XML Encryption Syntax and Processing Version 1.1

Abstract

This document specifies a process for encrypting data and representing the result in XML. The data may be in a variety of formats, including octet streams and other unstructured data, or structured data formats such as XML documents, an XML element, or XML element content. The result of encrypting data is an XML Encryption element that contains or references the cipher data.

URI	namespace prefix	XML internal entity
`http://www.w3.org/2001/04/xmlenc#`	default namespace, `xenc:`	`<!ENTITY xenc "http://www.w3.org/2001/04/xmlenc#">`
`http://www.w3.org/2009/xmlenc11#`	`xenc11:`	`<!ENTITY xenc11 "http://www.w3.org/2009/xmlenc11#">`

URI	namespace prefix	XML internal entity
`http://www.w3.org/2000/09/xmldsig#`	default namespace, `ds:`, `dsig:`	`<!ENTITY dsig "http://www.w3.org/2000/09/xmldsig#">`
`http://www.w3.org/2009/xmldsig11#`	`dsig11:`	`<!ENTITY dsig11 "http://www.w3.org/2009/xmldsig11#">`

2. Encryption Overview and Examples

This section is non-normative.

This section provides an overview and examples of XML Encryption syntax. The formal syntax is found in section 3. Encryption Syntax ; the specific processing is given in Processing Rules (section 4).

Expressed in shorthand form, the EncryptedData element has the following structure (where "?" denotes zero or one occurrence; "+" denotes one or more occurrences; "*" denotes zero or more occurrences; "|" denotes a choice; and the empty element tag means the element must be empty ):

The CipherData element envelopes or references the raw encrypted data. A CipherData element must have either a CipherValue or CipherReference child element. If enveloping, the raw encrypted data is the CipherValue element's content; if referencing, the CipherReference element's URI attribute points to the location of the raw encrypted data

2.1 Encryption Granularity

This section is non-normative.

Note: Examples in this document do not consider plaintext guessing attacks or other risks, and are only for illustrative purposes.

Consider the following fictitious payment information, which includes identification information and information appropriate to a payment method (e.g., credit card, money transfer, or electronic check):

This markup represents that John Smith is using his credit card with a limit of $5,000USD.

2.1.1 Encrypting an XML Element

This section is non-normative.

Smith's credit card number is sensitive information! If the application wishes to keep that information confidential, it can encrypt the CreditCard element:

By encrypting the entire CreditCard element from its start to end tags, the identity of the element itself is hidden. (An eavesdropper doesn't know whether he used a credit card or money transfer.) The CipherData element contains the encrypted serialization of the CreditCard element.

2.1.2 Encrypting XML Element Content (Elements)

As an alternative scenario, it may be useful for intermediate agents to know that John used a credit card with a particular limit, but not the card's number, issuer, and expiration date. In this case, the content (character data or children elements) of the CreditCard element can be encrypted:

2.1.3 Encrypting XML Element Content (Character Data)

Alternatively, consider the scenario in which all the information except the actual credit card number can be in the clear, including the fact that the Number element exists:

Both CreditCard and Number are in the clear, but the character data content of Number is encrypted.

2.1.4 Encrypting Arbitrary Data and XML Documents

If the application scenario requires all of the information to be encrypted, the whole document is encrypted as an octet sequence. This applies to arbitrary data including XML documents.

Where appropriate, such as in the case of encrypting an entire EXI stream, the Type attribute should be provided and indicate the use of EXI. The optional MimeType may be used to record the actual (non-EXI-encoded) type, but is not necessary and may be omitted, as in the following EXI encryption example:

2.1.5 Super-Encryption: Encrypting EncryptedData

An XML document may contain zero or more EncryptedData elements. EncryptedData cannot be the parent or child of another EncryptedData element. However, the actual data encrypted can be anything, including EncryptedData and EncryptedKey elements (i.e., super-encryption). During super-encryption of an EncryptedData or EncryptedKey element, one must encrypt the entire element. Encrypting only the content of these elements, or encrypting selected child elements is an invalid instance under the provided schema.

For example, consider the following:

A valid super-encryption of "//xenc:EncryptedData[@Id='ED1']" would be:

where the CipherValue content of 'newEncryptedData' is the base64 encoding of the encrypted octet sequence resulting from encrypting the EncryptedData element with Id='ED1'.

2.2 `EncryptedData` and `EncryptedKey` Usage

2.2.1 `EncryptedData` with Symmetric Key (`KeyName`)

[s1] The type of data encrypted may be represented as an attribute value to aid in decryption and subsequent processing. In this case, the data encrypted was an 'element'. Other alternatives include 'content' of an element, or an external octet sequence which can also be identified via the MimeType and Encoding attributes.

[s2] This (3DES CBC) is a symmetric key cipher.

[s4] The symmetric key has an associated name "John Smith".

[s6] CipherData contains a CipherValue, which is a base64 encoded octet sequence. Alternately, it could contain a CipherReference, which is a URI reference along with transforms necessary to obtain the encrypted data as an octet sequence

2.2.2 `EncryptedKey` (`ReferenceList`, `ds:RetrievalMethod`, `CarriedKeyName`)

The following EncryptedData structure is very similar to the one above, except this time the key is referenced using a ds:RetrievalMethod:

[t02] This (AES-128-CBC) is a symmetric key cipher.

[t04] ds:RetrievalMethod is used to indicate the location of a key with type xenc:EncryptedKey. The (AES) key is located at '#EK'.

[t05] ds:KeyName provides an alternative method of identifying the key needed to decrypt the CipherData. Either or both the ds:KeyName and ds:KeyRetrievalMethod could be used to identify the same key.

Within the same XML document, there existed an EncryptedKey structure that was referenced within [t04]:

[t09] The EncryptedKey element is similar to the EncryptedData element except that the data encrypted is always a key value.

[t10] The EncryptionMethod is the RSA public key algorithm.

[t12] ds:KeyName of "John Smith" is a property of the key necessary for decrypting (using RSA) the CipherData.

[t14] The CipherData's CipherValue is an octet sequence that is processed (serialized, encrypted, and encoded) by a referring encrypted object's EncryptionMethod. (Note, an EncryptedKey's EncryptionMethod is the algorithm used to encrypt these octets and does not speak about what type of octets they are.)

[t15-17] A ReferenceList identifies the encrypted objects (DataReference and KeyReference) encrypted with this key. The ReferenceList contains a list of references to data encrypted by the symmetric key carried within this structure.

[t18] The CarriedKeyName element is used to identify the encrypted key value which may be referenced by the KeyName element in ds:KeyInfo. (Since ID attribute values must be unique to a document,CarriedKeyName can indicate that several EncryptedKey structures contain the same key value encrypted for different recipients.)

3. Encryption Syntax

This section provides a detailed description of the syntax and features for XML Encryption. Features described in this section must be implemented unless otherwise noted. The syntax is defined via [XMLSCHEMA-1], [XMLSCHEMA-2] with the following XML preamble, declaration, internal entity, and import:

(Note: A newline has been added to the schemaLocation URI to fit on this page, but is not part of the URI.)

Additional markup defined in this specification uses the xenc11: namespace. The syntax is defined in an XML schema with the following preamble:

(Note: A newline has been added to the schemaLocation URI to fit on this page, but is not part of the URI.)

3.1 The `EncryptedType` Element

EncryptedType is the abstract type from which EncryptedData and EncryptedKey are derived. While these two latter element types are very similar with respect to their content models, a syntactical distinction is useful to processing. Implementations must generate laxly schema valid [XMLSCHEMA-1], [XMLSCHEMA-2] EncryptedData or EncryptedKey elements as specified by the subsequent schema declarations. (Note the laxly schema valid generation means that the content permitted by xsd:ANY need not be valid.) Implementations should create these XML structures (EncryptedType elements and their descendants/content) in Normalization Form C [NFC].

EncryptionMethod is an optional element that describes the encryption algorithm applied to the cipher data. If the element is absent, the encryption algorithm must be known by the recipient or the decryption will fail.

ds:KeyInfo is an optional element, defined by [XMLDSIG-CORE1], that carries information about the key used to encrypt the data. Subsequent sections of this specification define new elements that may appear as children of ds:KeyInfo.

CipherData is a mandatory element that contains the CipherValue or CipherReference with the encrypted data.

EncryptionProperties can contain additional information concerning the generation of the EncryptedType (e.g., date/time stamp).

Id is an optional attribute providing for the standard method of assigning a string id to the element within the document context.

Type is an optional attribute identifying type information about the plaintext form of the encrypted content. While optional, this specification takes advantage of it for processing described in section 4.4 Decryption. If the EncryptedData element contains data of Type 'element' or element 'content', and replaces that data in an XML document context, or contains data of Type 'EXI', it is strongly recommended the Type attribute be provided. Without this information, the decryptor will be unable to automatically restore the XML document to its original cleartext form.

MimeType is an optional (advisory) attribute which describes the media type of the data which has been encrypted. The value of this attribute is a string with values defined by [RFC2045]. For example, if the data that is encrypted is a base64 encoded PNG, the transfer Encoding may be specified as 'http://www.w3.org/2000/09/xmldsig#base64' and the MimeType as 'image/png'. This attribute is purely advisory; no validation of the MimeType information is required and it does not indicate the encryption application must do any additional processing. Note, this information may not be necessary if it is already bound to the identifier in the Type attribute. For example, the Element and Content types defined in this specification are always UTF-8 encoded text. In the case of Type EXI the MimeType attribute is not necessary, but if used should reflect the underlying type and not "EXI".

Encoding is an optional (advisory) attribute which describes the transfer encoding of the data that has been encrypted.

3.2 The `EncryptionMethod` Element

EncryptionMethod is an optional element that describes the encryption algorithm applied to the cipher data. If the element is absent, the encryption algorithm must be known to the recipient or the decryption will fail.

The permitted child elements of the EncryptionMethod are determined by the specific value of the Algorithm attribute URI, and the KeySize child element is always permitted. For example, the RSA-OAEP algorithm (section 5.5.2 RSA-OAEP) uses the ds:DigestMethod and OAEPparams elements, and may use the xenc11:MGF element when needed. (We rely upon the ANY schema construct because it is not possible to specify element content based on the value of an attribute.)

The presence of any child element under EncryptionMethod that is not permitted by the algorithm or the presence of a KeySize child inconsistent with the algorithm must be treated as an error. (All algorithm URIs specified in this document imply a key size but this is not true in general. Most popular stream cipher algorithms take variable size keys.)

3.3 The `CipherData` Element

The CipherData is a mandatory element that provides the encrypted data. It must either contain the encrypted octet sequence as base64 encoded text as element content of the CipherValue element, or provide a reference to an external location containing the encrypted octet sequence via the CipherReference element.

