We are about to start our Confluence→Fibery transfer for some of our existing projects, and I was wondering if there is a way to import CSV, but also re-attach images and attachments we had on the confluence page.
Basically, we have a database for projects, and an additional database for wiki pages. This allows us to link pages to projects and grant access to teams based on the projects a page is assigned to.
We plan to export from Confluence as MD/HTML, and restructure this into a CSV file so we can add metadata on the Project, as well as parent-pages for our wiki structure, and import it all from one CSV file.
The one thing we still need to figure out its: How to get the images back in.
Ideally we would just md link to the image/file, and get the importer to browse an attachments folder, but thats not possible with the current importer.
Unfortunately, it’s not easy to migrate from something like Confluence to Fibery, and still retain all inline images. When a rich text field in Fibery shows an image, this is either an image stored in the Fibery workspace, or an embed of an image stored elsewhere (which requires that image to be publicly accessible).
Confluence kinda does the same, so when you copy over the html, it is pointing to an image stored in Confluence (which Fibery can’t access, because it requires authentication).
As far as I understand, the only solution is to have a service which requests the images from Confluence, uploads them to Fibery, and then automatically re-maps the links in the HTML to point to the Fibery-stored image instead of a Confluence url.
I haven’t done this myself, but I assume it’s possible, if you have the coding skills.
Thank you for the reply. We expected that this was the case but its better to check before we engineer a crazy over-the-top version of an existing feature.
I will pass your answer over to the programmer focused on transferring the data and figure out the best solution.
@Wilco_Boode If you have the dev power for it, you could build a custom Confluence to Fibery integration and do it that way. The Notion one has the code avaliable to look at and could be a good inspiration to start from as it now also pulls the Rich Text and inline images from it.
I will note that the public version doesn’t include page content…
@ChrisG Is it possible to publish the updated Notion Integration to the public Notion - Fibery repository?
If you choose to go this way, be prepared for a certain amount of pain
I tried to develop one, and you will hit the issue of handling images, but also handling cross-references, and dealing with Confluence Page content which uses macros.
TLDR Confluence uses a proprietary XHTML storage format which does not adhere to HTML standard, and therefore you will always struggle to make Confluence content look/behave the same in Fibery.
@Wilco_Boode if you succeed, I’m sure other users (incl us!) would love to see/share the result
I think there’s a lot to learn from seeing how the Notion Integration works (to build your own custom one). And there has been updates to it recently, but the community one has not been updated since 2022.