Replacing style properties in multiple files

[PaoloT]

Hi,

As of now, I've been using captions around images as frames in a page layout program. This has helped greatly with localizations and change of text and object styles.

Now I have to move my captions, now consisting of reference numbers, directly into the image. Captions will be part of the main text, referred to part of the image by numbers.

I would do two things: (a) have the text in the image at the highest quality, and (b) be able to change the style of text and containing circle or box in batch.

As for (a), I guess the best solution is to place any raster image in a Designer layer, and the caption text in other (vector) layers. Should I export the resulting image as a PDF (or any other vector format), or can a TIFF file contain vector layers?

As for (b), is there a way to edit text and object styles in a Designer file, and then have these changes propagate to other files in a batch?

Paolo

 

[PaoloT]

Something else, about layers: when saving a raster image with text, you can leave text editable. I would suspect this is still "vector" text, that is not rasterized.

Paolo

 

[walt.farrell]

3 minutes ago, PaoloT said:

Something else, about layers: when saving a raster image with text, you can leave text editable. I would suspect this is still "vector" text, that is not rasteeized.

That depends on the output format. JPG, PNG, and TIFF, for example, are raster formats. They do not contain vector information.

Exception: Affinity applications can "Save Affinity layers" when Exporting a TIFF file. In essence, this embeds a native Affinity file (.afphoto, .afdesign, or .afpub) file within the TIFF file. Only Affinity applications can process that data. To all other applications the TIFF just has a rasterized image of the contents.

[PaoloT]

3 hours ago, walt.farrell said:

That depends on the output format. JPG, PNG, and TIFF, for example, are raster formats. They do not contain vector information.

So, unless using Publisher as the final page layout program, the only way to keep the captions included with the image file as vector text or curves, is to export the image as PDF.

Paolo

 

[walt.farrell]

31 minutes ago, PaoloT said:

So, unless using Publisher as the final page layout program, the only way to keep the captions included with the image file as vector text or curves, is to export the image as PDF.

 

Yes, but isn't that true even if you are using Publisher?

[PaoloT]

9 minutes ago, walt.farrell said:

Yes, but isn't that true even if you are using Publisher?

Since Publisher can accept images in the original format of Photo or Designer, I guess there is no need to first export them to PDF or TIFF.

Or does some sort of internal conversion happen in this case?

Paolo

 

[walt.farrell]

1 hour ago, PaoloT said:

Or does some sort of internal conversion happen in this case?

No, there shouldn't be any conversion.

[thomaso]

8 hours ago, PaoloT said:

the only way to keep the captions included with the image file as vector text or curves, is to export the image as PDF.

And SVG apparently. Advantage / difference to PDF:
– if it contains an image larger than the canvas (or nested /clipped) then the unclipped image gets exported.
– if placed as resource in APub you get direct access to its layers. (no need to "open")

svg text.svg

[PaoloT]

6 hours ago, thomaso said:

And SVG apparently. Advantage / difference to PDF:

Can SVG include raster images? This would be very important to me, having to mix screenshots, caption lines and caption text.

Paolo

 

[v_kyr]

1 hour ago, PaoloT said:

Can SVG include raster images?

Yes, these are then embedded as base64 encoded images.

[thomaso]

2 hours ago, PaoloT said:

Can SVG include raster images?

What do you see in the SVG that I attached in my post above if you place it in Apub? – Is there no raster image? No editable text?

[PaoloT]

51 minutes ago, thomaso said:

What do you see in the SVG that I attached in my post above if you place it in Apub? – Is there no raster image? No editable text?

That's true, sorry! Yes, editable text, over a layer containing a raster image.

Paolo

 

[PaoloT]

As far as I can see, what I'm trying to do (mass-replacing text and object styles in multiple files) is not possible. This complicates very much blending images are caption in the same image file. Any change would require hundred edits in a single project.

Importing is available for both types of styles. So, maybe if the Affinity suite becomes compatible with AppleScipt, this can be implemented as a script to be run on multiple files.

Paolo

 

[v_kyr]

17 minutes ago, PaoloT said:

As far as I can see, what I'm trying to do (mass-replacing text and object styles in multiple files) is not possible.

