3.org/2000/09/xmldsig#sha1"/> <DigestValue></DigestValue> </Reference> </Manifest> </Object> </Signature> </signatures>
This section is informative
An OCF ZIP Container is a physical single-file manifestation of an abstract container.
An OCF ZIP Container uses the ZIP format as specified by [ZIP APPNOTE], but with the following constraints and clarifications:
› The contents of the OCF ZIP Container must be a conforming abstract container.
› OCF ZIP Containers must not use the features in the ZIP application note that allow ZIP files to be split across multiple storage media. OCF Processors must treat any OCF files that specify that the ZIP file is split across multiple storage media as being in error.
› OCF ZIP Containers must include only uncompressed files or Deflate-compressed files within the ZIP archive. OCF Processors must treat any OCF Containers that use compression techniques other than Deflate as being in error.
› OCF ZIP Containers may use the ZIP64 extensions defined as "Version 1" in section V, subsection G of the application note at [ZIP APPNOTE] and should use only those extensions when the content requires them. OCF Processors must support the ZIP64 extensions defined as "Version 1".
› OCF ZIP Containers must not use the encryption features defined by the ZIP format; instead, encryption must be done using the features described in Encryption – META-INF/encryption.xml. OCF Processors must treat any other OCF ZIP Containers that use ZIP encryption features as being in error.
› It is not a requirement that OCF Processors preserve information from an OCF ZIP Container through load and save operations that is not defined within the OCF Abstract Container; in particular, an OCF Processor does not have to preserve CRC values, comment fields or fields that hold file system information corresponding to a particular operating system (e.g., External file attributes
and Extra field
).
› OCF ZIP Containers must encode File System Names using UTF-8 [Unicode].
The following constraints apply to particular fields in the OCF ZIP Container archive:
› In the local file header table, OCF ZIP Containers must set the version needed to extract
fields to the values 10
, 20
or 45
in order to match the maximum version level needed by the given file (e.g., 20
if Deflate is needed, 45
if ZIP64 is needed). OCF Processors must treat any other values as being in error.
› In the local file header table, OCF ZIP Containers must set the compression
method field to the values 0
or 8
. OCF Processors must treat any other values as being in error.
› OCF Processors must treat OCF ZIP Containers with an Archive decryption header
or an Archive extra data record
as being in error.
OCF ZIP Containers must include a mimetype
file as the first file in the Container, and the contents of this file must be the MIME type string application/epub+zip
.
The contents of the mimetype
file must not contain any leading padding or whitespace, must not begin with the Unicode signature (or Byte Order Mark), and the case of the MIME type string must be exactly as presented above. The mimetype
file additionally must be neither compressed nor encrypted, and there must not be an extra field in its ZIP header.
Refer to Appendix C, The application/epub+zip
Media Type for further information about the application/epub+zip
media type.
This section is informative
Since an OCF Zip Container is fundamentally a ZIP file, commonly available ZIP tools can be used to extract any unencrypted content stream from the package. On some systems, the contents of the ZIP file may appear like any other native container (e.g., a folder). While the ability to do this is quite useful, it can pose a problem for an Author who wishes to include a third-party font.
Many commercial fonts allow embedding, but embedding a font implies making it an integral part of the Publication, not providing the original font file along with the content. Since integrated ZIP support is so ubiquitous in modern operating systems, simply placing the font in the ZIP archive is insufficient to signify that the font is not intended to be reused in other contexts. This uncertainty can undermine the otherwise very useful font embedding capability of EPUB Publications.
In order to discourage reuse of the font, some font vendors may allow use of their fonts in EPUB Publications if those fonts are bound in some way to the Publication. That is, if the font file cannot be installed directly for use on an operating system with the built-in tools of that computing device, and it cannot be directly used by other EPUB Publications.