3.3.1 The `CipherReference` Element

If CipherValue is not supplied directly, the CipherReference identifies a source which, when processed, yields the encrypted octet sequence.

The actual value is obtained as follows. The CipherReference URI contains an identifier that is dereferenced. Should the CipherReference element contain an optional sequence of Transforms, the data resulting from dereferencing the URI is transformed as specified so as to yield the intended cipher value. For example, if the value is base64 encoded within an XML document; the transforms could specify an XPath expression followed by a base64 decoding so as to extract the octets.

The syntax of the URI and Transforms is defined in XML Signature [XMLDSIG-CORE1], however XML Encryption places the Transforms element in the XML Encryption namespace since it is used in XML Encryption to obtain an octet stream for decryption. In [XMLDSIG-CORE1] both generation and validation processing start with the same source data and perform that transform in the same order. In encryption, the decryptor has only the cipher data and the specified transforms are enumerated for the decryptor, in the order necessary to obtain the octets. Consequently, because it has different semantics Transforms is in the xenc: namespace.

For example, if the relevant cipher value is captured within a CipherValue element within a different XML document, the CipherReference might look as follows:

Implementations must support the CipherReference feature and the same URI encoding, dereferencing, scheme, and HTTP response codes as that of [XMLDSIG-CORE1]. The Transform feature and particular transform algorithms are optional.

3.4 The `EncryptedData` Element

The EncryptedData element is the core element in the syntax. Not only does its CipherData child contain the encrypted data, but it's also the element that replaces the encrypted element, or element content, or serves as the new document root.

3.5 Extensions to `ds:KeyInfo` Element

There are three ways that the keying material needed to decrypt CipherData can be provided:

The EncryptedData or EncryptedKey element specify the associated keying material via a child of ds:KeyInfo. All of the child elements of ds:KeyInfo specified in [XMLDSIG-CORE1] may be used as qualified:
1. Support for ds:KeyValue is optional and may be used to transport public keys, such as Diffie-Hellman Key Values (section 5.6.1 Diffie-Hellman Key Values). (Including the plaintext decryption key, whether a private key or a secret key, is obviously not recommended.)
2. Support of ds:KeyName to refer to an EncryptedKey CarriedKeyName is recommended.
3. Support for same document ds:RetrievalMethod is required.
In addition, we provide two additional child elements: applications must support EncryptedKey (section 3.5.1 The EncryptedKey Element) and may support AgreementMethod (section 5.6 Key Agreement).
A detached (not inside ds:KeyInfo) EncryptedKey element can specify the EncryptedData or EncryptedKey to which its decrypted key will apply via a DataReference or KeyReference ( section 3.6 The ReferenceList Element).
The keying material can be determined by the recipient by application context and thus need not be explicitly mentioned in the transmitted XML.

3.5.1 The `EncryptedKey` Element

Identifier: Type="http://www.w3.org/2001/04/xmlenc#EncryptedKey"
(This can be used within a ds:RetrievalMethod element to identify the referent's type.)

The EncryptedKey element is used to transport encryption keys from the originator to a known recipient(s). It may be used as a stand-alone XML document, be placed within an application document, or appear inside an EncryptedData element as a child of a ds:KeyInfo element. The key value is always encrypted to the recipient(s). When EncryptedKey is decrypted the resulting octets are made available to the EncryptionMethod algorithm without any additional processing.

ReferenceList is an optional element containing pointers to data and keys encrypted using this key. The reference list may contain multiple references to EncryptedKey and EncryptedData elements. This is done using KeyReference and DataReference elements respectively. These are defined below.

CarriedKeyName is an optional element for associating a user readable name with the key value. This may then be used to reference the key using the ds:KeyName element within ds:KeyInfo. The same CarriedKeyName label, unlike an ID type, may occur multiple times within a single document. The value of the key must be the same in all EncryptedKey elements identified with the same CarriedKeyName label within a single XML document. Note that because whitespace is significant in the value of the ds:KeyName element, whitespace is also significant in the value of the CarriedKeyName element.

Recipient is an optional attribute that contains a hint as to which recipient this encrypted key value is intended for. Its contents are application dependent.

The Type attribute inherited from EncryptedType can be used to further specify the type of the encrypted key if the EncryptionMethod Algorithm does not define a unambiguous encoding/representation. (Note, all the algorithms in this specification have an unambiguous representation for their associated key structures.)

3.5.2 The `DerivedKey` Element

Identifier: Type="http://www.w3.org/2009/xmlenc11#DerivedKey"
(This can be used within a ds:RetrievalMethod element to identify the referent's type.)

The DerivedKey element is used to transport information about a derived key from the originator to recipient(s). It may be used as a stand-alone XML document, be placed within an application document, or appear inside an EncryptedData or Signature element as a child of a ds:KeyInfo element. The key value itself is never sent by the originator. Rather, the originator provides information to the recipient(s) by which the recipient(s) can derive the same key value. When the key has been derived the resulting octets are made available to the EncryptionMethod or SignatureMethod algorithm without any additional processing.

KeyDerivationMethod is an optional element that describes the key derivation algorithm applied to the master (underlying) key material. If the element is absent, the key derivation algorithm must be known by the recipient or the recipient's key derivation will fail.

ReferenceList is an optional element containing pointers to data and keys encrypted using this key. The reference list may contain multiple references to EncryptedKey or EncryptedData elements. This is done using KeyReference and DataReference elements from XML Encryption.

The optional DerivedKeyName element is used to identify the derived key value. This element may then be referenced by the ds:KeyName element in ds:KeyInfo. The same DerivedKeyName label, unlike an ID type, may occur multiple times within a single document. Note that because whitespace is significant in the value of the ds:KeyName element, whitespace is also significant in the value of the DerivedKeyName element.

MasterKeyName is an optional element for associating a user readable name with the master key (or secret) value. The same MasterKeyName label, unlike an ID type, may occur multiple times within a single document. The value of the master key must be the same in all DerivedKey elements identified with the same MasterKeyName label within a single XML document. If no MasterKeyName is provided, the master key material must be known by the recipient or key derivation will fail.

Recipient is an optional attribute that contains a hint as to which recipient this derived key value is intended for. Its contents are application dependent.

The optional Id attribute provides for the standard method of assigning a string id to the element within the document context.

The Type attribute can be used to further specify the type of the derived key if the KeyDerivationMethod algorithm does not define an unambiguous encoding/representation.

3.5.3 The `ds:RetrievalMethod` Element

The ds:RetrievalMethod [XMLDSIG-CORE1] with a Type of 'http://www.w3.org/2001/04/xmlenc#EncryptedKey' provides a way to express a link to an EncryptedKey element containing the key needed to decrypt the CipherData associated with an EncryptedData or EncryptedKey element. The ds:RetrievalMethod [XMLDSIG-CORE1] with a Type of 'http://www.w3.org/2001/04/xmlenc#DerivedKey' provides a way to express a link to a DerivedKey element used to derive the key needed to decrypt the CipherData associated with an EncryptedData or EncryptedKey element. The ds:RetrievalMethod with one of these types is always a child of the ds:KeyInfo element and may appear multiple times. If there is more than one instance of a ds:RetrievalMethod in a ds:KeyInfo of this type, then the EncryptedKey objects referred to must contain the same key value, possibly encrypted in different ways or for different recipients.

3.6 The `ReferenceList` Element

ReferenceList is an element that contains pointers from a key value of an EncryptedKey or DerivedKey to items encrypted by that key value (EncryptedData or EncryptedKey elements).

DataReference elements are used to refer to EncryptedData elements that were encrypted using the key defined in the enclosing EncryptedKey or DerivedKey element. Multiple DataReference elements can occur if multiple EncryptedData elements exist that are encrypted by the same key.

KeyReference elements are used to refer to EncryptedKey elements that were encrypted using the key defined in the enclosing EncryptedKey or DerivedKey element. Multiple KeyReference elements can occur if multiple EncryptedKey elements exist that are encrypted by the same key.

For both types of references one may optionally specify child elements to aid the recipient in retrieving the EncryptedKey and/or EncryptedData elements. These could include information such as XPath transforms, decompression transforms, or information on how to retrieve the elements from a document storage facility. For example:

3.7 The `EncryptionProperties` Element

Identifier: Type="http://www.w3.org/2001/04/xmlenc#EncryptionProperties"
(This can be used within a ds:Reference element to identify the referent's type.)

Additional information items concerning the generation of the EncryptedData or EncryptedKey can be placed in an EncryptionProperty element (e.g., date/time stamp or the serial number of cryptographic hardware used during encryption). The Target attribute identifies the EncryptedType structure being described. anyAttribute permits the inclusion of attributes from the XML namespace to be included (i.e., xml:space, xml:lang, and xml:base).

4. Processing Rules

This section describes the operations to be performed as part of encryption and decryption processing by implementations of this specification. The conformance requirements are specified over the following roles:

Encryptor: An XML Encryption implementation with the role of encrypting data.
Decryptor: An XML Encryption implementation with the role of decrypting data.

Encryptor and Decryptor are invoked by the Application. This specification does not include normative definitions for application behavior. However, this specification does include conformance requirements on encrypted data that may only be achievable through appropriate behavior by all three parties. It is up to specific deployment contexts how this is achieved.

4.1 Intended Application Model

The processing rules for XML Encryption are designed around an intended application model that this version of the specification does not cover normatively.

In the intended processing model, XML Encryption is used to encrypt an octet-stream, an EXI stream, or a fragment of an XML document that matches either the content or element production from [XML10].

If XML Encryption is used with some octet-stream, the precise encoding and meaning of that octet-stream is up to the application, but treated as opaque by the Encryptor or Decryptor. The application may use the Type, Encoding and MimeType parameters to transport further information about the nature of that octet-stream. Hence, an unknown Type parameter is, in general, not treated as an error by either the Encryptor or Decryptor, but instead simply passed through, along with the other relevant parameters and the cleartext octet-stream.

If XML Encryption is used with an XML element or XML content, then Encryptors and Decryptors commonly perform type-specific processing:

If an element is encrypted, then the Encryptor will replace the element in question with an appropriately constructed EncryptedData element. The Decryptor will, conversely, replace the EncryptedData element with its cleartext.
If XML content is encrypted, then the Encryptor will likewise replace this content with an appropriately constructed EncryptedData element, and the Decryptor will reverse this operation.

Note that the intended Encryptor behavior will often cause the document with encrypted parts to become invalid with respect to its schema for the hosting XML format, unless that format is specifically prepared to be used with XML Encryption. An Encryptor or Decryptor that implements the intended processing model is not required to ensure that the resulting XML is schema-valid for the hosting XML format.

