| Internet-Draft | Constrained RESTful Application Language | January 2020 |
| Hartke | Expires 11 July 2020 | [Page] |
The Constrained RESTful Application Language (CoRAL) defines a data model and interaction model as well as two specialized serialization formats for the description of typed connections between resources on the Web ("links"), possible operations on such resources ("forms"), and simple resource metadata.¶
This note is to be removed before publishing as an RFC.¶
The issues list for this Internet-Draft can be found at <https://github.com/core-wg/coral/labels/coral>.¶
Companion material for this Internet-Draft can be found at <https://github.com/core-wg/coral>.¶
This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79.¶
Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet-Drafts is at https://datatracker.ietf.org/drafts/current/.¶
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."¶
This Internet-Draft will expire on 11 July 2020.¶
Copyright (c) 2020 IETF Trust and the persons identified as the document authors. All rights reserved.¶
This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (https://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Simplified BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Simplified BSD License.¶
The Constrained RESTful Application Language (CoRAL) is a language for the description of typed connections between resources on the Web ("links"), possible operations on such resources ("forms"), and simple resource metadata.¶
CoRAL is intended for driving automated software agents that navigate a Web application based on a standardized vocabulary of link relation types and operation types. It is designed to be used in conjunction with a Web transfer protocol, such as the Hypertext Transfer Protocol (HTTP) [RFC7230] or the Constrained Application Protocol (CoAP) [RFC7252].¶
This document defines the CoRAL data model and interaction model as well as two specialized CoRAL serialization formats.¶
The data model derives from the Web Linking model of RFC 8288 [RFC8288] and primarily consists of two elements: "links" that describe the relationship between two resources and the type of that relationship; and "forms" that describe a possible operation on a resource and the type of that operation.¶
Additionally, the data model can describe simple resource metadata in similarly to statements in the Resource Description Framework (RDF) [W3C.REC-rdf11-concepts-20140225]. In contrast to RDF, the focus of CoRAL, however, is not on the description of a resource graph, but on the discovery of possible future application states of a software agent.¶
The interaction model derives from HTML [W3C.REC-html52-20171214] and specifies how an automated software agent can change the application state, i.e., navigate between resources by following links and perform operations on resources by submitting forms.¶
The primary serialization format is a compact, binary encoding of links and forms in Concise Binary Object Representation (CBOR) [I-D.ietf-cbor-7049bis]. The format is intended for environments with constraints on power, memory, and processing resources [RFC7228] and shares many similarities with the message format of the Constrained Application Protocol (CoAP) [RFC7252]. For example, it uses numeric identifiers instead of verbose strings for link relation types and operation types, and pre-parses Uniform Resource Identifiers (URIs) [RFC3986] into (what CoAP considers to be) their components, which considerably simplifies URI processing for constrained nodes that already implement CoAP. As a result, link serializations in CoRAL are often much more compact and easier to process than equivalent serializations in CoRE Link Format [RFC6690].¶
The secondary serialization format is a lightweight, textual encoding of links and forms that is intended to be easy to read and to write for humans. The format is loosely inspired by the syntax of Turtle [W3C.REC-turtle-20140225] and is mainly intended for giving examples with precise semantics.¶
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in BCP 14 [RFC2119] [RFC8174] when, and only when, they appear in all capitals, as shown here.¶
Terms defined in this document appear in cursive where they are introduced.¶
The Constrained RESTful Application Language (CoRAL) is designed for building Web-based applications [W3C.REC-webarch-20041215] in which automated software agents navigate between resources by following links and perform operations on resources by submitting forms.¶
Borrowing from HTML 5 [W3C.REC-html52-20171214], each such agent maintains a browsing context in which the representations of Web resources are processed. (In HTML, the browsing context typically corresponds to a tab or window in a Web browser.)¶
At any time, one representation in a browsing context is designated the active representation.¶
A resource representation in one of the CoRAL serialization formats is called a CoRAL document. The URI that was used to retrieve such a document is called the document's retrieval context.¶
A CoRAL document consists of a list of zero or more links and forms, collectively called elements. CoRAL serialization formats may define additional types of elements for efficiency or convenience, such as a base for relative URI references.¶
A link describes a relationship between two resources on the Web [RFC8288]. As in RFC 8288, a link in CoRAL has a link context, a link relation type, and a link target. However, in CoRAL, links do not have target attributes. Instead, a link may have a list of zero or more nested elements. These enable both the description of resource metadata (which is done in RFC 8288 with target attributes) and the chaining of links (which can be done in RFC 8288 by setting the anchor of one link to the target of another).¶
A link can be viewed as a statement of the form "{link context} has a {link relation type} resource at {link target}" where the link target may be further described by nested elements.¶
The link relation type identifies the semantics of a link. While in HTML and RFC 8288 these are typically denoted by an IANA-registered name, such as "stylesheet" or "type", link relation types in CoRAL are denoted by an Internationalized Resource Identifier (IRI) [RFC3987], such as <http://www.iana.org/assignments/relation/stylesheet> or <http://www.w3.org/1999/02/22-rdf-syntax-ns#type>. This allows for the creation of new link relation types without the risk of collisions when from different organizations or domains of knowledge. IRIs can also lead to documentation, schema, and other information about a link relation type. In CoRAL documents, these IRIs are only used as identity tokens, though, and are compared with Simple String Comparison as specified in Section 5.3.1 of RFC 3987.¶
The link context and the link target can be both either a URI, a literal value, or an anonymous resource. If the scheme of a URI indicates a Web transfer protocol such as HTTP or CoAP, an agent can dereference the URI and navigate the browsing context to the link target; this is called following the link. A literal value can be either a Boolean value, an integer, a floating-point number, a date/time instant, a byte string, or a text string. An anonymous resource is a resource which is not identified by a URI and is not a literal.¶
