On Wed, 2017-04-26 at 11:26 +0200, Julian Reschke wrote:
> On 2017-04-26 11:11, Oleg Kalnichevski wrote:
> > On Tue, 2017-04-25 at 17:02 +0200, Julian Reschke wrote:
> > > On 2017-04-25 10:37, Oleg Kalnichevski wrote:
> > > > ...
> > > > I added an enum with standard methods to HttpClient 5.0. It is
> > > > still in
> > > > my private branch on GitHub though. Please just bear with me a
> > > > little
> > > > longer. I'll commit that code to SVN as soon as HttpCore 5.0a3
> > > > goes
> > > > out.
> > > > ...
> > > 
> > > FWIW, why have an enum for something that is extensible?
> > 
> > Enums are extensible, are they not? Just not at runtime.
> 
> Not for the *users* of the project, right?
> 

Right. Why should it even be? Those are methods we, as a project,
provide as standard (common, most widely applicable, whatever, feel
free to propose a better term).


> > > Or, phrased
> > > differently, what is your definition of "standard methods"?
> > > 
> > 
> > Standard might well be not the best term. Please do feel free to
> > suggest a better one. By 'standard' methods I meant those specified
> > in
> > RFC 7231, section 4.3 + PATCH.
> 
> I fail to see how these are special. 

They are not.

Oleg


> The HTTP spec is pretty clear that 
> methods are an extension point, and that the source of truth is the
> IANA 
> registry, not a specific RFC. If you want to draw a distinction, you 
> could check the standards level of the RFC defining the method (but 
> it'll turn out that pretty all of them currently are at "proposed"
> state 
> due to the weird way the IETF defines it's standards).
> 
> Best regards, Julian
> 
> 
> 
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [email protected]
> For additional commands, e-mail: [email protected]
> 

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to