If XML processing is handled inside the Encryptor and Decryptor, and the Type attribute values for element and content cleartext are used, then the Encryptor and Decryptor must ensure that the XML cleartext is serialized as UTF-8 before encryption, and -- if needed -- converted back to whatever other encoding might be used by the surrounding XML context.

If XML Encryption is used with an EXI stream [EXI], then Encryptors and Decryptors process content as for XML element or XML content processing, but taking into account EXI serialization. In particular, the encryptor will replace the XML element or XML fragment in question with an appropriately constructed EncryptedData element. The Decryptor will conversely replace the EncryptedData element with its cleartext XML element or XML fragment. Note that the XML document into which the EncryptedData element is embedded may be encoded using EXI and/or EXI may be used to encode the cleartext before encryption.

4.2 Well-known `Type` parameter values

For interoperability purposes, the following types must be implemented such that an implementation will be able to take as input and yield as output data matching the production rules 39 and 43 from [XML10]:

element 'http://www.w3.org/2001/04/xmlenc#Element': "[39] element ::= EmptyElemTag | STag content ETag"
content 'http://www.w3.org/2001/04/xmlenc#Content': "[43] content ::= CharData? ((element | Reference | CDSect | PI | Comment) CharData?)*"

Support for the following type is optional for Encryptors and Decryptors:

http://www.w3.org/2009/xmlenc11#EXI: Presence of this Type indicates that the cleartext is an EXI stream [EXI]. Encryptors and Decryptors that support this type may operate directly on (parts of) EXI streams.

Encryptors and Decryptors should handle unknown or empty Type attribute values as a signal that the cleartext is to be handled as an opaque octet-stream, whose specific processing is up to the invoking application. In this case, the Type, MimeType and Encoding parameters should be treated as opaque data whose appropriate processing is up to the application.

4.3 Encryption

The selection of the algorithm, parameters, and encryption keys is out of scope for this specification.

The cleartext data are assumed to be present as an octet stream. If the cleartext is of type element or content, the data must be serialized in UTF-8 as specified in [XML10], using Normal Form C [NFC].

For each data item to be encrypted as an EncryptedData or EncryptedKey element, the encryptor must:

Obtain (or derive) and (optionally) represent the key.
1. If the key is to be identified (via naming, URI, or included in a child element), construct the ds:KeyInfo as appropriate (e.g., ds:KeyName, ds:KeyValue, ds:RetrievalMethod, etc.)
2. If the key itself is to be encrypted, construct an EncryptedKey element by recursively applying this encryption process. The result may then be a child of ds:KeyInfo, or it may exist elsewhere and may be identified in the preceding step.
3. If the key was derived from a master key, construct a DerivedKey element with associated child elements. The result may, as in the EncryptedKey case, be a child of ds:KeyInfo, or it may exist elsewhere.
Encrypt the data:
1. Encrypt the octets using the algorithm and key.
2. Unless the decryptor will implicitly know the type of the encrypted data, the encryptor should set the Type to indicate the intended interpretation of the cleartext data. See section 4.2 Well-known Type parameter values for known parameter values.
  
  If the data is a simple octet sequence it may be described with the MimeType and Encoding attributes. For example, the data might be an XML document (MimeType="text/xml"), sequence of characters (MimeType="text/plain"), or binary image data (MimeType="image/png").
Build the EncryptedData or EncryptedKey structure:
An EncryptedData or EncryptedKey structure represents all of the information previously discussed including the type of the encrypted data, encryption algorithm, parameters, key, type of the encrypted data, etc.
1. If the encrypted octet sequence obtained in step 2 is to be stored in the CipherData element within the EncryptedData or EncryptedKey element, then the base64 representation of the encrypted octet sequence is inserted as the content of a CipherValue element.
2. If the encrypted octet sequence is stored externally to the EncryptedData or EncryptedKey element, then the URI and transforms (if any) required for the Decryptor to retrieve the encrypted octet sequence are described within a CipherReference element.

4.4 Decryption

For each EncryptedData or EncryptedKey to be decrypted, the decryptor must:

Determine the algorithm, parameters and key information to be used. This information may be obtained out-of-band, or determined according to a ds:KeyInfo element; see section 3.5 Extensions to ds:KeyInfo Element.
Decrypt the data contained in the CipherData element.
1. If a CipherValue child element is present, then the associated text value is retrieved and base64 decoded so as to obtain the encrypted octet sequence.
2. If a CipherReference child element is present, the URI and transforms (if any) are used to retrieve the encrypted octet sequence.
3. The encrypted octet sequence is decrypted using the algorithm, parameters and key value already determined from step 1.

4.5 XML Encryption

Encryption and decryption operations are operations on octets. The application is responsible for the marshalling XML such that it can be serialized into an octet sequence, encrypted, decrypted, and be of use to the recipient.

For example, if the application wishes to canonicalize its data or encode/compress the data in an XML packaging format, the application needs to marshal the XML accordingly and identify the resulting type via the EncryptedData Type attribute. The likelihood of successful decryption and subsequent processing will be dependent on the recipient's support for the given type. Also, if the data is intended to be processed both before encryption and after decryption (e.g., XML Signature [XMLDSIG-CORE1] validation or an XSLT transform) the encrypting application must be careful to preserve information necessary for that process's success.

The following sections contain specifications for decrypting, replacing, and serializing XML content (i.e., Type 'element' or element 'content') using the [XPATH] data model. These sections are non-normative and optional to implementers of this specification, but they may be normatively referenced by and be required by other specifications that require a consistent processing for applications, such as [XMLENC-DECRYPT].

4.5.1 A Decrypt Implementation (Non-normative)

Where P is the context in which the serialized XML should be parsed (a document node or element node) and O is the octet sequence representing UTF-8 encoded characters resulting from step 4.3 in section 4.4 Decryption. Y is node-set representing the decrypted content obtained by the following steps:

Let C be the parsing context of a child of P, which consists of the following items:
- Prefix and namespace name of each namespace that is in scope for P.
- Name and value of each general entity that is effective for the XML document causing P.
Wrap the decrypted octet stream O in the context C as specified in section 4.5.4 Text Wrapping.
Parse the wrapped octet stream as described in The Reference Processing Model (section 4.3.3.2) of [XMLDSIG-CORE1], resulting in a node-set.
Y is the node-set obtained by removing the root node, the wrapping element node, and its associated set of attribute and namespace nodes from the node-set obtained in Step 3.

4.5.2 A Decrypt and Replace Implementation (Non-normative)

Where X is the [XPATH] node set corresponding to an XML document and e is an EncryptedData element node in X.

Z is an [XPATH] node-set that identical to X except where the element node e is an EncryptedData element type. In which case:
1. Decrypt e in the context of its parent node as specified in the section 4.5.1 A Decrypt Implementation (Non-normative) yielding Y, an [XPATH] node set.
2. Include Y in place of e and its descendants in X. Since [XPATH] does not define methods of replacing node-sets from different documents, the result must be equivalent to replacing e with the octet stream resulting from its decryption in the serialized form of X and re-parsing the document. However, the actual method of performing this operation is left to the implementor.

4.5.3 Serializing XML (Non-normative)

4.5.3.1 Default Namespace Considerations

In section 4.3 Encryption (step 3.1), when serializing an XML fragment special care should be taken with respect to default namespaces. If the data will be subsequently decrypted in the context of a parent XML document then serialization can produce elements in the wrong namespace. Consider the following fragment of XML:

Serialization of the element ToBeEncrypted fragment via [XML-C14N] would result in the characters "<ToBeEncrypted></ToBeEncrypted>" as an octet stream. The resulting encrypted document would be:

Decrypting and replacing the EncryptedData within this document would produce the following incorrect result:

This problem arises because most XML serializations assume that the serialized data will be parsed directly in a context where there is no default namespace declaration. Consequently, they do not redundantly declare the empty default namespace with an xmlns="". If, however, the serialized data is parsed in a context where a default namespace declaration is in scope (e.g., the parsing context as described in section 4.5.1 A Decrypt Implementation (Non-normative)), then it may affect the interpretation of the serialized data.

To solve this problem, a canonicalization algorithm may be augmented as follows for use as an XML encryption serializer:

A default namespace declaration with an empty value (i.e., xmlns="") should be emitted where it would normally be suppressed by the canonicalization algorithm.

While the result may not be in proper canonical form, this is harmless as the resulting octet stream will not be used directly in a [XMLDSIG-CORE1] signature value computation. Returning to the preceding example with our new augmentation, the ToBeEncrypted element would be serialized as follows:

<ToBeEncrypted xmlns=""></ToBeEncrypted>

When processed in the context of the parent document, this serialized fragment will be parsed and interpreted correctly.

This augmentation can be retroactively applied to an existing canonicalization implementation by canonicalizing each apex node and its descendants from the node set, inserting xmlns="" at the appropriate points, and concatenating the resulting octet streams.

4.5.3.2 XML Attribute Considerations

Similar attention between the relationship of a fragment and the context into which it is being inserted should be given to the xml:base, xml:lang, and xml:space attributes as mentioned in the Security Considerations of [XML-EXC-C14N]. For example, if the element:

is taken from a context and serialized with no xml:base [XMLBASE] attribute and parsed in the context of the element:

the result will be:

Bongo's href is subsequently interpreted as "http://example.org/example.xml". If this is not the correct URI, Bongo should have been serialized with its own xml:base attribute.

Unfortunately, the recommendation that an empty value be emitted to divorce the default namespace of the fragment from the context into which it is being inserted cannot be made for the attributes xml:base, and xml:space. (Error 41 of the XML 1.0 Second Edition Specification Errata clarifies that an empty string value of the attribute xml:lang is considered as if, "there is no language information available, just as if xml:lang had not been specified".)The interpretation of an empty value for the xml:base or xml:space attributes is undefined or maintains the contextual value. Consequently, applications should ensure (1) fragments that are to be encrypted are not dependent on XML attributes, or (2) if they are dependent and the resulting document is intended to be valid [XML10], the fragment's definition permits the presence of the attributes and that the attributes have non-empty values.

4.5.4 Text Wrapping

This section specifies the process for wrapping text in a given parsing context. The process is based on the proposal by Richard Tobin [Tobin] for constructing the infoset [XML-INFOSET] of an external entity.

The process consists of the following steps:

If the parsing context contains any general entities, then emit a document type declaration that provides entity declarations.
Emit a dummy element start-tag with namespace declaration attributes declaring all the namespaces in the parsing context.
Emit the text.
Emit a dummy element end-tag.