Well it all depends on what and how you're trying to do that at all here. Meaning, Designer itself hasn't the possibilities to do so in a flexible multiple file manner. - Instead you would have to use an/some intermediate file format here (via another file format export/import workarounds) which allows to search/replace/modify file contents accordingly.

SVG would be probably your best choice here, since it's a XML based text description of vector graphics drawing commands, which you can work with on a textual basis with every better text editor. So if you have combined, ID named layer groups here (with seperate vector caption text, used SVG based style definitions -as textual commands here- and ID named raster/bitmap images parts), you could manipulate the whole to an image belonging text/style via plain textual search/replace operations.

You best show some example of what you are trying & wanting to do here!

[thomaso]

3 hours ago, PaoloT said:

As far as I can see, what I'm trying to do (mass-replacing text and object styles in multiple files) is not possible. This complicates very much blending images are caption in the same image file. Any change would require hundred edits in a single project.

Not sure about your task exactly, I wonder if Data Merge can support you here? Load an image file paths file + the captions text file in a prepared layout with text frame & text style > click generate > choose SVG for export.

If this work for you then you can use this layout document with same images but different text (or vice versa) repeatedly.

[PaoloT]

4 hours ago, v_kyr said:

SVG would be probably your best choice here, since it's a XML based text description of vector graphics drawing commands, which you can work with on a textual basis with every better text editor. So if you have combined, ID named layer groups here (with seperate vector caption text, used SVG based style definitions -as textual commands here- and ID named raster/bitmap images parts), you could manipulate the whole to an image belonging text/style via plain textual search/replace operations.

That's it! I see that the SVG image, exported from Designer, preserves all the original properties when opened in Inkscape. By using unique font and line definitions, or even the ID for groups as you suggest, one can easily change the styles of multiple files with a text editor.

Thank you for the hint! SVG be it!

Paolo

(Each time I open Inkscape, I have to ask myself how can people continue to recommend it over something like Designer…)

 

reference-numbers-ex.svg

[PaoloT]

1 hour ago, thomaso said:

I wonder if Data Merge can support you here? Load an image file paths file + the captions text file in a prepared layout with text frame & text style > click generate > choose SVG for export.

If using Publisher, this might be one of the ways to make captions. My case would be even easier, being the usual image + text captions around it + reference lines linking image details and text captions.

In this case, I would have to prepare the images and reference numbers/symbols highlighting the image's details in Photo and Designer, but then create the final layout with a different program, based on markup.

That's what the good guys at Serif have done to me, not letting Publisher import Pandoc!

Paolo

 

[PaoloT]

Ouch!

Quote

SVG files do not embed fonts directly

Is this true? Non a good thing, unless one only uses standard web fonts.

But it is true that web fonts are a lot, and for what I have to do I could just use the Noto family.

Paolo

 

[v_kyr]

50 minutes ago, PaoloT said:

Is this true?

Not really, there are several ways to deal with fonts in SVG, though not all tools and webbrowsers always do/can handle all of these. - Note that the Affinity tools are not a benchmark here, as they only/just support a rudimentary subset of the SVG 1.1 specification at all (SVG v1.1, SVG v2).

• Creating Embeddable Fonts as Data URIs
• How to use fonts in SVG
• Using Custom Fonts With SVG in an Image Tag
• svg-buddy: SVG Font Embedding and Optimization
• ... and so on, do a Google search ...

[PaoloT]

12 hours ago, v_kyr said:

there are several ways to deal with fonts in SVG

According to what I read in the links (and elsewhere), it is not real embedding in the sense it is in a PDF. While in this latter embedded fonts are part of the PDF file itself, with SVG files embedding means that the SVG file contains the commands to download the needed files.

This seems perfectly fine on the web, being careful to use fonts that are actually available. But what about printing or generating a PDF? If I have a project in Publisher or InDesign, containing SVG files, what will happen when generating a PDF file to be either printed or read onscreen? Will the needed fonts be embedded in the PDF when it is generated?

Paolo

 

[v_kyr]

2 hours ago, PaoloT said:

According to what I read in the links (and elsewhere), it is not real embedding in the sense it is in a PDF. While in this latter embedded fonts are part of the PDF file itself, with SVG files embedding means that the SVG file contains the commands to download the needed files.

