Web Bundles

1.1. Terminology and Conventions

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.¶

This specification uses the conventions and terminology defined in the Infra Standard ([INFRA]).¶

2.1. Operations

Bundle parsers support two primary operations:¶

1. They can load the bundle's metadata given a prefix of the bundle.¶
2. They can find a representation within the bundle given that representation's URL (Section 2.2).¶

2.2. Naming a representation

Representations within a bundle are named by their Content-Location (Section 8.7 of [I-D.ietf-httpbis-semantics]). This is also known as the representation's URL. It is either an absolute-URI or a partial-URI (Section 2.7 of [RFC7230]). In the latter case, the referenced URL is relative to the bundle's URL unless an additional section in the bundle specifies a different base URL.¶

This identifying information for each representation is stored in an index (Section 4.2.1) rather than in that representation's HTTP response message.¶

3.1. Random access

To load a resource when seeing a bundle for the first time, the implementation reads O(size of the metadata and resource index) before starting to return bytes of the resource.¶

TODO: Is big-O notation the right way to express expectations here?¶

3.2. Streaming

When sending a bundle over a stream, the implementation will need to wait until it has the sizes of all contained resources before starting to send the resource index.¶

When reading a bundle from a stream, the implementation starts returning bytes of a resource after receiving O(1) bytes of that resource, which comes after the O(# of resources) bytes of the index.¶

4.1. Top-level structure

A bundle is a CBOR array ([CBORbis]) with the following CDDL ([CDDL]) schema:¶

webbundle = [
  magic: h'F0 9F 8C 90 F0 9F 93 A6',
  version: bytes .size 4,
  section-lengths: bytes .cbor section-lengths,
  sections: [* any ],
  length: bytes .size 8,  ; Big-endian number of bytes in the bundle.
]

whatwg-url = tstr

When serialized, the bundle MUST satisfy the core deterministic encoding requirements from Section 4.2.1 of [CBORbis]. This format does not use floating point values or tags, so this specification does not add any deterministic encoding rules for them. If an item doesn't follow these requirements, or a byte-sequence being decoded as a CBOR item contains extra bytes, the parser MUST signal an error instead of any data it can extract from that item.¶

High-level fields in the bundle format are designed to provide their length-in-bytes before the field starts so that a recipient trying to stream a bundle from the network can always wait for a known number of bytes instead of needing to implement a streaming CBOR parser.¶

The magic number is "🌐📦" (U+1F310 U+1F4E6) encoded in UTF-8. With the CBOR initial bytes for the array and bytestring, this makes the format identifiable by looking for 8? 48 (in base 16) followed by that UTF-8 encoding. Parsers MUST only check the initial nibble of the initial 8? byte in order to accommodate any future version's change in the number of array elements (up to 15).¶

The version bytestring MUST be 31 00 00 00 in base 16 (an ASCII "1" followed by 3 0s) for this version of bundles. If the recipient doesn't support the version in this field, it MUST ignore the bundle.¶

The section-lengths and sections arrays contain the actual content of the bundle and are defined in Section 4.2. The section-lengths array is embedded in a byte string to facilitate reading it from a network. This byte string MUST be less than 8192 (8*1024) bytes long, and parsers MUST NOT load any data from a section-lengths item longer than this.¶

The bundle ends with an 8-byte integer holding the length of the whole bundle.¶

 +------------+-----+----+----+----+----+----+----+----+----+----+
 | first byte | ... | 48 | 00 | 00 | 00 | 00 | 00 | BC | 61 | 4E |
 +------------+-----+----+----+----+----+----+----+----+----+----+
             /       \
       0xBC614E-10=12345668 omitted bytes

4.2. Bundle sections

A bundle's content is in a series of sections, which can be accessed randomly using the information in the section-lengths CBOR item:¶

section-lengths = [* (section-name: tstr, length: uint) ],

This field lists the named sections in the bundle in the order they appear, with each section name followed by the length in bytes of the corresponding CBOR item in the sections array. This allows a random-access parser (Section 3) to jump directly to the section it needs. This specification defines the following sections:¶

• "index" (Section 4.2.1)¶
• "critical" (Section 4.2.2)¶
• "responses" (Section 4.3)¶

Future specifications can register new section names as described in Section 6.2, in order to extend the format without incrementing its version number.¶

The "responses" section MUST appear after the other three sections defined here, and parsers MUST NOT load any data if that is not the case.¶

The sections array contains the sections' content. The length of this array MUST be exactly half the length of the section-lengths array, and parsers MUST NOT load any data if that is not the case.¶

The bundle MUST contain the "index" and "responses" sections. All other sections are optional.¶

index = {* whatwg-url => [ location-in-responses ] }
location-in-responses = (offset: uint, length: uint)

critical = [*tstr]

4.3. Responses

responses = [*response]
response = [headers: bstr .cbor headers, payload: bstr]
headers = {* bstr => bstr}

The "responses" section holds the HTTP responses that represent the HTTP representations in the bundle. It consists of a CBOR array of responses, each of which is pointed to by one or more entries in the "index" section (Section 4.2.1).¶

The length of the headers byte string in a response MUST be less than 524288 (512*1024) bytes, and recipients MUST fail to load a response with longer headers.¶

When receiving a bundle in a stream, the recipient MAY process the headers before the payload has been received and MAY start processing the beginning of the payload before the end of the payload has been received.¶

The keys of the headers map MUST consist of lowercase ASCII as described in Section 8.1.2 of [RFC7540]. Response pseudo-headers (Section 8.1.2.4 of [RFC7540] are included in this headers map.¶

Each response's headers MUST include a :status pseudo-header with exactly 3 ASCII decimal digits and MUST NOT include any other pseudo-headers.¶

