Versioning XML Vocabularies
Editor's Note: An update to this article has been posted here on 10/27/04.
XML is designed for the creation of languages based upon self-describing markup. The inevitable evolution of these languages is called versioning. Versioning means adding, deleting, or changing parts of the language. Making versioning work in practice is one of the most difficult problems in computing, with a long history of failed attempts. Arguably one reason why the Web rose dramatically in popularity is because evolution and versioning were built into HTML and HTTP headers, each of which provides explict extensibility points and rules for understanding extensions that enabled their decentralized extension and versioning.
XML Namespaces provide an ideal mechanism for identifying versions of languages, and all XML schema languages -- such as W3C XML Schema -- provide for controlled extensibility.
This article describes techniques to achieve more effective loose coupling between systems by increasing the possibility for backwards- and forwards-compatible changes to occur when related systems evolve. These techniques are designed for compatible changes with or without schema propagation. A number of rules are described for versioning XML vocabularies, making use of XML Namespaces and XML Schema constructs. It includes rules for working with languages that provide an extensible container model, notably SOAP. The collective set of rules is called the "Must Ignore" pattern of extensibility. Strangely, the "Must Ignore" pattern for HTML tags and HTTP headers that significantly helped their adoption has not been widely adopted by XML practitioners. This article aims to rectify that situation for current Schema validation environments. Follow-on material will explore innovative relaxed schema validation environments.
FOLDOC [1] provides definitions of backwards and forwards compatibility. This article will reprise those definitions and focus on the actions of messages, specifically senders and receivers instead of clients and servers. Backwards compatibility means that a new version of a receiver can be rolled out so it does not break existing senders. This means that a sender can send an old version of a message to a receiver that understands the new version and still have the message successfully processed. Forwards compatibility means that an older version of a receiver can consume newer messages and not break. Of course the older version will not implement any new behavior, but a sender can send a newer version of a message and still have the message successfully processed.
In other words, backwards compatibility means that existing senders can use services that have been updated, and forwards compatibility means that newer senders can continue to use existing services.
Forwards-compatible changes typically involve adding optional element(s) and/or attribute(s). The costs associated with introducing changes that are not backwards- or forwards-compatible are often very high, typically requiring deployed software to be updated to accommodate the newer version. The rules described below optimize for the case of making backwards- and forwards-compatible changes. These article argues that this means not changing the namespace name or changing element names.
Compatibility is defined for the sender and receiver of an individual message. However, most web service specifications provide definitions of inputs and outputs. In these definitions of compatibility, a web service that updates its output schema is considered a newer sender. This simply reverses the sender/receiver terminology of input messages when applying compatibility definitions to output messages. If a receiver updates the schema of the output message, then it is "sending" a newer version of the message, hence it is considered a "sender".
Designing extensibility into languages has typically resulted in systems that are more loosely coupled. Extensibility allows senders to change instances without going through a centralized authority. Thus, the first rule relating to extensibility is
1. Allow Extensibility rule: Languages SHOULD be designed for extensibility.
A fundamental requirement for extensibility is to be able to determine the language of elements and attributes. XML Namespaces [14] provide a mechanism for associating a URI with an XML element or attribute name, thus specifying the language of the name. This also serves to prevent name collisions.
W3C XML Schema [15] provides a
mechanism called a wildcard, <xs:any>, for
controlling where elements from certain namespaces are allowed. The
wildcard indicates that elements in specified namespaces are
allowed in instance documents where the wildcard occurs. This
allows for later extension of a schema in a well-defined manner.
Receivers of extended documents can identify and, depending upon
the processing model for the extensions, safely ignore the
extensions they don't understand.
<xs:any> uses the namespace
attribute to control what namespaces extension elements can come
from. The main values for this attribute are ##any,
which means one can extend the schema using an element from any
possible namespace; ##other, which only allows
extension elements from namespaces other than the current one; and
##targetnamespace, which only allows extension
elements from the current namespace.
<xs:any>uses the processContents
attribute to control how a XML parser validates extended elements.
Permissible methods include "lax", allowing validation to occur,
"strict", requiring validation, and "skip", to skip validation.
This article uses "lax" validation, as it is the most flexible and
is the typical choice for web services specifications.