Read again and try out in a webbrowser! - As I've already said, there are several ways to deal with fonts in SVG, one way of completely embedding the fonts into a SVG file is, to use them the base64 encoded embedded way!

For example, try that use-embedded-tafelschrift.svgfile in FF ...

<svg xmlns="http://www.w3.org/2000/svg" width="450" height="150" font-size="24" text-anchor="middle">
<defs>
<style>
   @font-face {
    font-family: 'tafelschriftregular';
    src: url(data:font/truetype;charset=utf-8;base64,AAEAAAAQAQAABAAARkZUTXBBygsAAAEMAAAAH ... RkMAAAC4AAArugAAAAQABys=) format('truetype');
    font-weight: normal;
    font-style: normal;
   }
</style>
</defs>
  <text font-family="tafelschriftregular" x="190" y="32.92">
        This is the Tafelschrift embedded font
  </text>
  <text font-family="tafelschriftregular" x="190" y="82.92">
        The font is embedded base64 encoded
  </text>
  <text font-family="tafelschriftregular" x="190" y="132.92">
        Tafelschrift font
  </text>
</svg>

• use-embedded-tafelschrift.svg

Though Affinity apps can't deal with that, same as they can't with hundreds other SVG related things too! - Inkscape can't too in this case of embedded base64 font!

In order to generate woff/2 or base64 encoded fonts, the followwing sites are helpful here ...

• https://www.fontsquirrel.com/tools/webfont-generator
• https://www.htmlgoodies.com/webmaster/serving-up-base64-encoded-custom-fonts/  (related to the above shown URL)
• https://www.base64encode.org/enc/font/
• ... etc.

It's often easier to deal with woff/2 embedding of fonts from local files, here are some generated examples for both (woff & base64 encoded), see also the SVG samples inside, try out with FF etc.

• tafelschrift-woff-webfont.zip
• tafelschrift-base64-webfont.zip

2 hours ago, PaoloT said:

This seems perfectly fine on the web, being careful to use fonts that are actually available. But what about printing or generating a PDF? If I have a project in Publisher or InDesign, containing SVG files, what will happen when generating a PDF file to be either printed or read onscreen? Will the needed fonts be embedded in the PDF when it is generated?

The Affinity apps are too dumb therefor (...the handling of SVG embedded fonts), you will probably need other tools which are more capable in terms of overall SVG handling, capabilities & usage! - Can't tell/don't know, how InDesign overall handles SVG here.

You've to try out how Affinity apps do handle SVG referenced (linked) fonts inclusion, either for fonts from the net or local disk storage, aka if they then do embed those fonts when generating PDFs out of imported SVGs or not.

[PaoloT]

3 hours ago, v_kyr said:

For example, try that

Ok, I see that the font binaries have been embedded in the SVG file. Now that's clear.

I also understand that the Affinity apps (at least, Designer) can't embed fonts. Third-party utilities would be needed.

So, I'll file a feature request to Serif for embedding fonts when exporting to SVG.

In the meantime, I would think continuing to export to PDF is the savviest and fastest solution. Shame, because this would mean that all the PDF images will be converted to PNG when exporting to HTML, instead of remaining vector graphics.

Paolo

 

[v_kyr]

1 hour ago, PaoloT said:

So, I'll file a feature request to Serif for embedding fonts when exporting to SVG.

Well it's not just when exporting, in order to be more conform here, it would also be needed to import & parse in such SVGs the right way then. Meaning, what is the value of a functionality it can generate (export), but not parse in (read) itself then again? - So both directions export & import are important here to support then!

[PaoloT]

23 minutes ago, v_kyr said:

it would also be needed to import & parse

Just tried to import the SVG file with embedded fonts you attached above, and I confirm Designer 1.10 can't recognize the fonts (they are considered missing).

Paolo

 

[walt.farrell]

5 minutes ago, PaoloT said:

Just tried to import the SVG file with embedded fonts you attached above, and I confirm Designer 1.10 can't recognize the fonts (they are considered missing).

All fonts used by Affinity must be locally installed, or included as part of an afpackage file that you're working on.