Image diappears from ressource list

[fde101]

53 minutes ago, thomaso said:

So it's hardly comprehensible that "Convert to Curves" interrupts access to the existing image information.

The "Convert to Curves" feature does not produce a picture frame.  It produces a shape which happens to have a bitmap fill.

What you are really looking for here is an option to change the underlying picture frame into a shaped picture frame (or to change the shape of the picture frame) while keeping it as a picture frame, which Publisher currently does not support (but should).

If I create a shape and convert it to a picture frame (from the Layer menu), I can continue to use the node tool to adjust the properties of the shape.  This gives me what you are looking for: the resolution information of the image remains intact (as it is still an image and not a fill), and the frame is a manipulatable vector shape.

What I can't do is change what type of shape the frame is after it is created - I can't take the boring rectangular frame produced by the "Picture Frame Rectangle Tool" and convert it into a star without disrupting the contained image.  I can do that in QuarkXPress, but not in Affinity Publisher.

That is the major feature which is missing here - the ability to substitute a different shape for a picture frame without disrupting the contained image or the general size and position of the frame.  The same is true for text frames.

[thomaso]

1 hour ago, fde101 said:

(...) looking for: the resolution information of the image remains intact (as it is still an image and not a fill), and the frame is a manipulatable vector shape.

(...) That is the major feature which is missing here - the ability to substitute a different shape for a picture frame without disrupting the contained image or the general size and position of the frame.

I don't see a reason why the way of "thinking" would prevent the app from displaying existing info. There doesn't seem to be any purpose in this behavior. It's not that I don't recognize how (~why) it happens but rather for what goal / use / advantage. (Besides the fact that I am not really missing this info, different to the OP, I am just curious to understand the concept of Serif / the developers)

ps: You seem to mix the Affinity terms "image" vs. "picture frame". They are quite different – the image (frame) is more similar to a converted image layer (curve + fill) than to a picture frame, which can include both, a bitmap fill + an image (... besides additional content like further image(s) + vector object(s)).

[fde101]

45 minutes ago, thomaso said:

You seem to mix the Affinity terms "image" vs. "picture frame".

Sorry, I was being short with my explanation to cut down on wordiness.

A "picture frame" in Publisher is a vector layer which acts as a container for a (normally raster) Image layer.

Substituting a different picture frame (different shape) would in theory replace the vector layer but inherit the existing image layer, keeping its properties as an image intact.

 

The "Convert To Curves" feature as applied to an image layer currently replaces the image layer with a vector curve layer which has the image set as a bitmap fill.  This eliminates the image layer, removing it from the resource manger.

However, when applied to a picture frame, it actually converts the frame itself into one which can be manipulated with the node tool, leaving the image layer intact.

An alternative feature that would help with this (which was also discussed in other threads, as many of these other things have), would be a feature to wrap an existing image layer in a picture frame, keeping the image in place where it is - then you could apply "Convert To Curves" to the frame.

[thomaso]

Ironically we can convert a Picture Frame, containing a bitmap fill + an image layer, to curves, without losing its picture frame features (UI) and without creating/converting any existing content to a bitmap fill.

Instead the result appears buggy: Though the bitmap fill (before converting) still is reported at various UI spots to exists, it doesn't show in the layout, so now "Convert to Curves" seems someway to prevent a bitmap fill … and thus vice versa to a converted image, which forces a bitmap fill.
(not in the video: Via Fill Tool I can load a new bitmap to this converted Picture Frame – again without getting it displayed in the layout though the UI reports it in the context toolbar and the various color wells)