|
INTERNET-DRAFT draft-ietf-webdav-acl-12 Expires April 10, 2004 |
Geoffrey Clemm, IBM Anne Hopkins, Microsoft Corporation Eric Sedlar, Oracle Corporation Jim Whitehead, U.C. Santa Cruz
|
WebDAV Access Control Protocol
Status of this Memo
This document is an Internet-Draft and is subject to all provisions of Section 10 of RFC2026.
Internet-Drafts are working documents of the Internet Engineering Task Force (IETF), its areas, and its working groups. Note that other groups may also distribute working documents as Internet-Drafts.
Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress."
The list of current Internet-Drafts can be accessed at http://www.ietf.org/ietf/1id-abstracts.txt
The list of Internet-Draft Shadow Directories can be accessed at http://www.ietf.org/shadow.html.
Abstract
This document specifies a set of methods, headers, message bodies, properties, and reports that define Access Control extensions to the WebDAV Distributed Authoring Protocol. This protocol permits a client to read and modify access control lists that instruct a server whether to allow or deny operations upon a resource (such as HyperText Transfer Protocol (HTTP) method invocations) by a given principal. A lightweight representation of principals as Web resources supports integration of a wide range of user management repositories. Search operations allow discovery and manipulation of principals using human names.
This document is a product of the Web Distributed Authoring and Versioning (WebDAV) working group of the Internet Engineering Task Force. Comments on this draft are welcomed, and should be addressed to the acl@webdav.org mailing list. Other related documents can be found at http://www.example.com/acl/, and http://www.ics.uci.edu/pub/ietf/webdav/.
Table of Contents
3.7 DAV:read-current-user-privilege-set
Privilege
3.12 Aggregation of Predefined
Privileges
5.1.1 Example: Retrieving DAV:owner
5.1.2 Example: An Attempt to Set
DAV:owner
5.2 DAV:supported-privilege-set
5.2.1 Example: Retrieving a List of
Privileges Supported on a Resource
5.3 DAV:current-user-privilege-set
5.3.1 Example: Retrieving the User's
Current Set of Assigned Privileges
5.4.5 Example: Retrieving a Resource's
Access Control List
5.5.2 DAV:no-invert ACE Constraint
Example: Retrieving
DAV:acl-restrictions
5.7 DAV:principal-collection-set
5.7.1 Example: Retrieving
DAV:principal-collection-set
5.8 Example: PROPFIND to retrieve
access control properties
7 Access Control and
Existing Methods
8.1.3 Example: ACL method failure due to
protected ACE conflict
8.1.4 Example: ACL method failure due to an
inherited ACE conflict
8.1.5 Example: ACL method failure due to an
attempt to set grant and deny in a single ACE.
9.2 DAV:acl-principal-prop-set Report
9.2.1 Example: DAV:acl-principal-prop-set
Report
9.3 DAV:principal-match REPORT
9.3.1 Example: DAV:principal-match REPORT
9.4 DAV:principal-property-search
REPORT
9.4.2 Example: successful
DAV:principal-property-search REPORT
9.5 DAV:principal-search-property-set
REPORT
9.5.1 Example:
DAV:principal-search-property-set REPORT
11 Internationalization
Considerations
12.1 Increased Risk of Compromised Users
12.2 Risks of the DAV:read-acl and
DAV:current-user-privilege-set Privileges
12.3 No Foreknowledge of Initial ACL
19.1 WebDAV XML Document Type Definition
Addendum
19.2 WebDAV Method Privilege Table
(Normative)
The goal
of the WebDAV access control extensions is to provide an interoperable
mechanism for handling discretionary access control for content and metadata
managed by WebDAV servers. WebDAV
access control can be implemented on content repositories with security as
simple as that of a UNIX file system, as well as more sophisticated
models. The underlying principle of
access control is that who you are determines what operations you can perform
on a resource. The "who you are" is defined by a
"principal" identifier; users, client software, servers, and groups
of the previous have principal identifiers. The "operations you can
perform" are determined by a single "access control list" (ACL)
associated with a resource. An ACL
contains a set of "access control entries" (ACEs), where each ACE
specifies a principal and a set of privileges that are either granted or denied
to that principal. When a principal submits an operation (such as an HTTP or
WebDAV method) to a resource for execution, the server evaluates the ACEs in
the ACL to determine if the principal has permission for that operation.
Since
every ACE contains the identifier of a principal, client software operated by a
human must provide a mechanism for selecting this principal. This specification
uses http(s) scheme URLs to identify principals, which are represented as
WebDAV-capable resources. There is no guarantee that the URLs identifying
principals will be meaningful to a human. For example,
http://www.example.com/u/256432 and http://www.example.com/people/Greg.Stein
are both valid URLs that could be used to identify the same principal. To
remedy this, every principal resource has the DAV:displayname property
containing a human-readable name for the principal.
Since a
principal can be identified by multiple URLs, it raises the problem of
determining exactly which principal is being referenced in a given ACE. It is
impossible for a client to determine that an ACE granting the read privilege to
http://www.example.com/people/Greg.Stein also affects the principal at
http://www.example.com/u/256432. That is, a client has no mechanism for
determining that two URLs identify the same principal resource. As a result, this specification requires
clients to use just one of the many possible URLs for a principal when creating
ACEs. A client can discover which URL to use by retrieving the
DAV:principal-URL property (Section 4.2) from a principal resource. No matter which of the
principal's URLs is used with PROPFIND, the property always returns the same
URL.
With a
system having hundreds to thousands of principals, the problem arises of how to
allow a human operator of client software to select just one of these
principals. One approach is to use broad collection hierarchies to spread the
principals over a large number of collections, yielding few principals per
collection. An example of this is a two level hierarchy with the first level
containing 36 collections (a-z, 0-9), and the second level being another 36,
creating collections /a/a/, /a/b/, ..., /a/z/, such that a principal with last
name "Stein" would appear at /s/t/Stein. In effect, this pre-computes
a common query, search on last name, and encodes it into a hierarchy. The
drawback with this scheme is that it handles only a small set of predefined
queries, and drilling down through the collection hierarchy adds unnecessary
steps (navigate down/up) when the user already knows the principal's name.
While organizing principal URLs into a hierarchy is a valid namespace organization,
users should not be forced to navigate this hierarchy to select a principal.
This
specification provides the capability to perform substring searches over a
small set of properties on the resources representing principals. This permits
searches based on last name, first name, user name, job title, etc. Two
separate searches are supported, both via the REPORT method, one to search
principal resources (DAV:principal-property-search, Section 9.4), the other to determine which properties may be
searched at all (DAV:principal-search-property-set, Section 9.5).
Once a
principal has been identified in an ACE, a server evaluating that ACE must know
the identity of the principal making a protocol request, and must validate that
that principal is who they claim to be, a process known as authentication. This
specification intentionally omits discussion of authentication, as the HTTP
protocol already has a number of authentication mechanisms [RFC2617]. Some authentication mechanism (such as HTTP
Digest Authentication, which all WebDAV compliant implementations are required
to support) must be available to validate the identity of a principal.
The following issues are out of scope for this document:
·
Access
control that applies only to a particular property on a resource (excepting the
access control properties DAV:acl and DAV:current-user-privilege-set), rather
than the entire resource,
·
Role-based
security (where a role can be seen as a dynamically defined group of
principals),
·
Specification
of the ways an ACL on a resource is initialized,
·
Specification
of an ACL that applies globally to all resources, rather than to a particular
resource.
·
Creation
and maintenance of resources representing people or computational agents
(principals), and groups of these.
This
specification is organized as follows. Section 1.1 defines key concepts used throughout the
specification, and is followed by a more in-depth discussion of principals (Section
2), and privileges (Section 3). Properties defined on principals are specified in
Section 4, and access control properties for content resources
are specified in Section 5. The ways ACLs are to be evaluated is described in
section 6. Client discovery of access control capability using OPTIONS is
described in Section 7.1. Interactions between access control functionality
and existing HTTP and WebDAV methods are described in the remainder of Section 7. The access control setting method, ACL, is specified
in Section 8. Four reports that provide limited server-side
searching capabilities are described in Section 9. Sections on XML processing (Section 10), Internationalization considerations (Section 11), security considerations (Section 12), and authentication (Section 13) round out the specification. An appendix (Section 19.1) provides an XML Document Type Definition (DTD) for
the XML elements defined in the specification.
This draft uses the terms defined in HTTP [RFC2616] and WebDAV [RFC2518]. In addition, the following terms are defined:
principal
A "principal" is
a distinct human or computational actor that initiates access to network
resources. In this protocol, a
principal is an HTTP resource that represents such an actor.
group
A "group" is a principal that represents a set of other
principals.
privilege
A "privilege"
controls access to a particular set of HTTP operations on a resource.
aggregate privilege
An "aggregate privilege" is a privilege that contains a set of other privileges.
abstract privilege
The modifier "abstract", when applied to a privilege on a resource, means the privilege cannot be set in an access control element (ACE) on that resource .
access control list (ACL)
An "ACL" is a list of access control elements that define access control to a particular resource.
access control element (ACE)
An "ACE" either grants or denies a particular set of (non-abstract) privileges for a particular principal.
inherited ACE
An "inherited ACE" is an ACE that is dynamically shared from the ACL of another resource. When a shared ACE changes on the primary resource, it is also changed on inheriting resources.
protected property
A "protected property" is one whose value cannot be updated except by a method explicitly defined as updating that specific property. In particular, a protected property cannot be updated with a PROPPATCH request.
The augmented BNF used by this document to describe protocol elements is described in Section 2.1 of [RFC2616]. Because this augmented BNF uses the basic production rules provided in Section 2.2 of [RFC2616], those rules apply to this document as well.
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in [RFC2119].
Definitions of XML elements in this document use XML element type declarations (as found in XML Document Type Declarations), described in Section 3.2 of [REC-XML]. When an XML element type in the "DAV:" namespace is referenced in this document outside of the context of an XML fragment, the string "DAV:" will be prefixed to the element name.
A principal is a network resource
that represents a distinct human or computational actor that initiates access
to network resources. Users and groups are represented as principals in many
implementations; other types of principals are also possible. A URI of any
scheme MAY be used to identify a principal resource. However, servers
implementing this specification MUST expose principal resources at an http(s)
URL, which is a privileged scheme that points to resources that have additional
properties, as described in Section 4. So, a principal resource can have multiple URIs, one
of which has to be an http(s) scheme URL. Although an implementation SHOULD
support PROPFIND and MAY support PROPPATCH to access and modify information
about a principal, it is not required to do so.
A principal resource may be a
group, where a group is a principal that represents a set of other principals,
called the members of the group. If a
person or computational agent matches a principal resource that is a member of
a group, they also match the group. Membership in a group is recursive, so if a
principal is a member of group GRPA, and GRPA is a member of group GRPB, then
the principal is also a member of GRPB.
Ability to perform a given method
on a resource MUST be controlled by one or more privileges. Authors of protocol extensions that define
new HTTP methods SHOULD specify which privileges (by defining new privileges,
or mapping to ones below) are required to perform the method. A principal with no privileges to a resource
MUST be denied any HTTP access to that resource, unless the principal matches
an ACE constructed using the DAV:all, DAV:authenticated, or DAV:unauthenticated
pseudo-principals (see Section 5.4.1). Servers
MUST report a 403 "Forbidden" error if access is denied, except in
the case where the privilege restricts the ability to know the resource exists,
in which case 404 "Not Found" may be returned.
Privileges may be containers of
other privileges, in which case they are termed "aggregate privileges". If a principal is granted or denied an
aggregate privilege, it is semantically equivalent to granting or denying each
of the aggregated privileges individually.
For example, an implementation may define add-member and remove-member
privileges that control the ability to add and remove a member of a group. Since these privileges control the ability
to update the state of a group, these privileges would be aggregated by the DAV:write
privilege on a group, and granting the DAV:write privilege on a group would
also grant the add-member and remove-member privileges.
Privileges may be declared to be
"abstract" for a given resource, in which case they cannot be
set in an ACE on that resource. Aggregate and non-aggregate privileges are both
capable of being abstract. Abstract privileges are useful for modeling
privileges that otherwise would not be exposed via the protocol. Abstract
privileges also provide server implementations with flexibility in implementing
the privileges defined in this specification.