A link can occur as a top-level element in a document or as a nested element within a link. When a link occurs as a top-level element, the link context implicitly is the document's retrieval context. When a link occurs nested within a link, the link context of the inner link is the same resource as the link target of the outer link.¶
There are no restrictions on the cardinality of links; there can be multiple links to and from a particular target, and multiple links of the same or different types between a given link context and target. However, the nested data structure constrains the description of a resource graph to a tree: Links between linked resources can only be described by further nesting links.¶
A form provides instructions to an agent for performing an operation on a Web resource. A form has a form context, an operation type, a request method, and a submission target. Additionally, a form may be accompanied by a list of zero or more form fields.¶
A form can be viewed as an instruction of the form "To perform an {operation type} operation on {form context}, make a {request method} request to {submission target}" where the request may be further described by form fields.¶
The operation type identifies the semantics of the operation. Operation types are denoted (like link relation types) by an IRI.¶
Both the form context and the submission target are denoted by a URI. The form context is the resource on which an operation is ultimately performed. To perform the operation, an agent needs to construct a request with the specified method and the specified submission target as the request URI. Usually, the submission target is the same resource as the form context, but it may be a different resource. Constructing and sending the request is called submitting the form.¶
A form can occur as a top-level element in a document or as a nested element within a link. When a form occurs as a top-level element, the form context implicitly is the document's retrieval context. When a form occurs nested within a link, the form context is the same resource as the link target of the enclosing link.¶
Form fields can be used to provide more detailed instructions to agents for constructing the request when submitting a form. For example, form fields can instruct the agent to include a certain payload or certain header fields in the request. Form fields might describe a payload by identifying acceptable media types, referencing a schema, or listing a number of data items that need to be included. Form fields can be specific to the Web transfer protocol that is used for submitting the form.¶
A form field is a pair of a form field type and a form field value. Additionally, a form field may have a list of zero or more nested elements that further describe the form field value.¶
The form field type identifies the semantics of the form field. Form field types are denoted (like link relation types and operation types) by an IRI.¶
The form field value can be either a URI, a Boolean value, an integer, a floating-point number, a date/time instant, a byte string, a text string, or null. A null denotes the intentional absence of any form field value.¶
An agent begins interacting with an application by performing a GET request on an entry point URI. The entry point URI is the only URI that the agent is expected to know before interacting with the application. From then on, the agent is expected to make all requests by following links and submitting forms that are provided by servers in responses. The entry point URI can be obtained through some discovery process or manual configuration.¶
If dereferencing the entry point URI yields a CoRAL document (or any other representation that implements the CoRAL data and interaction model), then the agent makes this document the active representation in the browsing context and proceeds as follows:¶
The first step for the agent is to decide what to do next, i.e., which type of link to follow or type of form to submit, based on the link relation types and operation types it knows.¶
The agent then finds the link(s) or form(s) with the respective type in the active representation. This may yield one or more candidates, from which the agent will have to select the most appropriate one. The set of candidates can be empty, for example, when a transition is not supported or not allowed.¶
The agent selects one of the candidates based on the order of appearance in the document and the resource metadata associated with them in the form of nested elements and form fields. Examples for resource metadata include the indication of a content type for the target resource representation, the URI scheme of a link target, or the request method of a form.¶
The agent obtains the request URI from the link target or submission target. Link targets and submission targets may be denoted by relative URI references, which need to be resolved to obtain the request URI. Fragment identifiers are not part of the request URI and MUST be separated from the rest of the URI prior to the next step.¶
The agent constructs a new request with the request URI. If the agent is following a link, then the request method MUST be GET. If the agent is submitting a form, then the request method MUST be the one indicated by the form. An IRI may need to be converted to a URI (Section 3.1 of RFC 3987) for protocols that do not support IRIs.¶
The agent should set HTTP header fields and CoAP request options according to the nested elements of a link or form fields of a form (e.g., set the HTTP Accept header field or the CoAP Accept option when a media type for the target resource is provided). Depending on the operation type of a form, the agent may also have to include a request payload that matches the specifications of some form fields.¶
The agent sends the request and receives the response.¶
If a fragment identifier was separated from the request URI, the agent selects the fragment indicated by the fragment identifier within the received representation.¶
The agent updates the browsing context by making the (selected fragment of the) received representation the active representation.¶
Finally, the agent processes the representation according to the semantics of its content type. If the representation is a CoRAL document (or any other representation that implements the CoRAL data and interaction model), the agent has the choice of what to do next again -- and the cycle repeats.¶
Each browsing context has a session history that lists the resource representations that the agent has processed, is processing, or will process. The maximum length of the session history is up for the agent to decide and may be zero.¶
An entry in the session history consists of a resource representation and the request URI that was used to retrieve the representation. New entries are added to the session history as the agent navigates from resource to resource.¶
In addition to following links and submitting forms, an agent can decide tp navigate a browsing context by traversing the session history. For example, when an agent receives a representation that does not contain any further links or forms, it can set the active representation back to one it has visited earlier.¶
