Not requiring schemes in Gemini requests has led to lots of confusion. I propose that Gemini requests must contain a scheme. This will make things simpler for server authors, who will not have to check for the scheme "gemini" as well as check for an empty scheme. Some servers do not properly handle this anyways. Linking to a URL without a scheme should still work. It would be up to the client to resolve the URL reference before making the request. This functionality should already be provided by whatever library the client uses to handle relative URLs.
On 2020-11-24 (Tuesday) at 21:44, Adnan Maolood <me at adnano.co> wrote: > Not requiring schemes in Gemini requests has led to lots of confusion. > I propose that Gemini requests must contain a scheme. > > This will make things simpler for server authors, who will not have to > check for the scheme "gemini" as well as check for an empty scheme. > Some servers do not properly handle this anyways. > > Linking to a URL without a scheme should still work. It would be up to > the client to resolve the URL reference before making the request. This > functionality should already be provided by whatever library the client > uses to handle relative URLs. > I very much agree with this, and there's recently been a long discussion on the very matter: gemini://gemi.dev/gemini-mailing-list/messages/003361.gmi I don't recall it being resolved, however. I think the Gemini spec should be revised to be compliant with the URL RFCs. -- ~ acdw acdw.net | breadpunk.club/~breadw
I also agree. I just finished implementing this in my gemini server and it was a pain. I had to switch from a URL parser to a URI parser.
It was thus said that the Great Ben Goldberg once stated: > I also agree. I just finished implementing this in my gemini server and > it was a pain. I had to switch from a URL parser to a URI parser. Agree. As I mentioned earlier [1], 0.2% of all requests were schemeless to my server, and only three clients accounted for that traffic. Make the Gemini request a full URL. -spc [1] gemini://gemi.dev/gemini-mailing-list/messages/003392.gmi
On Wed Nov 25, 2020 at 3:57 AM CET, Sean Conner wrote: > > Agree. As I mentioned earlier [1], 0.2% of all requests were schemeless > to my server, and only three clients accounted for that traffic. Make > the > Gemini request a full URL. I think I'm happy with this, does anybody object? Cheers, Solderpunk
> I think I'm happy with this, does anybody object? No, this is a great idea. It's also been confusing a lot of people (iirc) about what is a valid URL without a scheme (e.g. //example.com) so making it absolutely clearly is a win. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 833 bytes Desc: not available URL: <https://lists.orbitalfox.eu/archives/gemini/attachments/20201127/9765 1335/attachment.sig>
It was thus said that the Great Arav K. once stated: > > I think I'm happy with this, does anybody object? > > No, this is a great idea. It's also been confusing a lot of people > (iirc) about what is a valid URL without a scheme (e.g. //example.com) > so making it absolutely clearly is a win. I've gone ahead and removed the schemeless support from GLV-1.12556 in the hopes this pushes other people to do the same, and get Solderpunk to update the specification. -spc
---
Previous Thread: [ANN] gmitohtml, a Gemini to HTML proxy written in Go