Another errornous image extraction...

[Tokolino]

I have one note which translates to a weird image extraction behaviour. The image itself is extracted and named properly, but it's linking in the note is totally strange - I see parts of the image name, but also a Base64 encoding of the image. Very weird.

An export with this note is attached.
Debug.zip

[akosbalasko]

Ehh.. yeah, webclips, another solution for the same stuff, now it is a link ("A" in HTML), but it's href is the base64-encoded pure content of the image, while within the tag the innerHTML contains an en-media tag with the detailed properties of the given resource.
It will take a bit longer to solve, sorry for the inconviences.

[Tokolino]

Hi @akosbalasko,
unfortunately this problem got back with version 6.0.5...

[akosbalasko]

Hi @Tokolino ,
Huh, okay, yes yes, it is still Open, I didn't solve it.

[akosbalasko]

Hi @Tokolino ,
I fixed it in version 6.0.6.

[Tokolino]

Hello @akosbalasko,
the result of your fix is that there are now two versions of the images in the resources folder: One under its original name (without link to the note) and one with the name "embedded.jpg". Is this as intended? So if you create one from the BASE64 data, then there is no need to extract the same again, so it could be deleted.
My goal would be that after conversion there are no unlinked resources.

[akosbalasko]

Hi @Tokolino ,
ehh, that's true. It is because these embedded images come from base64 image of the href attribute of a link tag, and as they don't have filename attribute or something else, I generated these embedded+index filenames.
While the others with the correct filenames are from the resource tags stored in a different place. The next step would be to check how they can be connected together.

[akosbalasko]

I think I found a workaround, released in v6.0.7