"It's never been safe to replace DW's Markdown implementation, because Markdown is low-key a super diabolical language...."
I love how every single reference to markdown in this conversation has been from someone who is a markdown user and also has an association between markdown and hell
1c) formats will be divided into current (eg, the ones you can use in the text editor) and legacy (which will only be used for rendering older posts) so that old posts can maintain their correct formatting without ending up too many different formatting options available
1c1) if you want to get around this, all legacy options will be secretly supported for posts created via email or the posting api for Reasons
2) this is what happens when you try to maintain a codebase old enough to drink
Full discussion at https://dw-dev.dreamwidth.org/222342.html and https://github.com/dreamwidth/dw-free/pull/2680
@Satsuma holy fuck have I been out of the DW loop. Markdown? A friggin github repo??? 👀👀👀👀
@nursemchurt markdown has existed for a while, but only secretly (the api allows for it if you compose dw posts offsite for whatever reason and post by email supports it as well) so a lot of people don’t know about it haha
And yeah!!! The codes on github now! They’re working on making the site more mobile friendly right now too, the first set of tweaks just went to beta I think
So yeah definitely subscribe to the dw-dev & dw-beta comms if you’re interested in this sort of thing!
@Satsuma
Did not know this was going in behind the scenes.
The language around markdown is making me chuckle.
@hit_the_books it’s not official yet—the code’s not finished and then they’ll prob put it in dw-beta for a while after that
But yeah it’s being discussed, and it sounds like it’ll be useful & cleaner so I’m excited
Anyway summary of proposed changes & discussion:
1) all the formats DW posts can be formatted in will now have names!
1a) anything without a named format (eg, older posts, certain syndicated posts) will get assigned a format on a best guess basis and then parsed accordingly instead of the complicated mess of parsing logic that is currently going on
1b) the new RTE, when it finally emerges from beta, should get it's own name even though its output is functionally identical to HTML0