The main goal of the "Must Ignore" pattern of extensibility is
to allow backwards- and forwards-compatible changes to documents. At
a minimum, this means neither changing nor adding namespace names,
nor changing element names. Adding element names in the target
namespace can only be done with the ##any namespace or
a combination of ##other namespace and the target
namespace of the vocabulary.
A number of examples illustrate the application of these rules. Imagine that a Purchase Order is being sent from one machine to another. The processing of the purchase order results in a "shipped" message. But that message could be sent quite some time after the Purchase order is received. It would be undesirable for the sending software to have to wait for a potentially arbitrary amount of time for the response (synchronous messaging). A preferable model is for the receiver to be able to send the "shipped" message under its own control, without the sender waiting. The receiver "calls back" to the initial sender, hence the term "callback". The sender provides an address to the receiver in a Callback address. This indicates the address that the receiver should use for sending any subsequent messages to the sender. In web services, this Callback would typically be sent as a SOAP Header block.
Our preference would be to have an extensible style of
##any. A callback type that uses this model is:
<s:complexType name="CallbackType">
<s:sequence>
<s:element name="callbackLocation" type="s:anyURI"
minOccurs="1" maxOccurs="1"/>
<s:any processContents="lax" minOccurs="0"
maxOccurs="unbounded" namespace="##any"/>
</s:sequence>
<s:anyAttribute/>
</s:complexType>Example 1 -- A schema using ##any for extensibility
However, the determinism constraint of W3C XML Schema, described in more detail later, prevents this model from working. The problem arises when a subsequent version of the callback adds an optional element. A timeout is a good example of this. The timeout of a callback is an informative piece of information for a receiver. Receivers can continue processing if they don't understand the timeout. The following schema is roughly what is desired using wildcards, but it is illegal because of the determinism constraint:
<s:complexType name="CallbackType">
<s:sequence>
<s:element name="callbackLocation" type="s:anyURI"
minOccurs="1" maxOccurs="1"/>
<s:element name="expires" type="s:dateTime"
minOccurs="0" maxOccurs="1"/>
<s:any processContents="lax" minOccurs="0"
maxOccurs="unbounded" namespace="##any"/>
</s:sequence>
<s:anyAttribute/>
</s:complexType>Example 2 -- An illegal schema
Since this pattern does not work, we need to create a design pattern that enables roughly the equivalent in order to achieve the original goals. For allowing new extensions in the same namespace, the author must create an extension type that allows extensions in the same namespace. The extension type should be used only for future compatible extensions in the same namespace. We need two more rules to allow proper versioning of XML language definitions. First the rule for namespaces:
2. Any Namespace rule: The extensibility point SHOULD allow for extension in any namespace. For XML Schema applications, the extensibility point SHOULD be an element that allows for extension in the target namespace and a wildcard that allows for extension in any other namespace.
The rule for allowing extensibility:
3. Full Extensibility rule: All XML Elements SHOULD allow for element extensibility after element definitions, and allow any attributes.
An example of a Callback type that follows these rules:
<s:complexType name="CallbackType">
<s:sequence>
<s:element name="callbackLocation" type="s:anyURI"
minOccurs="1" maxOccurs="1"/>
<s:element name="Extension" type="wscb:ExtensionType"
minOccurs="0" maxOccurs="1"/>
<s:any processContents="lax" minOccurs="0"
maxOccurs="unbounded" namespace="##other"/>
</s:sequence>
<s:anyAttribute/>
</s:complexType>
<s:complexType name="ExtensionType">
<s:sequence>
<s:any processContents="lax" minOccurs="1"
maxOccurs="unbounded" namespace="##targetnamespace"/>
</s:sequence>
<s:anyAttribute/>
</s:complexType>Example 3 -- A Callback type with extensibility
Because each extension in the targetnamespace is inside an Extension element, each subsequent target namespace extension will increase nesting by another layer. While this layer of nesting per extension is not desirable, it is what can be accomplished today when applying strict W3C XML Schema validation. It seems that having multiple nested elements is worthwhile if multiple compatible revisions can be made to a language. This technique allows validation of extensions in the targetnamespace, while retaining validation of the targetnamespace itself.
In general an extension can be defined by a new specification that makes a normative reference to the earlier specification and then defines the new element. No permission should be needed from the authors of the specification to make such an extension. In fact the major design point of XML namespaces is to allow decentralized extensions. The corollary is that permission is required for extensions in the same namespace. A namespace has an owner; non-owners changing the meaning of something can be harmful.