For example, if a server is incapable of separating the read resource
capability from the read ACL capability, it can still model the DAV:read and
DAV:read-acl privileges defined in this specification by declaring them
abstract, and containing them within a non-abstract aggregate privilege (say,
read-all) that holds DAV:read, and DAV:read-acl. In this way, it is possible to
set the aggregate privilege, read-all, thus coupling the setting of DAV:read
and DAV:read-acl, but it is not possible to set DAV:read, or DAV:read-acl
individually. Since aggregate privileges can be abstract, it is also possible
to use abstract privileges to group or organize
non-abstract privileges. Privilege containment loops are not allowed;
therefore, a privilege MUST NOT contain itself. For example, DAV:read cannot
contain DAV:read.
The set of privileges that apply to
a particular resource may vary with the DAV:resourcetype of the resource, as
well as between different server implementations. To promote interoperability, however, this specification defines
a set of well-known privileges (e.g. DAV:read, DAV:write, DAV:read-acl,
DAV:write-acl, DAV:read-current-user-privilege-set, and DAV:all), which can at
least be used to classify the other privileges defined on a particular
resource. The access permissions on null resources (defined in [RFC2518],
Section 3) are solely those they inherit (if any), and they are not
discoverable (i.e., the access control properties specified in Section 5 are not defined on null resources). On the transition
from null to stateful resource, the initial access control list is set by the
server's default ACL value policy (if any).
Server implementations MAY define
new privileges beyond those defined in this specification. Privileges defined
by individual implementations MUST NOT use the DAV: namespace, and instead
should use a namespace that they control, such as an http scheme URL.
The read privilege controls methods
that return information about the state of the resource, including the
resource's properties. Affected methods include GET and PROPFIND. Any implementation-defined privilege that
also controls access to GET and PROPFIND must be aggregated under DAV:read—if
an ACL grants access to DAV:read, the client may expect that no other privilege
needs to be granted to have access to GET and PROPFIND. Additionally, the read privilege MUST
control the OPTIONS method.
<!ELEMENT read EMPTY>
The write privilege controls
methods that lock a resource or modify the content, dead properties, or (in the
case of a collection) membership of the resource, such as PUT and
PROPPATCH. Note that state modification
is also controlled via locking (see section 5.3 of [WEBDAV]), so effective
write access requires that both write privileges and write locking requirements
are satisfied. Any
implementation-defined privilege that also controls access to methods modifying
content, dead properties or collection membership must be aggregated under
DAV:write, e.g. if an ACL grants access to DAV:write, the client may expect
that no other privilege needs to be granted to have access to PUT and
PROPPATCH.
<!ELEMENT write EMPTY>
The
DAV:write-properties privilege controls methods that modify the dead properties
of the resource, such as PROPPATCH.
Whether this privilege may be used to control access to any live
properties is determined by the implementation. Any implementation-defined privilege that also controls
access to methods modifying dead properties must be aggregated under
DAV:write-properties—e.g. if an ACL grants access to DAV:write-properties, the
client can safely expect that no other privilege needs to be granted to have
access to PROPPATCH.
<!ELEMENT write-properties
EMPTY>
The
DAV:write-content privilege controls methods that modify the content
or (in the case of a collection) membership of the resource, such as PUT and
DELETE. Any implementation-defined
privilege that also controls access to content or
alteration of collection membership must be aggregated under
DAV:write-content—e.g. if an ACL grants access to DAV:write-content, the client
can safely expect that no other privilege needs to be granted to have access to
PUT or DELETE.
<!ELEMENT write-content
EMPTY>
The DAV:unlock privilege controls the use of the UNLOCK method by a principal other than the lock owner (the principal that created a lock can always perform an UNLOCK). While the set of users who may lock a resource is most commonly the same set of users who may modify a resource, servers may allow various kinds of administrators to unlock resources locked by others. Any privilege controlling access by non-lock owners to UNLOCK MUST be aggregated under DAV:unlock.
A lock owner can always remove a lock by issuing an UNLOCK with the correct lock token and authentication credentials. That is, even if a principal does not have DAV:unlock privilege, they can still remove locks they own. Principals other than the lock owner can remove a lock only if they have DAV:unlock privilege and they issue an UNLOCK with the correct lock token. Lock timeout is not affected by the DAV:unlock privilege.
<!ELEMENT unlock EMPTY>
The DAV:read-acl privilege controls
the use of PROPFIND to retrieve the DAV:acl property of the resource.
<!ELEMENT read-acl EMPTY>
The
DAV:read-current-user-privilege-set privilege controls the use of PROPFIND to
retrieve the DAV:current-user-privilege-set property of the resource.
Clients are intended to use this
property to visually indicate in their UI items that are dependent on the
permissions of a resource, for example, by graying out resources that are not
writeable.
This privilege is separate from
DAV:read-acl because there is a need to allow most users access to the
privileges permitted the current user (due to its use in creating the UI),
while the full ACL contains information that may not be appropriate for the
current authenticated user. As a result, the set of users who can view the full
ACL is expected to be much smaller than those who can read the current user
privilege set, and hence distinct privileges are needed for each.
<!ELEMENT read-current-user-privilege-set
EMPTY>
The DAV:write-acl privilege
controls use of the ACL method to modify the DAV:acl property of the resource.
<!ELEMENT write-acl EMPTY>
The
DAV:bind privilege allows a method to add a new member URL to the specified
collection (for example via PUT or MKCOL).
It is ignored for resources that are not collections.
<!ELEMENT
bind EMPTY>
The
DAV:unbind privilege allows a method to remove a member URL from the specified collection
(for example via DELETE or MOVE). It is
ignored for resources that are not collections.
<!ELEMENT
unbind EMPTY>
DAV:all is an aggregate privilege
that contains the entire set of privileges that can be applied to the resource.
<!ELEMENT all EMPTY>
Server implementations are free to
aggregate the predefined privileges (defined above in Sections 3.1-3.9) subject to the following limitations:
DAV:read-acl MUST NOT contain
DAV:read, DAV:write, DAV:write-acl, DAV:write-properties, DAV:write-content, or
DAV:read-current-user-privilege-set.
DAV:write-acl MUST NOT contain
DAV:write, DAV:read, DAV:read-acl, or DAV:read-current-user-privilege-set.
DAV:read-current-user-privilege-set
MUST NOT contain DAV:write, DAV:read, DAV:read-acl, or DAV:write-acl.
DAV:write MUST NOT contain
DAV:read, DAV:read-acl, or DAV:read-current-user-privilege-set.
DAV:read MUST NOT contain
DAV:write, DAV:write-acl, DAV:write-properties, or
DAV:write-content.
DAV:write
MUST contain DAV:write-properties and DAV:write-content.
Principals are manifested to
clients as a WebDAV resource, identified by a URL. A principal MUST have a non-empty DAV:displayname property
(defined in Section 13.2 of [RFC2518]), and a DAV:resourcetype property
(defined in Section 13.9 of [RFC2518]).
Additionally, a principal MUST report the DAV:principal XML element in
the value of the DAV:resourcetype property.
The element type declaration for DAV:principal is:
<!ELEMENT principal EMPTY>
This protocol defines the following
additional properties for a principal. Since it can be expensive for a server
to retrieve access control information, the name and value of these properties
SHOULD NOT be returned by a PROPFIND allprop request (as defined in Section
12.14.1 of [RFC2518]).
This protected property, if non-empty, contains the URIs of network resources with additional descriptive information about the principal. This property identifies additional network resources (i.e., it contains one or more URIs) that may be consulted by a client to gain additional knowledge concerning a principal. One expected use for this property is the storage of an LDAP [RFC2255] scheme URL. A user-agent encountering an LDAP URL could use LDAP [RFC2589] to retrieve additional machine-readable directory information about the principal, and display that information in its user interface. Support for this property is REQUIRED, and the value is empty if no alternate URI exists for the principal.
<!ELEMENT alternate-URI-set (href*)>
A principal may have many URLs, but there must be one "principal URL" that clients can use to uniquely identify a principal. This protected property contains the URL that MUST be used to identify this principal in an ACL request. Support for this property is REQUIRED.
<!ELEMENT principal-URL (href)>
This property of a group principal identifies the principals that are direct members of this group. Since a group may be a member of another group, a group may also have indirect members (i.e. the members of its direct members). A URL in the DAV:group-member-set for a principal MUST be the DAV:principal-URL of that principal.
<!ELEMENT group-member-set (href*)>
This protected property identifies the groups in which the principal is directly a member. Note that a server may allow a group to be a member of another group, in which case the DAV:group-membership of those other groups would need to be queried in order to determine the groups in which the principal is indirectly a member. Support for this property is REQUIRED.
<!ELEMENT group-membership (href*)>
This specification defines a number
of new properties for WebDAV resources.
Access control properties may be retrieved just like other WebDAV
properties, using the PROPFIND method.
Since it is expensive, for many servers, to retrieve access control
information, a PROPFIND allprop request (as defined in Section 12.14.1 of
[RFC2518]) SHOULD NOT return the names and values of the properties defined in
this section.
Access control properties
(especially DAV:acl and DAV:inherited-acl-set) are defined on the resource
identified by the Request-URI of a PROPFIND request. A direct consequence is
that if the resource is accessible via multiple URI, the value of access
control properties is the same across these URI.
HTTP resources that support the
WebDAV Access Control Protocol MUST contain the following properties. Null
resources (described in Section 3 of [RFC2518]) MUST NOT contain the following
properties.
This protected
property identifies a particular principal as being the "owner" of
the resource. Since the owner of a resource often has special access control
capabilities (e.g., the owner frequently has permanent DAV:write-acl
privilege), clients might display the resource owner in their user interface.
<!ELEMENT owner (href)>
This example shows a client request
for the value of the DAV:owner property from a collection resource with URL
http://www.example.com/papers/. The principal making the request is
authenticated using Digest authentication. The value of DAV:owner is the URL
http://www.example.com/acl/users/gstein, wrapped in the DAV:href XML element.
>> Request <<
PROPFIND /papers/ HTTP/1.1
Host: www.example.com
Content-type: text/xml; charset="utf-8"
Content-Length: xxx
Depth: 0
Authorization: Digest
username="jim",
realm="jim@webdav.org", nonce="...",
uri="/papers/", response="...",
opaque="..."
<?xml version="1.0"
encoding="utf-8" ?>
<D:propfind
xmlns:D="DAV:">
<D:prop>
<D:owner/>
</D:prop>
</D:propfind>
>> Response <<
HTTP/1.1 207 Multi-Status
Content-Type: text/xml;
charset="utf-8"
Content-Length: xxx
<?xml version="1.0" encoding="utf-8" ?>
<D:multistatus xmlns:D="DAV:">
<D:response>
<D:href>http://www.example.com/papers/</D:href>
<D:propstat>
<D:prop>
<D:owner>
<D:href>http://www.example.com/acl/users/gstein</D:href>
</D:owner>
</D:prop>
<D:status>HTTP/1.1 200
OK</D:status>
</D:propstat>
</D:response>
</D:multistatus>
The following example shows a
client request to modify the value of the DAV:owner property on the resource
with URL <http://www.example.com/papers>. Since DAV:owner is a protected
property, the server responds with a 207 (Multi-Status) response that contains
a 403 (Forbidden) status code for the act of setting DAV:owner. Section 8.2.1
of [RFC2518] describes PROPPATCH status code information, and Section 11 of
[RFC2518] describes the Multi-Status response.
>>
Request <<
PROPPATCH /papers/ HTTP/1.1
Host: www.example.com
Content-type: text/xml; charset="utf-8"
Content-Length: xxx
Depth: 0
Authorization: Digest
username="jim",
realm="jim@webdav.org", nonce="...",
uri="/papers/", response="...",
opaque="..."
<?xml
version="1.0" encoding="utf-8" ?>
<D:propertyupdate xmlns:D="DAV:">
<D:set>
<D:prop>
<D:owner>
<D:href>http://www.example.com/acl/users/jim</D:href>
</D:owner>
</D:prop>
</D:set>
</D:propertyupdate>
>> Response <<
HTTP/1.1 207 Multi-Status
Content-Type: text/xml;
charset="utf-8"
Content-Length: xxx
<?xml version="1.0" encoding="utf-8" ?>
<D:multistatus xmlns:D="DAV:">
<D:response>
<D:href>http://www.example.com/papers/</D:href>
<D:propstat>
<D:prop><D:owner/></D:prop>
<D:status>HTTP/1.1 403
Forbidden</D:status>
<D:responsedescription>
Failure to set protected property (DAV:owner)
</D:responsedescription>
</D:propstat>
</D:response>
</D:multistatus>
This is a protected property that
identifies the privileges defined for the resource.