Traversing the history should take advantage of caches to avoid new requests. An agent MAY reissue a safe request (e.g., a GET request) when it does not have a fresh representation in its cache. An agent MUST NOT reissue an unsafe request (e.g., a PUT or a POST request) unless it intends to perform that operation again.¶
This section defines the encoding of documents in the CoRAL binary format.¶
A document in the binary format is encoded in Concise Binary Object Representation (CBOR) [I-D.ietf-cbor-7049bis]. The encoding MUST satisfy the Core Deterministic Encoding Requirements specified in Section XX of RFC XXXX [I-D.ietf-cbor-7049bis].¶
The CBOR structure of a document is presented in the Concise Data Definition Language (CDDL) [RFC8610]. All CDDL rules not defined in this document are defined in Appendix D of RFC 8610 [RFC8610].¶
The media type of documents in the binary format is "application/coral+cbor".¶
The data structure of a document in the binary format is made up of three kinds of elements: links, forms, and (as an extension to the CoRAL data model) directives. Directives provide a way to encode URI references with a common base more efficiently.¶
A document in the binary format is encoded as a CBOR array that contains zero or more elements. An element is either a link, a form, or a directive.¶
The elements are processed in the order they appear in the document. Document processors need to maintain an environment while iterating an array of elements. The environment consists of two variables: the current context and the current base. Both the current context and the current base are initially set to the document's retrieval context.¶
Directives provide the ability to manipulate the environment while processing elements.¶
There is a single type of directives available: the Base directive.¶
directive = base-directive¶
A Base directive is encoded as a CBOR array that contains the unsigned integer 1 and a base URI.¶
base-directive = [1, baseURI]¶
The base URI is denoted by a Constrained Resource Identifier (CoRI) reference [I-D.ietf-core-href]. The CoRI reference MUST be resolved against the current context (not the current base).¶
The directive is processed by resolving the CoRI reference against the current context and assigning the result to the current base.¶
IRIs in links and forms are encoded as CoRI references.¶
IRI = CoRI¶
A CoRI reference is processed by resolving it to an IRI as specified in Section XX of RFC XXXX [I-D.ietf-core-href] using the current base.¶
A link is encoded as a CBOR array that contains the unsigned integer 2, the link relation type, the link target, and, optionally, an array of zero or more nested elements.¶
link = [2, relation-type, link-target, ?[*element]]¶
The link relation type is encoded as a text string that conforms to the syntax of an IRI [RFC3987].¶
relation-type = text¶
The link target is either an IRI, a literal value, or null.¶
The nested elements, if any, MUST be processed in a fresh environment. Both the current context and current base in this environment are initially set to the link target of the enclosing link.¶
A form is encoded as a CBOR array that contains the unsigned integer 3, the operation type, the submission target, and, optionally, an array of zero or more form fields.¶
form = [3, operation-type, submission-target, ?[*form-field]]¶
The operation type is encoded as a text string that conforms to the syntax of an IRI [RFC3987].¶
operation-type = text¶
The submission target is an IRI.¶
submission-target = IRI¶
The request method is either implied by the operation type or encoded as a form field. If both are given, the form field takes precedence over the operation type. Either way, the method MUST be applicable to the Web transfer protocol identified by the scheme of the submission target.¶
The form fields, if any, MUST be processed in a fresh environment. Both the current context and the current base in this environment are initially set to the submission target of the enclosing form.¶
A form field is encoded as a CBOR sequence that consists of a form field type, a form field value, and, optionally, an array of zero or more nested elements.¶
form-field = (form-field-type, form-field-value, ?[*element])¶
The form field type is encoded as a text string that conforms to the syntax of an IRI [RFC3987].¶
form-field-type = text¶
The form field value is either an IRI, a literal value, or null.¶
form-field-value = IRI / literal / null¶
The nested elements, if any, MUST be processed in a fresh environment. Both the current context and current base in this environment are initially set to the form field value of the enclosing form field.¶
A document in the binary format can reference values from an external dictionary to reduce representation size and processing cost. Dictionary references can be used in place of link relation types, link targets, operation types, submission targets, form field types, and form field values.¶
A dictionary reference is encoded as an unsigned integer. Where a dictionary reference cannot be expressed unambiguously, the unsigned integer is tagged with CBOR tag TBD6, as follows:¶
relation-type /= uint¶
link-target /= #6.TBD6(uint)¶
operation-type /= uint¶
submission-target /= #6.TBD6(uint)¶
form-field-type /= uint¶
form-field-value /= #6.TBD6(uint)¶
A dictionary reference MUST NOT refer to a dictionary value that is otherwise not allowed. For example, a dictionary reference that is used in place of a link relation type is not allowed to refer to a Boolean value.¶
The "application/coral+cbor" media type for documents in the binary format is defined to have a "dictionary" parameter that specifies the dictionary in use. The dictionary is identified by a URI [RFC3986]. For example, a CoRAL document that uses the dictionary identified by the URI <http://example.com/dictionary> can use the following content type:¶
The URI serves only as an identifier; it does not necessarily have to be dereferencable (or even use a dereferencable URI scheme). It is permissible, though, to use a dereferencable URI and to serve a representation that provides information about the dictionary in a machine- or human-readable way. (The representation format and security considerations of such a representation are outside the scope of this document.)¶
For simplicity, a CoRAL document can reference values only from one dictionary; the value of the "dictionary" parameter MUST be a single URI. The "dictionary" parameter is OPTIONAL. If it is absent, the default dictionary specified in Appendix B of this document is assumed.¶
Once a dictionary has made an assignment, the assignment MUST NOT be changed or removed. A dictionary, however, may contain additional information about an assignment, which may change over time.¶
In CoAP [RFC7252], media types (including specific values for their parameters) are encoded as an unsigned integer called the "content format" of a representation. For use with CoAP, each new CoRAL dictionary therefore needs to have a new content format registered with IANA in the CoAP Content-Formats Registry.¶
