[spec] nitpicking

Omar Polo <op (a) omarpolo.com>

Hello,

I was re-reading the spec to clarify I doubt I had about relative vs
absolute URLs and where they're allowed, while I stumbled on this part:

> 2 Gemini Request
> [...]
> <URL> is a UTF-8 encoded absolute URL, including a scheme, of maximum length
> 1024 bytes.
>
> Sending an absolute URL instead of only a path or selector is effectively
> equivalent to building in a HTTP "Host" header. [...]

If I'm understanding correctly, the "instead of only a path or selector"
is redundant, since relative URLs aren't allowed.  This is probably a
leftover from when the spec allowed both relative and absolute URLs in
the request, or I'm missing something?

Thanks,

Omar Polo

Link to individual message.

John Cowan <cowan (a) ccil.org>

On Wed, Jan 20, 2021 at 11:36 AM Omar Polo <op at omarpolo.com> wrote:


> If I'm understanding correctly, the "instead of only a path or selector"
> is redundant, since relative URLs aren't allowed.


I think it is meant to emphasize that this is how Gemini does virtual
hosting (and also proxying), rather than a combination of a relative
reference in the GET line and a Host: header, as HTTP/1.1 does.  So this
sentence is informative rather than normative.



John Cowan          http://vrici.lojban.org/~cowan        cowan at ccil.org
                Charles li reis, nostre emperesdre magnes,
                Set anz totz pleinz ad ested in Espagnes.

Link to individual message.

Omar Polo <op (a) omarpolo.com>


John Cowan <cowan at ccil.org> writes:

> On Wed, Jan 20, 2021 at 11:36 AM Omar Polo <op at omarpolo.com> wrote:
>
>
>> If I'm understanding correctly, the "instead of only a path or selector"
>> is redundant, since relative URLs aren't allowed.
>
>
> I think it is meant to emphasize that this is how Gemini does virtual
> hosting (and also proxying), rather than a combination of a relative
> reference in the GET line and a Host: header, as HTTP/1.1 does.  So this
> sentence is informative rather than normative.

Your interpretation seems right.  I initially read it from the
prospective of someone who knew relative URLs were allowed, not as
someone approaching a new protocol.

Thanks,

Omar Polo

>
>
> John Cowan          http://vrici.lojban.org/~cowan        cowan at ccil.org
>                 Charles li reis, nostre emperesdre magnes,
>                 Set anz totz pleinz ad ested in Espagnes.

Link to individual message.

---

Previous Thread: [users] Simple Gemini server for testing in the current directory?

Next Thread: Gemini works great for mobile content