Conversation

Notices

1. silverpill (silverpill@mitra.social)'s status on Tuesday, 09-Apr-2024 22:46:23 JST silverpill

   • iced quinnsmas :blobcatsanta:
   • r
   • Nimda
   • frogzone
   • p

   @r @icedquinn @nimda @frogzone @p

   >Parsing the HTML and looking for the links with the "hashtag" class is another solution, but I don't want to do that either.

   I think this is the correct solution. Hashtag links should always have .hashtag class or rel=tag attribute. In my JavaScript client I attach event handlers to them, and on a server side I would rewrite href attributes.

   • frogzone@wizard.casa's status on Sunday, 28-Apr-2024 22:12:52 JST frogzone

     in reply to

     • iced quinnsmas :blobcatsanta:
     • r
     • Nimda
     • p

     @r @nimda @silverpill @icedquinn @p I liked your above toot a week ago, and I get where u were coming frm at the time, but now i'm seeing where else i'm being possibly affected by this resistance to use a parser and i'm thinking that maybe(?) its time to parse????

     its re "RE:" links. See this very recent convo i was in if you weren't including in it already: https://wizard.casa/objects/018eea1b-82a5-f024-ddff-315f8f8f2f18

     People often post them and when i click em they are useless to me. they try to go to their actual website rather than try to find the thread on the server that i'm using, long story short i'd rather it did the latter, even if it resulted in errors like "thread not currently stored on this server". Ideally it would not just work for boost-toots, but MAYBE(?) for any toots that are linked to (dunno tho, might not be trivial computationally to detect and account for all the ways fediservers arrange toot urls).

   • r (r@freesoftwareextremist.com)'s status on Sunday, 28-Apr-2024 22:12:54 JST r

     in reply to

     • iced quinnsmas :blobcatsanta:
     • Nimda
     • frogzone
     • p
     @silverpill @icedquinn @p @frogzone @nimda It would've been easy if the URLs in the tags array also contained the remote URLs, then it'd be just a simple string replacements. What's confusing is that that's how it is for mentions, both the links inside the post HTML and the mentions object array contain the remote link.
     So far I've avoided using and HTML parser and I don't think the hashtags are important enough to worth the extra bloat.
   • r (r@freesoftwareextremist.com)'s status on Sunday, 28-Apr-2024 22:13:23 JST r

     in reply to

     • iced quinnsmas :blobcatsanta:
     • Nimda
     • frogzone
     • p
     @frogzone @icedquinn @silverpill @p @nimda Ah, it's a quote post. Showing quote post doesn't require an HTML parser, the JSON response includes all the necessary data to render it. But I personally don't like the quote posts because they break the normal flow of the conversation, which is why it's not supported in bloat. But there are some forks of bloat that support it.