-
Notifications
You must be signed in to change notification settings - Fork 19
Move image handler into pipeline step before "meta" #706
Comments
An alternative approach would be to move the "meta" step after the "html" step (thereby slightly breaking the extension point API) and extract the metadata from the VDOM. |
@trieloff was there a specific reason you added the image-helixalizer as part of the mdast->VDOM transformation and not as an operation on the mdast ? |
If I remember correctly, you said that you wouldn’t do this in word2md, so that the markdown would stay consistent. I might have concluded that this was a transformation that is specific to the HTML conversion. We can still pull this out into a separate pipeline step. |
yes. not in word2md. we could:
or
|
Right. But all other 'special' operations are individual steps. |
I very much prefer your second option.
|
? |
^^ that second option |
Or the one already mentioned in the description ^^ 😉 |
🎉 This issue has been resolved in version 7.7.0 🎉 The release is available on: Your semantic-release bot 📦🚀 |
Helix Pages now uses
context.content.image
in the metadata, but in case of Azure blob store images it unfortunately doesn't contain the/hlx_*
URL.@tripodsan suggested to move the image handler, which is currently rewriting blob store links on the VDOM as part of the "html" step, into its own "image" step and insert it before "meta", so the image URLs are already rewritten in the mdast when metadata is being extracted.
The text was updated successfully, but these errors were encountered: