]> xyzzy

Hamburg Germany hmdmhdfmhdjmzdtjmzdtzktdkztdjz@gmail.com http://purl.net/xyzzy/

This memo specifies the 'news' and 'nntp' Uniform Resource Identifier (URI) schemes that were originally defined in RFC 1738. The purpose of this document is to allow RFC 1738 to be made obsolete while keeping the information about these schemes on standards track. In the collected ABNF the NEWS in RFC NEWS should be replaced by the RFC number for , and the ABNF in RFC ABNF should be replaced by the RFC number for . In RFC&rfc.number; is a placeholder for this memo. This editorial note and the document history should be removed before publication.

The first definition for many URI schemes appeared in . This memo extracts the 'news' and 'nntp' URI schemes from it to allow that material to remain on standards track if is moved to "historic" status. It belongs to a series of similar documents like , , , and discussed on the list. The definitions for the 'news' and 'nntp' URI schemes given here are updates from based on modern usage of these schemes. specifies how to define schemes for URIs, it also explains the term "Uniform Resource Locator" (URL). The Network News Transfer Protocol (NNTP) is specified in . The Netnews Article Format is defined in . The key word "MUST" in this memo is to be interpreted as described in . UTF-8 is specified in . The syntax uses the ABNF defined in .

The 'news' and 'nntp' URI schemes identify resources on an NNTP server, individual articles, individual newsgroups, or sets of newsgroups. User agents like Web browsers supporting these schemes use the NNTP protocol to access the corresponding resources. The details how they do this, e.g. employing a separate or integrated newsreader, depend on the implementation. The default <port> associated with NNTP in is 119.

The 'nntp' URI scheme identifies articles in a newsgroup on a specific NNTP server. In terminology this means that 'nntp' URIs have a non-empty <authority> component, there is no default <host> as for the 'file' or 'news' URI schemes. Netnews is typically distributed among several news servers, using the same newsgroup names, but local article numbers. An article available as number 10 in group example on server news.example.com has most likely a different number on any other server where the same article is still available. Users allowed to read and post articles on "their" server are generally not allowed to access articles on an "arbitrary" server specified in an 'nntp' URI. For these reasons the use of the 'nntp' URI scheme is limited, and it's less widely supported by user agents than the similar 'news' URI scheme.

The 'news' URI scheme identifies articles by their worldwide unique Message-ID, independent of the server and the newsgroup. Newsreaders support access on articles by their Message-ID, without the overhead for an URI scheme. In simple cases they do this directly as NNTP client of a default or currently used server as configured by the user. More general user agents use the 'news' URI scheme to distinguish Message-IDs from similar constructs like other URI schemes in contexts like a plain text message body. The 'news' URI scheme also allows to identify newsgroups or sets of newsgroups independent of a specific server. For Netnews a group example has the same name on any server carrying this group, exotic cases involving gateways not withstanding. To distinguish Message-IDs and newsgroup names the 'news' URI scheme uses the @ between local part (left hand side) and domain part (right hand side) of Message-IDs. offered only one wildcard for sets of newsgropus in 'news' URIs, a * used to refer to "all available newsgroups". In common practice this was extended to varying degrees by some user agents, an NNTP extension known as <wildmat> specified in and now part of the base NNTP specification allows pattern matching in the style of the find command. For the purpose of this memo this means that some additional special characters have to be allowed in 'news' URIs, some of them percent-encoded as required by the overall URI syntax. User agents and NNTP servers might not (yet) implement all parts of this new feature. Another commonly supported addition to the syntax is the optional specification of a server at the begin of 'news' URIs. This optional <authority> component follows the overall syntax preceded by a double slash // and terminated by the next slash /, question mark ?, number sign #, or the end of the URI.

This memo doesn't specify a query part introduced by a question mark ? or a fragment introduced by a number sign for the 'news' and 'nntp' URI schemes, but some implementations are known to use query parts internally to address parts of a composite media type in Multipurpose Internet Mail Extensions (MIME). There are no special . or .. path segments in 'news' and 'nntp' URLs. Please note that . and .. are no valid <newsgroup-name>s. URI producers have to percent-encode some characters as specified below, otherwise they MUST treat a Message-ID without angle brackets for 'news' URLs as is, i.e. case-sensitive, preserving quoted pairs and quoted strings.