In the above steps, the document type declaration and dummy element tags must be encoded in UTF-8.

Consider the following document containing an EncryptedData element:

If the EncryptedData element is decrypted to the text "<One><foo:Two/></One>", then the wrapped form is as follows:

5. Algorithms

This section discusses algorithms used with the XML Encryption specification. Entries contain the identifier to be used as the value of the Algorithm attribute of the EncryptionMethod element or other element representing the role of the algorithm, a reference to the formal specification, definitions for the representation of keys and the results of cryptographic operations where applicable, and general applicability comments.

5.1 Algorithm Identifiers and Implementation Requirements

All algorithms listed below have implicit parameters depending on their role. For example, the data to be encrypted or decrypted, keying material, and direction of operation (encrypting or decrypting) for encryption algorithms. Any explicit additional parameters to an algorithm appear as content elements within the element. Such parameter child elements have descriptive element names, which are frequently algorithm specific, and should be in the same namespace as this XML Encryption specification, the XML Signature specification, or in an algorithm specific namespace. An example of such an explicit parameter could be a nonce (unique quantity) provided to a key agreement algorithm.

This specification defines a set of algorithms, their URIs, and requirements for implementation. Levels of requirement specified, such as "required" or "optional", refer to implementation, not use. Furthermore, the mechanism is extensible, and alternative algorithms may be used.

5.1.1 Table of Algorithms

The table below lists the categories of algorithms. Within each category, a brief name, the level of implementation requirement, and an identifying URI are given for each algorithm.

Block Encryption

required TRIPLEDES
http://www.w3.org/2001/04/xmlenc#tripledes-cbc
required AES-128
http://www.w3.org/2001/04/xmlenc#aes128-cbc
required AES-256
http://www.w3.org/2001/04/xmlenc#aes256-cbc
required AES128-GCM
http://www.w3.org/2009/xmlenc11#aes128-gcm
optional AES-192
http://www.w3.org/2001/04/xmlenc#aes192-cbc
optional AES192-GCM
http://www.w3.org/2009/xmlenc11#aes192-gcm
optional AES256-GCM
http://www.w3.org/2009/xmlenc11#aes256-gcm

Note: Use of AES GCM is strongly recommended over any CBC block encryption algorithms as recent advances in cryptanalysis [XMLENC-CBC-ATTACK][XMLENC-CBC-ATTACK-COUNTERMEASURES] have cast doubt on the ability of CBC block encryption algorithms to protect plain text when used with XML Encryption. Other mitigations should be considered when using CBC block encryption, such as conveying the encrypted data over a secure channel such as TLS. The CBC block encryption algorithms that are listed as required remain so for backward compatibility.

Stream Encryption

none
Syntax and recommendations are given below to support user specified algorithms.

Key Derivation

required ConcatKDF
http://www.w3.org/2009/xmlenc11#ConcatKDF
optional PBKDF2
http://www.w3.org/2009/xmlenc11#pbkdf2

Key Transport

required RSA-OAEP (including MGF1 with SHA1)
http://www.w3.org/2001/04/xmlenc#rsa-oaep-mgf1p
Optional RSA-OAEP
http://www.w3.org/2009/xmlenc11#rsa-oaep
optional RSA-v1.5 (see RSA-v1.5 security note)
http://www.w3.org/2001/04/xmlenc#rsa-1_5

Key Agreement

required Elliptic Curve Diffie-Hellman (Ephemeral-Static mode)
http://www.w3.org/2009/xmlenc11#ECDH-ES
optional Diffie-Hellman Key Agreement (Ephemeral-Static mode) with Legacy Key Derivation Function
http://www.w3.org/2001/04/xmlenc#dh
optional Diffie-Hellman Key Agreement (Ephemeral-Static mode) with explicit Key Derivation Functions
http://www.w3.org/2009/xmlenc11#dh-es

Symmetric Key Wrap

required TRIPLEDES KeyWrap
http://www.w3.org/2001/04/xmlenc#kw-tripledes
required AES-128 KeyWrap
http://www.w3.org/2001/04/xmlenc#kw-aes128
required AES-256 KeyWrap
http://www.w3.org/2001/04/xmlenc#kw-aes256
optional AES-192 KeyWrap
http://www.w3.org/2001/04/xmlenc#kw-aes192

Message Digest

required SHA1 (Use is DISCOURAGED; see below).
http://www.w3.org/2000/09/xmldsig#sha1
required SHA256
http://www.w3.org/2001/04/xmlenc#sha256
optional SHA384
http://www.w3.org/2001/04/xmlenc#sha384
optional SHA512
http://www.w3.org/2001/04/xmlenc#sha512
optional RIPEMD-160
http://www.w3.org/2001/04/xmlenc#ripemd160

Canonicalization

optional Canonical XML 1.0 (omit comments)
http://www.w3.org/TR/2001/REC-xml-c14n-20010315
optional Canonical XML 1.0 (with comments)
http://www.w3.org/TR/2001/REC-xml-c14n-20010315#WithComments
optional Canonical XML 1.1 (omit comments)
http://www.w3.org/2006/12/xml-c14n11
optional Canonical XML 1.1 (with comments)
http://www.w3.org/2006/12/xml-c14n11#WithComments
optional Exclusive XML Canonicalization 1.0 (omit comments)
http://www.w3.org/2001/10/xml-exc-c14n#
optional Exclusive XML Canonicalization 1.0 (with comments)
http://www.w3.org/2001/10/xml-exc-c14n#WithComments

Encoding

required base64 (*note)
http://www.w3.org/2000/09/xmldsig#base64

Transforms

required base64 (*note)
http://www.w3.org/2000/09/xmldsig#base64

*note: The same URI is used to identify base64 both in "encoding" context (e.g. when used with the Encoding attribute of an EncryptedKey element, see section 3.1 The EncryptedType Element) as well as in "transform" context (when identifying a base64 transform for a CipherReference, see section 3.3.1 The CipherReference Element).

5.2 Block Encryption Algorithms

Block encryption algorithms are designed for encrypting and decrypting data in fixed size, multiple octet blocks. Their identifiers appear as the value of the Algorithm attributes of EncryptionMethod elements that are children of EncryptedData.

Note: CBC block encryption algorithms should not be used without consideration of possibly severe security risks.

Block encryption algorithms take, as implicit arguments, the data to be encrypted or decrypted, the keying material, and their direction of operation. For all of these algorithms specified below, an initialization vector (IV) is required that is encoded with the cipher text. For user specified block encryption algorithms, the IV, if any, could be specified as being with the cipher data, as an algorithm content element, or elsewhere.

The IV is encoded with and before the cipher text for the algorithms below for ease of availability to the decryption code and to emphasize its association with the cipher text. Good cryptographic practice requires that a different IV be used for every encryption.

5.2.1 Padding

Since the data being encrypted is an arbitrary number of octets, it may not be a multiple of the block size. This is solved by padding the plain text up to the block size before encryption and unpadding after decryption. The padding algorithm is to calculate the smallest non-zero number of octets, say N, that must be suffixed to the plain text to bring it up to a multiple of the block size. We will assume the block size is B octets so N is in the range of 1 to B. Pad by suffixing the plain text with N-1 arbitrary pad bytes and a final byte whose value is N. On decryption, just take the last byte and, after sanity checking it, strip that many bytes from the end of the decrypted cipher text.

For example, assume an 8 byte block size and plain text of 0x616263. The padded plain text would then be 0x616263????????05 where the "??" bytes can be any value. Similarly, plain text of 0x2122232425262728 would be padded to 0x2122232425262728??????????????08.

5.2.2 Triple DES

Identifier:: http://www.w3.org/2001/04/xmlenc#tripledes-cbc

NIST SP800-67 [SP800-67] specifies three sequential FIPS 46-3 [DES] operations. The XML Encryption TRIPLEDES consists of a DES encrypt, a DES decrypt, and a DES encrypt used in the Cipher Block Chaining (CBC) mode with 192 bits of key and a 64 bit Initialization Vector (IV). Of the key bits, the first 64 are used in the first DES operation, the second 64 bits in the middle DES operation, and the third 64 bits in the last DES operation.

Note: Each of these 64 bits of key contain 56 effective bits and 8 parity bits. Thus there are only 168 operational bits out of the 192 being transported for a TRIPLEDES key. (Depending on the criterion used for analysis, the effective strength of the key may be thought to be 112 bits (due to meet in the middle attacks) or even less.)

The resulting cipher text is prefixed by the IV. If included in XML output, it is then base64 encoded. An example TRIPLEDES EncryptionMethod is as follows:

Note: CBC block encryption algorithms should not be used without consideration of possibly severe security risks.

5.2.3 AES

Identifier:: http://www.w3.org/2001/04/xmlenc#aes128-cbc; http://www.w3.org/2001/04/xmlenc#aes192-cbc; http://www.w3.org/2001/04/xmlenc#aes256-cbc

[AES] is used in the Cipher Block Chaining (CBC) mode with a 128 bit initialization vector (IV). The resulting cipher text is prefixed by the IV. If included in XML output, it is then base64 encoded. An example AES EncryptionMethod is as follows:

Note: CBC block encryption algorithms should not be used without consideration of possibly severe security risks.

5.2.4 AES-GCM

Identifier:: http://www.w3.org/2009/xmlenc11#aes128-gcm; http://www.w3.org/2009/xmlenc11#aes192-gcm; http://www.w3.org/2009/xmlenc11#aes256-gcm

AES-GCM [SP800-38D] is an authenticated encryption mechanism. It is equivalent to doing these two operations in one step - AES encryption followed by HMAC signing.

AES-GCM is very attractive from a performance point of view because the cost of AES-GCM is similar to regular AES-CBC encryption, yet it achieves the same result as encryption and HMAC signing. Also AES-GCM can be pipelined so it is amenable to hardware acceleration.

For the purposes of this specification, AES-GCM shall be used with a 96 bit Initialization Vector (IV) and a 128 bit Authentication Tag (T). The cipher text contains the IV first, followed by the encrypted octets and finally the Authentication tag. No padding should be used during encryption. During decryption the implementation should compare the authentication tag computed during decryption with the specified Authentication Tag, and fail if they don't match. For details on the implementation of AES-GCM, see [SP800-38D].

5.3 Stream Encryption Algorithms

Simple stream encryption algorithms generate, based on the key, a stream of bytes which are XORed with the plain text data bytes to produce the cipher text on encryption and with the cipher text bytes to produce plain text on decryption. They are normally used for the encryption of data and are specified by the value of the Algorithm attribute of the EncryptionMethod child of an EncryptedData element.