<!ELEMENT supported-privilege-set (supported-privilege*)>
Each privilege appears as an XML element, where aggregate privileges list as sub-elements all of the privileges that they aggregate.
<!ELEMENT supported-privilege
(privilege, abstract?, description, supported-privilege*)>
<!ELEMENT privilege ANY>
An abstract privilege MUST NOT be used in an ACE for that resource. Servers MUST fail an attempt to set an abstract privilege.
<!ELEMENT abstract EMPTY>
A description is a human-readable description of what this privilege controls access to. Servers MUST indicate the human language of the description using the xml:lang attribute and SHOULD consider the HTTP Accept-Language request header when selecting one of multiple available languages.
<!ELEMENT description #PCDATA>
It is envisioned that a WebDAV
ACL-aware administrative client would list the supported privileges in a dialog
box, and allow the user to choose non-abstract privileges to apply in an
ACE. The privileges tree is useful
programmatically to map well-known privileges (defined by WebDAV or other
standards groups) into privileges that are supported by any particular server
implementation. The privilege tree also
serves to hide complexity in implementations allowing large number of
privileges to be defined by displaying aggregates to the user.
This example shows a client request
for the DAV:supported-privilege-set property on the resource
http://www.example.com/papers/. The value of the DAV:supported-privilege-set
property is a tree of supported privileges (using "[XML Namespace ,
localname]" to identify each privilege):
[DAV:, all] (aggregate, abstract)
|
+-- [DAV:, read] (aggregate)
|
+-- [DAV:, read-acl] (abstract)
+-- [DAV:, read-current-user-privilege-set] (abstract)
|
+-- [DAV:, write] (aggregate)
|
+-- [DAV:, write-acl] (abstract)
+-- [DAV:, write-properties]
+-- [DAV:, write-content]
|
+-- [DAV:, unlock]
This privilege tree is not normative
(except that it reflects the normative aggregation rules given in Section 3.12), and many possible privilege trees are possible.
>>
Request <<
PROPFIND /papers/ HTTP/1.1
Host: www.example.com
Content-type: text/xml; charset="utf-8"
Content-Length: xxx
Depth: 0
Authorization: Digest
username="gclemm",
realm="gclemm@webdav.org", nonce="...",
uri="/papers/", response="...",
opaque="..."
<?xml version="1.0"
encoding="utf-8" ?>
<D:propfind
xmlns:D="DAV:">
<D:prop>
<D:supported-privilege-set/>
</D:prop>
</D:propfind>
>> Response <<
HTTP/1.1 207 Multi-Status
Content-Type: text/xml;
charset="utf-8"
Content-Length: xxx
<?xml version="1.0" encoding="utf-8" ?>
<D:multistatus xmlns:D="DAV:">
<D:response>
<D:href>http://www.example.com/papers/</D:href>
<D:propstat>
<D:prop>
<D:supported-privilege-set>
<D:supported-privilege>
<D:privilege> <D:all/> </D:privilege>
<D:abstract/>
<D:description xml:lang="en">Any operation</D:description>
<D:supported-privilege>
<D:privilege> <D:read/> </D:privilege>
<D:description xml:lang="en">Read any object</D:description>
<D:supported-privilege>
<D:privilege> <D:read-acl/> </D:privilege>
<D:abstract/>
<D:description xml:lang="en">Read ACL</D:description>
</D:supported-privilege>
<D:supported-privilege>
<D:privilege>
<D:read-current-user-privilege-set/>
</D:privilege>
<D:abstract/>
<D:description xml:lang="en">Read current user privilege set property</D:description>
</D:supported-privilege>
</D:supported-privilege>
<D:supported-privilege>
<D:privilege> <D:write/> </D:privilege>
<D:description xml:lang="en">Write any object</D:description>
<D:supported-privilege>
<D:privilege> <D:write-acl/> </D:privilege>
<D:description xml:lang="en">Write ACL</D:description>
<D:abstract/>
</D:supported-privilege>
<D:supported-privilege>
<D:privilege> <D:write-properties/> </D:privilege>
<D:description xml:lang="en">Write properties</D:description>
</D:supported-privilege>
<D:supported-privilege>
<D:privilege> <D:write-content/> </D:privilege>
<D:description xml:lang="en">Write resource content</D:description>
</D:supported-privilege>
</D:supported-privilege>
<D:supported-privilege>
<D:privilege> <D:unlock/> </D:privilege>
<D:description xml:lang="en">Unlock resource</D:description>
</D:supported-privilege>
</D:supported-privilege>
</D:supported-privilege-set>
</D:prop>
<D:status>HTTP/1.1 200 OK</D:status>
</D:propstat>
</D:response>
</D:multistatus>
DAV:current-user-privilege-set
is a protected property containing the exact set of privileges (as computed by
the server) granted to the currently authenticated HTTP user. Aggregate
privileges and their contained privileges are listed. A user-agent can use the
value of this property to adjust its user interface to make actions
inaccessible (e.g., by graying out a menu item or button) for which the current
principal does not have permission. This property is also useful for
determining what operations the current principal can perform, without having
to actually execute an operation.
<!ELEMENT current-user-privilege-set (privilege*)>
<!ELEMENT privilege ANY>
If the current
user is granted a specific privilege, that privilege must belong to the set of
privileges that may be set on this resource. Therefore, each element in the
DAV:current-user-privilege-set property MUST identify a non-abstract privilege
from the DAV:supported-privilege-set property.
Continuing the example from Section 5.2.1, this example shows a client requesting the DAV:current-user-privilege-set property from the resource with URL http://www.example.com/papers/. The username of the principal making the request is "khare", and Digest authentication is used in the request. The principal with username "khare" has been granted the DAV:read privilege. Since the DAV:read privilege contains the DAV:read-acl and DAV:read-current-user-privilege-set privileges (see Section 5.2.1), the principal with username "khare" can read the ACL property, and the DAV:current-user-privilege-set property. However, the DAV:all, DAV:read-acl, DAV:write-acl and DAV:read-current-user-privilege-set privileges are not listed in the value of DAV:current-user-privilege-set, since (for this example) they are abstract privileges. DAV:write is not listed since the principal with username "khare" is not listed in an ACE granting that principal write permission.
>>
Request <<
PROPFIND /papers/ HTTP/1.1
Host: www.example.com
Content-type: text/xml; charset="utf-8"
Content-Length: xxx
Depth: 0
Authorization: Digest
username="khare",
realm="khare@webdav.org", nonce="...",
uri="/papers/", response="...",
opaque="..."
<?xml version="1.0"
encoding="utf-8" ?>
<D:propfind
xmlns:D="DAV:">
<D:prop>
<D:current-user-privilege-set/>
</D:prop>
</D:propfind>
>> Response <<
HTTP/1.1 207 Multi-Status
Content-Type: text/xml;
charset="utf-8"
Content-Length: xxx
<?xml version="1.0" encoding="utf-8" ?>
<D:multistatus xmlns:D="DAV:">
<D:response>
<D:href>http://www.example.com/papers/</D:href>
<D:propstat>
<D:prop>
<D:current-user-privilege-set>
<D:privilege> <D:read/> </D:privilege>
</D:current-user-privilege-set>
</D:prop>
<D:status>HTTP/1.1 200 OK</D:status>
</D:propstat>
</D:response>
</D:multistatus>
This is a
protected property that specifies the list of access control entries (ACEs),
which define what principals are to get what privileges for this resource.
<!ELEMENT acl (ace*) >
Each DAV:ace
element specifies the set of privileges to be either granted or denied to a
single principal. If the DAV:acl
property is empty, no principal is granted any privilege.
<!ELEMENT ace ((principal | invert), (grant | deny), protected?, inherited?)>
The
DAV:principal element identifies the principal to which this ACE applies.
<!ELEMENT principal (href | all | authenticated | unauthenticated
| property | self)>
The current user matches DAV:href only if that user is authenticated as being (or being a member of) the principal identified by the URL contained by that DAV:href.
The current user always matches
DAV:all.
<!ELEMENT all EMPTY>
The current user matches
DAV:authenticated only if authenticated.
<!ELEMENT authenticated EMPTY>
The current user matches
DAV:unauthenticated only if not authenticated.
<!ELEMENT unauthenticated EMPTY>
DAV:all is the union of
DAV:authenticated, and DAV:unauthenticated. For a given request, the user
matches either DAV:authenticated, or DAV:unauthenticated, but not both (that
is, DAV:authenticated and DAV:unauthenticated are disjoint sets).
The current user matches a
DAV:property principal in a DAV:acl property of a resource only if the value of
the identified property of that resource contains at most one DAV:href XML
element, the URI value of DAV:href identifies a principal, and the current user
is authenticated as being (or being a member of) that principal. For example, if the DAV:property element
contained <DAV:owner/>, the current user would match the DAV:property
principal only if the current user is authenticated as matching the principal
identified by the DAV:owner property of the resource.
<!ELEMENT property ANY>
The current user matches DAV:self
in a DAV:acl property of the resource only if that resource is a principal and
that principal matches the current user or, if the principal is a group, a
member of that group matches the current user.
<!ELEMENT self EMPTY>
Some servers
may support ACEs applying to those users NOT matching the current principal,
e.g. all users not in a particular group.
This can be done by wrapping the DAV:principal element with DAV:invert.
<!ELEMENT invert principal>
Each DAV:grant
or DAV:deny element specifies the set of privileges to be either granted or
denied to the specified principal. A
DAV:grant or DAV:deny element of the DAV:acl of a resource MUST only contain
non-abstract elements specified in the DAV:supported-privilege-set of that
resource.
<!ELEMENT grant (privilege+)>
<!ELEMENT deny (privilege+)>
<!ELEMENT privilege ANY>
A server
indicates an ACE is protected by including the DAV:protected element in the
ACE. If the ACL of a resource contains an ACE with a DAV:protected element, an
attempt to remove that ACE from the ACL MUST fail.
<!ELEMENT protected EMPTY>
The presence
of a DAV:inherited element indicates that this ACE is inherited from another
resource that is identified by the URL contained in a DAV:href element. An inherited ACE cannot be modified
directly, but instead the ACL on the resource from which it is inherited must
be modified.
Note that ACE inheritance is not
the same as ACL initialization. ACL
initialization defines the ACL that a newly created resource will use (if not
specified). ACE inheritance refers to
an ACE that is logically shared - where an update to the resource containing an
ACE will affect the ACE of each resource that inherits that ACE. The method by which ACLs are initialized or
by which ACEs are inherited is not defined by this document.
<!ELEMENT inherited (href)>
Continuing the example from Sections 5.2.1 and 5.3.1, this example shows a client requesting the DAV:acl property from the resource with URL http://www.example.com/papers/. There are two ACEs defined in this ACL:
ACE #1: The group identified by URL http://www.example.com/acl/groups/maintainers (the group of site maintainers) is granted DAV:write privilege. Since (for this example) DAV:write contains the DAV:write-acl privilege (see Section 5.2.1), this means the "maintainers" group can also modify the access control list.
ACE #2: All principals (DAV:all) are granted the DAV:read privilege. Since (for this example) DAV:read contains DAV:read-acl and DAV:read-current-user-privilege-set, this means all users (including all members of the "maintainers" group) can read the DAV:acl property and the DAV:current-user-privilege-set property.
>> Request <<
PROPFIND /papers/ HTTP/1.1
Host: www.example.com
Content-type: text/xml; charset="utf-8"
Content-Length: xxx
Depth: 0
Authorization: Digest
username="masinter",
realm="webdav.org", nonce="...",
uri="/papers/", response="...",
opaque="..."
<D:propfind
xmlns:D="DAV:">
<D:prop>
<D:acl/>
</D:prop>
</D:propfind>
>>
Response <<
HTTP/1.1 207 Multi-Status
Content-Type: text/xml;
charset="utf-8"
Content-Length: xxx
<D:multistatus xmlns:D="DAV:">
<D:response>
<D:href>http://www.example.com/papers/</D:href>
<D:propstat>
<D:prop>
<D:acl>
<D:ace>
<D:principal>
<D:href>http://www.example.com/acl/groups/maintainers</D:href>
</D:principal>
<D:grant>
<D:privilege> <D:write/> </D:privilege>
</D:grant>
</D:ace>
<D:ace>
<D:principal>
<D:all/>
</D:principal>
<D:grant>
<D:privilege> <D:read/> </D:privilege>
</D:grant>
</D:ace>
</D:acl>
</D:prop>
<D:status>HTTP/1.1
200 OK</D:status>
</D:propstat>
</D:response>
</D:multistatus>
This protected property defines the
types of ACLs supported by this server, to avoid clients needlessly getting
errors. When a client tries to set an
ACL via the ACL method, the server may reject the attempt to set the ACL as
specified. The following properties
indicate the restrictions the client must observe before setting an ACL:
<grant-only> Deny
ACEs are not supported
<no-invert> Inverted
ACEs are not supported
<deny-before-grant> All deny ACEs must occur before
any grant ACEs
<required-principal> Indicates which principals are
required to be present
<!ELEMENT acl-restrictions (grant-only?, no-invert?, deny-before-grant?, required-principal?)>
This element indicates that ACEs with deny clauses are not allowed.