This section defines the syntax of documents in the CoRAL textual format using two grammars: The lexical grammar defines how Unicode characters are combined to form line terminators, white space, comments, and tokens. The syntactic grammar defines how tokens are combined to form documents. Both grammars are presented in Augmented Backus-Naur Form (ABNF) [RFC5234].¶
A document in the textual format is a Unicode string in a Unicode encoding form [UNICODE]. The media type for such documents is "text/coral". The "charset" parameter of textual media types [RFC6657] is not used; instead, charset information is transported inside the document in the form of an OPTIONAL Byte Order Mark (BOM). The use of the UTF-8 encoding scheme [RFC3629] without a BOM is RECOMMENDED.¶
The lexical structure of a document in the textual format is made up of four basic elements: line terminators, white space, comments, and tokens. Of these, only tokens are significant in the syntactic grammar. There are three kinds of tokens: identifier tokens, literal tokens, and punctuator tokens.¶
token = identifier / IRIref / boolean / integer / float¶
token =/ datetime / bytes / text / null / punctuator¶
When several lexical grammar rules match a sequence of characters in a document, the longest match takes priority.¶
Line terminators divide text into lines. A line terminator is any Unicode character with Line_Break class BK, CR, LF, or NL. However, any CR character that immediately precedes a LF character is ignored. (This affects only the numbering of lines in error messages.)¶
White space is a sequence of one or more white space characters. A white space character is any Unicode character with the White_Space property.¶
Comments are sequences of characters that are ignored when parsing text into tokens. Single-line comments begin with the characters "//" and extend to the end of the line. Delimited comments begin with the characters "/*" and end with the characters "*/". Delimited comments can occupy a portion of a line, a single line, or multiple lines.¶
Comments do not nest. The character sequences "/*" and "*/" have no special meaning within a single-line comment; the character sequences "//" and "/*" have no special meaning within a delimited comment.¶
An identifier token is a user-defined symbolic name. The rules for identifiers correspond to those recommended by the Unicode Standard Annex #31 [UAX31] with the following profile:¶
identifier = START *CONTINUE *(MEDIAL 1*CONTINUE)¶
START = <Any character with the XID_Start property>¶
CONTINUE = <Any character with the XID_Continue property>¶
MEDIAL = "-" / "." / "~" / %x58A / %xF0B¶
MEDIAL =/ %x2010 / %x2027 / %x30A0 / %x30FB¶
All identifiers MUST be converted into Unicode Normalization Form C (NFC), which is defined in Unicode Standard Annex #15 [UAX15]. Comparison of identifiers is based on NFC and case-sensitive (unless otherwise noted).¶
A literal token is a textual representation of a value.¶
IRI reference tokens denote references to resources on the Web.¶
An IRI reference literal consists of a Unicode string that conforms to the syntax defined in RFC 3987 [RFC3987]. An IRI reference is either an IRI or a relative reference. IRI references are enclosed in angle brackets ("<" and ">").¶
The case-insensitive tokens "true" and "false" denote the Boolean values true and false, respectively.¶
boolean = "true" / "false"¶
Integer literal tokens denote an integer value of unspecified precision. By default, integer literals are expressed in decimal, but they can also be specified in an alternate base using a prefix: Binary literals begin with "0b", octal literals begin with "0o", and hexadecimal literals begin with "0x".¶
Decimal literals contain the digits "0" through "9". Binary literals contain "0" and "1", octal literals contain "0" through "7", and hexadecimal literals contain "0" through "9" as well as "A" through "F" in upper- or lowercase.¶
Negative integers are expressed by prepending a minus sign ("-").¶
Floating-point literal tokens denote a floating-point number of unspecified precision.¶
Floating-point literals consist of a sequence of decimal digits followed by a fraction, an exponent, or both. The fraction consists of a decimal point (".") followed by a sequence of decimal digits. The exponent consists of the letter "e" in upper- or lowercase, followed by an optional sign and a sequence of decimal digits that indicate a power of 10 by which the value preceding the "e" is multiplied.¶
Negative floating-point values are expressed by prepending a minus sign ("-").¶
float = ["+" / "-"] 1*DIGIT [fraction] [exponent]¶
fraction = "." 1*DIGIT¶
exponent = (%x45 / %x65) ["+" / "-"] 1*DIGIT¶
A floating-point literal can additionally denote either the special "Not-a-Number" (NaN) value, positive infinity, or negative infinity. The NaN value is produced by the case-insensitive token "NaN". The two infinite values are produced by the case-insensitive tokens "+Infinity" (or simply "Infinity") and "-Infinity".¶
Date/time literal tokens denote an instant in time.¶
A date/time literal consists of the prefix "dt" and a sequence of Unicode characters in Internet Date/Time Format [RFC3339], enclosed in single quotes.¶
Byte string literal tokens denote an ordered sequence of bytes.¶
A byte string literal consists of a prefix and zero or more bytes encoded in Base16, Base32, or Base64 [RFC4648], enclosed in single quotes. Byte string literals encoded in Base16 begin with "h" or "b16", byte string literals encoded in Base32 begin with "b32", and byte string literals encoded in Base64 begin with "b64".¶
Text string literal tokens denote a Unicode string.¶
A text string literal consists of zero or more Unicode characters enclosed in double quotes. It can include simple escape sequences (such as \t for the tab character) as well as hexadecimal and Unicode escape sequences.¶
text = DQUOTE *(char / %x5C escape) DQUOTE¶
char = <Any character except DQUOTE, %x5C, and line terminators>¶
escape = simple-escape / hexadecimal-escape / unicode-escape¶
simple-escape = %x30 / %x62 / %x74 / %x6E / %x76¶
simple-escape =/ %x66 / %x72 / %x22 / %x27 / %x5C¶
hexadecimal-escape = (%x78 / %x58) 2HEXDIG¶
unicode-escape = %x75 4HEXDIG / %x55 8HEXDIG¶
DQUOTE = %x22¶
An escape sequence denotes a single Unicode code point. For hexadecimal and Unicode escape sequences, the code point is expressed by the hexadecimal number following the "\x", "\X", "\u", or "\U" prefix. Simple escape sequences indicate the code points listed in Table 1.¶
| Escape Sequence | Code Point | Character Name |
|---|---|---|
| \0 | U+0000 | Null |
| \b | U+0008 | Backspace |
| \t | U+0009 | Character Tabulation |
| \n | U+000A | Line Feed |
| \v | U+000B | Line Tabulation |
| \f | U+000C | Form Feed |
| \r | U+000D | Carriage Return |
| \" | U+0022 | Quotation Mark |
| \' | U+0027 | Apostrophe |
| \\ | U+005C | Reverse Solidus |
The case-insensitive tokens "null" and "_" denote the intentional absence of any value.¶
null = "null" / "_"¶
Punctuator tokens are used for grouping and separating.¶
punctuator = "#" / ":" / "=" / "@" / "[" / "]" / "{" / "}" / "->"¶
The syntactic structure of a document in the textual format is made up of three kinds of elements: links, forms, and (as an extension to the CoRAL data model) directives. Directives provide a way to make documents easier to read and write by setting a base for relative IRI references and introducing shorthands for IRIs.¶
