💾 Archived View for gemi.dev › gemini-mailing-list › 000486.gmi captured on 2024-03-21 at 17:43:13. Gemini links have been rewritten to link to archived content

View Raw

More Information

⬅️ Previous capture (2023-12-28)

-=-=-=-=-=-=-

Lightweight subscription to Gemini pages

1. Solderpunk (solderpunk (a) posteo.net)

Howdy Geminauts,

As per recent discussions, I have now published a "companion
specification" for lightweight subscription to Gemini pages which do not
have Atom feeds.  Please see:

gemini://gemini.circumlunar.space/docs/companion/subscription.gmi

I ended up wording some things slightly differently from my earlier
proposal.  Link items now represent feed entires if the first 10
characters of their label match the YYYY-MM-DD format, and their title
is derived from everything after the first whitespace-separated
component (with mention that implementations may clean up separating
hyphens, etc.).  This is slightly more flexible and means that all of
the following label formats will be recognised as entries:

2020-11-22: A post title
2020-11-22  A post title
2020-11-22 - A post title
2020-11-22T17:45:20: A post title
2020-11-22T17:45:20 A post title
2020-11-22T17:45:20 - A post title

Note that the timestamps in the latter three formats will not be
interpreted according to this specification, but since I have actually
seen at least one use of full-blown ISO 8601 timestamps in the wild
it seemed wise to permit them to be recognised as entries.  I don't
expect or encourage people to adopt this style as it results in ugly
pages from the perspective of human readers.

Cheers,
Solderpunk

Link to individual message.

2. colecmac (a) protonmail.com (colecmac (a) protonmail.com)

Hello Solderpunk,

Thanks a lot for posting a full spec on this. I will be supporting this format
in Amfora's upcoming subscription feature. Hopefully this will put this
discussion to rest.

> Link items now represent feed entires if the first 10
> characters of their label match the YYYY-MM-DD format, and their title
> is derived from everything after the first whitespace-separated
> component (with mention that implementations may clean up separating
> hyphens, etc.).

Thanks for this fix!


Cheers,
makeworld

Link to individual message.

3. Alex // nytpu (alex (a) nytpu.com)

Hi,

I have a minor suggestion to make when parsing the feed title from
headers.

I use ASCII art as headers for all my sites, with alt text with a
plaintext header. If you tried to use a feed reader on my gemlog
(gemini://nytpu.com/gemlog/) it would consider the title to be ?2020?
rather than the ?yet another tech gemlog? indicated by the alt text in
my header. I suggest that a reader should get the title from *either*
the first header on the page, or the first preformatted text line. For
example, with regex you could search for "^[`#]+\s*(.*)$" (with the
capture group being the title).

A minor change, but I know there's a few other pages I've seen around
that also use a header in the same way, so I'd think it'd be helpful.

nytpu

-- 
Alex // nytpu
alex at nytpu.com
GPG Key: https://www.nytpu.com/files/pubkey.asc
Key fingerprint: 43A5 890C EE85 EA1F 8C88 9492 ECCD C07B 337B 8F5B
https://useplaintext.email/
-------------- 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/20201122/2c73
1f57/attachment.sig>

Link to individual message.

4. Adnan Maolood (me (a) adnano.co)

On Sun Nov 22, 2020 at 3:54 PM EST, Alex // nytpu wrote:
> Hi,
>
> I have a minor suggestion to make when parsing the feed title from
> headers.
>
> I use ASCII art as headers for all my sites, with alt text with a
> plaintext header. If you tried to use a feed reader on my gemlog
> (gemini://nytpu.com/gemlog/) it would consider the title to be
> ?2020?
> rather than the ?yet another tech gemlog? indicated by the alt text
> in
> my header. I suggest that a reader should get the title from *either*
> the first header on the page, or the first preformatted text line. For
> example, with regex you could search for "^[`#]+\s*(.*)$" (with the
> capture group being the title).
>
> A minor change, but I know there's a few other pages I've seen around
> that also use a header in the same way, so I'd think it'd be helpful.

I would recommend putting an actual header along with the preformatted
text. That would make it clear what the intention is.

Link to individual message.

5. colecmac (a) protonmail.com (colecmac (a) protonmail.com)

> it would consider the title to be ?2020?

It would not. The spec only sets level one header lines as
the title. Level two headers can be the subtitle, but that's only if
there's already a level one header.

See
gemini://gemini.circumlunar.space/docs/companion/subscription.gmi
for more details.

I agree with Adnan that you should just provide a header line.


makeworld

Link to individual message.

---

Previous Thread: robots.txt for Gemini formalised

Next Thread: Capping CAPCOM, calling for more aggregators