NOTE: It is critical that each simple stream encryption key (or key and initialization vector (IV) if an IV is also used) be used once only. If the same key (or key and IV) is ever used on two messages then, by XORing the two cipher texts, you can obtain the XOR of the two plain texts. This is usually very compromising.

No specific stream encryption algorithms are specified herein but this section is included to provide general guidelines.

Stream algorithms typically use the optional KeySize explicit parameter. In cases where the key size is not apparent from the algorithm URI or key source, as in the use of key agreement methods, this parameter sets the key size. If the size of the key to be used is apparent and disagrees with the KeySize parameter, an error must be returned. Implementation of any stream algorithms is optional. The schema for the KeySize parameter is as follows:

5.4 Key Derivation

Key derivation is a well-established mechanism for generating new cryptographic key material from some existing, original ("master") key material and potentially other information. Derived keys are used for a variety of purposes including data encryption and message authentication. The reason for doing key derivation itself is typically a combination of a desire to expand a given, but limited, set of original key material and prudent security practices of limiting use (exposure) of such key material. Key separation (such as avoiding use of the same key material for multiple purposes) is an example of such practices.

The key derivation process may be based on passphrases agreed upon or remembered by users, or it can be based on some shared "master" cryptographic keys (and be intended to reduce exposure of such master keys), etc. Derived keys themselves may be used in XML Signature and XML Encryption as any other keys; in particular, they may be used to compute message authentication codes (e.g. digital signatures using symmetric keys) or for encryption/decryption purposes.

5.4.1 ConcatKDF

Identifier:: http://www.w3.org/2009/xmlenc11#ConcatKDF

The ConcatKDF key derivation algorithm, defined in Section 5.8.1 of NIST SP 800-56A [SP800-56A] (and equivalent to the KDF3 function defined in ANSI X9.44-2007 [ANSI-X9-44-2007] when the contents of the OtherInfo parameter is structured as in NIST SP 800-56A), takes several parameters. These parameters are represented in the xenc11:ConcatKDFParamsType:

The ds:DigestMethod element identifies the digest algorithm used by the KDF. Compliant implementations must support SHA-256 and SHA-1 (support for SHA-1 is present only for backwards-compatibility reasons). Support for SHA-384 and SHA-512 is optional.

The AlgorithmID, PartyUInfo, PartyVInfo, SuppPubInfo and SuppPrivInfo attributes are as defined in [SP800-56A]. Their presence is optional but AlgorithmID, PartyVInfo and PartyUInfo must be present for applications that need to comply with [SP800-56A]. Note: The PartyUInfo component shall include a nonce when ConcatKDF is used in conjunction with a static-static Diffie-Hellman (or static-static ECDH) key agreement scheme; see further [SP800-56A].

In [SP800-56A], AlgorithmID, PartyUInfo, PartyVInfo, SuppPubInfo and SuppPrivInfo attributes are all defined as arbitrary-length bitstrings, thus they may need to be padded in order to be encoded into hexBinary for XML Encryption. The following padding and encoding method must be used when encoding bitstring values for the AlgorithmID, PartyUInfo, PartyVInfo, SuppPubInfo and SuppPrivInfo:

The bitstring is divided into octets using big-endian encoding. If the length of the bitstring is not a multiple of 8 then add padding bits (value 0) as necessary to the last octet to make it a multiple of 8.
Prepend one octet to the octets string from step 1. This octet shall identify (in a big-endian representation) the number of padding bits added to the last octet in step 1.
Encode the octet string resulting from step 2 as a hexBinary string.

Example: the bitstring 11011, which is 5 bits long, gets 3 additional padding bits to become the bitstring 11011000 (or D8 in hex). This bitstring is then prepended with one octet identifying the number of padding bits to become the octet string (in hex) 03D8, which then finally is encoded as a hexBinary string value of "03D8".

Note that as specified in [SP800-56A], these attributes shall be concatenated to form a bit string “OtherInfo” that is used with the key derivation function. The concatenation shall be done using the original, unpadded bit string values.” Applications must also verify that these attributes, in an application-specific way not defined in this document, identify algorithms and parties in accordance with NIST SP800-56.

An example of an xenc11:DerivedKey element with this key derivation algorithm given below. In this example, the bitstring value of AlgorithmID is 00000000, the bitstring value of PartyUInfo is 11011 and the bitstring value of PartyVInfo is 11010:

Note

While any bit string can be used with ConcatKDF, it is recommended to keep byte aligned for greatest interoperability.

5.4.2 PBKDF2

Identifier:: http://www.w3.org/2009/xmlenc11#pbkdf2

The PBKDF2 key derivation algorithm and the ASN.1 type definitions for its parameters are defined in PKCS #5 v2.0 [PKCS5]. The XML schema definitions for the parameters is defined in [PKCS5Amd1] and the same can be specified by enclosing them within an xenc11:PBKDF2-params child element of the xenc11:KeyDerivationMethod element.

(Note: A newline has been added to the Algorithm attribute to fit on this page, but is not part of the URI.)

The PBKDF2-params element and its child elements have the same names and meaning as the corresponding components of the PBKDF2-params ASN.1 type in [PKCS5]. Note, in case of ConcatKDF and the Diffie Hellman legacy KDF, KeyLength is an implied parameter and needs to be inferred from the context, but in the case of PBKDF2 the KeyLength child element has to be specified, as it has been made a mandatory parameter to be consistent with PKCS5. For PBKDF2, the inferred key length must match the specified key length, otherwise it is an error condition.

The AlgorithmIdentifierType corresponds to the AlgorithmIdentifier type of [PKCS5] and carries the algorithm identifier in the Algorithm attribute. Algorithm specific parameters, where applicable, can be specified using the Parameters element.

The PRFAlgorithmIdentifierType is derived from the AlgorithmIdentifierType and constrains the choice of algorithms to those contained in the PBKDF2-PRFs set defined in [PKCS5]. This type is used to specify a pseudorandom function (PRF) for PBKDF2. Whereas HMAC-SHA1 is the default PRF algorithm in [PKCS5], use of HMAC-SHA256 is recommended by this specification (see [XMLDSIG-CORE1], [HMAC]).

An example of an xenc11:DerivedKey element with this key derivation algorithm is:

5.5 Key Transport

Key Transport algorithms are public key encryption algorithms especially specified for encrypting and decrypting keys. Their identifiers appear as Algorithm attributes to EncryptionMethod elements that are children of EncryptedKey. EncryptedKey is in turn the child of a ds:KeyInfo element. The type of key being transported, that is to say the algorithm in which it is planned to use the transported key, is given by the Algorithm attribute of the EncryptionMethod child of the EncryptedData or EncryptedKey parent of this ds:KeyInfo element.

(Key Transport algorithms may optionally be used to encrypt data in which case they appear directly as the Algorithm attribute of an EncryptionMethod child of an EncryptedData element. Because they use public key algorithms directly, Key Transport algorithms are not efficient for the transport of any amounts of data significantly larger than symmetric keys.)

5.5.1 RSA Version 1.5

Identifier:: http://www.w3.org/2001/04/xmlenc#rsa-1_5

The RSAES-PKCS1-v1_5 algorithm, specified in RFC 3447 [PKCS1], takes no explicit parameters. An example of an RSA Version 1.5 EncryptionMethod element is:

The CipherValue for such an encrypted key is the base64 [RFC2045] encoding of the octet string computed as per RFC 3447 [PKCS1], section 7.2.1: Encryption operation]. As specified in the EME-PKCS1-v1_5 function RFC 3447 [PKCS1], section 7.2.1, the value input to the key transport function is as follows:

Example 1

CRYPT ( PAD ( KEY ))

where the padding is of the following special form:

Example 2

02 | PS* | 00 | key

where "|" is concatenation, "02" and "00" are fixed octets of the corresponding hexadecimal value, PS is a string of strong pseudo-random octets [RANDOM] at least eight octets long, containing no zero octets, and long enough that the value of the quantity being CRYPTed is one octet shorter than the RSA modulus, and "key" is the key being transported. The key is 192 bits for TRIPLEDES and 128, 192, or 256 bits for AES.

Implementations must support this key transport algorithm for transporting 192-bit TRIPLEDES keys. Support of this algorithm for transporting other keys is optional. RSA-OAEP is recommended for the transport of AES keys.

The resulting base64 [RFC2045] string is the value of the child text node of the CipherData element, e.g.

(Note: A newline has been added to the CipherValue to fit on this page, but is not part of value.)

Note: Implementation of RSA v1.5 is not recommended due to security risks associated with the algorithm.

5.5.2 RSA-OAEP

Identifier:: http://www.w3.org/2001/04/xmlenc#rsa-oaep-mgf1p (including MGF1 with SHA1 mask generation function)
Identifier:: http://www.w3.org/2009/xmlenc11#rsa-oaep

The RSAES-OAEP-ENCRYPT algorithm, as specified in RFC 3447 [PKCS1], has options that define the message digest function and mask generation function, as well as an optional PSourceAlgorithm parameter. Default values defined in RFC 3447 are SHA1 for the message digest and MGF1 with SHA1 for the mask generation function. Both the message digest and mask generation functions are used in the EME-OAEP-ENCODE operation as part of RSAES- OAEP-ENCRYPT.

The http://www.w3.org/2001/04/xmlenc#rsa-oaep-mgf1p identifier defines the mask generation function as the fixed value of MGF1 with SHA1. In this case the optional xenc11:MGF element of the xenc:EncryptionMethod element must not be provided.

The http://www.w3.org/2009/xmlenc11#rsa-oaep identifier defines the mask generation function using the optional xenc11:MGF element of the xenc:EncryptionMethod element. If not present, the default of MGF1 with SHA1 is to be used.

The following URIs define the various mask generation function URI values that may be used. These correspond to the object identifiers defined in RFC 4055 [RFC4055]:

MGF1 with SHA1: http://www.w3.org/2009/xmlenc11#mgf1sha1
MGF1 with SHA224: http://www.w3.org/2009/xmlenc11#mgf1sha224
MGF1 with SHA256: http://www.w3.org/2009/xmlenc11#mgf1sha256
MGF1 with SHA384: http://www.w3.org/2009/xmlenc11#mgf1sha384
MGF1 with SHA512: http://www.w3.org/2009/xmlenc11#mgf1sha512

Otherwise the two identifiers define the same usage of the RSA-OAEP algorithm, as follows.

The message digest function should be specified using the Algorithm attribute of the ds:DigestMethod child element of the xenc:EncryptionMethod element. If it is not specified, the default value of SHA1 is to be used.

The optional RSA-OAEP PSourceAlgorithm parameter value may be explicitly provided by placing the base64 encoded octets in the xenc:OAEPparams XML element.