Senders should ideally be able to extend existing XML documents with new elements without receivers having to change existing implementations. Extensibility is one step toward this goal, but achieving compatibility also requires a processing model for the extensions. The behavior of software when it encounters an extension should be clear. Thus, we introduce the next rule:
4. Provide Processing Model Rule: Languages SHOULD specify a processing model for dealing with extensions.
The simplest processing model that enables compatible changes is to ignore content that is not understood. This rule is
5. Must Ignore Rule: Document receivers MUST ignore any XML attributes or elements in a valid XML document that they do not recognize.
This rule does not require that the elements be physically removed, only ignored for processing purposes. There is a great deal of historic usage of the Must Ignore rule. HTML 1.1, 2.0, and 3.2 follow the Must Ignore rule; they specify that any unknown start tags or end tags are mapped to nothing during tokenization. HTTP 1.1 [6] specifies that a receiver should ignore any headers it doesn't understand: "Unrecognized header fields SHOULD be ignored by the recipient and MUST be forwarded by transparent proxies." The Must Ignore rule for XML was first introduced in 1997 by the WebDAV working group [19], standardized in the WebDAV specification RFC 2518 [5] section 14 and later separately published as the Flexible XML Processing Profile [2].
There are two broad types of vocabularies relating to dealing with extensions. The types are data oriented and presentation (or document) applications. For data oriented applications, such as web services, the rule is
6. Must Ignore All Rule: The Must Ignore rule applies to unrecognized elements and their descendants.
For example, if a message is received with unrecognized elements in a SOAP header block, they must be ignored unless marked as "mustUnderstand" (see Rule 10 below), but it is reasonable to expect that the unrecognized elements might be written to a log file.
Document oriented vocabularies may need a different rule as the application will typically want to present the content of an unknown element. The rule for document-oriented applications is
7. Must Ignore Container Rule: The Must Ignore rule applies only to unrecognized elements
This retains the element descendants, such as for display purposes.
Instead of ignoring unrecognized components, a language may provide a different model for handling extensions. One model is that a receiver will generate a fault if it finds a component it doesn't understand. An example might be a security specification where a receiver must understand any extension. This suffers from the significant drawback that it does not allow compatible changes to occur in the language; changes can't be ignored. Another model is a fallback model, where alternate elements are provided if the receiver does not understand the extension. XSLT 2.0 provides such a model.
|
When a new version of a language is required, and it is backwards compatible with the older language, then the author must make a decision about the namespace name for names in the new language. There are two choices: create a new namespace name or reuse the existing namespace name. We argue that reusing is more efficient, and we will explore the problems with option #1 in the "new namespace" section. The reusing namespace rule is
8. Re-use namespace names Rule: If a backwards compatible change can be made to a specification, then the old namespace name SHOULD be used in conjunction with XML's extensibility model.
An important conclusion is that a new namespace name is only required when an incompatible change is made.
9. New namespaces to break Rule: A new namespace name is used when backwards compatibility is not permitted, that is software MUST break if it does not understand the new language components.
Non-backwards compatible changes typically occur in two ways: a required information item is added or the semantics of an existing information item are changed.
The reuse namespace names rule requires the previous Must Ignore and Any Namespace rules be followed. If these rules are not followed, then a language designer is precluded from making compatible changes and reusing the namespace name.
We've articulated that reusing namespace names for compatible
extensions are good practice. The counter position is that the
namespace owner could use a new namespace for the compatible
changes by providing extensibility points allowing other namespaces
-- <xs:any namespace="##other">. This technique
suffers from the problem that an extension in a different namespace
means that the combined schema cannot be fully validated.
Specifically, there is no way to create a new schema that constrains
the wildcard. For example, imagine that ns1 contains foo and bar. It
is not possible to take the SOAP schema -- an example of a schema with
a wildcard -- and require that ns1:foo element must be a child of the
header element and ns1:bar must not be a child of the header element
using just W3C XML Schema constructs. Indeed, the need for this
functionality spawned some of the WSDL functionality. The new
namespace name approach results in specifications and namespaces that
are inappropriately factored, as related constructs will be in
separate namespaces. Further, the reuse of the same namespace has
better tooling support. Many applications use a single schema to
create the equivalent programming constructs. These tools often work
best with single namespace support for the "generated" constructs. The
reuse of the namespace name allows at least the namespace author to
make changes to the namespace and perform validation of the
extensions.
