I just tried with Masto - maybe there’s different versions, but it didn’t work with the one I tried.
Screenshot:
It’s probably for the best that this PR doesn’t also convert inline Markdown into an attachment to send out for Mastodon’s benefit, because then there would be the danger of apps that understand both showing two images. It’d be better if Mastodon did the translation when receiving stuff, but Mastodon doesn’t seem as good as MBIN when it comes to co-operating with Lemmy.
(edit: how that screenshot shows on MBIN is a bit disappointing though - at least looking at on the web)
For comments: not at all. If a Mastodon user tried to do what I did, with the inline image, nothing would show.
We could do what I think you’ve done, and regex the details of the attachment into ! [] () Markdown and add it to the text. There’s also a DB relationship between comments and images that isn’t used, but could be, I suppose.
I’ve never actually seen a Mastodon user try to add an image to something that ended up as a Lemmy comment, tbh, so it’s not something I’ve thought too much about.
We could do what I think you’ve done, and regex the details of the attachment into ! [] ()
To be clear, this pull request doesn’t use regex, it’s just JSON deserialisation and string interpolation.
I’ve never actually seen a Mastodon user try to add an image to something that ended up as a Lemmy comment, tbh, so it’s not something I’ve thought too much about.
The pull request actually includes one, the main KDE account tags !kde@lemmy.kde.social and includes pictures in their threads regularly. It’s just hard to tell from our side as you can’t see what’s missing.
I just tried with Masto - maybe there’s different versions, but it didn’t work with the one I tried.
Screenshot:
It’s probably for the best that this PR doesn’t also convert inline Markdown into an attachment to send out for Mastodon’s benefit, because then there would be the danger of apps that understand both showing two images. It’d be better if Mastodon did the translation when receiving stuff, but Mastodon doesn’t seem as good as MBIN when it comes to co-operating with Lemmy.
(edit: how that screenshot shows on MBIN is a bit disappointing though - at least looking at on the web)
This is Mastodon’s HTML sanitiser, you can see they stipe out
<img>tags.How does Piefed handle image attachments, btw?
For comments: not at all. If a Mastodon user tried to do what I did, with the inline image, nothing would show.
We could do what I think you’ve done, and regex the details of the attachment into
! [] ()Markdown and add it to the text. There’s also a DB relationship between comments and images that isn’t used, but could be, I suppose.I’ve never actually seen a Mastodon user try to add an image to something that ended up as a Lemmy comment, tbh, so it’s not something I’ve thought too much about.
To be clear, this pull request doesn’t use regex, it’s just JSON deserialisation and string interpolation.
The pull request actually includes one, the main KDE account tags !kde@lemmy.kde.social and includes pictures in their threads regularly. It’s just hard to tell from our side as you can’t see what’s missing.
Ah, great, thank you. It’s been added as an Issue for PF now, with a link to this post, so that’ll be handy.
(I was likely misusing the term ‘regex’).