On 14 May 2017, at 6:03 pm, Julian Reschke <julian.reschke@xxxxxx> wrote: >>> Actually, one popular browser also looks for the stylesheet link relation in the "Link" header field... >> >> True, but that's not specified behaviour, AIUI. > > <https://www.w3.org/TR/html5/document-metadata.html#the-link-element>: > > "HTTP Link: headers, if supported, must be assumed to come before any links in the document, in the order that they were given in the HTTP message. These headers are to be processed according to the rules given in the relevant specifications. [HTTP] [WEBLINK]" > > So this refers to RFC 5988, which delegates to the IANA registry, which in turn points back to the HTML spec for the definition of the "stylesheet" link relation. > > To me this sounds like it is well-defined, just not required ("if supported"). https://github.com/mnot/I-D/commit/8e3d4c5167e > Just added to the issue: > > - "_" is not legal in a production name > - use "/" instead of "|" Now in the repo, thanks. >>> 4.2. Link Relation Type Registry >>> >>> Each published document will be at a URL mutually agreed to by IANA >>> and the Expert(s), and IANA will set HTTP response headers on them as >>> (reasonably) requested by the Expert(s). >>> >>> Wow. Does IANA know that we want to configure their web server? >> >> We're talking about it. Data, metadata -- what's the difference? > > Operationally? We'll see. >>> 1. Let "links" be an empty list. >>> >>> 2. Create "link_strings" by splitting "field_value" on "," >>> characters, excepting "," characters within quoted strings as per >>> >>> I have my doubts that people will be able to translate it into actually correct code. To detect whether something is inside a quoted-string essentially requires running a parser; the prose here suggests that this is not necessary. >> >> I reluctantly agree. I'll need to rework the algorithm to address this; please stand by. > > The only way to properly split list supporting field values is to parse them according to the grammar for the list items; I don't believe in the existence of a shortcut that actually works correctly. Suggesting otherwise IMHO will lead to broken implementations for certain edge cases. +1 >>> It also should mention that there might be multiple instances of field_value. >> >> The algorithm is for parsing a single field-value... > > Yes, but should we tell the audience that they need to parse *all* field values, and combine the result? OK. -- Mark Nottingham https://www.mnot.net/