Given adoption of the Must Ignore rule, it is often the case that the creator of an extension wants to require that the receiver understand the extension, overriding the Must Ignore rule.
10. Provide mustUnderstand Rule: Container languages SHOULD provide a "mustUnderstand" model for dealing with optionality of extensions that override a default Must Ignore Rule.
This rule and the Must Ignore rule work together to provide a
stable and flexible processing model for extensions. Arguably the
simplest and most flexible override technique is a
mustUnderstand flag that indicates whether the item
must be understood. The SOAP [7], WSDL
[8], and WS-Policy [10] attributes and values for specifying understand
are respectively: soap:mustUnderstand="1",
wsdl:required="1",
wsp:Usage="wsp:Required". SOAP is probably the most
common case of a container that provides a
mustUnderstand model. The default value is
0, which is effectively the Must Ignore rule.
A mustUnderstand flag allows the sender to insert extensions into the container and use the mustUnderstand attribute to override the must Ignore rule. This allows senders to extend messages without changing the extension element's parent's namespace, retaining backwards compatibility. Obviously the receiver must be extended to handle new extensions, but there is now a loose coupling between the language's processing model and the extension's processing model.
There are other techniques possible, such as providing an element that indicates which extension namespaces must be understood.
In some cases a language does not provide a mustUnderstand mechanism. In the absence of a mustUnderstand model, there is no way to force receivers to reject a message if they don't understand the extension namespace.
XML DTDs and W3C XML Schema have a rule that requires schemas to have deterministic content models. From the XML 1.0 specification,
For example, the content model((b, c) | (b, d))is non-deterministic, because given an initialbthe XML processor cannot know whichbin the model is being matched without looking ahead to see which element follows theb.
The use of ##any means there are some schemas that
we might like to express, but that aren't allowed.
##any, where
minOccurs does not equal maxOccurs, are
not allowed before an element declaration. An instance of the
element would be valid for the ##any or the element.
##other could be used.maxOccurs equals its minOccurs. If
these were different, say minOccurs="1" and
maxOccurs="2", then the optional occurrences could
match either the element definition or the ##any. As a
result of this rule, the minOccurs must be greater
than zero.any must be avoided. A derived type might add
an element definition after the wildcard, then an instance of the
added element definition could match either the wildcard or the
derived element definition.11. Be
Deterministic rule: Use of wildcards MUST be
deterministic. Location of wildcards, namespace of wildcard
extensions, minOccurs and maxOccurs values
are constrained, and type restriction is controlled.
As shown earlier, a common design pattern is to provide an
extensibility point -- not an element -- allowing any namespace at the
end of a type. This is typically done with <xs:any
namespace="##any">.
Determinism makes this unworkable as a complete solution in many
cases. Firstly, the extensibility point can only occur after required
elements in the original schema, limiting the scope of extensibility
in the original schema. Secondly, backwards compatible changes require
that the added element is optional, which means
a minOccurs="0". Determinism prevents us from placing
a minOccurs="0" before an extensibility point of
##any. Thus, when adding an element at an extensibility point, the
author can make the element optional and lose the extensibility point,
or the author can make the element required and lose backwards
compatibility.
We've shown that using XML and W3C XML Schema to achieve loose coupling via compatible changes that fully utilize yet do not require new schema definitions is hard. Following these extensibility rules leads to W3C XML Schema documents that are more cumbersome and at the same time less expressive than one might like. The structural limitations introduced by W3C XML Schema's handling of extensibility are a consequence of W3C XML Schema's design and are not an inherent limitation of schema-based structures.
With respect to W3C XML Schema, it would useful to be able to add
elements into arbitrary places, such as before other elements, but the
determinism constraint constrains this. A less restrictive type of
deterministic model could be employed, such as the "greedy" algorithm
defined in the URI specification [4]. This
would allow optional elements before wildcards and removing the need
for the Extension type we introduced. This still does not allow
wildcards before elements, as the wildcard would match the elements
instead. Further, this still does not allow wildcards and type
extension of the type to coexist. A "priority" wildcard model, where
an element that could be matched by a wildcard or an element would
match with an element if possible would allow wildcards before and
after element declarations. Additionally, a wildcard that only allowed
elements that had not been defined -- effectively other namespaces
plus anything not defined in the target namespace -- is another useful
model. These changes would also allow cleaner mixing of inheritance
and wildcards. But that still means that the author has to sprinkle
wildcards throughout their types. A type-level any
element combined with the aforementioned wildcard changes is
needed. One potential solution is that the sequence declaration could
have an attribute specifying that extensions be allowed in any place,
then a commensurate attributes specifying namespaces, elements, and
validation rules.