If a response's payload is not empty, its headers MUST include a Content-Type header (Section 8.3 of [I-D.ietf-httpbis-semantics]). The client MUST interpret the following payload as this specified media type instead of trying to sniff a media type from the bytes of the payload, for example by appending an artificial X-Content-Type-Options: nosniff header field ([FETCH]) to downstream protocols.¶

4.4. Serving constraints

When served over HTTP, a response containing an application/webbundle payload MUST include at least the following response header fields, to reduce content sniffing vulnerabilities (Section 5.2):¶

• Content-Type: application/webbundle¶
• X-Content-Type-Options: nosniff¶

5.1. Version skew

Bundles currently have no mechanism for ensuring that any signed exchanges they contain constitute a consistent version of those resources. Even if a website never has a security vulnerability when resources are fetched at a single time, an attacker might be able to combine a set of resources pulled from different versions of the website to build a vulnerable site. While the vulnerable site could have occurred by chance on a client's machine due to normal HTTP caching, bundling allows an attacker to guarantee that it happens. Future work in this specification might allow a bundle to constrain its resources to come from a consistent version.¶

5.2. Content sniffing

While modern browsers tend to trust the Content-Type header sent with a resource, especially when accompanied by X-Content-Type-Options: nosniff, plugins will sometimes search for executable content buried inside a resource and execute it in the context of the origin that served the resource, leading to XSS vulnerabilities. For example, some PDF reader plugins look for %PDF anywhere in the first 1kB and execute the code that follows it.¶

The application/webbundle format defined above includes URLs and request headers early in the format, which an attacker could use to cause these plugins to sniff a bad content type.¶

To avoid vulnerabilities, in addition to the response header requirements in Section 4.4, servers are advised to only serve an application/webbundle resource from a domain if it would also be safe for that domain to serve the bundle's content directly, and to follow at least one of the following strategies:¶

1. Only serve bundles from dedicated domains that don't have access to sensitive cookies or user storage.¶
2. Generate bundles "offline", that is, in response to a trusted author submitting content or existing signatures reaching a certain age, rather than in response to untrusted-reader queries.¶

3. Do all of:¶

   1. If the bundle's contained URLs (e.g. in the index) are derived from the request for the bundle, percent-encode ([URL]) any bytes that are greater than 0x7E or are not URL code points ([URL]) in these URLs. It is particularly important to make sure no unescaped nulls (0x00) or angle brackets (0x3C and 0x3E) appear.¶
   2. Similarly, if the request headers for any contained resource are based on the headers sent while requesting the bundle, only include request header field names and values that appear in a static allowlist. Keep the set of allowed request header fields smaller than 24 elements to prevent attackers from controlling a whole CBOR length byte.¶
   3. Restrict the number of items a request can direct the server to include in a bundle to less than 12, again to prevent attackers from controlling a whole CBOR length byte.¶
   4. Do not reflect request header fields into the set of response headers.¶

If the server serves responses that are written by a potential attacker but then escaped, the application/webbundle format allows the attacker to use the length of the response to control a few bytes before the start of the response. Any existing mechanisms that prevent polyglot documents probably keep working in the face of this new attack, but we don't have a guarantee of that.¶

To encourage servers to include the X-Content-Type-Options: nosniff header field, clients SHOULD reject bundles served without it.¶

6.1. Internet Media Type Registration

IANA is requested to register the MIME media type ([IANA.media-types]) for web bundles, application/webbundle, as follows:¶

• Type name: application¶
• Subtype name: webbundle¶

• Required parameters:¶

  • v: A string denoting the version of the file format. ([RFC5234] ABNF: version = 1*(DIGIT/%x61-7A)) The version defined in this specification is 1.¶

    Note: RFC EDITOR PLEASE DELETE THIS NOTE; Implementations of drafts of this specification MUST NOT use simple integers to describe their versions, and MUST instead define implementation-specific strings to identify which draft is implemented.¶

• Optional parameters: N/A¶
• Encoding considerations: binary¶
• Security considerations: See Section 5 of this document.¶
• Interoperability considerations: N/A¶
• Published specification: This document¶
• Applications that use this media type: None yet, but it is expected that web browsers will use this format.¶
• Fragment identifier considerations: N/A¶

• Additional information:¶

  • Deprecated alias names for this type: N/A¶
  • Magic number(s): 86 48 F0 9F 8C 90 F0 9F 93 A6¶
  • File extension(s): .wbn¶
  • Macintosh file type code(s): N/A¶
• Person & email address to contact for further information: See the Author's Address section of this specification.¶
• Intended usage: COMMON¶
• Restrictions on usage: N/A¶
• Author: See the Author's Address section of this specification.¶
• Change controller: The IESG iesg@ietf.org¶
• Provisional registration? Yes.¶

6.2. Web Bundle Section Name Registry

IANA is directed to create a new registry with the following attributes:¶

Name: Web Bundle Section Names¶

Review Process: Specification Required¶

Initial Assignments:¶

Requirements on new assignments:¶

Section Names MUST be encoded in UTF-8.¶

A section's specification MAY say that, if it is present, another section is not processed.¶

A.1. Bundle format

webbundle = [
  magic: h'F0 9F 8C 90 F0 9F 93 A6',
  version: bytes .size 4,
  section-lengths: bytes .cbor section-lengths,
  sections: [* any ],
  length: bytes .size 8, ; Big-endian number of bytes in the bundle.
]

section-lengths = [* (section-name: tstr, length: uint) ]

A.2. Core sections

index = {* whatwg-url => [ location-in-responses ] }
whatwg-url = tstr
location-in-responses = (offset: uint, length: uint)

critical = [*tstr]

responses = [*response]
response = [headers: bstr .cbor headers, payload: bstr]
headers = {* bstr => bstr}