It is beyond the scope of this document to provide a digital rights management or enforcement system for font files. It instead defines a method of obfuscation that will require additional work on the part of the final OCF recipient to gain general access to any included fonts. It is the hope of the IDPF that this will meet the requirements of most font vendors. No claim is made in this document or by the IDPF, that this constitutes encryption, nor does it guarantee that the font file will be secure from copyright infringement. The defined mechanism will simply provide a stumbling block for those who are unaware of the license details of the supplied font. It will not prevent a determined user from gaining full access to the font. Given an OCF Container, it is possible to apply the algorithms defined to extract the raw font file. Whether this satisfies the requirements of individual font licenses remains a question for the licensor and licensee.
The algorithm employed to obfuscate the font file consists of modifying the first 1040 bytes (~1KB) of the font file. In the unlikely event that the file is less than 1040 bytes, then the entire file will be modified. The key for the algorithm is generated using the instructions as given in the section Generating the Obfuscation Key. To obfuscate the original data, the result of performing a logical exclusive or (XOR) on the first byte of the raw file and the first byte of the key is stored as the first byte of the embedded font file. This process is repeated with the next byte of source and key, until all bytes in the key have been used. At this point, the process continues starting with the first byte of the key and 21st byte of the source. Once 1040 bytes have been encoded in this way (or the end of the source is reached), any remaining data in the source is directly copied to the destination. In pseudo-code, this is the algorithm:
set source to font file set destination to obfuscated file set keyData to key for font set outer to 0 while outer < 52 and not (source at EOF) set inner to 0 while inner < 20 and not (source at EOF) read 1 byte from source //Assumes read advances file position set sourceByte to result of read set keyByte to byte inner of keyData set obfuscatedByte to (sourceByte XOR keyByte) write obfuscatedByte to destination increment inner end while increment outer end while if not (source at EOF) then read source to EOF write result of read to destination end if
To get the original font data back, the process is simply reversed. That is, the source file becomes the obfuscated data and the destination file will contain the raw font data.
The key used in the obfuscation algorithm is derived from unique identifer(s) of the Publication(s) in the Container, as required by the EPUB Publications 3.0 specification and detailed in Unique Identifier [Publications30]. In order to create the key, the unique identifiers of all Publications contained in the container must be concatenated in the order that the Publications appear in container.xml
and a space (Unicode code point U+0020
) inserted between each identifier. Before generating this string, all whitespace characters as defined by the XML 1.0 specification [XML], section 2.3 are removed from the individual identifiers. Specifically the Unicode code points U+0020
, U+0009
, U+000D
and U+000A
must be stripped from each identifier before it is added to the concanenated space-delimited string. An SHA-1 digest of the UTF-8 representation of this string should be generated as specified by the Secure Hash Standard [SHA-1]. This digest is then directly used as the key for the algorithm described in Obfuscation Algorithm.
All encrypted data in an OCF Abstract Container must have an entry in the encryption.xml
file accompanying the Publication (see Encryption – META-INF/encryption.xml), which includes fonts obfuscated using the method described here. For such obfuscated fonts, in the encryption.xml
file, the EncryptionMethod
element child of the EncryptedData
must have an Algorithm
attribute with the value http://www.idpf.org/2008/embedding
. The presence of this attribute signals the use of the algorithm described in this specification. All resources that have been obfuscated using this approach must be listed in the CipherData
element.
An example encryption.xml
file might look like this:
<encryption xmlns="urn:oasis:names:tc:opendocument:xmlns:container" xmlns:enc="http://www.w3.org/2001/04/xmlenc#"> <enc:EncryptedData> <enc:EncryptionMethod Algorithm="http://www.idpf.org/2008/embedding"/> <enc:CipherData> <enc:CipherReference URI="OEBPS/Fonts/BKANT.TTF"/> </enc:CipherData> </enc:EncryptedData> </encryption>
To prevent trivial copying of the embedded font to other Publications, the explicit key must not be provided in the encryption.xml
file. Reading systems must derive the key from the package's Unique Identifier.