A document in the textual format consists of a sequence of zero or more elements. An element is either a link, a form, or a directive.¶
The elements are processed in the order they appear in the document. Document processors need to maintain an environment while iterating a sequence of elements. The environment consists of three variables: the current context, the current base, and the current mapping from identifiers to IRIs. Both the current context and the current base are initially set to the document's retrieval context. The current mapping from identifiers to IRIs is initially empty.¶
Directives provide the ability to manipulate the environment while processing elements.¶
All directives start with a number sign ("#") followed by an identifier. The identifier is case-insensitive and restricted to Unicode characters in the Basic Latin block.¶
The following two types of directives are available: the Base directive and the Using directive.¶
directive = base-directive / using-directive¶
A Base directive consists of a number sign ("#"), followed by the case-insensitive token "base", followed by a base IRI.¶
base-directive = "#" "base" baseIRI¶
The base IRI is denoted by an IRI reference. The IRI reference MUST be resolved against the current context (not the current base).¶
baseIRI = IRIref¶
The directive is processed by resolving the IRI reference against the current context and assigning the result to the current base.¶
A Using directive consists of a number sign ("#"), followed by the case-insensitive token "using", optionally followed by an identifier and an equals sign ("="), finally followed by an IRI. If the identifier is not specified, it is assumed to be the empty string.¶
using-directive = "#" "using" [identifier "="] IRIref¶
The directive is processed by adding the specified identifier and IRI to the current mapping from identifiers to IRIs. It is an error if the identifier is already present in the mapping or if the IRI is not an IRI but a relative reference.¶
IRIs in links and forms can be either denoted by an IRI reference or looked up from a mapping from identifiers to IRIs using names. There are three kinds of names: simple names, qualified names, and predefined names.¶
IRI = IRIref / simple-name / qualified-name / predefined-name¶
Both IRI references and names are processed by resolving them to an IRI, as described in the following sub-sections.¶
An IRI reference is resolved to an IRI as specified in Section 6.5 of RFC 3987 [RFC3987] using the current base.¶
A simple name consists of an identifier.¶
simple-name = identifier¶
A simple name is resolved to an IRI by looking up the empty string in the current mapping from identifiers to IRIs and appending the given identifier to the result. It is an error if the empty string is not present in the mapping.¶
A qualified name consists of two identifiers separated by a colon (":").¶
qualified-name = identifier ":" identifier¶
A qualified name is resolved to an IRI by looking up the identifier given on the left hand side in the current mapping from identifiers to IRIs. The identifier given on the right hand side is appended to the result. It is an error if the identifier on the left hand side is not present in the mapping.¶
A predefined name consists of a commercial at sign ("@") followed by an identifier. The identifier is case-insensitive and restricted to Unicode characters in the Basic Latin block.¶
predefined-name = "@" identifier¶
A predefined name is resolved to an IRI by looking up the identifier in Table 2. It is an error if the identifier is not present in the table.¶
| Identifier | IRI |
|---|---|
| direction | <http://coreapps.org/base#direction> |
| language | <http://coreapps.org/base#language> |
A link consists of the link relation type, followed by the link target, optionally followed by a sequence of zero or more nested elements enclosed in curly brackets ("{" and "}").¶
link = relation-type link-target ["{" *element "}"]¶
The link relation type is an IRI.¶
relation-type = IRI¶
The link target is either an IRI, a literal value, or null.¶
The nested elements, if any, MUST be processed in a fresh environment. Both the current context and current base in this environment are initially set to the link target of the enclosing link. The mapping from identifiers to IRIs is initially set to a copy of the mapping from identifiers to IRIs in the current environment.¶
A form consists of the operation type, followed by a "->" token and the submission target, optionally followed by a sequence of zero or more form fields enclosed in square brackets ("[" and "]").¶
form = operation-type "->" submission-target ["[" *form-field "]"]¶
The operation type is an IRI.¶
operation-type = IRI¶
The submission target is an IRI.¶
submission-target = IRI¶
The request method is either implied by the operation type or encoded as a form field. If both are given, the form field takes precedence over the operation type. Either way, the method MUST be applicable to the Web transfer protocol identified by the scheme of the submission target.¶
The form fields, if any, MUST be processed in a fresh environment. Both the current context and the current base in this environment are initially set to the submission target of the enclosing form. The mapping from identifiers to IRIs is initially set to a copy of the mapping from identifiers to IRIs in the current environment.¶
A form field consists of a form field type, followed by a form field value, optionally followed by a sequence of zero or more nested elements enclosed in curly brackets ("{" and "}").¶
form-field = form-field-type form-field-value ["{" *element "}"]¶
The form field type is an IRI.¶
form-field-type = IRI¶
The form field value is either an IRI, a literal value, or null.¶
form-field-value = IRI / literal / null¶
The nested elements, if any, MUST be processed in a fresh environment. Both the current context and current base in this environment are initially set to the form field value of the enclosing form field. The mapping from identifiers to IRIs is initially set to a copy of the mapping from identifiers to IRIs in the current environment.¶
By default, a CoRAL document is a representation that captures the current state of a resource. The meaning of a CoRAL document changes when it is submitted in a request. Depending on the request method, the CoRAL document can capture the intended state of a resource (PUT) or be subject to application-specific processing (POST).¶
A PUT request with a CoRAL document enclosed in the request payload requests that the state of the target resource be created or replaced with the state described by the CoRAL document. A successful PUT of a CoRAL document generally means that a subsequent GET on that same target resource would result in an equivalent document being sent in a success response.¶
An origin server SHOULD verify that a submitted CoRAL document is consistent with any constraints the server has for the target resource. When a document is inconsistent with the target resource, the origin server SHOULD either make it consistent (e.g., by removing inconsistent elements) or respond with an appropriate error message containing sufficient information to explain why the document is unsuitable.¶
