Creating an irregular shape for wrapped text

[CaroleA]

I know I can create a frame in any shape using the Pen tool to then include some Framed Text.

If I need to follow an intricate shape to create a frame, is there a way to create it with the Selection tools? It seems like it would be easier to do.

Here is an example of what I mean:

or this

[MikeTO]

Affinity does not include an auto trace feature to automatically create a vector path from the subject of an image.

[CaroleA]

3 minutes ago, MikeTO said:

Affinity does not include an auto trace feature to automatically create a vector path from the subject of an image.

I was not looking for an auto trace. The images above were screenshots taken from projects in PaintShop Pro. It does not have an auto-trace either. I just needed to make a selection with the Magic Wand, contract a bit (to give some padding on the edges) and then, I was able to add the text inside.

It is the making of the frame that is convoluted if using the Pen tool, so I was wondering if something could be done starting with a selection instead.

I guess not?

[MikeTO]

26 minutes ago, CaroleA said:

I was not looking for an auto trace. The images above were screenshots taken from projects in PaintShop Pro. It does not have an auto-trace either. I just needed to make a selection with the Magic Wand, contract a bit (to give some padding on the edges) and then, I was able to add the text inside.

It is the making of the frame that is convoluted if using the Pen tool, so I was wondering if something could be done starting with a selection instead.

I thought you were asking for a feature similar to InDesign which uses an autotrace function to detect the subject of an image and allow you to wrap text around or inside it. Publisher doesn't have that feature.

There are two new machine learning features coming in Affinity 2.6 which will likely do what you want. See below. For now, you can either create a path with the Pen tool and convert that to a text frame which you mentioned above, or edit a text wrap outline to match what you want it to wrap about.

Cheers

 

[CaroleA]

1 hour ago, MikeTO said:

There are two new machine learning features coming in Affinity 2.6 which will likely do what you want. See below. For now, you can either create a path with the Pen tool and convert that to a text frame which you mentioned above, or edit a text wrap outline to match what you want it to wrap about.

Yes, I saw that before. But I don't need the automation as a single click with a Magic Wand would be enough. My issue is just to use that SELECTION as a FRAME for the Framed Text tool (and yes, I should have specified that I was talking about Affinity Photo).

If it does not exist and is not in the "known" pipeline, I might add it as a suggestion since it would be faster for the user in many instances, than the Pen tool.

[R C-R]

3 hours ago, CaroleA said:

It is the making of the frame that is convoluted if using the Pen tool...

Understood, but how closely do you really have to follow the outline of some even complex shape you want to use for a text frame? It should usually be good enough to just create the shape, using the shape as a rough guide. To keep the text from looking too broken up & hard to read, you would most likely need to tweak the shape afterwards anyway, even if the contour exactly matched the edges of the shape.

[CaroleA]

1 minute ago, R C-R said:

Understood, but how closely do you really have to follow the outline of some even complex shape you want to use for a text frame? It should usually be good enough to just create the shape, using the shape as a rough guide. To keep the text from looking too broken up & hard to read, you would most likely need to tweak the shape afterwards anyway, even if the contour exactly matched the edges of the shape.

I thought that the example of a silhouette would illustrate what I would hope to achieve. In that example, I can use 1 click to select the area, 1 click to contract (to give padding), and if there are some tiny segments that are in the way, using the freehand selection tool to remove them would be just another click or two. To me, it is faster than making a shape with the Pen tool that might have 10+ nodes.

I understand that I might not need to have an exact duplicate of the original shape, but it seems faster to select the area using Selection tools than the Pen tool.

So if it is not available, I think it is worth adding it as a suggestion.

[MikeTO]

There are two similar features but I don't think you need to request anything given what is coming in 2.6.

• Autotrace: This has been a feature of design apps since the '90s - it could handle both of your examples. InDesign bundles this feature with Text Wrap so that you can click on the subject like you want and it generates an editable outline. I know you said you didn't want Autotrace, but I think this is the feature you actually want.
• Machine-learning subject selection: Autotrace is great for simple images - it started with images on a solid coloured background but it doesn't do well with complex photographs which is probably why Serif hasn't bothered adding it. With machine-learning, you can take any photograph, click on a person, and it will select the person, even if their long hair is blowing in the wind. This is the feature that is in coming in version 2.6 and which is in beta now. The beta is publicly available so you can try it yourself, or you can just click the links I shared above and watch the video that shows how it works.

Cheers

[CaroleA]

@MikeTO I'll look at that. I do have 2.6 so I'll see if it does work the way I wish.

[R C-R]

19 minutes ago, CaroleA said:

I do have 2.6 so I'll see if it does work the way I wish.

Keep in mind from Mike's link the Compatibility requirements:

Quote

COMPATIBILITY: Sorry but due to the functions we need to run the Object Selection and Subject Selection functionality, the Machine Learning features only available on Apple Silicon iPads and macOS devices (and those need to be running on a recent macOS, as Catalina, Big Sur and Monterey do not support the required calls). Machine Learning will run on both Windows x64 and Arm64 hardware running Windows 10 or Windows 11.

The necessary files you must download are also quite large so that may be an issue if you do not have a lot of available drive space.

EDIT: It is also not clear to me if the ML selected areas of pixel images can be converted to shapes that can then be used for text frames. (I don't have a compatible Mac to test with so could someone that does see if that works?)

[thomaso]

4 hours ago, MikeTO said:

Affinity does not include an auto trace feature to automatically create a vector path from the subject of an image.