<!ELEMENT grant-only EMPTY>
This element indicates that ACEs with the <invert> element are not allowed.
<!ELEMENT no-invert EMPTY>
This element indicates that all deny ACEs must precede all grant ACEs.
<!ELEMENT deny-before-grant EMPTY>
The required principal elements
identify which principals must have an ACE defined in the ACL.
<!ELEMENT required-principal
(all? | authenticated? | unauthenticated? | self? | href* | property*)>
For example, the following element
requires that the ACL contain a DAV:owner property ACE:
<D:required-principal xmlns:D="DAV:">
<D:property> <D:owner/> </D:property>
</D:required-principal>
In this example, the client requests the value of the DAV:acl-restrictions property. Digest authentication provides credentials for the principal operating the client.
>>
Request <<
PROPFIND /papers/ HTTP/1.1
Host: www.example.com
Content-type: text/xml; charset="utf-8"
Content-Length: xxx
Depth: 0
Authorization: Digest
username="srcarter",
realm="srcarter@webdav.org", nonce="...",
uri="/papers/", response="...",
opaque="..."
<?xml version="1.0"
encoding="utf-8" ?>
<D:propfind
xmlns:D="DAV:">
<D:prop>
<D:acl-restrictions/>
</D:prop>
</D:propfind>
>> Response <<
HTTP/1.1 207 Multi-Status
Content-Type: text/xml;
charset="utf-8"
Content-Length: xxx
<?xml version="1.0" encoding="utf-8" ?>
<D:multistatus xmlns:D="DAV:">
<D:response>
<D:href>http://www.example.com/papers/</D:href>
<D:propstat>
<D:prop>
<D:acl-restrictions>
<D:grant-only/>
<D:required-principal>
<D:all/>
</D:required-principal>
</D:acl-restrictions>
</D:prop>
<D:status>HTTP/1.1 200 OK</D:status>
</D:propstat>
<D:response>
</D:multistatus>
This
protected property contains a set of URLs that identify other resources that
also control the access to this resource.
To have a privilege on a resource, not only must the ACL on that
resource (specified in the DAV:acl property of that resource) grant the
privilege, but so must the ACL of each resource identified in the
DAV:inherited-acl-set property of that resource. Effectively, the privileges granted by the current ACL are ANDed
with the privileges granted by each inherited ACL.
<!ELEMENT inherited-acl-set (href*)>
This protected property of a resource contains a set of URLs that identify the root collections that contain the principals that are available on the server that implements this resource. A WebDAV Access Control Protocol user agent could use the contents of DAV:principal-collection-set to retrieve the DAV:displayname property (specified in Section 13.2 of [RFC2518]) of all principals on that server, thereby yielding human-readable names for each principal that could be displayed in a user interface.
<!ELEMENT principal-collection-set (href*)>
Since different servers can control different parts of the URL namespace, different resources on the same host MAY have different DAV:principal-collection-set values. The collections specified in the DAV:principal-collection-set MAY be located on different hosts from the resource. The URLs in DAV:principal-collection-set SHOULD be http or https scheme URLs. For security and scalability reasons, a server MAY report only a subset of the entire set of known principal collections, and therefore clients should not assume they have retrieved an exhaustive listing. Additionally, a server MAY elect to report none of the principal collections it knows about, in which case the property value would be empty.
The value of DAV:principal-collection-set gives the scope of the DAV:principal-property-search REPORT (defined in Section 9.4). Clients use the DAV:principal-property-search REPORT to populate their user interface with a list of principals. Therefore, servers that limit a client's ability to obtain principal information will interfere with the client's ability to manipulate access control lists, due to the difficulty of getting the URL of a principal for use in an ACE.
In this example, the client requests the value of the DAV:principal-collection-set property on the collection resource identified by URL http://www.example.com/papers/. The property contains the two URLs, http://www.example.com/acl/users/ and http://www.example.com/acl/groups/, both wrapped in DAV:href XML elements. Digest authentication provides credentials for the principal operating the client.
The client might reasonably follow this request with two separate PROPFIND requests to retrieve the DAV:displayname property of the members of the two collections (/acl/users and /acl/groups). This information could be used when displaying a user interface for creating access control entries.
>>
Request <<
PROPFIND /papers/ HTTP/1.1
Host: www.example.com
Content-type: text/xml; charset="utf-8"
Content-Length: xxx
Depth: 0
Authorization: Digest username="yarong",
realm="yarong@webdav.org", nonce="...",
uri="/papers/", response="...",
opaque="..."
<?xml version="1.0"
encoding="utf-8" ?>
<D:propfind
xmlns:D="DAV:">
<D:prop>
<D:principal-collection-set/>
</D:prop>
</D:propfind>
>> Response <<
HTTP/1.1 207 Multi-Status
Content-Type: text/xml;
charset="utf-8"
Content-Length: xxx
<?xml version="1.0" encoding="utf-8" ?>
<D:multistatus xmlns:D="DAV:">
<D:response>
<D:href>http://www.example.com/papers/</D:href>
<D:propstat>
<D:prop>
<D:principal-collection-set>
<D:href>http://www.example.com/acl/users/</D:href>
<D:href>http://www.example.com/acl/groups/</D:href>
</D:principal-collection-set>
</D:prop>
<D:status>HTTP/1.1 200 OK</D:status>
</D:propstat>
</D:response>
</D:multistatus>
The following example shows how
access control information can be retrieved by using the PROPFIND method to
fetch the values of the DAV:owner, DAV:supported-privilege-set,
DAV:current-user-privilege-set, and DAV:acl properties.
>> Request <<
PROPFIND
/top/container/ HTTP/1.1
Host:
www.example.com
Content-type: text/xml; charset="utf-8"
Content-Length: xxx
Depth: 0
Authorization: Digest
username="ejw",
realm="users@foo.org", nonce="...",
uri="/top/container/", response="...",
opaque="..."
<?xml version="1.0"
encoding="utf-8" ?>
<D:propfind
xmlns:D="DAV:">
<D:prop>
<D:owner/>
<D:supported-privilege-set/>
<D:current-user-privilege-set/>
<D:acl/>
</D:prop>
</D:propfind>
>> Response <<
HTTP/1.1 207 Multi-Status
Content-Type: text/xml;
charset="utf-8"
Content-Length: xxx
<?xml version="1.0" encoding="utf-8" ?>
<D:multistatus
xmlns:D="DAV:"
xmlns:A="http://www.example.com/acl/"> <D:response>
<D:href>http://www.example.com/top/container/</D:href>
<D:propstat>
<D:prop>
<D:owner>
<D:href>http://www.example.com/users/gclemm</D:href>
</D:owner>
<D:supported-privilege-set>
<D:supported-privilege>
<D:privilege> <D:all/> </D:privilege>
<D:abstract/>
<D:description xml:lang="en">Any operation</D:description>
<D:supported-privilege>
<D:privilege> <D:read/> </D:privilege>
<D:description xml:lang="en">Read any object</D:description>
</D:supported-privilege>
<D:supported-privilege>
<D:privilege> <D:write/> </D:privilege>
<D:abstract/>
<D:description xml:lang="en">Write any object</D:description>
<D:supported-privilege>
<D:privilege> <A:create/> </D:privilege>
<D:description xml:lang="en">Create an object</D:description>
</D:supported-privilege>
<D:supported-privilege>
<D:privilege> <A:update/> </D:privilege>
<D:description xml:lang="en">Update an object</D:description>
</D:supported-privilege>
<D:supported-privilege>
<D:privilege> <A:unbind/> </D:privilege>
<D:description xml:lang="en">Remove binding to an object</D:description>
</D:supported-privilege>
</D:supported-privilege>
<D:supported-privilege>
<D:privilege> <D:read-acl/> </D:privilege>
<D:description xml:lang="en">Read the ACL</D:description>
</D:supported-privilege>
<D:supported-privilege>
<D:privilege> <D:write-acl/> </D:privilege>
<D:description xml:lang="en">Write the ACL</D:description>
</D:supported-privilege>
</D:supported-privilege>
</D:supported-privilege-set>
<D:current-user-privilege-set>
<D:privilege> <D:read/> </D:privilege>
<D:privilege> <D:read-acl/> </D:privilege>
</D:current-user-privilege-set>
<D:acl>
<D:ace>
<D:principal>
<D:href>http://www.example.com/users/esedlar</D:href>
</D:principal>
<D:grant>
<D:privilege> <D:read/> </D:privilege>
<D:privilege> <D:write/> </D:privilege>
<D:privilege> <D:read-acl/> </D:privilege>
</D:grant>
</D:ace>
<D:ace>
<D:principal>
<D:href>http://www.example.com/groups/marketing</D:href>
</D:principal>
<D:deny>
<D:privilege> <D:read/> </D:privilege> </D:deny>
</D:ace>
<D:ace>
<D:principal>
<D:property> <D:owner/>
</D:property> </D:principal>
<D:grant>
<D:privilege> <D:read-acl/> </D:privilege>
<D:privilege> <D:write-acl/> </D:privilege>
</D:grant>
</D:ace>
<D:ace>
<D:principal> <D:all/> </D:principal>
<D:grant>
<D:privilege> <D:read/> </D:privilege></D:grant>
<D:inherited>
<D:href>http://www.example.com/top</D:href>
</D:inherited>
</D:ace> </D:acl>
</D:prop>
<D:status>HTTP/1.1 200 OK</D:status>
</D:propstat> </D:response> </D:multistatus>
The value of the DAV:owner property is a single
DAV:href XML element containing the URL of the principal that owns this
resource.
The value of the DAV:supported-privilege-set
property is a tree of supported privileges (using "[XML Namespace ,
localname]" to identify each privilege):
[DAV:, all] (aggregate, abstract)
|
+-- [DAV:, read]
+-- [DAV:, write] (aggregate, abstract)
|
+-- [http://www.example.com/acl, create]
+-- [http://www.example.com/acl, update]
+-- [http://www.example.com/acl, delete]
+-- [DAV:, read-acl]
+-- [DAV:, write-acl]
The DAV:current-user-privilege-set property
contains two privileges, DAV:read, and DAV:read-acl. This indicates that the
current authenticated user only has the ability to read the resource, and read
the DAV:acl property on the resource.
The DAV:acl property contains a set of four
ACEs:
ACE #1: The principal identified by the URL
http://www.example.com/users/esedlar is granted the DAV:read, DAV:write, and
DAV:read-acl privileges.
ACE #2: The principals identified by the URL
http://www.example.com/groups/marketing are denied the DAV:read privilege. In this example, the principal URL identifies
a group.
ACE #3: In this ACE, the principal is a
property principal, specifically the DAV:owner property. When evaluating this
ACE, the value of the DAV:owner property is retrieved, and is examined to see
if it contains a DAV:href XML element. If so, the URL within the DAV:href
element is read, and identifies a principal. In this ACE, the owner is granted
DAV:read-acl, and DAV:write-acl privileges.
ACE #4: This ACE grants the DAV:all principal
(all users) the DAV:read privilege. This ACE is inherited from the resource
http://www.example.com/top, the parent collection of this resource.
WebDAV ACLs are evaluated in
similar manner as ACLs on Windows NT and in NFSv4 [NFSV4]). An ACL is evaluated to determine whether or
not access will be granted for a WebDAV request. ACEs are maintained in a particular order, and are evaluated
until all of the permissions required by the current request have been granted,
at which point the ACL evaluation is terminated and access is granted. If, during ACL evaluation, a <deny>
ACE (matching the current user) is encountered for a privilege which has not
yet been granted, the ACL evaluation is terminated and access is denied. Failure to have all required privileges
granted results in access being denied.