The schemas in this Appendix are normative.
container.xml
The schema for container.xml
files is available at http://www.idpf.org/epub/30/schema/ocf-container-30.rnc.
encryption.xml
The schema for encryption.xml
files is available at http://www.idpf.org/epub/30/schema/ocf-encryption-30.rnc. It is based on schemas in [XML Sec RNG Schemas].
signatures.xml
The schema for signatures.xml
files is available at http://www.idpf.org/epub/30/schema/ocf-signatures-30.rnc. It is based on schemas in [XML Sec RNG Schemas].
The following example demonstrates the use of this OCF format to contain a signed and encrypted EPUB Publication within a ZIP Container.
Example B.1. Ordered list of files in the ZIP Container
mimetype META-INF/container.xml META-INF/signatures.xml META-INF/encryption.xml OEBPS/As You Like It.opf OEBPS/book.html OEBPS/nav.html OEBPS/toc.ncx OEBPS/images/cover.png
Example B.3. The contents of the META-INF/container.xml
file
<?xml version="1.0"?> <container version="1.0" xmlns="urn:oasis:names:tc:opendocument:xmlns:container"> <rootfiles> <rootfile full-path="OEBPS/As You Like It.opf" media-type="application/oebps-package+xml" /> </rootfiles> </container>
Example B.4. The contents of the META-INF/signatures.xml
file
<signatures xmlns="urn:oasis:names:tc:opendocument:xmlns:container"> <Signature Id="AsYouLikeItSignature" xmlns="http://www.w3.org/2000/09/xmldsig#"> <!-- SignedInfo is the information that is actually signed. In this case --> <!-- the SHA1 algorithm is used to sign the canonical form of the XML --> <!-- documents enumerated in the Object element below --> <SignedInfo> <CanonicalizationMethod Algorithm="http://www.w3.org/TR/2001/REC-xml-c14n-20010315"/> <SignatureMethod Algorithm="http://www.w3.org/2000/09/xmldsig#dsa-sha1"/> <Reference URI="#AsYouLikeIt"> <DigestMethod Algorithm="http://www.w3.org/2000/09/xmldsig#sha1"/> <DigestValue>…</DigestValue> </Reference> </SignedInfo> <!-- The signed value of the digest above using the DSA algorithm --> <SignatureValue>…</SignatureValue> <!-- The key to use to validate the signature --> <KeyInfo> <KeyValue> <DSAKeyValue> <P>…</P> <Q>…</Q> <G>…</G> <Y>…</Y> </DSAKeyValue> </KeyValue> </KeyInfo> <!-- The list documents to sign. Note that the canonical form of XML --> <!-- documents is signed while the binary form of the other documents --> <!-- is used --> <Object> <Manifest Id="AsYouLikeIt"> <Reference URI="OEBPS/As You Like It.opf"> <Transforms> <Transform Algorithm="http://www.w3.org/TR/2001/REC-xml-c14n-20010315"/> </Transforms> <DigestMethod Algorithm="http://www.w3.org/2000/09/xmldsig#sha1"/> <DigestValue></DigestValue> </Reference> <Reference URI="OEBPS/book.html"> <Transforms> <Transform Algorithm="http://www.w3.org/TR/2001/REC-xml-c14n-20010315"/> </Transforms> <DigestMethod Algorithm="http://www.w3.org/2000/09/xmldsig#sha1"/> <DigestValue></DigestValue> </Reference> <Reference URI="OEBPS/images/cover.png"> <DigestMethod Algorithm="http://www.w3.org/2000/09/xmldsig#sha1"/> <DigestValue></DigestValue> </Reference> </Manifest> </Object> </Signature> </signatures>
Example B.5. The contents of the META-INF/encryption.xml
file