For text wrap in Affinity a vector path is not required, nor is AI. – Since V1 Affinity has enough intelligence to create a text wrap path from a pixel layer. Accordingly @CaroleA can use the desired pixel selection tool to create text wrapping for her existing images.

[CaroleA]

7 minutes ago, thomaso said:

For text wrap in Affinity a vector path is not required, nor is AI. – Since V1 Affinity has enough intelligence to create a text wrap path from a pixel layer. Accordingly @CaroleA can use the desired pixel selection tool to create text wrapping for her existing images.

If I am not mistaken, you are using another app than Photo, correct?

Can you do the same thing ONLY in A Photo?

[thomaso]

Just now, CaroleA said:

If I am not mistaken, you are using another app than Photo, correct?

This was APub, switched to its Photo Persona to create the pixel selection.

[CaroleA]

2 minutes ago, thomaso said:

This was APub, switched to its Photo Persona to create the pixel selection.

Good to know it is possible with that workaround. I wonder how easy/hard it would be to add that functionality to Photo since not everyone will have both apps.

[thomaso]

1 minute ago, CaroleA said:

Good to know it is possible with that workaround. I wonder how easy/hard it would be to add that functionality to Photo since not everyone will have both apps.

It is not a workaround but the steps to create text wrapping for a pixel layer in a free shape with a pixel selection as you requested in your initial post. Switching to the Photo Persona is just a switch of tools, more of an expansion of the interface.

Since the code is present in Affinity, it is merely a marketing decision by Serif that Photo does not have the text wrapping options.

[MikeTO]

45 minutes ago, thomaso said:

For text wrap in Affinity a vector path is not required, nor is AI. – Since V1 Affinity has enough intelligence to create a text wrap path from a pixel layer. Accordingly @CaroleA can use the desired pixel selection tool to create text wrapping for her existing images.

Thank you, I didn't know this feature existed. It appears to work only with transparent backgrounds which is why I thought it didn't exist.

It seems a bit odd that white backgrounds are treated differently since it wouldn't matter to the code.

Perhaps someday the new ML feature could be integrated into the Text Wrap feature to save a step, too.

[thomaso]

1 hour ago, MikeTO said:

It seems a bit odd that white backgrounds are treated differently since it wouldn't matter to the code.

True. White or any plain colour. And transparency caused by a modified blend range curve could also be automatically used for the creation of a text wrapping curve.

[CaroleA]

I do have Publisher also installed but how do I access it when I am working with Photo?

I will be teaching Affinity Photo so, although I can explain this option if they have Publisher too, how would we go from working on a Photo project to the Publisher app for this particular task?

[thomaso]

Each of the 3 Affinity apps can open each of the 3 Affinity file types.

Additionally in each app the menu "File" offers "Edit in …" options which closes the document in the current app and opens it in the other.

[CaroleA]

2 minutes ago, thomaso said:

Additionally in each app the menu "File" offers "Edit in …" options which closes the document in the current app and opens it in the other.

Great! thanks.

[Alfred]

6 hours ago, thomaso said:

Each of the 3 Affinity apps can open each of the 3 Affinity file types.

There is only one Affinity document type. The three different filename extensions merely serve to identify the app that was originally used to create the document.

[thomaso]

6 hours ago, Alfred said:

There is only one Affinity document type.

Sounds philosophical. It depends on how "file type"/"file format" is defined. If the definition distinguishes by file extension (suffix) or by file header, then they are three different file types. The fact that any Affinity app is able to open any Affinity suffix is not sufficient to conclude that they create identical files and file types. (Compare: text files like .txt, .xml, .htm, etc. are not the same file types although they may contain the exactly identical contents (and encoding), also .PDF is not just another suffix for the .AI or the Affinity file formats.)

Although .afpub | .afdesign | .aphoto are considered interchangeable and/or use the same encoding ("language", not content), they still differ in certain aspects (content) when edited with an app other than the one they were originally created with. For certain edits, one app may require a destructive change of attributes but another may not, for example if it forces the change of "artboards" to "spreads". [furthermore: without resulting in actually facing spreads but single, separated pages].

While certain Affinity features for content editing are simply only accessible/available in certain Affinity apps (e.g. "Text Wrap"), there are features that have more impact on the content of the document than just using a specific tool.

The document dimension/orientation/arrangement ('layout') seem to be the main difference: In APhoto you cannot edit the document dimensions of an opened .afpub file, but you can rotate an opened spread. When saved and reopened in APub, this will not open as a rotated spread (with the spine rotated from vertical to horizontal), but with changed (squeezed/stretched) page dimensions in a different aspect ratio, which is of course very different from a spread/page rotation in APub (i.e. changing from landscape to portrait) and identical, compatible file types.

1.  APub layout, 2 facing portrait pages / 1 landscape spread:

 

2.  … opened in APhoto -> rotated:

3. … and reopened in APub: 2 facing portrait pages / 1 portrait spread with very different aspect ratio:

[R C-R]

2 hours ago, thomaso said:

Although .afpub | .afdesign | .aphoto are considered interchangeable and/or use the same encoding ("language", not content), they still differ in certain aspects (content) when edited with an app other than the one they were originally created with.

??? The content is the same; it is only the tools & features of the apps that differ.

If nothing else, consider that you can manually change the extension among .afdesign, .afphoto, & .afpub without changing the content of the file in any way.

[R C-R]

8 hours ago, Alfred said:

There is only one Affinity document type. The three different filename extensions merely serve to identify the app that was originally used to create the document.

Technically, the 3 extensions only determine which of the 3 apps an Affinity document file will open in by default, like if you double-click on it in Finder or Explorer.