Note that the semantics of many
other existing ACL systems may be represented via this mechanism, by mixing
deny and grant ACEs. For example,
consider the standard "rwx" privilege scheme used by UNIX. In this scheme, if the current user is the
owner of the file, access is granted if the corresponding privilege bit is set
and denied if not set, regardless of the permissions set on the file’s group
and for the world. An ACL for UNIX
permissions of "r--rw-r--"might be constructed like:
<D:acl>
<D:ace>
<D:principal><D:property><D:owner/></D:property></D:principal>
<D:grant><D:privilege><D:read/></D:privilege></D:grant>
</D:ace>
<D:ace>
<D:principal><D:property><D:owner/></D:property></D:principal>
<D:deny><D:privilege><D:all/></D:privilege></D:deny>
</D:ace>
<D:ace>
<D:principal><D:property><D:group/></D:property></D:principal>
<D:grant><D:privilege><D:read/></D:privilege>
<D:privilege><D:write/></D:privilege></D:grant>
</D:ace>
<D:ace>
<D:principal><D:property><D:group/></D:property></D:principal>
<D:deny><D:privilege><D:all/></D:privilege></D:deny>
</D:ace>
<D:ace>
<D:principal><D:all></D:principal>
<D:grant><D:privilege><D:read/></D:privilege></D:grant>
</D:ace>
</D:acl>
and the <acl-restrictions>
would be defined as:
<D:no-invert/>
<D:required-principal>
<D:all/>
<D:property><D:owner/></D:property>
<D:property><D:group/><D:group/>
</D:required-principal>
Note that the client can still get
errors from a UNIX server in spite of obeying the <acl-restrictions>,
including <D:allowed-principal> (adding an ACE specifying a principal
other than the ones in the ACL above) or <D:ace-conflict> (by trying to reorder
the ACEs in the example above), as these particular implementation semantics
are too complex to be captured with the simple (but general) declarative
restrictions.
This section defines the impact of access control functionality on existing methods.
The WebDAV ACL mechanism requires the usage of HTTP method "preconditions" as described in section 1.6 of RFC3253 for ALL HTTP methods. All HTTP methods have an additional precondition called DAV:need-privileges. If an HTTP method fails due to insufficient privileges, the response body to the "403 Forbidden" error MUST contain the <DAV:error> element, which in turn contains the <DAV:need-privileges> element, which contains one or more <DAV:resource> elements indicating which resource had insufficient privileges, and what the lacking privileges were:
<!ELEMENT need-privileges (resource)* >
<!ELEMENT resource ( href , privilege ) >
Since some methods require multiple permissions on multiple
resources, this information is needed to resolve any ambiguity. There is no requirement that all privilege
violations be reported—for implementation reasons, some servers may only report
the first privilege violation. For example:
>> Request <<
MOVE /a/b/ HTTP/1.1
Host: www.example.com
Destination: http://www.example.com/c/d
>> Response <<
HTTP/1.1 403 Forbidden
Content-Type: text/xml;
charset="utf-8"
Content-Length: xxx
<D:error xmlns:D="DAV:">
<D:need-privileges>
<D:resource>
<D:href>/a</D:href>
<D:privilege><D:unbind/></D:privilege>
</D:resource>
<D:resource>
<D:href>/c</D:href>
<D:privilege><D:bind/></D:privilege>
</D:resource>
</D:need-privileges>
</D:error>
If the server supports access control, it MUST return "access-control" as a field in the DAV response header from an OPTIONS request on any resource implemented by that server. A value of "access-control" in the DAV header MUST indicate that the server supports all MUST level requirements and REQUIRED features specified in this document.
>> Request <<
OPTIONS /foo.html HTTP/1.1
Host: www.example.com
Content-Length: 0
>> Response <<
HTTP/1.1 200 OK
DAV: 1, 2, access-control
Allow: OPTIONS, GET, PUT, PROPFIND, PROPPATCH, ACL
In this example, the OPTIONS response indicates that the server supports access control and that /foo.html can have its access control list modified by the ACL method.
When a resource is moved from one location to another due to a MOVE request, the non-inherited and non-protected ACEs in the DAV:acl property of the resource MUST NOT be modified, or the MOVE request fails. Handling of inherited and protected ACEs is intentionally undefined to give server implementations flexibility in how they implement ACE inheritance and protection.
The DAV:acl property on the resource at the destination of a COPY MUST be the same as if the resource was created by an individual resource creation request (e.g. MKCOL, PUT). Clients wishing to preserve the DAV:acl property across a copy need to read the DAV:acl property prior to the COPY, then perform an ACL operation on the new resource at the destination to restore, insofar as this is possible, the original access control list.
A lock on a resource ensures that only the lock owner can modify ACEs that are not inherited and not protected (these are the only ACEs that a client can modify with an ACL request). A lock does not protect inherited or protected ACEs, since a client cannot modify them with an ACL request on that resource.
The ACL method modifies the access
control list (which can be read via the DAV:acl property) of a resource. Specifically, the ACL method only permits
modification to ACEs that are not inherited, and are not protected. An ACL
method invocation modifies all non-inherited and non-protected ACEs in a
resource's access control list to exactly match the ACEs contained within in
the DAV:acl XML element (specified in Section 5.4) of the request body. An ACL request body MUST
contain only one DAV:acl XML element. Unless the non-inherited and
non-protected ACEs of the DAV:acl property of the resource can be updated to be
exactly the value specified in the ACL request, the ACL request MUST fail.
It is possible that the ACEs
visible to the current user in the DAV:acl property may only be a portion of
the complete set of ACEs on that resource. If this is the case, an ACL request
only modifies the set of ACEs visible to the current user, and does not affect
any non-visible ACE.
In order to avoid overwriting
DAV:acl changes by another client, a client SHOULD acquire a WebDAV lock on the
resource before retrieving the DAV:acl property of a resource that it intends
on updating.
Implementation Note: Two common
operations are to add or remove an ACE from an existing access control list. To
accomplish this, a client uses the PROPFIND method to retrieve the value of the
DAV:acl property, then parses the returned access control list to remove all
inherited and protected ACEs (these ACEs are tagged with the DAV:inherited and
DAV:protected XML elements). In the remaining set of non-inherited,
non-protected ACEs, the client can add or remove one or more ACEs before
submitting the final ACE set in the request body of the ACL method.
An implementation MUST enforce the
following constraints on an ACL request.
If the constraint is violated, a 403 (Forbidden) or 409 (Conflict)
response MUST be returned and the indicated XML element MUST be returned as a
child of a top level DAV:error element in an XML response body.
Though these status elements are
generally expressed as empty XML elements (and are defined as EMPTY in the
DTD), implementations MAY return additional descriptive XML elements as
children of the status element. Clients MUST be able to accept children of
these status elements. Clients that do not understand the additional XML
elements should ignore them.
(DAV:no-ace-conflict): The ACEs
submitted in the ACL request MUST NOT conflict with each other. This is a catchall error code indicating
that an implementation-specific ACL restriction has been violated.
(DAV:no-protected-ace-conflict):
The ACEs submitted in the ACL request MUST NOT conflict with the protected ACEs
on the resource. For example, if the resource has a protected ACE granting
DAV:write to a given principal, then it would not be consistent if the ACL
request submitted an ACE denying DAV:write to the same principal.
(DAV:no-inherited-ace-conflict):
The ACEs submitted in the ACL request MUST NOT conflict with the inherited ACEs
on the resource. For example, if the resource inherits an ACE from its parent
collection granting DAV:write to a given principal, then it would not be
consistent if the ACL request submitted an ACE denying DAV:write to the same
principal. Note that reporting of this error will be implementation-dependent.
Implementations MUST either report this error or allow the ACE to be set, and
then let normal ACE evaluation rules determine whether the new ACE has any
impact on the privileges available to a specific principal.
(DAV:limited-number-of-aces): The
number of ACEs submitted in the ACL request MUST NOT exceed the number of ACEs
allowed on that resource. However,
ACL-compliant servers MUST support at least one ACE granting privileges to a
single principal, and one ACE granting privileges to a group.
(DAV:deny-before-grant): All
non-inherited deny ACEs MUST precede all non-inherited grant ACEs.
(DAV:grant-only): The ACEs submitted in the ACL request MUST NOT
include a deny ACE. This precondition
applies only when the ACL restrictions of the resource include the
DAV:grant-only constraint (defined in Section 5.5.1).
(DAV:no-invert): The ACL request MUST NOT include a
DAV:invert element. This precondition
applies only when the ACL semantics of the resource includes the DAV:no-invert
constraint (defined in Section 6.3.4).
(DAV:no-abstract): The ACL request
MUST NOT attempt to grant or deny an abstract privilege (see Section 5.2).
(DAV:not-supported-privilege): The
ACEs submitted in the ACL request MUST be supported by the resource.
(DAV:missing-required-principal):
The result of the ACL request MUST have at least one ACE for each principal
identified in a DAV:required-principal XML element in the ACL semantics of that
resource (see Section 5.5.4).
(DAV:recognized-principal): Every
principal URL in the ACL request MUST identify a principal resource.
(DAV:allowed-principal): The
principals specified in the ACEs submitted in the ACL request MUST be allowed
as principals for the resource. For example, a server where only authenticated
principals can access resources would not allow the DAV:all or
DAV:unauthenticated principals to be used in an ACE, since these would allow
unauthenticated access to resources.
In the following example, user
"fielding", authenticated by information in the Authorization header,
grants the principal identified by the URL http://www.example.com/users/esedlar (i.e., the user "esedlar") read
and write privileges, grants the owner of the resource read-acl and write-acl
privileges, and grants everyone read privileges.
>>
Request <<
ACL /top/container/ HTTP/1.1
Host: www.example.com
Content-Type: text/xml;
charset="utf-8"
Content-Length: xxxx
Authorization: Digest
username="fielding",
realm="users@foo.org", nonce="...",
uri="/top/container/", response="...",
opaque="..."
<?xml version="1.0"
encoding="utf-8" ?>
<D:acl
xmlns:D="DAV:">
<D:ace>
<D:principal>
<D:href>http://www.example.com/users/esedlar</D:href>
</D:principal>
<D:grant>
<D:privilege> <D:read/> </D:privilege>
<D:privilege> <D:write/> </D:privilege>
</D:grant>
</D:ace>
<D:ace>
<D:principal>
<D:property> <D:owner/> </D:property>
</D:principal>
<D:grant>
<D:privilege> <D:read-acl/> </D:privilege>
<D:privilege> <D:write-acl/> </D:privilege>
</D:grant>
</D:ace>
<D:ace>
<D:principal> <D:all/> </D:principal>
<D:grant>
<D:privilege> <D:read/> </D:privilege>
</D:grant>
</D:ace> </D:acl>
>>
Response <<
HTTP/1.1 200 OK
In the following request, user
"fielding", authenticated by information in the Authorization header,
attempts to deny the principal identified by the URL http://www.example.com/users/esedlar (i.e., the user "esedlar") write
privileges. Prior to the request, the DAV:acl property on the resource
contained a protected ACE (see Section 5.4.3) granting DAV:owner the DAV:read and DAV:write
privileges. The principal identified by URL
http://www.example.com/users/esedlar is the owner of the resource. The ACL
method invocation fails because the submitted ACE conflicts with the protected
ACE, thus violating the semantics of ACE protection.
>>
Request <<
ACL /top/container/ HTTP/1.1
Host: www.example.com
Content-Type: text/xml;
charset="utf-8"
Content-Length: xxxx
Authorization: Digest
username="fielding",
realm="users@foo.org", nonce="...",
uri="/top/container/", response="...", opaque="..."
<?xml version="1.0"
encoding="utf-8" ?>
<D:acl
xmlns:D="DAV:">
<D:ace>
<D:principal>
<D:href>http://www.example.com/users/esedlar</D:href>
</D:principal>
<D:deny>
<D:privilege> <D:write/> </D:privilege>
</D:deny>
</D:ace>
</D:acl>
>>
Response <<
HTTP/1.1 403 Forbidden
Content-Type: text/xml;
charset="utf-8"
Content-Length: xxx
<?xml version="1.0" encoding="utf-8" ?>
<D:error xmlns:D="DAV:">
<D:no-protected-ace-conflict/>
</D:error>
In the following request, user "ejw", authenticated by information in the Authorization header, tries to change the access control list on the resource http://www.example.com/top/index.html. This resource has two inherited ACEs.
Inherited ACE #1 grants the principal identified by URL http://www.example.com/users/ejw (i.e., the user "ejw") http://www.example.com/privs/write-all and DAV:read-acl privileges. On this server, http://www.example.com/privs/write-all is an aggregate privilege containing DAV:write, and DAV:write-acl.
Inherited ACE #2 grants principal DAV:all the DAV:read privilege.