The retrieval context and the base URI of a CoRAL document in a PUT are the request URI of the request.¶
A POST request with a CoRAL document enclosed in the request payload requests that the target resource process the CoRAL document according to the resource's own specific semantics.¶
The retrieval context of a CoRAL document in a POST is defined by the target resource's processing semantics; it can be an unspecified URI. The base URI of the document is the request URI of the request.¶
In a response, the meaning of a CoRAL document changes depending on the request method and the response status code. For example, a CoRAL document in a successful response to a GET represents the current state of the target resource, whereas a CoRAL document in a successful response to a POST might represent either the processing result or the new resource state. A CoRAL document in an error response represents the error condition, usually describing the error state and what next steps are suggested for resolving it.¶
Success responses have a response status code that indicates that the client's request was successfully received, understood, and accepted (2xx in HTTP, 2.xx in CoAP). When the representation in a success response does not describe the state of the target resource, it describes result of processing the request. For example, when a request has been fulfilled and has resulted in one or more new resources being created, a CoRAL document in the response can link to and describe the resource(s) created.¶
The retrieval context and the base URI of a CoRAL document representing the current state of a resource are the request URI of the request.¶
The retrieval context of a CoRAL document representing a processing result is an unspecified URI that refers to the processing result itself. The base URI of the document is the request URI of the request.¶
Redirection responses have a response status code that indicates that further action needs to be taken by the agent (3xx in HTTP). A redirection response, for example, might indicate that the target resource is available at a different URI or the server offers a choice of multiple matching resources, each with its own specific URI.¶
In the latter case, the representation in the response might contain a list of resource metadata and URI references (i.e., links) from which the agent can choose the most preferred one.¶
The retrieval context of a CoRAL document representing such multiple choices in a redirection response is an unspecified URI that refers to the redirection itself. The base URI of the document is the request URI of the request.¶
Error response have a response status code that indicates that either the request cannot be fulfilled or the server failed to fulfill an apparently valid request (4xx or 5xx in HTTP, 4.xx or 5.xx in CoAP). A representation in an error response describes the error condition.¶
The retrieval context of a CoRAL document representing such an error condition is an unspecified URI that refers to the error condition itself. The base URI of the document is the request URI of the request.¶
This section discusses some considerations in creating CoRAL-based applications and vocabularies.¶
CoRAL-based applications naturally implement the Web architecture [W3C.REC-webarch-20041215] and thus are centered around orthogonal specifications for identification, interaction, and representation:¶
Specifications for CoRAL-based applications need to list the specific components used in the application interface and their identifiers. This should include the following items:¶
URIs [RFC3986] are a cornerstone of Web-based applications. They enable the uniform identification of resources and are used every time a client interacts with a server or a resource representation needs to refer to another resource.¶
URIs often include structured application data in the path and query components, such as paths in a filesystem or keys in a database. It is a common practice in HTTP-based application programming interfaces (APIs) to make this part of the application specification, i.e., to prescribe fixed URI templates that are hard-coded in implementations. However, there are a number of problems with this practice [I-D.nottingham-rfc7320bis].¶
In CoRAL-based applications, resource names are therefore not part of the application specification -- they are an implementation detail. The specification of a CoRAL-based application MUST NOT mandate any particular form of resource name structure.¶
BCP 190 [I-D.nottingham-rfc7320bis] describes the problematic practice of fixed URI structures in more detail and provides some acceptable alternatives.¶
This document places no restrictions on the number of elements in a CoRAL document or the depth of nested elements. Applications using CoRAL (in particular those running in constrained environments) may limit these numbers and give specific implementation limits that an application implementation must at least support to be interoperable.¶
Applications may also mandate the following and other restrictions:¶
New link relation types, operation types, and form field types can be minted by defining an IRI [RFC3987] that uniquely identifies the item. Although the IRI can point to a resource that contains a definition of the semantics, clients SHOULD NOT automatically access that resource to avoid overburdening its server. The IRI SHOULD be under the control of the person or party defining it, or be delegated to them.¶
To avoid interoperability problems, it is RECOMMENDED that only IRIs are minted that are normalized according to Section 5.3 of RFC 3987. Non-normalized forms that are best avoided include:¶
IRIs that identify vocabulary do not need to be registered. The inclusion of domain names in IRIs allows for the decentralized creation of new IRIs without the risk of collisions.¶
However, IRIs can be relatively verbose and impose a high overhead on a representation. This can be a problem in constrained environments [RFC7228]. Therefore, CoRAL alternatively allows the use of unsigned integers to reference CBOR data items from a dictionary, as specified in Section 3.2. These impose a much smaller overhead but instead need to be assigned by an authority to avoid collisions.¶
Link relation types registered in the IANA Link Relations Registry, such as "collection" [RFC6573] or "icon" [W3C.REC-html52-20171214], can be used in CoRAL by appending the registered name to the IRI <http://www.iana.org/assignments/relation/>:¶
#using iana = <http://www.iana.org/assignments/relation/> iana:collection </items> iana:icon </favicon.png>¶
The convention of appending the relation type name to the prefix "http://www.iana.org/assignments/relation/" to form IRIs is adopted from Atom [RFC4287]; see also Appendix A.2 of RFC 8288 [RFC8288].¶
Note that registered relation type names are required to be lowercase ASCII letters (Section 3.3 of RFC 8288).¶
An RDF statement [W3C.REC-rdf11-concepts-20140225] says that some relationship, indicated by a predicate, holds between two resources. Existing RDF vocabularies can therefore be good source for link relation types that describe resource metadata. For example, a CoRAL document could use the FOAF vocabulary [FOAF] to describe the person or software that made it:¶