An 'nntp' URI identifies an article by its number in a given newsgroup of a specified server, or it identifies the newsgroup without article number.

nntpURL = "nntp:" server "/" group [ "/" article-number ] server = "//" authority ; see RFC 3986 group = 1*( group-char / pct-encoded ) article-number = 1*16DIGIT ; see RFC 3977 group-char = ALPHA / DIGIT / "-" / "+" / "_" / "."

In the form with an <article-number> the URL corresponds roughly to the content of an <xref> header field as specified in , replacing its more general <article&nbhy;locator> by the <article&nbhy;number> used with NNTP. A <newsgroup-name> as specified in consists of dot-separated components. Each component contains one or more letters, digits, - (hyphen-minus), +, or _ (underscore). These characters can be directly used in a segment of a path in a URI, no percent-encoding is necessary. Example:

nntp://news.server.example/example.group.this/12345

A <wildmat-exact> newsgroup name as specified in allows (in theory) any <UTF8-non-ascii> and most printable US&nbhy;ASCII characters excluding !, *, ,, ?, [, \, and ]. To keep the syntax here simple all additional characters in <wildmat-exact> not (yet) allowed in are covered by <pct-encoded> as defined in , although percent-encoding is not strictly necessary for some of these additional characters like :, ;, and ~. Most of the additional characters have to be percent-encoded, example:

nntp://wild.server.example/example.group.n%2Fa/12345

In the form without <article-number> the URL identifies a single group on the specified server. This is also possible with an equivalent 'news' URL, and the latter is better supported by user agents, example:

nntp://news.server.example/example.group.this news://news.server.example/example.group.this

A 'news' URI identifies an article by its unique Message-ID, or it identifies a set of newsgroups. Additionally it can specify a server, without it a configured default server for Netnews access is used.

newsURL = "news:" [ server "/" ] ( article / newsgroups ) article = mid-left "@" mid-right newsgroups = *( group-char / pct-encoded / "*" ) mid-left = 1*( mid-atext / "." ) / ; <dot-atom-text> ( "%22" mid-quote "%22" ) ; <no-fold-quote> mid-quote = 1*( mid-atext / "." / ; <mqtext> incl. mid-special / ; '\"' / "[" / "]" "%5C%22" / "%5B" / "%5D" ) mid-right = 1*( mid-atext / "." ) / ; <dot-atom-text> ( "%5B" mid-literal "%5D" ) ; <no-fold-literal> mid-literal = 1*( mid-atext / "." / ; <mdtext> incl. mid-special / ; '"' / "\[" / "\]" "%22" / "%5C%5B" / "%5C%5D" ) mid-special = "(" / ")" / "," / ":" / ";" / "%3C" / "%40" / "%5C%5C" ; "<" / "@" / "\\" mid-atext = ALPHA / DIGIT / ; RFC 2822 <atext> "!" / "$" / "&" / "'" / ; allowed sub-delims "*" / "+" / "=" / ; allowed sub-delims "-" / "_" / "~" / ; allowed unreserved "%23" / "%25" / "%2F" / ; "#" / "%" / "/" "%3F" / "%5E" / "%60" / ; "?" / "^" / "`" "%7B" / "%7C" / "%7D" ; "{" / "|" / "}"

The form identifying an <article> corresponds to the <msg-id-core> construct in , it's a Message-ID without angle brackets. Characters not directly allowed in this part of an URI have to be percent-encoded, minimally anything that is not <unreserved>, no : (colon), and doesn't belong to the <sub&nbhy;delims>. Several details of a canonical <msg-id-core> are omitted here, e.g. leading, adjacent, or trailing dots are not allowed in <dot&nbhy;atom&nbhy;text>. The syntax mainly shows which characters MUST be percent-encoded in a <mid-left> (local part) or <mid-right> (domain part). Please note that %20 (space) and %3E (>) are not allowed. A %5C (backslash \) can only occur in four combinations as shown above. Examples:

news://server.example/ab.cd@example.com news:%22do..ts%22@example.com news:ab.cd@%5B2001:DB8::CD30%5D