The request attempts to set a (non-inherited) ACE, denying the principal identified by the URL http://www.example.com/users/ejw (i.e., the user "ejw") DAV:write permission. This conflicts with inherited ACE #1. Note that the decision to report an inherited ACE conflict is specific to this server implementation. Another server implementation could have allowed the new ACE to be set, and then used normal ACE evaluation rules to determine whether the new ACE has any impact on the privileges available to a principal.
>> Request <<
ACL /top/index.html HTTP/1.1
Host: www.example.com
Content-Type: text/xml;
charset="utf-8"
Content-Length: xxxx
Authorization: Digest
username="ejw",
realm="users@foo.org", nonce="...",
uri="/top/index.html", response="...",
opaque="..."
<?xml version="1.0"
encoding="utf-8" ?>
<D:acl xmlns:D="DAV:"
xmlns:F="http://www.example.com/privs/">
<D:ace>
<D:principal>
<D:href>http://www.example.com/users/ejw</D:href>
</D:principal>
<D:grant><D:write/></D:grant>
</D:ace>
</D:acl>
>> Response <<
HTTP/1.1 403
Forbidden
Content-Type:
text/xml; charset="utf-8"
Content-Length:
xxx
<?xml version="1.0" encoding="utf-8" ?>
<D:error xmlns:D="DAV:">
<D:no-inherited-ace-conflict xmlns:D="DAV:"/>
</D:error>
In this example, user "ygoland", authenticated by information in the Authorization header, tries to change the access control list on the resource http://www.example.com/diamond/engagement-ring.gif. The ACL request includes a single, syntactically and semantically incorrect ACE, which attempts to grant the group identified by the URL http://www.example.com/users/friends DAV:read privilege and deny the principal identified by URL http://www.example.com/users/ygoland-so (i.e., the user "ygoland-so") DAV:read privilege. However, it is illegal to have multiple principal elements, as well as both a grant and deny element in the same ACE, so the request fails due to poor syntax.
>> Request <<
ACL /diamond/engagement-ring.gif
HTTP/1.1
Host: www.example.com
Content-Type: text/xml;
charset="utf-8"
Content-Length: xxxx
Authorization: Digest
username="ygoland",
realm="users@foo.org", nonce="...",
uri="/diamond/engagement-ring.gif", response="...",
opaque="..."
<?xml version="1.0"
encoding="utf-8" ?>
<D:acl
xmlns:D="DAV:">
<D:ace>
<D:principal>
<D:href>http://www.example.com/users/friends</D:href>
</D:principal>
<D:grant><D:read/></D:grant>
<D:principal>
<D:href>http://www.example.com/users/ygoland-so</D:href>
</D:principal>
<D:deny><D:read/></D:deny>
</D:ace>
</D:acl>
>>
Response <<
HTTP/1.1 400 Bad Request
Content-Length: 0
Note that if the request had been
divided into two ACEs, one to grant, and one to deny, the request would have
been syntactically well formed.
The REPORT method (defined in Section 3.6 of [RFC3253]) provides an extensible mechanism for obtaining information about a resource. Unlike the PROPFIND method, which returns the value of one or more named properties, the REPORT method can involve more complex processing. REPORT is valuable in cases where the server has access to all of the information needed to perform the complex request (such as a query), and where it would require multiple requests for the client to retrieve the information needed to perform the same request.
A server that supports the WebDAV Access Control Protocol MUST support the DAV:expand-property report (defined in Section 3.8 of [RFC3253]).
The DAV:acl-principal-prop-set
report returns, for all principals in the DAV:acl property (of the Request-URI)
that are identified by http(s) URLs or by a DAV:property principal, the value
of the properties specified in the REPORT request body. In the case where a
principal URL appears multiple times, the DAV:acl-principal-prop-set report
MUST return the properties for that principal only once. Support for this
report is REQUIRED.
One expected use of this report is
to retrieve the human readable name (found in the DAV:displayname property) of
each principal found in an ACL. This is useful for constructing user interfaces
that show each ACE in a human readable form.
Marshalling
The
request body MUST be a DAV:acl-principal-prop-set
XML element.
<!ELEMENT acl-principal-prop-set ANY>
ANY value: a sequence of one or more elements, with at most one DAV:prop element.
prop: see RFC 2518, Section 12.11
This report is only defined when the Depth header has value "0"; other values result in a 400 (Bad Request) error response. Note that [RFC3253], Section 3.6, states that if the Depth header is not present, it defaults to a value of "0".
The response body for a successful request MUST be a DAV:multistatus XML element (i.e., the response uses the same format as the response for PROPFIND). In the case where there are no response elements, the returned multistatus XML element is empty.
multistatus: see RFC 2518, Section 12.9
The response body for a successful DAV:acl-principal-prop-set REPORT request MUST contain a DAV:response element for each principal identified by an http(s) URL listed in a DAV:principal XML element of an ACE within the DAV:acl property of the resource identified by the Request-URI.
Postconditions:
(DAV:number-of-matches-within-limits): The number of matching principals must fall within server-specific, predefined limits. For example, this condition might be triggered if a search specification would cause the return of an extremely large number of responses.
Resource
http://www.example.com/index.html has an ACL with three ACEs:
ACE #1: All principals (DAV:all)
have DAV:read and DAV:read-current-user-privilege-set access.
ACE #2: The principal identified by
http://www.example.com/people/gstein (the user "gstein") is granted
DAV:write, DAV:write-acl, DAV:read-acl
privileges.
ACE #3: The group identified by
http://www.example.com/groups/authors (the "authors" group) is
granted DAV:write and DAV:read-acl privileges.
The following example shows a
DAV:acl-principal-prop-set report requesting the DAV:displayname property. It
returns the value of DAV:displayname for resources
http://www.example.com/people/gstein and http://www.example.com/groups/authors
, but not for DAV:all, since this is not an http(s) URL.
>> Request <<
REPORT /index.html HTTP/1.1
Host: www.example.com
Content-Type: text/xml; charset="utf-8"
Content-Length: xxxx
Depth: 0
<?xml version="1.0" encoding="utf-8" ?>
<D:acl-principal-prop-set xmlns:D="DAV:">
<D:prop>
<D:displayname/>
</D:prop>
</D:acl-principal-prop-set>
>> Response <<
HTTP/1.1 207 Multi-Status
Content-Type: text/xml; charset="utf-8"
Content-Length: xxxx
<?xml version="1.0" encoding="utf-8" ?>
<D:multistatus xmlns:D="DAV:">
<D:response>
<D:href>http://www.example.com/people/gstein</D:href>
<D:propstat>
<D:prop>
<D:displayname>Greg Stein</D:displayname>
</D:prop>
<D:status>HTTP/1.1 200 OK</D:status>
</D:propstat>
</D:response>
<D:response>
<D:href>http://www.example.com/groups/authors</D:href>
<D:propstat>
<D:prop>
<D:displayname>Site authors</D:displayname>
</D:prop>
<D:status>HTTP/1.1 200 OK</D:status>
</D:propstat>
</D:response>
</D:multistatus>
The DAV:principal-match REPORT is used to identify all members (at any depth) of the collection identified by the Request-URI that are principals and that match the current user. In particular, if the collection contains principals, the report can be used to identify all members of the collection that match the current user. Alternatively, if the collection contains resources that have a property that identifies a principal (e.g. DAV:owner), the report can be used to identify all members of the collection whose property identifies a principal that matches the current user. For example, this report can return all of the resources in a collection hierarchy that are owned by the current user. Support for this report is REQUIRED.
Marshalling:
The request body MUST be a DAV:principal-match XML element.
<!ELEMENT principal-match ((principal-property | self), prop?)>
<!ELEMENT principal-property ANY>
ANY value: an element whose value identifies a property. The expectation is the value of the named property typically contains an href element that contains the URI of a principal
<!ELEMENT self EMPTY>
prop: see RFC 2518, Section 12.11
This report is only defined when the Depth header has value "0"; other values result in a 400 (Bad Request) error response. Note that [RFC3253], Section 3.6, states that if the Depth header is not present, it defaults to a value of "0".
The response body for a successful request MUST be a DAV:multistatus XML element. In the case where there are no response elements, the returned multistatus XML element is empty.
multistatus: see RFC 2518, Section 12.9
The response body for a successful DAV:principal-match REPORT request MUST contain a DAV:response element for each member of the collection that matches the current user. When the DAV:principal-property element is used, a match occurs if the current user is matched by the principal identified by the URI found in the DAV:href element of the property identified by the DAV:principal-property element. When the DAV:self element is used in a DAV:principal-match report issued against a group, it matches the group if a member identifies the same principal as the current user.
If DAV:prop is specified in the request body, the properties specified in the DAV:prop element MUST be reported in the DAV:response elements.
The following example identifies the members of the collection identified by the URL http://www.example.com/doc that are owned by the current user. The current user ("gclemm") is authenticated using Digest authentication.
>> Request <<
REPORT /doc/ HTTP/1.1
Host: www.example.com
Authorization: Digest
username="gclemm",
realm="gclemm@webdav.org", nonce="...",
uri="/papers/", response="...",
opaque="..."
Content-Type: text/xml; charset="utf-8"
Content-Length: xxxx
Depth: 0
<?xml version="1.0" encoding="utf-8" ?>
<D:principal-match xmlns:D="DAV:">
<D:principal-property>
<D:owner/>
</D:principal-property>
</D:principal-match>
>> Response <<
HTTP/1.1 207 Multi-Status
Content-Type: text/xml; charset="utf-8"
Content-Length: xxxx
<?xml version="1.0" encoding="utf-8" ?>
<D:multistatus xmlns:D="DAV:">
<D:response>
<D:href>http://www.example.com/doc/foo.html</D:href>
<D:status>HTTP/1.1 200 OK</D:status>
</D:response>
<D:response>
<D:href>http://www.example.com/doc/img/bar.gif</D:href>
<D:status>HTTP/1.1 200 OK</D:status>
</D:response>
</D:multistatus>
The DAV:principal-property-search REPORT performs a search for all principals whose properties contain character data that matches the search criteria specified in the request. One expected use of this report is to discover the URL of a principal associated with a given person or group by searching for them by name. This is done by searching over DAV:displayname, which is defined on all principals.
The actual search method (exact matching vs. substring matching vs, prefix-matching, case-sensitivity) deliberately is left to the server implementation to allow implementation on a wide set of possible user management systems. In cases where the implementation of DAV:principal-property-search is not constrained by the semantics of an underlying user management repository, preferred default semantics are caseless substring matches.
For implementation efficiency, servers do not typically support searching on all properties. A search requesting properties that are not searchable for a particular principal will not match that principal.
Support for the DAV:principal-property-search report is REQUIRED.
Implementation Note: The value of a WebDAV property is a sequence of well-formed XML, and hence can include any character in the Unicode/ISO-10646 standard, that is, most known characters in human languages. Due to the idiosyncrasies of case mapping across human languages, implementation of case-insensitive matching is non-trivial. Implementors of servers that do perform substring matching are strongly encouraged to consult [CaseMap], especially Section 2.3 ("Caseless Matching"), for guidance when implementing their case-insensitive matching algorithms.
Implementation Note: Some implementations of this protocol will use an LDAP repository for storage of principal metadata. The schema describing each attribute (akin to a WebDAV property) in an LDAP repository specifies whether it supports case-sensitive or caseless searching. One of the benefits of leaving the search method to the discretion of the server implementation is the default LDAP attribute search behavior can be used when implementing the DAV:principal-property-search report.
Marshalling:
The request body MUST be a DAV:principal-property-search XML element containing a search specification and an optional list of properties. For every principal that matches the search specification, the response will contain the value of the requested properties on that principal.
<!ELEMENT principal-property-search
((property-search+), prop?, apply-to-principal-collection-set?) >
By default, the report searches all members (at any depth) of the collection identified by the Request-URI. If DAV:apply-to-principal-collection-set is specified in the request body, the request is applied instead to each collection identified by the DAV:prinicipal-collection-set property of the resource identified by the Request-URI.
The DAV:property-search element contains a prop element enumerating the properties to be searched and a match element, containing the search string.
<!ELEMENT property-search (prop, match) >
prop: see RFC 2518, Section 12.11
<!ELEMENT match #PCDATA >
Multiple property-search elements or multiple elements within a DAV:prop element will be interpreted with a logical AND.
This report is only defined when the Depth header has value "0"; other values result in a 400 (Bad Request) error response. Note that [RFC3253], Section 3.6, states that if the Depth header is not present, it defaults to a value of "0".