The problem with this last approach is that with a specific schema it is sometimes necessary to apply the same schema in a strict or relaxed fashion in different parts of a system. A long-standing rule for the Internet is the Robustness Principle, articulated in the Internet Protocol [3], as "In general, an implementation must be conservative in its sending behavior, and liberal in its receiving behavior". In schema validation terms, a sender can apply a schema in a strict way while a receiver can apply a schema in a relaxed way. In this case, the degree of strictness is not an attribute of the schema, but of how it is used. A solution that appears to solve these problems is to define a form of schema validation that permits an open content model that is used when schemas are versioned. We call this model validation "by projection", and it works by ignoring, rather than rejecting, component names that appear in a message that are not explicitly defined by the schema. We plan to explore this relaxed validation model in the future.
A final comment on W3C XML Schema extensibility is that there is still the unmet need to define schemas that validate known extensions while retaining extensibility. An author will want to create a schema based upon an extensible schema but mix in other known schemas in particular wildcards while retaining the wildcard extensibility. We encounter this difficulty in areas like describing SOAP header blocks. The topic of composing schemas from many schemas is difficult yet pressing.
Leaving the topic of wildcard extensibility, the use of type
extension over the Web might be more palatable if the instance
document could express a base type if the receiver does not understand
the extension type, as in xsi:basetype="". The receiver
could then fallback to using the basetype if it did not understand the
base type's extension.
Another area for architectural improvement is that XML -- or
even W3C XML Schema -- could have provided a
mustUnderstand model. As things stand, each vocabulary
that provides a mustUnderstand model reinvents the mU
wheel. XML could have provided an xml:mustUnderstand
attribute and model that each language could use. Tim Berners-Lee
articulated the need for this in XML in his design note on
mandatory extensions in Feb 2000[18],
but neither XML 1.0 nor 1.1 included this model.
Finally, there is ambiguity in compliance testing for W3C XML
Schema implementations. The W3C XML Schema test collection [16]
does not test some of the more common cases that have been precluded
here. For example, the wildcard tests cover a different style, which
is
xs:any inside a complex type. These do not cover some
of the non-deterministic cases, typically achieved by combining
minOccurs/maxOccurs variations with
##any or combining inheritance with
##any. Thus, some implementations do not correctly test
for non-determinism, which may yield non-interoperable documents.
One common concern is about implementation support for these features and combinations. These samples have been tried in many different schema parsers and toolkits, such as XML Beans, SQC, and JAX-RPC. While it's impossible to know whether all implementations support these rules, there seems to be good support for what was tested. The author is certainly interested in hearing about toolkits that don't support these rules.
The W3C TAG decided that the topic of versioning and extensibility is important enough to web architecture to work on a finding [20] and to include material into the Web Architecture document [21]. While this article provided a starting point for the TAG material, that material will cover a broader scope and progress in a more interactive and iterative fashion than an article can. Readers can follow the TAG material for an ongoing treatment of the area of extensibility and versioning.
This article describes a number of rules for using XML, W3C XML Schema, and XML Namespaces in language construction and extension. The main goal of the set of rules is to allow language designers to make backwards- and forwards-compatible changes to their languages in order to achieve loose coupling between systems.
To a certain degree, the technique described herein is a
combination of the ##any and ##other
designs with well-known rules to produce a design that achieves the
goals of compatible extensibility and versioning with validation
using W3C XML Schema. The namespace name owner can add backwards-
and forwards-compatible changes into the extensibility element
while retaining the ability to validate all components, and other
authors can add their changes at the ##other wildcard
location.
The author thanks the many reviewers that have contributed to the article, particularly David Bau, William Cox, Edd Dumbill, Chris Ferris, Yaron Goland, Hal Lockhart, Mark Nottingham, Jeffrey Schlimmer, Cliff Schmidt, and Norman Walsh. This article borrows, with permission of the authors, examples and some text from WS-Callback [9].
XML.com Copyright © 1998-2006 O'Reilly Media, Inc.