#using rdf = <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
#using foaf = <http://xmlns.com/foaf/0.1/>
foaf:maker null {
rdf:type <http://xmlns.com/foaf/0.1/Person>
foaf:familyName "Hartke"
foaf:givenName "Klaus"
foaf:mbox <mailto:klaus.hartke@ericsson.com>
}
¶
Text strings that are the target of a link can be associated with a language tag [RFC5646] and a base text direction (i.e., right-to-left or left-to-right) by nesting links of type <http://coreapps.org/base#language> and <http://coreapps.org/base#direction> under that link, respectively:¶
#using <http://coreapps.org/base#>
#using iana = <http://www.iana.org/assignments/relation/>
iana:terms-of-service </tos> {
title "Nutzungsbedingungen" {
language "de"
direction "ltr"
}
title "Terms of use" {
language "en-US"
direction "ltr"
}
}
¶
The link relation types <http://coreapps.org/base#language> and <http://coreapps.org/base#direction> are defined in Appendix A.¶
When a document links to many Web resources and an agent needs a representation of each of them, it can be inefficient to retrieve each representations individually. To minimize round-trips, documents can embed representations of resources.¶
A representation can be embedded in a document by including a link of type <http://coreapps.org/base#representation>:¶
#using <http://coreapps.org/base#>
#using http = <http://coreapps.org/http#>
#using iana = <http://www.iana.org/assignments/relation/>
iana:icon </favicon.gif> {
representation
b64'R0lGODlhAQABAAAAACH5BAEKAAEALAAAAAABAAEAAAIAOw==' {
http:type "image/gif"
}
}
¶
An embedded representation should have a nested link of type <http://coreapps.org/http#type> or <http://coreapps.org/coap#type> that indicates the content type of the representation.¶
The link relation types <http://coreapps.org/base#representation>, <http://coreapps.org/http#type>, and <http://coreapps.org/coap#type> are defined in Appendix A.¶
Data items in the CoRAL binary format (Section 3) may be embedded in other CBOR structures [I-D.ietf-cbor-7049bis]. Specifications using CDDL [RFC8610] can reference the following CDDL definitions for this purpose:¶
For each embedded document, link, and form, the specification for the embedding CBOR structure needs to specify the document retrieval context, the link context, and the form context, respectively.¶
CoRAL document processors need to be fully prepared for all types of hostile input that may be designed to corrupt, overrun, or achieve control of the agent processing the document. For example, hostile input may be constructed to overrun buffers, allocate very big data structures, or exhaust the stack depth by setting up deeply nested elements. Processors need to have appropriate resource management to mitigate these attacks.¶
CoRAL serialization formats intentionally do not feature the equivalent of XML entity references so as to preclude the entire class of attacks relating to them, such as exponential XML entity expansion ("billion laughs") [CAPEC-197] and malicious XML entity linking [CAPEC-201].¶
Implementers of the CoRAL binary format need to consider the security aspects of decoding CBOR. See Section XX of RFC XXXX [I-D.ietf-cbor-7049bis] for security considerations relating to CBOR. In particular, different number encodings for the same numeric value are not equivalent in CoRAL (e.g., a floating-point value of 0.0 is not the same as the integer 0).¶
Implementers of the CoRAL textual format need to consider the security aspects of handling Unicode input. See Unicode Technical Report #36 [UTR36] for security considerations relating to visual spoofing and misuse of character encodings. See Section 10 of RFC 3629 [RFC3629] for security considerations relating to UTF-8. See Unicode Technical Standard #39 [UTS39] for security mechanisms that can be used to detect possible security problems relating to Unicode.¶
CoRAL makes extensive use of resource identifiers. See Section 7 of RFC 3986 [RFC3986] for security considerations relating to URIs. See Section 8 of RFC 3987 [RFC3987] for security considerations relating to IRIs. See Section XX of RFC XXXX [I-D.ietf-core-href] for security considerations relating to CoRIs.¶
The security of applications using CoRAL can depend on the proper preparation and comparison of internationalized strings. For example, such strings can be used to make authentication and authorization decisions, and the security of an application could be compromised if an entity providing a given string is connected to the wrong account or online resource based on different interpretations of the string. See RFC 6943 [RFC6943] for security considerations relating to identifiers in IRIs and other strings.¶
CoRAL is intended to be used in conjunction with a Web transfer protocol like HTTP or CoAP. See Section 9 of RFC 7230 [RFC7230], Section 9 of RFC 7231 [RFC7231], etc., for security considerations relating to HTTP. See Section 11 of RFC 7252 [RFC7252] for security considerations relating to CoAP.¶
CoRAL does not define any specific mechanisms for protecting the confidentiality and integrity of CoRAL documents. It relies on security mechanisms on the application layer or transport layer for this, such as Transport Layer Security (TLS) [RFC8446].¶
CoRAL documents and the structure of a web of resources revealed from automatically following links can disclose personal information and other sensitive information. Implementations need to prevent the unintentional disclosure of such information. See Section 9 of RFC 7231 [RFC7231] for additional considerations.¶
Applications using CoRAL ought to consider the attack vectors opened by automatically following, trusting, or otherwise using links and forms in CoRAL documents. See Section 5 of RFC 8288 [RFC8288] for related considerations.¶
In particular, when a CoRAL document is the representation of a resource, the server that is authoritative for that resource may not necessarily be authoritative for nested elements in the document. In this case, unless an application defines specific rules, any link or form where the link/form context and the document's retrieval context do not share the same Web origin [RFC6454] should be discarded ("same-origin policy").¶
This document registers the media type "application/coral+cbor" according to the procedures of BCP 13 [RFC6838].¶
Deprecated alias names for this type: N/A¶
Magic number(s): N/A¶
File extension(s): .coral.cbor¶
Macintosh file type code(s): N/A¶
This document registers the media type "text/coral" according to the procedures of BCP 13 [RFC6838] and guidelines of RFC 6657 [RFC6657].¶
Deprecated alias names for this type: N/A¶
Magic number(s): N/A¶
File extension(s): .coral¶
Macintosh file type code(s): N/A¶
This document registers CoAP content formats for the content types "application/coral+cbor" and "text/coral" according to the procedures of RFC 7252 [RFC7252].¶
[[NOTE TO RFC EDITOR: Please replace all occurrences of "TBD3" and "TBD4" in this document with the code points assigned by IANA.]]¶
[[NOTE TO IMPLEMENTERS: Experimental implementations can use content format ID 65087 for "application/coral+cbor" and content format ID 65343 for "text/coral" until IANA has assigned code points.]]¶