The response body for a successful request MUST be a DAV:multistatus XML element. In the case where there are no response elements, the returned multistatus XML element is empty.
multistatus: see RFC 2518, Section 12.9
The response body for a successful DAV:principal-property-search REPORT request MUST contain a DAV:response element for each principal whose property values satisfy the search specification given in DAV:principal-property-search.
The response body for an unsuccessful DAV:principal-property-search REPORT request MUST contain, after the XML element indicating the failed precondition or postcondition, a DAV:prop element containing the property that caused the pre/postcondition to fail.
If DAV:prop is specified in the request body, the properties specified in the DAV:prop element MUST be reported in the DAV:response elements.
Preconditions:
None
Postconditions:
(DAV:number-of-matches-within-limits):
The number of matching principals must fall within server-specific, predefined
limits. For example, this condition might be triggered if a search
specification would cause the return of an extremely large number of responses.
There are several cases to consider when matching strings. The easiest case is when a property value is "simple" and has only character information item content (see [REC-XML-INFOSET]). For example, the search string "julian" would match the DAV:displayname property with value "Julian Reschke". Note that the on-the-wire marshalling of DAV:displayname in this case is:
<D:displayname xmlns:D="DAV:">Julian Reschke</D:displayname>
The name of the property is encoded into the XML element information item, and the character information item content of the property is "Julian Reschke".
A more complicated case occurs when properties have mixed content (that is, compound values consisting of multiple child element items, other types of information items, and character information item content). Consider the property "aprop" in the namespace "http://www.example.com/props/", marshalled as:
<W:aprop
xmlns:W="http://www.example.com/props/">
{cdata 0}<W:elem1>{cdata 1}</W:elem1>
<W:elem2>{cdata 2}</W:elem2>{cdata
3}
</W:aprop>
In this case, matching is performed on each individual contiguous sequence of character information items. In the example above, a search string would be compared to the four following strings:
{cdata
0}
{cdata 1}
{cdata 2}
{cdata 3}
That is, four individual matches would be performed, one each for {cdata 0}, {cdata 1}, {cdata 2}, and {cdata 3}.
In this example, the client requests the principal URLs of all users whose DAV:displayname property contains the substring "doE" and whose "title" property in the namespace "http://BigCorp.com/ns/" (that is, their professional title) contains "Sales". In addition, the client requests five properties to be returned with the matching principals:
In
the DAV: namespace: displayname
In the http://www.example.com/ns/ namespace: department, phone, office, salary
The response shows that two principal resources meet the search specification, "John Doe" and "Zygdoebert Smith". The property "salary" in namespace "http://www.example.com/ns/" is not returned, since the principal making the request does not have sufficient access permissions to read this property.
>>
Request <<
REPORT /users/ HTTP/1.1
Host: www.example.com
Content-Type: text/xml; charset=utf-8
Content-Length: xxxx
Depth: 0
<?xml version="1.0" encoding="utf-8" ?>
<D:principal-property-search xmlns:D="DAV:">
<D:property-search>
<D:prop>
<D:displayname/>
</D:prop>
<D:match>doE</D:match>
</D:property-search>
<D:property-search>
<D:prop xmlns:B="http://www.example.com/ns/">
<B:title/>
</D:prop>
<D:match>Sales</D:match>
</D:property-search>
<D:prop xmlns:B="http://www.example.com/ns/">
<D:displayname/>
<B:department/>
<B:phone/>
<B:office/>
<B:salary/>
</D:prop>
</D:principal-property-search>
>>
Response <<
HTTP/1.1 207 Multi-Status
Content-Type: text/xml; charset=utf-8
Content-Length: xxxx
<?xml version="1.0" encoding="utf-8" ?>
<D:multistatus xmlns:D="DAV:" xmlns:B="http://BigCorp.com/ns/">
<D:response>
<D:href>http://www.example.com/users/jdoe</D:href>
<D:propstat>
<D:prop>
<D:displayname>John Doe</D:displayname>
<B:department>Widget Sales</B:department>
<B:phone>234-4567</B:phone>
<B:office>209</B:office>
</D:prop>
<D:status>HTTP/1.1 200 OK</D:status>
</D:propstat>
<D:propstat>
<D:prop>
<B:salary/>
</D:prop>
<D:status>HTTP/1.1 403 Forbidden</D:status>
</D:propstat>
</D:response>
<D:response>
<D:href>http://www.example.com/users/zsmith</D:href>
<D:propstat>
<D:prop>
<D:displayname>Zygdoebert Smith</D:displayname>
<B:department>Gadget Sales</B:department>
<B:phone>234-7654</B:phone>
<B:office>114</B:office>
</D:prop>
<D:status>HTTP/1.1 200 OK</D:status>
</D:propstat>
<D:propstat>
<D:prop>
<B:salary/>
</D:prop>
<D:status>HTTP/1.1 403 Forbidden</D:status>
</D:propstat>
</D:response>
</D:multistatus>
The DAV:principal-search-property-set REPORT identifies those properties that may be searched using the DAV:principal-property-search REPORT (defined in Section 9.4).
Servers MUST support the DAV:principal-search-property-set REPORT on all collections identified in the value of a DAV:principal-collection-set property.
An access control protocol user agent could use the results of the DAV:principal-search-property-set REPORT to present a query interface to the user for retrieving principals.
Support for this report is
REQUIRED.
Implementation Note: Some clients will have only limited screen real estate for the display of lists of searchable properties. In this case, a user might appreciate having the most frequently searched properties be displayed on-screen, rather than having to scroll through a long list of searchable properties. One mechanism for signaling the most frequently searched properties is to return them towards the start of a list of properties. A client can then preferentially display the list of properties in order, increasing the likelihood that the most frequently searched properties will appear on-screen, and will not require scrolling for their selection.
Marshalling:
The request body MUST be an empty DAV:principal-search-property-set XML element.
This report is only defined when the Depth header has value "0"; other values result in a 400 (Bad Request) error response. Note that [RFC3253], Section 3.6, states that if the Depth header is not present, it defaults to a value of "0".
The response body MUST be a DAV:principal-search-property-set XML element, containing a DAV:principal-search-property XML element for each property that may be searched with the DAV:principal-property-search REPORT. A server MAY limit its response to just a subset of the searchable properties, such as those likely to be useful to an interactive access control client.
<!ELEMENT principal-search-property-set (principal-search-property*) >
Each DAV:principal-search-property XML element contains exactly one searchable property, and a description of the property.
<!ELEMENT principal-search-property (prop, description) >
The DAV:prop element contains one principal property on which the server is able to perform a DAV:principal-property-search REPORT.
prop: see RFC 2518, Section 12.11
The description element is a human-readable description of what information this property represents. Servers MUST indicate the human language of the description using the xml:lang attribute and SHOULD consider the HTTP Accept-Language request header when selecting one of multiple available languages.
<!ELEMENT description #PCDATA >
In this example, the client determines the set of searchable principal properties by requesting the DAV:principal-search-property-set REPORT on the root of the server's principal URL collection set, identified by http://www.example.com/users/.
>>
Request <<
REPORT /users/ HTTP/1.1
Host: www.example.com
Content-Type: text/xml; charset="utf-8"
Content-Length: xxx
Accept-Language: en, de
Authorization: BASIC d2FubmFtYWs6cGFzc3dvcmQ=
Depth: 0
<?xml version="1.0" encoding="utf-8" ?>
<D:principal-search-property-set xmlns:D="DAV:"/>
>>
Response <<
HTTP/1.1 200 OK
Content-Type: text/xml; charset="utf-8"
Content-Length: xxx
<?xml version="1.0" encoding="utf-8" ?>
<D:principal-search-property-set xmlns:D="DAV:">
<D:principal-search-property>
<D:prop>
<D:displayname/>
</D:prop>
<D:description xml:lang="en">Full name</D:description>
</D:principal-search-property>
<D:principal-search-property>
<D:prop xmlns:B="http://BigCorp.com/ns/">
<B:title/>
</D:prop>
<D:description xml:lang="en">Job title</D:description>
</D:principal-search-property>
</D:principal-search-property-set>
Implementations of this
specification MUST support the XML element ignore rule, as specified in Section
23.3.2 of [RFC2518], and the XML Namespace recommendation [REC-XML-NAMES].
Note that use of the DAV namespace
is reserved for XML elements and property names defined in a standards-track or
Experimental IETF RFC.
In this specification, the only human-readable content can be found in the description XML element, found within the DAV:supported-privilege-set property. This element contains a human-readable description of the capabilities controlled by a privilege. As a result, the description element must be capable of representing descriptions in multiple character sets. Since the description element is found within a WebDAV property, it is represented on the wire as XML [REC-XML], and hence can leverage XML's language tagging and character set encoding capabilities. Specifically, XML processors at minimum must be able to read XML elements encoded using the UTF-8 [UTF-8] encoding of the ISO 10646 multilingual plane. XML examples in this specification demonstrate use of the charset parameter of the Content-Type header, as defined in [RFC3023], as well as the XML "encoding" attribute, which together provide charset identification information for MIME and XML processors. Futhermore, this specification requires server implementations to tag description fields with the xml:lang attribute (see Section 2.12 of [REC-XML]), which specifies the human language of the description. Additionally, server implementations should take into account the value of the Accept-Language HTTP header to determine which description string to return.
For XML elements other than the description element, it is expected that implementations will treat the property names, privilege names, and values as tokens, and convert these tokens into human-readable text in the user's language and character set when displayed to a person. Only a generic WebDAV property display utility would display these values in their raw form to a human user.
For error reporting, we follow the convention of HTTP/1.1 status codes, including with each status code a short, English description of the code (e.g., 200 (OK)). While the possibility exists that a poorly crafted user agent would display this message to a user, internationalized applications will ignore this message, and display an appropriate message in the user's language and character set.
Further internationalization considerations for this protocol are described in the WebDAV Distributed Authoring protocol specification [RFC2518].
Applications and users of this access control protocol should be aware of several security considerations, detailed below. In addition to the discussion in this document, the security considerations detailed in the HTTP/1.1 specification [RFC2616], the WebDAV Distributed Authoring Protocol specification [RFC2518], and the XML Media Types specification [RFC3023] should be considered in a security analysis of this protocol.
In the absence of a mechanism for remotely manipulating access control lists, if a single user's authentication credentials are compromised, only those resources for which the user has access permission can be read, modified, moved, or deleted. With the introduction of this access control protocol, if a single compromised user has the ability to change ACLs for a broad range of other users (e.g., a super-user), the number of resources that could be altered by a single compromised user increases. This risk can be mitigated by limiting the number of people who have write-acl privileges across a broad range of resources.
The ability to read the access privileges (stored in the DAV:acl property), or the privileges permitted the currently authenticated user (stored in the DAV:current-user-privilege-set property) on a resource may seem innocuous, since reading an ACL cannot possibly affect the resource's state. However, if all resources have world-readable ACLs, it is possible to perform an exhaustive search for those resources that have inadvertently left themselves in a vulnerable state, such as being world-writeable. In particular, the property retrieval method PROPFIND, executed with Depth infinity on an entire hierarchy, is a very efficient way to retrieve the DAV:acl or DAV:current-user-privilege-set properties. Once found, this vulnerability can be exploited by a denial of service attack in which the open resource is repeatedly overwritten. Alternately, writeable resources can be modified in undesirable ways.
To reduce this risk, read-acl privileges should not be granted to unauthenticated principals, and restrictions on read-acl and read-current-user-privilege-set privileges for authenticated principals should be carefully analyzed when deploying this protocol. Access to the current-user-privilege-set property will involve a tradeoff of usability versus security. When the current-user-privilege-set is visible, user interfaces are expected to provide enhanced information concerning permitted and restricted operations, yet this information may also indicate a vulnerability that could be exploited. Deployment of this protocol will need to evaluate this tradeoff in light of the requirements of the deployment environment.
In an effort to reduce protocol complexity, this protocol specification intentionally does not address the issue of how to manage or discover the initial ACL that is placed upon a resource when it is created. The only way to discover the initial ACL is to create a new resource, then retrieve the value of the DAV:acl property. This assumes the principal creating the resource also has been granted the DAV:read-acl privilege.
As a result, it is possible that a principal could create a resource, and then discover that its ACL grants privileges that are undesirable. Furthermore, this protocol makes it possible (though unlikely) that the creating principal could be unable to modify the ACL, or even delete the resource. Even when the ACL can be modified, there will be a short period of time when the resource exists with the initial ACL before its new ACL can be set.