<?xml version="1.0"?> <encryption xmlns="urn:oasis:names:tc:opendocument:xmlns:container" xmlns:enc="http://www.w3.org/2001/04/xmlenc#" xmlns:ds="http://www.w3.org/2000/09/xmldsig#"> <!-- The RSA encrypted AES-128 symmetric key used to encrypt the data --> <enc:EncryptedKey Id="EK"> <enc:EncryptionMethod Algorithm="http://www.w3.org/2001/04/xmlenc#rsa-1_5"/> <ds:KeyInfo> <ds:KeyName>John Smith</ds:KeyName> </ds:KeyInfo> <enc:CipherData> <enc:CipherValue>xyzabc…</enc:CipherValue> </enc:CipherData> </enc:EncryptedKey> <!-- Each EncryptedData block identifies a single document that has been --> <!-- encrypted using the AES-128 algorithm. The data remains stored in it’s --> <!-- encrypted form in the original file within the container. --> <enc:EncryptedData Id="ED1"> <enc:EncryptionMethod Algorithm="http://www.w3.org/2001/04/xmlenc#kw-aes128"/> <ds:KeyInfo> <ds:RetrievalMethod URI="#EK" Type="http://www.w3.org/2001/04/xmlenc#EncryptedKey"/> </ds:KeyInfo> <enc:CipherData> <enc:CipherReference URI="OEBPS/book.html"/> </enc:CipherData> </enc:EncryptedData> <enc:EncryptedData Id="ED2"> <enc:EncryptionMethod Algorithm="http://www.w3.org/2001/04/xmlenc#kw-aes128"/> <ds:KeyInfo> <ds:RetrievalMethod URI="#EK" Type="http://www.w3.org/2001/04/xmlenc#EncryptedKey"/> </ds:KeyInfo> <enc:CipherData> <enc:CipherReference URI="OEBPS/images/cover.png"/> </enc:CipherData> </enc:EncryptedData> </encryption>
Example B.6. The contents of the OEBPS/As You Like It.opf
file
<?xml version="1.0"?> <package version="3.0" xml:lang="en" xmlns="http://www.idpf.org/2007/opf" unique-identifier="pub-id"> <metadata xmlns:dc="http://purl.org/dc/elements/1.1/"> <dc:identifier id="pub-id">urn:uuid:B9B412F2-CAAD-4A44-B91F-A375068478A0</dc:identifier> <meta refines="#pub-id" property="identifier-type" scheme="xsd:string">uuid</meta> <dc:language>en</dc:language> <dc:title>As You Like It</dc:title> <dc:creator id="creator">William Shakespeare</dc:creator> <meta refines="#creator" property="role" scheme="marc:relators">aut</meta> <meta property="dcterms:modified">2000-03-24T00:00:00Z</meta> <dc:publisher>Project Gutenberg</dc:publisher> <dc:date>2000-03-24</dc:date> <meta property="dcterms:dateCopyrighted">9999-01-01</meta> <dc:identifier id="isbn13">urn:isbn:9780741014559</dc:identifier> <meta refines="#isbn13" property="identifier-type" scheme="onix:codelist5">15</meta> <dc:identifier id="isbn10">0-7410-1455-6</dc:identifier> <meta refines="#isbn10" property="identifier-type" scheme="onix:codelist5">2</meta> <link rel="xml-signature" href="../META-INF/signatures.xml#AsYouLikeItSignature"/> </metadata> <manifest> <item id="r4915" href="book.html" media-type="application/xhtml+xml"/> <item id="r7184" href="images/cover.png" media-type="image/png"/> <item id="nav" href="nav.html" media-type="application/xhtml+xml" properties="nav"/> <item id="ncx" href="toc.ncx" media-type="application/x-dtbncx+xml"/> </manifest> <spine toc="ncx"> <itemref idref="r4915"/> </spine> </package>
application/epub+zip
Media TypeThis appendix registers the media type application/epub+zip
for the EPUB Open Container Format (OCF).
An OCF file is a container technology based on the ZIP archive format. It is used to encapsulate EPUB Publications and optional alternate renditions thereof. OCF and its related standards are maintained and defined by the International Digital Publishing Forum (IDPF).
application
epub+zip
None.
None.
OCF files are binary files in ZIP (http://www.iana.org/assignments/media-types/application/zip) format.
All processors that read OCF files should rigorously check the size and validity of data retrieved.