This document registers a CBOR tag for dictionary references according to the procedures of RFC XXXX [I-D.ietf-cbor-7049bis].¶
[[NOTE TO RFC EDITOR: Please replace all occurrences of "TBD6" in this document with the code point assigned by IANA.]]¶
This section defines the core vocabulary for CoRAL: a set of link relation types, operation types, and form field types.¶
Link Relation Types:¶
Indicates that the link's context is an instance of the class specified as the link's target, as defined by RDF Schema [W3C.REC-rdf-schema-20140225].¶
Indicates that the link target is a human-readable label (e.g., a menu entry).¶
The link target MUST be a text string. The text string SHOULD be annotated with a language and text direction using nested links of type <http://coreapps.org/base#language> and <http://coreapps.org/base#direction>, respectively.¶
Indicates that the link target is a language tag [RFC5646] that specifies the language of the link context.¶
The link target MUST be a text string in the format specified in Section 2.1 of RFC 5646 [RFC5646].¶
Indicates that the link target is a base text direction (right-to-left or left-to-right) that specifies the text directionality of the link context.¶
The link target MUST be either the text string "rtl" or the text string "ltr".¶
Indicates that the link target is a representation of the link context.¶
The link target MUST be a byte string.¶
The representation may be a full, partial, or inconsistent version of the representation served from the URI of the resource.¶
A link with link relation type can occur as a top-level element in a document or as a nested element within a link. When it occurs as a top-level element, it provides an alternate representation of the document's retrieval context. When it occurs nested within a link, it provides a representation of link target of the enclosing link.¶
Operation Types:¶
Indicates that the state of the form's context can be replaced with the state described by a representation submitted to the server.¶
This operation type defaults to the PUT method [RFC7231] [RFC7252] for both HTTP and CoAP. Typical overrides by a form field include the PATCH method [RFC5789] [RFC8132] for HTTP and CoAP and the iPATCH method [RFC8132] for CoAP.¶
Indicates that the form's context can be searched by submitting a search query.¶
This operation type defaults to the POST method [RFC7231] for HTTP and the FETCH method [RFC8132] for CoAP. Typical overrides by a form field include the POST method [RFC7252] for CoAP.¶
Link Relation Types:¶
Indicates that the link's context is a collection and that the link's target is a member of that collection, as defined in Section 2.1 of RFC 6573 [RFC6573].¶
Indicates that the link's target is a collection and that the link's context is a member of that collection, as defined in Section 2.2 of RFC 6573 [RFC6573].¶
Operation Types:¶
Indicates that the form's context is a collection and that a new item can be created in that collection with the state defined by a representation submitted to the server.¶
This operation type defaults to the POST method [RFC7231] [RFC7252] for both HTTP and CoAP.¶
Indicates that the form's context is a member of a collection and that the form's context can be removed from that collection.¶
This operation type defaults to the DELETE method [RFC7231] [RFC7252] for both HTTP and CoAP.¶
Form Field Types:¶
Specifies the HTTP method for the request.¶
The form field value MUST be a text string in the format defined in Section 4.1 of RFC 7231 [RFC7231]. The set of possible values is maintained in the IANA HTTP Method Registry.¶
A form field of this type MUST NOT occur more than once in a form. If absent, it defaults to the request method implied by the form's operation type.¶
Specifies an acceptable HTTP content type for the request payload. There may be multiple form fields of this type. If a form does not include a form field of this type, the server accepts any or no request payload, depending on the operation type.¶
The form field value MUST be a text string in the format defined in Section 3.1.1.1 of RFC 7231 [RFC7231]. The possible set of media types and their parameters are maintained in the IANA Media Types Registry.¶
Link Relation Types:¶
Specifies the HTTP content type of the link context.¶
The link target MUST be a text string in the format defined in Section 3.1.1.1 of RFC 7231 [RFC7231]. The possible set of media types and their parameters are maintained in the IANA Media Types Registry.¶
A link of this type MUST NOT occur more than once for the link context. If absent, its value defaults to the content type "application/octet-stream".¶
Form Field Types:¶
Specifies the CoAP method for the request.¶
The form field value MUST be an integer identifying one of the CoAP request methods maintained in the IANA CoAP Method Codes Registry (e.g., the integer 2 for the POST method).¶
A form field of this type MUST NOT occur more than once in a form. If absent, it defaults to the request method implied by the form's operation type.¶
Specifies an acceptable CoAP content format for the request payload. There may be multiple form fields of this type. If a form does not include a form field of this type, the server accepts any or no request payload, depending on the operation type.¶
The form field value MUST be an integer identifying one of the content formats maintained in the IANA CoAP Content-Formats Registry.¶
Link Relation Types:¶
Specifies the CoAP content format of the link context.¶
The link target MUST be an integer identifying one of the content formats maintained in the IANA CoAP Content-Formats Registry.¶
A link of this type MUST NOT occur more than once for the link context. If absent, it defaults to content format 42 (i.e., the content type "application/octet-stream" without a content coding).¶
This section defines a default dictionary that is assumed when the "application/coral+cbor" media type is used without a "dictionary" parameter.¶
| Key | Value |
|---|---|
| 0 | <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> |
| 1 | <http://www.iana.org/assignments/relation/item> |
| 2 | <http://www.iana.org/assignments/relation/collection> |
| 3 | <http://coreapps.org/collections#create> |
| 4 | <http://coreapps.org/base#update> |
| 5 | <http://coreapps.org/collections#delete> |
| 6 | <http://coreapps.org/base#search> |
| 7 | <http://coreapps.org/coap#accept> |
| 8 | <http://coreapps.org/coap#type> |
| 9 | <http://coreapps.org/base#language> |
| 10 | <http://coreapps.org/coap#method> |
| 11 | <http://coreapps.org/base#direction> |
| 12 | "ltr" |
| 13 | "rtl" |
| 14 | <http://coreapps.org/base#representation> |
This section is to be removed before publishing as an RFC.¶
Changes from -01 to -02:¶
Changes from -00 to -01:¶
CoRAL is heavily inspired by Mike Kelly's JSON Hypertext Application Language [HAL].¶
The recommendations for minting IRIs have been adopted from RDF 1.1 Concepts and Abstract Syntax [W3C.REC-rdf11-concepts-20140225] to ease the interoperability between RDF predicates and link relation types.¶
Thanks to Christian Amsüss, Carsten Bormann, Thomas Fossati, Jaime Jiménez, Jim Schaad, Sebastian Käbisch, Ari Keranen, Michael Koster, Matthias Kovatsch, and Niklas Widell for helpful comments and discussions that have shaped the document.¶