Several
factors mitigate this risk. Human principals are often aware of the default
access permissions in their editing environments and take this into account
when writing information. Furthermore, default privilege policies are usually
very conservative, limiting the privileges granted by the initial ACL.
Authentication mechanisms defined for use with HTTP and WebDAV also apply to this WebDAV Access Control Protocol, in particular the Basic and Digest authentication mechanisms defined in [RFC2617]. Implementation of the ACL spec requires that Basic authentication, if used, MUST only be supported over secure transport such as TLS.
This document uses the namespace defined by [RFC2518] for XML elements. That is, this specification uses the "DAV:" URI namespace, previously registered in the URI schemes registry. All other IANA considerations mentioned in [RFC2518] are also applicable to this specification.
The following notice is copied from RFC 2026, section 10.4, and describes the position of the IETF concerning intellectual property claims made against this document.
The IETF takes no position regarding the validity or scope of any intellectual property or other rights that might be claimed to pertain to the implementation or use other technology described in this document or the extent to which any license under such rights might or might not be available; neither does it represent that it has made any effort to identify any such rights. Information on the IETF's procedures with respect to rights in standards-track and standards-related documentation can be found in BCP-11. Copies of claims of rights made available for publication and any assurances of licenses to be made available, or the result of an attempt made to obtain a general license or permission for the use of such proprietary rights by implementers or users of this specification can be obtained from the IETF Secretariat.
The IETF invites any interested party to bring to its attention any copyrights, patents or patent applications, or other proprietary rights that may cover technology that may be required to practice this standard. Please address the information to the IETF Executive Director.
This protocol is the collaborative product of the WebDAV ACL design team: Bernard Chester, Geoff Clemm, Anne Hopkins, Barry Lind, Sean Lyndersay, Eric Sedlar, Greg Stein, and Jim Whitehead. The authors are grateful for the detailed review and comments provided by Jim Amsden, Dylan Barrell, Gino Basso, Murthy Chintalapati, Lisa Dusseault, Stefan Eissing, Tim Ellison, Yaron Goland, Dennis Hamilton, Laurie Harper, Eckehard Hermann, Ron Jacobs, Chris Knight, Remy Maucherat, Larry Masinter, Joe Orton, Peter Raymond, Julian Reschke, and Keith Wannamaker. We thank Keith Wannamaker for the initial text of the principal property search sections. Prior work on WebDAV access control protocols has been performed by Yaron Goland, Paul Leach, Lisa Dusseault, Howard Palmer, and Jon Radoff. We would like to acknowledge the foundation laid for us by the authors of the DeltaV, WebDAV and HTTP protocols upon which this protocol is layered, and the invaluable feedback from the WebDAV working group.
[RFC2119] S.Bradner, "Key words for use in RFCs to Indicate Requirement Levels." RFC 2119, BCP 14, March, 1997.
[REC-XML] T. Bray, J. Paoli, C.M. Sperberg-McQueen, "Extensible Markup Language (XML)." World Wide Web Consortium Recommendation REC-xml.http://www.w3.org/TR/REC-xml
[REC-XML-NAMES] T. Bray, D. Hollander, A. Layman, "Name Spaces in XML" World Wide Web Consortium Recommendation REC-xml-names. http://www.w3.org/TR/REC-xml-names/
[RFC3253] G. Clemm, J. Amsden, T. Ellison, C. Kaler, J. Whitehead, "Versioning Extensions to WebDAV." RFC 3253, March 2002.
[REC-XML-INFOSET] J. Cowan, R. Tobin, "XML Information Set." World Wide Web Consortium Recommendation REC-xml-infoset. http://www.w3.org/TR/xml-infoset/
[RFC2616] R. Fielding, J. Gettys, J. C. Mogul, H. Frystyk, L. Masinter, P. Leach, and T. Berners-Lee, "Hypertext Transfer Protocol -- HTTP/1.1." RFC 2616, June, 1999.
[RFC2617] J. Franks, P. Hallam-Baker, J. Hostetler, S. Lawrence, P. Leach, A. Luotonen, L. Stewart, "HTTP Authentication: Basic and Digest Access Authentication." RFC 2617, June, 1999.
[RFC2518] Y. Goland, E. Whitehead, A. Faizi, S. R. Carter, D. Jensen, "HTTP Extensions for Distributed Authoring -- WEBDAV." RFC 2518, February, 1999.
[RFC2368] P. Hoffman, L. Masinter, J. Zawinski, "The mailto URL scheme." RFC 2368, July, 1998.
[RFC3023] M. Murata, S. St.Laurent, D. Kohn, "XML Media Types." RFC 3023, January, 2001.
[RFC3010] S. Shepler, B. Callaghan, D. Robinson, R. Thurlow, C. Beame, M. Eisler, D.Noveck "NFS version 4 Protocol." RFC 3010, December 2000.
[UTF-8] F. Yergeau, "UTF-8, a transformation format of Unicode and ISO 10646." RFC 2279, January, 1998.
[RFC2026] S.Bradner, "The Internet Standards Process - Revision 3." RFC 2026, BCP 9. Harvard, October, 1996.
[RFC2255] T. Howes, M. Smith, "The LDAP URL Format." RFC 2255. Netscape, December, 1997.
[RFC2251] M. Wahl, T. Howes, S. Kille, "Lightweight Directory Access Protocol (v3)." RFC 2251. Critical Angle, Netscape, Isode, December, 1997.
[CaseMap] M. Davis, "Case Mappings", Unicode Standard Annex #21, March 26, 2001. http://www.unicode.org/unicode/reports/tr21
Geoffrey Clemm
IBM
20 Maguire Road
Lexington, MA 02421
Email: geoffrey.clemm@us.ibm.com
Anne Hopkins
Microsoft Corporation
One Microsoft Way
Redmond, WA 98052
Email: annehop@microsoft.com
Eric Sedlar
Oracle Corporation
500 Oracle Parkway
Redwood Shores, CA 94065
Email: eric.sedlar@oracle.com
Jim Whitehead
U.C. Santa Cruz
Dept. of Computer Science
Baskin Engineering
1156 High Street
Santa Cruz, CA 95064
Email: ejw@cse.ucsc.edu
All XML elements defined in this Document Type Definition (DTD) belong to the DAV namespace. This DTD should be viewed as an addendum to the DTD provided in [RFC2518], section 23.1.
<!-- Privileges (Section 3) -->
<!ELEMENT read EMPTY>
<!ELEMENT write EMPTY>
<!ELEMENT write-properties EMPTY>
<!ELEMENT
write-content EMPTY>
<!ELEMENT
unlock EMPTY>
<!ELEMENT read-acl EMPTY>
<!ELEMENT read-current-user-privilege-set EMPTY>
<!ELEMENT write-acl EMPTY>
<!ELEMENT
bind EMPTY>
<!ELEMENT
unbind EMPTY>
<!ELEMENT all EMPTY>
<!-- Principal Properties (Section 4) -->
<!ELEMENT principal EMPTY>
<!ELEMENT alternate-URI-set (href*)>
<!ELEMENT principal-URL (href)>
<!ELEMENT group-member-set (href*)>
<!ELEMENT group-membership (href*)>
<!-- Access Control Properties (Section 5) -->
<!-- DAV:owner Property (Section 5.1) -->
<!ELEMENT owner (href)>
<!-- DAV:supported-privilege-set Property (Section 5.2) -->
<!ELEMENT supported-privilege-set (supported-privilege*)>
<!ELEMENT supported-privilege
(privilege, abstract?, description, supported-privilege*)>
<!ELEMENT privilege ANY>
<!ELEMENT abstract EMPTY>
<!ELEMENT description #PCDATA>
<!--
DAV:current-user-privilege-set Property (Section 5.3) -->
<!ELEMENT current-user-privilege-set (privilege*)>
<!-- DAV:acl Property (Section 5.4) -->
<!ELEMENT acl (ace)* >
<!ELEMENT ace ((principal|invert), (grant|deny), protected?, inherited?)>
<!ELEMENT principal (href | all | authenticated | unauthenticated
| property | self)>
<!ELEMENT all EMPTY>
<!ELEMENT authenticated EMPTY>
<!ELEMENT unauthenticated EMPTY>
<!ELEMENT property ANY>
<!ELEMENT self EMPTY>
<!ELEMENT invert principal>
<!ELEMENT grant (privilege+)>
<!ELEMENT deny (privilege+)>
<!ELEMENT privilege ANY>
<!ELEMENT protected EMPTY>
<!ELEMENT inherited (href)>
<!-- DAV:acl-restrictions Property (Section 5.5) -->
<!ELEMENT acl-restrictions (grant-only?, no-invert?, deny-before-grant?, required-principal?)>
<!ELEMENT grant-only EMPTY>
<!ELEMENT no-invert EMPTY>
<!ELEMENT deny-before-grant EMPTY>
<!ELEMENT required-principal
(all? | authenticated? | unauthenticated? | self? | href* |property*)>
<!-- DAV:inherited-acl-set Property (Section 5.6) -->
<!ELEMENT inherited-acl-set (href*)>
<!-- DAV:principal-collection-set Property (Section 5.7) -->
<!ELEMENT principal-collection-set (href*)>
<!-- Access Control and Existing Methods (Section 7) -->
<!ELEMENT need-privileges (resource)* >
<!ELEMENT resource ( href , privilege ) >
<!-- ACL method preconditions (Section 8) -->
<!ELEMENT
no-ace-conflict EMPTY>
<!ELEMENT
no-protected-ace-conflict EMPTY>
<!ELEMENT
no-inherited-ace-conflict EMPTY>
<!ELEMENT
limited-number-of-aces EMPTY>
<!ELEMENT
deny-before-grant EMPTY>
<!ELEMENT
grant-only EMPTY>
<!ELEMENT
no-invert EMPTY>
<!ELEMENT
no-abstract EMPTY>
<!ELEMENT
not-supported-privilege EMPTY>
<!ELEMENT
missing-required-principal EMPTY>
<!ELEMENT
recognized-principal EMPTY>
<!ELEMENT
allowed-principal EMPTY>
<!-- REPORTs (Section 9) -->
<!ELEMENT acl-principal-prop-set ANY>
ANY value: a sequence of one or more elements, with at most one DAV:prop element.
<!ELEMENT principal-match ((principal-property | self), prop?)>
<!ELEMENT principal-property ANY>
ANY value: an element whose value identifies a property. The expectation is the value of the named property typically contains an href element that contains the URI of a principal
<!ELEMENT self EMPTY>
<!ELEMENT principal-property-search ((property-search+), prop?) >
<!ELEMENT property-search (prop, match) >
<!ELEMENT match #PCDATA >
<!ELEMENT principal-search-property-set (principal-search-property*) >
<!ELEMENT principal-search-property (prop, description) >
<!ELEMENT description #PCDATA >
The following table of WebDAV methods (as defined in RFC 2518, 2616, and 3253) clarifies which privileges are required for access for each method. Note that the privileges listed, if denied, MUST cause access to be denied. However, given that a specific implementation MAY define an additional custom privilege to control access to existing methods, having all of the indicated privileges does not mean that access will be granted. Note that lack of the indicated privileges does not imply that access will be denied, since a particular implementation may use a sub-privilege aggregated under the indicated privilege to control access. Privileges required refer to the current resource being processed unless otherwise specified.
METHOD
PRIVILEGES
GET <D:read>
HEAD <D:read>
OPTIONS <D:read>
PUT (target exists) <D:write-content>
on target resource
PUT (no target exists) <D:bind>
on parent collection of target
PROPPATCH <D:write-properties>
ACL <D:write-acl>
PROPFIND <D:read> (plus
<D:read-acl> and
<D:read-current-user-privilege-set>
as needed)
COPY (target exists) <D:read>,
<D:write-content> and <D:write-properties> on target resource
COPY (no target exists) <D:read>,
<D:bind> on target collection
MOVE (no target exists) <D:unbind>
on source collection and <D:bind> on target collection
MOVE (target exists) As above, plus
<D:unbind> on the target collection
DELETE <D:unbind>
on parent collection
LOCK (target exists) <D:write-content>
LOCK (no target exists) <D:bind>
on parent collection
MKCOL <D:bind>
on parent collection
UNLOCK <D:unlock>
CHECKOUT <D:write-properties >
CHECKIN <D:write-properties >
REPORT <D:read>
(on all referenced resources)
VERSION-CONTROL <D:write-properties>
MERGE <D:write-content>
MKWORKSPACE <D:write-content> on parent
collection
BASELINE-CONTROL <D:write-properties>
and <D:write-content>
MKACTIVITY <D:write-content> on parent
collection