The XML Encryption 1.0 schema definition and description for the EncryptionMethod element is in section 3.2 The EncryptionMethod Element. The following shows the XML Encryption 1.1 addition for the MGF type:

An example of an RSA-OAEP element is:

Example 3

<EncryptionMethod Algorithm="http://www.w3.org/2001/04/xmlenc#rsa-oaep-mgf1p">
            <OAEPparams>9lWu3Q==</OAEPparams>
            <ds:DigestMethod Algorithm="http://www.w3.org/2000/09/xmldsig#sha1"/>
            <EncryptionMethod>

Another example is:

The CipherValue for an RSA-OAEP encrypted key is the base64 [RFC2045] encoding of the octet string computed as per RFC 3447 [PKCS1], section 7.1.1: Encryption operation. As described in the EME-OAEP-ENCODE function RFC 3447 [PKCS1], section 7.1.1, the value input to the key transport function is calculated using the message digest function and string specified in the DigestMethod and OAEPparams elements and using either the mask generator function specified with the xenc11:MGF element or the default MGF1 with SHA1 specified in RFC 3447. The desired output length for EME-OAEP-ENCODE is one byte shorter than the RSA modulus.

The transported key size is 192 bits for TRIPLEDES and 128, 192, or 256 bits for AES. Implementations must implement RSA-OAEP for the transport of all key types and sizes that are mandatory to implement for symmetric encryption. They may implement RSA-OAEP for the transport of other keys.

5.6 Key Agreement

A Key Agreement algorithm provides for the derivation of a shared secret key based on a shared secret computed from certain types of compatible public keys from both the sender and the recipient. Information from the originator to determine the secret is indicated by an optional OriginatorKeyInfo parameter child of an AgreementMethod element while that associated with the recipient is indicated by an optional RecipientKeyInfo. A shared key is derived from this shared secret by a method determined by the Key Agreement algorithm.

Note: XML Encryption does not provide an online key agreement negotiation protocol. The AgreementMethod element can be used by the originator to identify the keys and computational procedure that were used to obtain a shared encryption key. The method used to obtain or select the keys or algorithm used for the agreement computation is beyond the scope of this specification.

The AgreementMethod element appears as the content of a ds:KeyInfo since, like other ds:KeyInfo children, it yields a key. This ds:KeyInfo is in turn a child of an EncryptedData or EncryptedKey element. The Algorithm attribute and KeySize child of the EncryptionMethod element under this EncryptedData or EncryptedKey element are implicit parameters to the key agreement computation. In cases where this EncryptionMethod algorithm URI is insufficient to determine the key length, a KeySize must have been included.

Key derivation algorithms (with associated parameters) may be explicitly declared by using the xenc11:KeyDerivationMethod element. This element will then be placed at the extensibility point of the xenc:AgreementMethodType (see below).

In addition, the sender may place a KA-Nonce element under AgreementMethod to assure that different keying material is generated even for repeated agreements using the same sender and recipient public keys. For example:

If the agreed key is being used to wrap a key, rather than data as above, then AgreementMethod would appear inside a ds:KeyInfo inside an EncryptedKey element.

The Schema for AgreementMethod is as follows:

5.6.1 Diffie-Hellman Key Values

Identifier:: http://www.w3.org/2001/04/xmlenc#DHKeyValue

Diffie-Hellman keys can appear directly within KeyValue elements or be obtained by ds:RetrievalMethod fetches as well as appearing in certificates and the like. The above identifier can be used as the value of the Type attribute of Reference or ds:RetrievalMethod elements.

As specified in [ESDH], a DH public key consists of up to six quantities, two large primes p and q, a "generator" g, the public key, and validation parameters "seed" and "pgenCounter". These relate as follows: The public key = ( g**x mod p ) where x is the corresponding private key; p = j*q + 1 where j >= 2. "seed" and "pgenCounter" are optional and can be used to determine if the Diffie-Hellman key has been generated in conformance with the algorithm specified in [ESDH]. Because the primes and generator can be safely shared over many DH keys, they may be known from the application environment and are optional. The schema for a DHKeyValue is as follows:

5.6.2 Diffie-Hellman Key Agreement

The Diffie-Hellman (DH) key agreement protocol [ESDH] involves the derivation of shared secret information based on compatible DH keys from the sender and recipient. Two DH public keys are compatible if they have the same prime and generator. If, for the second one, Y = g**y mod p, then the two parties can calculate the shared secret ZZ = ( g**(x*y) mod p ) even though each knows only their own private key and the other party's public key. Leading zero bytes must be maintained in ZZ so it will be the same length, in bytes, as p. The size of p must be at least 512 bits and g at least 160 bits. There are numerous other complex security considerations in the selection of g, p, and a random x as described in [ESDH].

The Diffie-Hellman shared secret zz is used as the input to a KDF to produce a secret key. XML Signature 1.0 defined a specific KDF to be used with Diffie-Hellman; that KDF is now known as the "Legacy KDF" and is defined in Section 5.6.2.2. Use of Diffie-Hellman with explicit KDFs is described in Section 5.6.2.1.

Implementation of Diffie-Hellman key agreement is optional. However, if implemented, such implementations must support the Legacy Key Derivation Function and should support Diffie-Hellman with explicit Key Derivation Functions

An example of a DH AgreementMethod element using the Legacy Key Derivation Function (Section 5.6.2.2) is as follows:

5.6.2.1 Diffie-Hellman Key Agreement with Explicit Key Derivation Functions

Identifier:: http://www.w3.org/2009/xmlenc11#dh-es

It is recommended that the shared key material for a Diffie-Hellman key agreement be calculated from the Diffie-Hellman shared secret using a key derivation function (KDF) in accordance with Section 5.4.

An example of a DH AgreementMethod element using an explicit key derivation function is as follows:

5.6.2.2 Diffie-Hellman Key Agreement with Legacy Key Derivation Function

Identifier:: http://www.w3.org/2001/04/xmlenc#dh

XML Signature 1.0 defined a specific KDF for use with Diffie-Hellman key agreement. In order to guarantee interoperability, implementations that choose to implement Diffie-Hellman must support the use of the Diffie-Hellman Legacy KDF defined in this section.

Assume that the Diffie-Hellman shared secret is the octet sequence ZZ. The Diffie-Hellman Legacy KDF calculates the shared keying material as follows:

Example 4

Keying Material = KM(1) | KM(2) | ...

where "|" is byte stream concatenation and

Example 5

KM(counter) = DigestAlg ( ZZ | counter | EncryptionAlg |
            KA-Nonce | KeySize )

DigestAlg: The message digest algorithm specified by the DigestMethod child of AgreementMethod.
EncryptionAlg: The URI of the encryption algorithm, including possible key wrap algorithms, in which the derived keying material is to be used ("Example:Block/Alg" in the example above), not the URI of the agreement algorithm. This is the value of the Algorithm attribute of the EncryptionMethod child of the EncryptedData or EncryptedKey grandparent of AgreementMethod.
KA-Nonce: The base64 decoding the content of the KA-Nonce child of AgreementMethod, if present. If the KA-Nonce element is absent, it is null.
Counter: A one byte counter starting at one and incrementing by one. It is expressed as two hex digits where letters A through F are in upper case.
KeySize: The size in bits of the key to be derived from the shared secret as the UTF-8 string for the corresponding decimal integer with only digits in the string and no leading zeros. For some algorithms the key size is inherent in the URI. For others, such as most stream ciphers, it must be explicitly provided.

For example, the initial (KM(1)) calculation for the EncryptionMethod of the Key Agreement example (section 5.5) would be as follows, where the binary one byte counter value of 1 is represented by the two character UTF-8 sequence 01, ZZ is the shared secret, and "foo" is the base64 decoding of "Zm9v".

Example 6

SHA-1 ( ZZ01Example:Block/Algfoo80 )

Assuming that ZZ is 0xDEADBEEF, that would be

Example 7

SHA-1( 0xDEADBEEF30314578616D706C653A426C6F636B2F416C67666F6F3830 )

whose value is

Example 8

0x534C9B8C4ABDCB50038B42015A181711068B08C1

Each application of DigestAlg for successive values of Counter will produce some additional number of bytes of keying material. From the concatenated string of one or more KM's, enough leading bytes are taken to meet the need for an actual key and the remainder discarded. For example, if DigestAlg is SHA-1 which produces 20 octets of hash, then for 128 bit AES the first 16 bytes from KM(1) would be taken and the remaining 4 bytes discarded. For 256 bit AES, all of KM(1) suffixed with the first 12 bytes of KM(2) would be taken and the remaining 8 bytes of KM(2) discarded.

5.6.3 Elliptic Curve Diffie-Hellman (ECDH) Key Values

Identifier:: http://www.w3.org/2009/xmldsig11#ECKeyValue

ECDH has identical public key parameters as ECDSA and can be represented with the ECKeyValue element [XMLDSIG-CORE1]. Note that if the curve parameters are explicitly stated using the ECParameters element, then the Cofactor element must be included.

As with Diffie-Hellman keys, Elliptic Curve Key Values can appear directly within KeyValue elements or be obtained by ds:RetrievalMethod fetches as well as appearing in certificates and the like. The above identifier can be used as the value of the Type attribute of Reference or ds:RetrievalMethod elements.

5.6.4 Elliptic Curve Diffie-Hellman (ECDH) Key Agreement (Ephemeral-Static Mode)

Identifier:: http://www.w3.org/2009/xmlenc11#ECDH-ES

ECDH is the elliptic curve analogue to the Diffie-Hellman key agreement algorithm. Details of the ECDH primitive can be found in [ECC-ALGS]. When ECDH is used in Ephemeral-Static (ES) mode, the recipient has a static key pair, but the sender generates a ephemeral key pair for each message. The same ephemeral key may be used when there are multiple recipients that use the same curve parameters.

Compliant implementations are required to support ECDH-ES key agreement using the P-256 prime curve specified in Section D.2.3 of FIPS 186-3 [FIPS-186-3]. (This is the same curve that is required in XML Signature 1.1 to be supported for the ECDSAwithSHA256 algorithm.) It is further recommended that implementations also support the P-384 and P-521 prime curves for ECDH-ES; these curves are defined in Sections D.2.4 and D.2.5 of FIPS 186-3, respectively.

The shared key material is calculated from the Diffie-Hellman shared secret using a key derivation function (KDF). While applications may define other KDFs, compliant implementations must implement ConcatKDF (see section 5.4.1 ConcatKDF). An example of xenc:EncryptedData using the ECDH-ES key agreement algorithm with the ConcatKDF key derivation algorithm is as follows:

5.7 Symmetric Key Wrap

Symmetric Key Wrap algorithms are shared secret key encryption algorithms especially specified for encrypting and decrypting symmetric keys. When wrapped keys are used, then an EncryptedKey element will appear as a child of a ds:KeyInfo element. This EncryptedKey element will have an EncryptionMethod child whose Algorithm attribute in turn identifies the key wrap algorithm.

