Struggling with HTML import

Question

I am trying to import html pages into a new space in Confluence, but this does not work as expected. The html pages are in a zip file like described on https://support.atlassian.com/confluence-cloud/docs/faq-import-data-from-html-to-confluence/, this is:

* Contains a folder with the name of the space
* html files in that folder
* For each html file there is a folder with the same name with the images

The result is a new space with a page for each html file, but the pages are empty.

Note that the importer also tells that the process can take hours and an e-mail will be send when done, but I never received an e-mail so far.

Can anyone provide assistance please?

Dirk Desmet · Answer

Thank you all for your input.

I conclude that the html import is not yet usable. It is probably still beta for that reason.

As plan B, I created a Rovo agent to help me importing the content.

Dirk Desmet · Answer

I created my 232 files by exporting the pages from another online documentation system to an xml file, and then converted that with a python script to html files, one html per page, and an image folder with the same name as the html file. Then all added that folder to a zip.

A simple html with only text works.
When I add an image, I see a replacement image with 'Preview unavailable' and 'External media file'.

I also checked my html files. They look very simple: head, title, body (with h1, p, src and inline styling).

I then tested with 2 html's: the simple one with 1 image and 1 of my 232. Result: the simple one shows the text and a replacement image. The other one is empty.

Still received not a single notification, not by e-mail, not in Confluence.

Kris Klima _K15t_ · Answer

Hi @Dirk Desmet

Importing anything to anything else is always a hit or miss. A couple of months ago I tried to import various HTML doc sets into Confluence and the results were subpar. From empty pages, to everything crammed into a single page... and formatting was all over the place. Some images were there, some were missing.

This is not your fault, or Confluence's fault, or the HTML's fault. There's just way to many variables in the entire route that inconsistencies will occur.

Me and a fellow Atlassian Champion @Matt Reiner _K15t_ had a live webinar on importing content into Confluence. We're both veterans of multiple content (mostly technical documentation) migrations from various tools into Confluence and we're sharing the experience.

You may find it useful.