In addition, because of the various content types that can be embedded in OCF files, it is possible that application/epub+zip
may describe content that has security implications beyond those described here. However, only in the case where the processor recognizes and processes the additional content, or where further processing of that content is dispatched to other processors, would security issues potentially arise. And in that case, they would fall outside the domain of this registration document.
Security considerations that apply to application/zip also apply to OCF files.
None.
This media type registration is for the EPUB Open Container Format (OCF), as described by the EPUB Open Container Format (OCF) 3.0 specification located at http://www.idpf.org/epub/30/spec/epub30-ocf.html.
The EPUB OCF 3.0 specification supersedes the Open Container Format 2.0.1 specification, which is located at http://www.idpf.org/doc_library/epub/OCF_2.0.1_draft.doc and which also uses the application/epub+zip
media type.
This media type is in wide use for the distribution of ebooks in the EPUB format. The following list of applications is not exhaustive.
Adobe Digital Editions
Aldiko
Azardi
Apple iBooks
Barnes & Noble Nook
Calibre
Google Books
Ibis Reader
MobiPocket reader
Sony Reader
Stanza
0: PK 0x03 0x04
, 30: mimetype
, 38: application/epub+zip
OCF files are most often identified with the extension .epub
.
ZIP
The IDPF maintains a registry of linking schemes at http://idpf.org/epub/linking/. Some of these schemes define custom fragment identifiers that resolve to application/epub+zip
and application/oebps-package+xml
documents.
William McCoy, [email protected]
COMMON
International Digital Publishing Forum (http://www.idpf.org)
This appendix is informative
EPUB has been developed by the International Digital Publishing Forum in a cooperative effort, bringing together publishers, vendors, software developers, and experts in the relevant standards.
The EPUB 3 specifications were prepared by the International Digital Publishing Forum’s EPUB Maintenance Working Group, operating under a charter approved by the membership in May, 2010 under the leadership of:
Active members of the working group included:
IDPF Members
Invited Experts/Observers
For more detailed acknowledgements and information about contributors to each version of EPUB, refer to Acknowledgements and Contributors [EPUB3Overview].
[ContentDocs30] EPUB Content Documents 3.0 .
[MediaOverlays30] EPUB Media Overlays 3.0 .
[OCF2] Open Container Format 2.0.1 .
[OCF3] Open Container Format 3.0 .
[Publications30] EPUB Publications 3.0 .
[RFC2119] Key words for use in RFCs to Indicate Requirement Levels (RFC 2119) . March 1997.
[RFC3986] Uniform Resource Identifier (URI): Generic Syntax (RFC 3986) . January 2005.
[RFC3987] Internationalized Resource Identifiers (IRIs) (RFC 3987) . January 2005.
[SHA-1] Federal Information Processing Standards Publication 180-3: Secure Hash Standard (SHS) . October 2008.
[TR15] Unicode Normalization Forms . 17 September 2010.
[Unicode] The Unicode Consortium. The Unicode Standard, Version 5.0.0, defined by: The Unicode Standard, Version 5.0 (Boston, MA, Addison-Wesley, 2007. ISBN 0-321-48091-0).
[XML] Extensible Markup Language (XML) 1.0 (Fifth Edition) . 26 November 2008.
[XML DSIG Core] XML-Signature Syntax and Processing Version 1.1 . 3 March 2011.
[XML ENC Core] XML Encryption Syntax and Processing Version 1.1 . 3 March 2011.
[XML SIG Decrypt] Decryption Transform for XML Signature . 10 December 2002.
[XML Sec RNG Schemas] XML Security RELAX NG Schemas .
[XMLNS] Namespaces in XML (Third Edition) . 8 December 2009.
[ZIP APPNOTE] ZIP File Format Specification . September 28, 2007. PKWARE, Inc..
[EPUB3Changes] EPUB 3 Differences from EPUB 2.0.1 .
[EPUB3Overview] EPUB 3 Overview .
[ODF] ODF Open Document Format .