The form identifying <newsgroups> corresponds to the <wildmat&nbhy;pattern>, a newsgroup name with wildcards * and ?. Any ? has to be be percent-encoded as %3F in this part of an URI. Examples, the first two are equivalent:

news://news.server.example/* news://news.server.example/ news://wild.server.example/example.group.th%3Fse news:example.group.* news:example.group.this

Without wildcards this form of the URL identifies a single group if it's not empty, and user agents would typically try to present an overview of the articles available in this group, probably somehow limiting this overview to the newest unread articles up to a configured maximum. With wildcards user agents could try to list matching group names on the specified or default server. Some user agents support only a specific <group> without wildcards, or an optional single *.

The Internet Drafts by Alfred S. Gilman published 1998 introduced additions to the original 'news' URI scheme. Some of these ideas are now widely supported and reflected by the revised 'news' URI scheme specified here. Paul Hoffman started the series of Internet Drafts allowing to replace . Charles Lindsey and Russ Allbery contributed all missing details about the 'nntp' URI scheme. Roy Fielding helped to figure out what is and what is not a <pchar>. Nicolas Krebs proposed to stick to the revised syntax supporting 'news' URLs without <newgroups>, as in many other URI schemes allowing empty paths. Ken Murchison and Chris Newman helped to improve the 'snews' registration template. Bill Fenner's xml2rfc validator and ABNF checker were a great help in the creation of (not only) this memo. Henry Spencer was the driving force to adopt MIME also in NetNews, he registered the MIME 'message/external-body' access type 'news-message-ID' discussed below 1993 for .

The URI schemes were updated to support percent-encoded UTF-8 characters in NNTP newsgroup names as specified in and . The Netnews Article Format in does not yet allow UTF-8 in <newsgroup-name>s, therefore well-known Unicode and UTF-8 security considerations are not listed below. For an overview see and . The work on E-mail Address Internationalization (EAI) started in most likely won't change the syntax of a Message-ID. The work on a successor of might end up with a significantly simplified syntax for at least the local part (right hand side) of a Message-ID.

There are many security considerations for URI schemes discussed in . The NNTP protocol may use passwords in the clear for authentication, or offer no privacy, both of which are considered extremely unsafe in current practice. Alternatives and further security considerations with respect to NNTP are discussed in and . The syntax for the 'news' and 'nntp' URI schemes contains the general <authority> construct with an optional <userinfo> defined in . As noted in the user:password form of a <userinfo> is deprecated. Articles on NNTP servers typically expire after some time. After that time corresponding 'news' and 'nntp' URLs won't work anymore depending on the server. While a Message-ID is supposed to be worldwide unique forever the NNTP protocol does not guarantee this. Under various conditions depending on the servers the same Message-ID could be used for different articles, and attackers could try to distribute malicious content for known 'news' or 'nntp' URLs. Compare for similar security considerations.

The IANA registry of URI schemes could be updated to point to this memo instead of for the 'news' and 'nntp' URI schemes.

This section contains the template for the registration of the historical 'snews' scheme specified in . snews historical Same as for 'news' Syntactically equivalent to 'news', but using NNTP over SSL/TLS (SSL/TLS with security layer is negotiated immediately after establishing the TCP connection) with a default port of 563, registered as nntps Same as for 'news' For some user agents 'snews' URLs trigger the use of nntps instead of NNTP for their access on Netnews This URI scheme was not widely deployed, its further use is deprecated in favour of ordinary 'news' URLs in conjunction with NNTP servers supporting See , the use of a dedicated port for secure variants of a protocol was discouraged in (URI mailing list) IETF RFC&rfc.number;, ,

This section contains the template for the registration of the 'nntp' URI scheme with the Dynamic Delegation Discovery System.

Key: nntp Authority: RFC&rfc.number; Record: nntp IN NAPTR 0 0 "" "" "!^nntp://([^/?#]*@)?([^:/?#]*).*$!\\2!i" .