The algorithm for which the encrypted key is intended depends on the context of the ds:KeyInfo element: ds:KeyInfo can occur as a child of either an EncryptedData or EncryptedKey element; in both cases, ds:KeyInfo will have an EncryptionMethod sibling that identifies the algorithm.

5.7.1 CMS Triple DES Key Wrap

Identifiers:: http://www.w3.org/2001/04/xmlenc#kw-tripledes

XML Encryption implementations must support TRIPLEDES wrapping of 168 bit keys as described in [CMS-WRAP] and may optionally support TRIPLEDES wrapping of other keys.

An example of a TRIPLEDES Key Wrap EncryptionMethod element is as follows:

5.7.2 AES KeyWrap

Identifiers:: http://www.w3.org/2001/04/xmlenc#kw-aes128; http://www.w3.org/2001/04/xmlenc#kw-aes192; http://www.w3.org/2001/04/xmlenc#kw-aes256

Implementation of AES key wrap is described in [AES-WRAP]. It provides for confidentiality and integrity. This algorithm is defined only for inputs which are a multiple of 64 bits. The information wrapped need not actually be a key. The algorithm is the same whatever the size of the AES key used in wrapping, called the key encrypting key or KEK. The implementation requirements are indicated below.

128 bit AES Key Encrypting Key: Implementation of wrapping 128 bit keys required.
Wrapping of other key sizes optional.
192 bit AES Key Encrypting Key: All support optional.
256 bit AES Key Encrypting Key: Implementation of wrapping 256 bit keys required.
Wrapping of other key sizes optional.

5.8 Message Digest

Message digest algorithms can be used in AgreementMethod as part of the key derivation, within RSA-OAEP encryption as a hash function, and in connection with the HMAC message authentication code method [HMAC] as described in [XMLDSIG-CORE1].) Use of SHA-256 is strongly recommended over SHA-1 because recent advances in cryptanalysis (see e.g. [SHA-1-Analysis], [SHA-1-Collisions] ) have cast doubt on the long-term collision resistance of SHA-1. Therefore, SHA-1 support is required in this specification only for backwards-compatibility reasons.

5.8.1 SHA1

Identifier:: http://www.w3.org/2000/09/xmldsig#sha1

The SHA-1 algorithm [FIPS-180-3] takes no explicit parameters. An example of an SHA-1 DigestMethod element is:

A SHA-1 digest is a 160-bit string. The content of the DigestValue element shall be the base64 encoding of this bit string viewed as a 20-octet octet stream. For example, the DigestValue element for the message digest:

Example 9

A9993E36 4706816A BA3E2571 7850C26C 9CD0D89D

from Appendix A of the SHA-1 standard would be:

5.8.2 SHA256

Identifier:: http://www.w3.org/2001/04/xmlenc#sha256

The SHA-256 algorithm [FIPS-180-3] takes no explicit parameters. An example of an SHA-256 DigestMethod element is:

A SHA-256 digest is a 256-bit string. The content of the DigestValue element shall be the base64 encoding of this bit string viewed as a 32-octet octet stream.

5.8.3 SHA384

Identifier:: http://www.w3.org/2001/04/xmlenc#sha384

The SHA-384 algorithm [FIPS-180-3] takes no explicit parameters. An example of an SHA-384 DigestMethod element is:

A SHA-384 digest is a 384-bit string. The content of the DigestValue element shall be the base64 encoding of this bit string viewed as a 48-octet octet stream.

5.8.4 SHA512

Identifier:: http://www.w3.org/2001/04/xmlenc#sha512

The SHA-512 algorithm [FIPS-180-3] takes no explicit parameters. An example of an SHA-512 DigestMethod element is:

A SHA-512 digest is a 512-bit string. The content of the DigestValue element shall be the base64 encoding of this bit string viewed as a 64-octet octet stream.

5.8.5 RIPEMD-160

Identifier:: http://www.w3.org/2001/04/xmlenc#ripemd160

The RIPEMD-160 algorithm [RIPEMD-160] takes no explicit parameters. An example of an RIPEMD-160 DigestMethod element is:

A RIPEMD-160 digest is a 160-bit string. The content of the DigestValue element shall be the base64 encoding of this bit string viewed as a 20-octet octet stream.

5.9 Canonicalization

A Canonicalization of XML is a method of consistently serializing XML into an octet stream as is necessary prior to encrypting XML.

5.9.1 Inclusive Canonicalization

Identifiers:: http://www.w3.org/TR/2001/REC-xml-c14n-20010315; http://www.w3.org/TR/2001/REC-xml-c14n-20010315#WithComments; http://www.w3.org/2006/12/xml-c14n11; http://www.w3.org/2006/12/xml-c14n11#WithComments

Canonical XML [XML-C14N11] is a method of serializing XML which includes the in scope namespace and xml namespace attribute context from ancestors of the XML being serialized.

If XML is to be encrypted and then later decrypted into a different environment and it is desired to preserve namespace prefix bindings and the value of attributes in the "xml" namespace of its original environment, then the canonical XML with comments version of the XML should be the serialization that is encrypted.

5.9.2 Exclusive Canonicalization

Identifiers:: http://www.w3.org/2001/10/xml-exc-c14n#; http://www.w3.org/2001/10/xml-exc-c14n#WithComments

Exclusive XML Canonicalization [XML-EXC-C14N] serializes XML in such a way as to include to the minimum extent practical the namespace prefix binding and xml namespace attribute context inherited from ancestor elements.

It is the recommended method where the outer context of a fragment which was signed and then encrypted may be changed. Otherwise the validation of the signature over the fragment may fail because the canonicalization by signature validation may include unnecessary namespaces into the fragment.

6. Security Considerations

6.1 Chosen-Ciphertext Attacks

A number of chosen-ciphertext attacks against implementations of this specification have been published and demonstrated. They all involve the following elements:

The attacker knows about the format of the cleartext.
The attacker is able to submit substantial numbers of ciphertext messages.
The attacker is able to send arbitrary ciphertext, based on previous results.
The attacker is able to force the server to use the same key (secret key by CBC-based attacks and server's private key by PKCS#1.5 attacks) for processing of the adapted ciphertext.
The server attempting to decrypt the ciphertext in some way signals whether the decrypted text is well-formed or not.

The attacker uses the knowledge of the format and the information about well-formedness to construct a series of ciphertext guesses which reveal the plaintext with much less work than brute force. Attacks of this type have been demonstrated against symmetric encryption using CBC mode [XMLENC-CBC-ATTACK][XMLENC-CBC-ATTACK-COUNTERMEASURES] and on PKCS#1 v1.5. Other future attacks can be expected whenever these conditions are met.

6.1.1 Attacks against the encrypted data (`<EncryptedData>` part)

Using the CBC-based chosen-ciphertext attacks, the attacker sends to the server an XML document with modified encrypted data in the symmetric part (<EncryptedData>). After a few requests, the attacker is able to get the whole cleartext without knowledge of the symmetric key.

It would seem that these attacks can be countered by by disrupting any of the conditions, however in practice only preventing condition 3 (sending arbitrary ciphertext) is fully effective. To counter condition 3, it is necessary for the decrypting system to require authenticated integrity protection over the ciphertext. However, unless the mechanism used is bound to the encryption key, there will no way to be sure that the signer is not attempting to recover the plaintext. The simplest and most efficient way to do this is to use an authenticating block mode, such as GCM. An alternative would be an HMAC based on the encryption key over the ciphertext, but it is less efficient and provides no advantages.

Other countermeasures are not likely to be effective. Limiting the number of messages presented or the number of messages using the same key is not practical in large server farms. Attackers can spread their attempts over different servers and long or short periods of time, to foil attempts to detect attacks in progress or determine the location of the attacker.

Signaling well-formedness can occur by emitting different messages for distinct security errors or by exhibiting timing differences. Implementations should avoid these practices, however that is not sufficient to prevent such attacks in an XML protocol environment, such as SOAP. Using a technique called encryption wrapping, the attacker can insert the ciphertext in some schema-legal part of the message. If the decryption code notices a format error, an error will be returned, but if not the message will be passed to the application which will ignore the bogus plaintext and ultimately respond with an application level success or failure message.

6.1.2 Attacks against the encrypted key (Bleichenbacher's Million question attack on PKCS#1.5)

The goal of the attacker applying the Bleichenbacher's attack is to get the symmetric secret key, which is encrypted in the <EncryptedKey> part. Afterward, he would be able to decrypt the whole data carried in the <EncryptedData> part.

The basic idea of this attack is to modify the data in the <EncryptedKey> part, send the document to the server, and observe if the modified ciphertext contains PKCS#1.5 conformant data. This can be done by:

Observing fault messages of the server notifying directly that the request was not PKCS#1.5 conformant (this should not happen).
Enlarging the data in the <EncryptedData> part and observing the timing differences between inclusion of PKCS-valid and PKCS-invalid keys: if the key is PKCS-valid, the session key is extracted, and the large data is decrypted. Otherwise, the session key cannot be extracted and the large data is not processed, which yields a timing difference.
Making specific modifications of the <EncryptedData> part based on CBC and padding-properties.

These problems are described in detail in RFC 3218 [RFC3218].

The most effective countermeasure against the timing attack (2) is to generate a random secret key every time when the decrypted data was not PKCS#1-conformant. This way, the attacker would not get any timing side-channel.

Please note however that this is not a valid countermeasure against the specific modification of the <EncryptedData> described in part (3). The attacker could still use a few millions of requests to decrypt the encrypted symmetric key. Therefore, we recommend the usage of RSA-OAEP. RSA-OAEP also has a risk of a chosen ciphertext attack [OAEP-ATTACK] which can be mitigated in security library implementations.

6.1.3 Backwards Compatibility Attacks

Use of state-of-the-art and secure encryption algorithms such as RSA-OAEP and AES-GCM can become insecure when the adversary can force the server to process eavesdropped ciphertext with legacy algorithms such as RSA-PKCS#1 v1.5 or AES-CBC [XMLENC-BACKWARDS-COMP]:

The attacker may be able to break the security of an AES-GCM ciphertext if he is able to force the server to process the ciphertext with AES-CBC and the same symmetric key.
The attacker may be able to decrypt an RSA-OAEP ciphertext if he is able to force the server to process the ciphertext with RSA-PKCS#1 v1.5 and the same asymmetric key.
The attacker may be able to forge valid server signatures if the server decrypts RSA-PKCS#1 v1.5 ciphertexts and the signatures are computed with the same asymmetric key pair.

