Improve importing and exporting capability

[xplosionmind]

So far, except for Matter, the only way to import stuff is via the API.

It would be extremely useful if importing and exporting functionality could be added within the (web)app.

Furthermore, not only a mere list of links should be imported, but also:

• tags
• location (Archive, Inbox, favorites)
• date added
• date last opened
• …?

Personally, I am interested in importing from Wallabag.

[jacksonh]

I think what we can do is start to develop a standard CSV format, and then write adapters to get things like Wallabag and Matter into that format.

[thiswillbeyourgithub]

Fyi related to #2094 Would love that so much!

[sopoforic]

The homepage promises "Reading is a lifetime activity, and you shouldn't have to worry you'll lose your library after you've spent years building it. Our open-source platform ensures your reading won't be held prisoner in a proprietary system.", but as far as I can see the only part that can be exported is the highlights--so if Omnivore-the-service disappeared, maybe I have got a recent backup of my highlights (though--can you only back them up one at a time without using the API? a bulk export is needed, and I couldn't see anywhere you could do that), but without the documents they're attached to, they're of much less value. It's needed to be able to export both the saved document and the highlights so I can do with it as I choose, including importing into a successor system (self-hosted or not).

[sopoforic]

I want to note something that I've just realized, in connection with this: if you use the email feature of Omnivore to subscribe to newsletters, then you cannot even rely on the wayback machine + highlight export in case omnivore vanishes, and you won't have the newsletters in your own email, either, so this is a bit more serious for those cases.

[jacksonh]

I want to note something that I've just realized, in connection with this: if you use the email feature of Omnivore to subscribe to newsletters, then you cannot even rely on the wayback machine + highlight export in case omnivore vanishes, and you won't have the newsletters in your own email, either, so this is a bit more serious for those cases.

Yeah when we build an exporter tool we definitely wont make it URL only, it should export the full content (original + readable). You can do this now with the api-demo app, it just all needs to be packaged up into a standalone application.

[watsonbox]

Indeed, I discovered Omnivore myself recently, and it's surprisingly close to being a perfect read-it-later solution, save for the lack of export. I've started with the free hosted service which is great, but plan to self-host in future. Okay, I could put together a script to move data between the instances using the API, but it's a shame that's not available out of the box.

This is a matter of data portability. In the unlikely event of Omnivore disappearing all of a sudden, I'm satisfied that the Obsidian sync feature (also wonderful) provides a good solution to data backup should it ever get to that. That's just my use-case of course, as not all users will be able to take advantage of the notes sync facility.

[Cito]

I have provided a simple export script here: https://github.com/Cito/omnivore-export

[whatfuturedotlol]

hey, i'd also love a raw export of everything. my current flow backs up all the original content in a dropbox and also on my nas. @Cito am i correct in thinking your script only backs up the links, not the actual content itself?

[Cito]

your script only backs up the links, not the actual content itself

There is now a flag that also allows to export the content.