The MIME 'news-message-ID' access type was erroneously listed as subtype. IANA should remove 'news-message-ID' from the application subtype registry, and add it to the access type registry defined in : . requires an RFC for the access types registry. Because was never published as RFC the following paragraph quotes the relevant definition: NOTE: In the specific case where it is desired to essentially make another article PART of the current one, e.g. for annotation of the other article, MIME's "message/external-body" convention can be used to do so without actual inclusion. "news-message-ID" was registered as a standard external-body access method, with a mandatory NAME parameter giving the message ID and an optional SITE parameter suggesting an NNTP site that might have the article available (if it is not available locally), by IANA 22 June 1993. Please note that 'news' URLs offer a very similar and (today) more common way to access articles by their Message-ID, compare .

&rfc2119; &rfc3977; &rfc3986; &rfcABNF; &rfcNEWS; &rfc1738; &rfc2017; &rfc2046; &rfc2595; &rfc2822; &rfc2980; &rfc3405; &rfc3629; &rfc3987; &rfc4156; &rfc4157; &rfc4248; &rfc4266; &rfc4289; &rfc4395; &rfc4642; &rfc4643; &rfc4952; &rfcARCH;

In addition to the syntax given above this appendix also lists the sources of terms used in comments and the prose:

nntpURL = "nntp:" server "/" group [ "/" article-number ] server = "//" authority ; see RFC 3986 group = 1*( group-char / pct-encoded ) article-number = 1*16DIGIT ; see RFC 3977 group-char = ALPHA / DIGIT / "-" / "+" / "_" / "." newsURL = "news:" [ server "/" ] ( article / newsgroups ) article = mid-left "@" mid-right newsgroups = *( group-char / pct-encoded / "*" ) mid-left = 1*( mid-atext / "." ) / ; <dot-atom-text> ( "%22" mid-quote "%22" ) ; <no-fold-quote> mid-quote = 1*( mid-atext / "." / ; <mqtext> incl. mid-special / ; '\"' / "[" / "]" "%5C%22" / "%5B" / "%5D" ) mid-right = 1*( mid-atext / "." ) / ; <dot-atom-text> ( "%5B" mid-literal "%5D" ) ; <no-fold-literal> mid-literal = 1*( mid-atext / "." / ; <mdtext> incl. mid-special / ; '"' / "\[" / "\]" "%22" / "%5C%5B" / "%5C%5D" ) mid-special = "(" / ")" / "," / ":" / ";" / "%3C" / "%40" / "%5C%5C" ; "<" / "@" / "\\" mid-atext = ALPHA / DIGIT / ; RFC 2822 <atext> "!" / "$" / "&" / "'" / ; allowed sub-delims "*" / "+" / "=" / ; allowed sub-delims "-" / "_" / "~" / ; allowed unreserved "%23" / "%25" / "%2F" / ; "#" / "%" / "/" "%3F" / "%5E" / "%60" / ; "?" / "^" / "`" "%7B" / "%7C" / "%7D" ; "{" / "|" / "}" authority = <see RFC 3986 Section 3.2> host = <see RFC 3986 Section 3.2.2> pct-encoded = <see RFC 3986 Section 2.1> port = <see RFC 3986 Section 3.2.3> sub-delims = <see RFC 3986 Section 2.2> unreserved = <see RFC 3986 Section 2.3> userinfo = <see RFC 3986 Section 3.2.1> UTF8-non-ascii = <see RFC 3977 Section 9.8> wildmat = <see RFC 3977 Section 4.1> wildmat-exact = <see RFC 3977 Section 4.1> wildmat-pattern = <see RFC 3977 Section 4.1> ALPHA = <see RFC ABNF Appendix B.1> DIGIT = <see RFC ABNF Appendix B.1> atext = <see RFC 2822 Section 3.2.4> dot-atom-text = <see RFC 2822 Section 3.2.4> article-locator = <see RFC NEWS Section 3.2.14> mdtext = <see RFC NEWS Section 3.1.3> mqtext = <see RFC NEWS Section 3.1.3> msg-id-core = <see RFC NEWS Section 3.1.3> newsgroup-name = <see RFC NEWS Section 3.1.4> no-fold-literal = <see RFC NEWS Section 3.1.3> no-fold-quote = <see RFC NEWS Section 3.1.3> xref = <see RFC NEWS Section 3.2.14>