Accordingly, in situations where an attacker may be able to mount chosen-ciphertext attacks, we recommend the following to implementers:

Implementations should always use a different public key pair for data confidentiality and for data integrity functionality.
Implementations using symmetric keys should not use the same key material for different algorithms, even if serving the same purpose. Key derivation based on a single key and the algorithm identifier can be used to accomplish this, for example.
Implementations that plan to use the same symmetric key for both confidentiality and integrity functions should use it as the basis for a key derivation producing different keys for those functions.
Implementations should restrict algorithm usage to algorithms known to be secure in the face of chosen-ciphertext attacks (RSA-OAEP, AES-GCM). In that case, documents containing RSA-PKCS#1 v1.5 [XMLENC-PKCS15-ATTACK] and AES-CBC [XMLENC-CBC-ATTACK] ciphertexts should be rejected without decryption.

6.2 Relationship to XML Digital Signatures

The application of both encryption and digital signatures over portions of an XML document can make subsequent decryption and signature verification difficult. In particular, when verifying a signature one must know whether the signature was computed over the encrypted or unencrypted form of elements.

A separate, but important, issue is introducing cryptographic vulnerabilities when combining digital signatures and encryption over a common XML element. Hal Finney has suggested that encrypting digitally signed data, while leaving the digital signature in the clear, may allow plaintext guessing attacks. This vulnerability can be mitigated by using secure hashes and the nonces in the text being processed.

In accordance with the requirements document [XML-ENCRYPTION-REQ] the interaction of encryption and signing is an application issue and out of scope of the specification. However, we make the following recommendations:

When data is encrypted, any digest or signature over that data should be encrypted. This satisfies the first issue in that only those signatures that can be seen can be validated. It also addresses the possibility of a plaintext guessing vulnerability, though it may not be possible to identify (or even know of) all the signatures over a given piece of data.
Employ the "decrypt-except" signature transform [XMLENC-DECRYPT]. It works as follows: during signature transform processing, if you encounter a decrypt transform, decrypt all encrypted content in the document except for those excepted by an enumerated set of references.

Additionally, while the following warnings pertain to incorrect inferences by the user about the authenticity of information encrypted, applications should discourage user misapprehension by communicating clearly which information has integrity, or is authenticated, confidential, or non-repudiable when multiple processes (e.g., signature and encryption) and algorithms (e.g., symmetric and asymmetric) are used:

When an encrypted envelope contains a signature, the signature does not necessarily protect the authenticity or integrity of the ciphertext [Davis].
While the signature secures plaintext it only covers that which is signed, recipients of encrypted messages must not infer integrity or authenticity of other unsigned information (e.g., headers) within the encrypted envelope, see [XMLDSIG-CORE1], section 8.1.1 Only What is Signed is Secure].

6.3 Information Revealed

Where a symmetric key is shared amongst multiple recipients, that symmetric key should only be used for the data intended for all recipients; even if one recipient is not directed to information intended (exclusively) for another in the same symmetric key, the information might be discovered and decrypted.

Additionally, application designers should be careful not to reveal any information in parameters or algorithm identifiers (e.g., information in a URI) that weakens the encryption.

6.4 Nonce and IV (Initialization Value or Vector)

An undesirable characteristic of many encryption algorithms and/or their modes is that the same plaintext when encrypted with the same key has the same resulting ciphertext. While this is unsurprising, it invites various attacks which are mitigated by including an arbitrary and non-repeating (under a given key) data with the plaintext prior to encryption. In encryption chaining modes this data is the first to be encrypted and is consequently called the IV (initialization value or vector).

Different algorithms and modes have further requirements on the characteristic of this information (e.g., randomness and secrecy) that affect the features (e.g., confidentiality and integrity) and their resistance to attack.

Given that XML data is redundant (e.g., Unicode encodings and repeated tags ) and that attackers may know the data's structure (e.g., DTDs and schemas) encryption algorithms must be carefully implemented and used in this regard.

For the Cipher Block Chaining (CBC) mode used by this specification, the IV must not be reused for any key and should be random, but it need not be secret. Additionally, under this mode an adversary modifying the IV can make a known change in the plain text after decryption. This attack can be avoided by securing the integrity of the plain text data, for example by signing it.

Note: CBC block encryption algorithms should not be used without consideration of possibly severe security risks.

For the Galois/Counter Mode (GCM) used by this specification, the IV must not be reused for any key and should be random, but it need not be secret.

6.5 Denial of Service

This specification permits recursive processing. For example, the following scenario is possible: EncryptedKey A requires EncryptedKey B to be decrypted, which itself requires EncryptedKey A! Or, an attacker might submit an EncryptedData for decryption that references network resources that are very large or continually redirected. Consequently, implementations should be able to restrict arbitrary recursion and the total amount of processing and networking resources a request can consume.

6.6 Unsafe Content

XML Encryption can be used to obscure, via encryption, content that applications (e.g., firewalls, virus detectors, etc.) consider unsafe (e.g., executable code, viruses, etc.). Consequently, such applications must consider encrypted content to be as unsafe as the unsafest content transported in its application context. Consequently, such applications may choose to (1) disallow such content, (2) require access to the decrypted form for inspection, or (3) ensure that arbitrary content can be safely processed by receiving applications.

6.7 Error Messages

Implementations should not provide detailed error responses related to security algorithm processing. Error messages should be limited to a generic error message to avoid providing information to a potential attacker related to the specifics of the algorithm implementation. For example, if an error occurs in decryption processing the error response should be a generic message providing no specifics on the details of the processing error.

6.8 Timing Attacks

It has been known for some time that it is feasible for an attacker to recover keys or cleartext by repeatedly sending chosen ciphertext and measuring the time required to process different requests with different types of errors. It has been demonstrated that attacks of this type are practical even when communicating over large and busy networks, especially if the receiver is willing to process large numbers of ciphertext blocks.

Implementers should ensure that distinct errors detected during security algorithm processing do not consume systematically different amounts of processing time from each other. Implementers should consult the technical literature for more details on specific attacks and recommended countermeasures.

Deployments should treat as suspect inputs when a large number of security algorithm processing errors are detected within a short period of time, especially in messages from the same origin.

6.9 CBC Block Encryption Vulnerability

Note: CBC block encryption algorithms should not be used without consideration of possibly severe security risks.

XML Encryption Syntax and Processing Version 1.1

W3C Proposed Recommendation 24 January 2013

Abstract

Status of This Document

Table of Contents

1. Introduction

1.1 Editorial and Conformance Conventions

1.2 Design Philosophy

1.3 Versions, Namespaces, URIs, and Identifiers

1.4 Acknowledgements

2. Encryption Overview and Examples

2.1 Encryption Granularity

2.1.1 Encrypting an XML Element

2.1.2 Encrypting XML Element Content (Elements)

2.1.3 Encrypting XML Element Content (Character Data)

2.1.4 Encrypting Arbitrary Data and XML Documents

2.1.5 Super-Encryption: Encrypting EncryptedData

2.2 EncryptedData and EncryptedKey Usage

2.2.1 EncryptedData with Symmetric Key (KeyName)

2.2.2 EncryptedKey (ReferenceList, ds:RetrievalMethod, CarriedKeyName)

3. Encryption Syntax

3.1 The EncryptedType Element

3.2 The EncryptionMethod Element

3.3 The CipherData Element

3.3.1 The CipherReference Element

3.4 The EncryptedData Element

3.5 Extensions to ds:KeyInfo Element

3.5.1 The EncryptedKey Element

3.5.2 The DerivedKey Element

3.5.3 The ds:RetrievalMethod Element

3.6 The ReferenceList Element

3.7 The EncryptionProperties Element

4. Processing Rules

4.1 Intended Application Model

4.2 Well-known Type parameter values

4.3 Encryption

4.4 Decryption

4.5 XML Encryption

4.5.1 A Decrypt Implementation (Non-normative)

4.5.2 A Decrypt and Replace Implementation (Non-normative)

4.5.3 Serializing XML (Non-normative)

4.5.3.1 Default Namespace Considerations

4.5.3.2 XML Attribute Considerations

4.5.4 Text Wrapping

5. Algorithms

5.1 Algorithm Identifiers and Implementation Requirements

5.1.1 Table of Algorithms

5.2 Block Encryption Algorithms

5.2.1 Padding

5.2.2 Triple DES

5.2.3 AES

5.2.4 AES-GCM

5.3 Stream Encryption Algorithms

5.4 Key Derivation

5.4.1 ConcatKDF

5.4.2 PBKDF2

5.5 Key Transport

5.5.1 RSA Version 1.5

5.5.2 RSA-OAEP

5.6 Key Agreement

5.6.1 Diffie-Hellman Key Values

5.6.2 Diffie-Hellman Key Agreement

5.6.2.1 Diffie-Hellman Key Agreement with Explicit Key Derivation Functions

5.6.2.2 Diffie-Hellman Key Agreement with Legacy Key Derivation Function

5.6.3 Elliptic Curve Diffie-Hellman (ECDH) Key Values

5.6.4 Elliptic Curve Diffie-Hellman (ECDH) Key Agreement (Ephemeral-Static Mode)

5.7 Symmetric Key Wrap

5.7.1 CMS Triple DES Key Wrap

5.7.2 AES KeyWrap

5.8 Message Digest

5.8.1 SHA1

5.8.2 SHA256

5.8.3 SHA384

5.8.4 SHA512

5.8.5 RIPEMD-160

5.9 Canonicalization

5.9.1 Inclusive Canonicalization

5.9.2 Exclusive Canonicalization

6. Security Considerations

6.1 Chosen-Ciphertext Attacks

2.2 `EncryptedData` and `EncryptedKey` Usage

2.2.1 `EncryptedData` with Symmetric Key (`KeyName`)

2.2.2 `EncryptedKey` (`ReferenceList`, `ds:RetrievalMethod`, `CarriedKeyName`)

3.1 The `EncryptedType` Element

3.2 The `EncryptionMethod` Element

3.3 The `CipherData` Element

3.3.1 The `CipherReference` Element

3.4 The `EncryptedData` Element

3.5 Extensions to `ds:KeyInfo` Element

3.5.1 The `EncryptedKey` Element

3.5.2 The `DerivedKey` Element

3.5.3 The `ds:RetrievalMethod` Element

3.6 The `ReferenceList` Element

3.7 The `EncryptionProperties` Element

4.2 Well-known `Type` parameter values

6.1.1 Attacks against the encrypted data (`<EncryptedData>` part)