URI Template(tools.ietf.org)

74 pointsby soheilprovor 13 Jahren20 comments

tools.ietf.org

http://tools.ietf.org/html/rfc6570

10 comments

buro9vor 13 Jahren

This is one of two proposed standards I've been following with some interest, the other being a sane(ish) way to implement PATCH for JSON objects:

http://tools.ietf.org/html/rfc6902

nicwolffvor 13 Jahren

There's a mature Perl module for this as well: http://search.cpan.org/~bricas/URI-Template-0.17/lib/URI/Tem...

zimbatmvor 13 Jahren

For rubyists, the addressable gem has support for it: https://github.com/sporkmonger/addressable/tree/#uri-templat...

soheilprovor 13 Jahren

And for Node.js: https://github.com/bramstein/url-template

spankaleevor 13 Jahren

It's pretty great to have a standard for templates to generate URIs, but it's equally important to parse URIs and extract the variables. It looks like this RFC skimps on that part:

In Section 1.4:

    Some URI Templates can be used in reverse for the purpose of variable
    matching: comparing the template to a fully formed URI in order to
    extract the variable parts from that URI and assign them to the named
    variables.  Variable matching only works well if the template
    expressions are delimited by the beginning or end of the URI or by
    characters that cannot be part of the expansion, such as reserved
    characters surrounding a simple string expression.  In general,
    regular expression languages are better suited for variable matching.

It'd be better IMO to have a stricter definition of parseable templates and include test cases so that implementations that try to validate templates for parseability don't disagree.

steveklabnikvor 13 Jahren

GitHub uses URI templates extensively: https://api.github.com/

specialistvor 13 Jahren

Very helpful, thanks for link. I'm working on my own web app "framework" and had rolled my own URL/URI router. This will reduce my reinvention quotient.

alanhovor 13 Jahren

Ruby Gems for this https://github.com/sporkmonger/addressable

vy8vWJlcovor 13 Jahren

Slightly related (if you hold your head at the right angle) is the (commercial) platform called NetKernel. It brings URIs into conventional programming scenarios, presenting something like a RESTful programming language. (Whatever that might mean...)

I haven't done anything with it, but I find it fascinating just to read the documentation:

http://docs.netkernel.org/book/view/book:quickstart/doc:quic...

ubersoldat2k7vor 13 Jahren

Why do RFC's have to be so ugly?

byrootvor 13 Jahren

It's just a formating issue. Nothing prevent you to reformat them: http://pretty-rfc.herokuapp.com/RFC2616

r00fusvor 13 Jahren

This is very cool. Why is the RFC source not available and what does it take to get it?

byrootvor 13 Jahren

The home page states:

  Note: most RFCs can't be prettified at the moment, as their
  source files are unavailable to the public and need to be 
  requested from rfc-editor.org. I'm working on this.

I suppose that it's just a matter of time.

treerexvor 13 Jahren

Because they originate from a time where plain text was what you had, and the format was created to be consistent and easy to format by hand. I still prefer my specifications to be in plain text so that I can view them easily and quickly.

jrochkind1vor 13 Jahren

It would be easy enough to at least format them in HTML, they use such consistent ascii 'typesetting', it could easily be automated.

I mean, they are formatted in HTML now, but as a giant block of monospace text. They could use <hx>s for the headers, let the text flow with browser resize for small screens, heck use bootstrap for good readable web typesetting, etc.

I am not sure why the IETF isn't interested in better web typesetting for em.

At one point someone had a site that did that, I think it was advertised and discussed here on HN, but I forget what it was called, and I think it faded into non-supported obscurity (since it was obviously a labor of love, unpaid).

nfozvor 13 Jahren

I still prefer my everything in plaintext....

twstedvor 13 Jahren

I really love them. This probably reveals my age...

icebrainingvor 13 Jahren

Nope, I'm <25 and I love them too.

seliopouvor 13 Jahren

I think this[1] is to blame for it.

[1]: http://tools.ietf.org/tools/rfcmarkup/

zacharyvoasevor 13 Jahren

I wonder if it'd be worth implementing this in my URLObject library (https://github.com/zacharyvoase/urlobject).