Here's an example of a mail to the list with Message-ID p0624081dc30b8699bf9b@[10.20.30.108]. is one of the various list archives, it converts mails into Netnews articles. The header of this article contains the following fields (among others):

Message-ID: p0624081dc30b8699bf9b@[10.20.30.108] Xref: news.gmane.org gmane.ietf.tools:742 Archived-At: <http://permalink.gmane.org/gmane.ietf.tools/742>

The Xref roughly indicates the 742nd article in newsgroup on this server. An 'nntp' URL might be . For details about the Archived-At URL see . The list software and list subscribers reading the list elsewhere can't predict a server specific article number 742 in this archive. If they know this server they can however guess the corresponding URL. In theory the list software could use the guessed 'news' URL in an Archived-At header field, but if a list tries this it would likely use . Using domain literals in a Message-ID could cause collisions. A collision might force the mail2news gateway in this example to invent a new Message-ID, and an attempt to guess the future URL on this server would then fail.

Changes in version 06: References to RFC 4234 replaced by 4234bis (waiting for its RFC and STD numbers). Reference to added. Added a security consideration proposed by Chris Newman for the Archived-At header field also here. Added an appendix with a detailed 'news', 'nntp', Xref, and Archived-At example. Use a more reliable e-mail address (and thanks for the feedback from somebody who figured the new address out following obscure contact or rev=made" links ;-) The USEFOR WG did not adopt this draft as work item, they are busy to complete a document blocking the publication of . IANA revived the list mentioned in . Added a note about EAI and its most likely unmodified Message-ID concept. Changes in version 05: Added an attempt to cleanup the erroneous MIME application subtype 'news-message-ID' registration. This was meant to be a MIME 'message/external-body' access type as published in . The 'news-message-ID' review request was posted 2007-02-19. Changes in version 04: Minor editorial fixes. Just in case waiting for the IESG approval of . The 'snews' URI review request was posted 2006-11-10. Two reviewers of the 'snews' registration template are now apparently satisfied with the 'snews' URI scheme semantics. Changes in version 03: The 'snews' semantics was improved after discussions with Chris Newman and Ken Murchinson. Various editorial fixes proposed by Alfred Hönes. Changes in version 02: The referenced NNTP specifications got their RFC numbers, NNTP TLS added for info to the security considerations. The ABNF for an <article> was further simplified by extracting the <mid-special> characters used on both sides of the @, i.e. within a quoted string <mid-quote> for the unique part (left hand side) or within a literal in square brackets for the domain part (right hand side). Now it's obvious that the differences between both sides are limited to '"', [, and ] as expected. Removed the dubious 1 at the begin of the <newsgroups> rule based on an observation by Nicolas Krebs. Created a proper informative reference for the historical . The IETF archive offers only -01, a copy of -02 covering 'snews' is now available below . Other minor changes include the addition of a reference to , and the collected ABNF. The IANA registration template for the historical 'snews' URI scheme was added. The IANA registration template for an nntp.uri.arpa NAPTR record was added. If that record is correct the existing ftp.uri.arpa and http.uri.arpa records could be updated, apparently they don't remove the optional <userinfo> at the moment. Changes in version 01: References of RFC 977 and RFC 2980 replaced by the now approved NNTP base document . Security considerations updated with a reference to the now approved NNTP Auth document . References of RFC 1036 and replaced by the last called . References of RFC 2396 removed, the jumps from to and from RFC 1036 to are interesting enough without talking about intermediate steps. has no <range> for the 'nntp' URI scheme, and this memo isn't the place to invent new tricks for a rarely used scheme. Changes in version 00: Derived from [I-D.hoffman-news-nntp-uri-04] after discussions on the URI list. At this time what's now known as the Netnews Article Format was still far from ready, and RFC977bis not yet finished.