Jump to content
You must now use your email address to sign in [click for more info] ×

Bug report - Publisher 1.9.1: When exporting to PDF with Rasterise: Everything, inserted Passthrough PDFs are missing


Tonda

Recommended Posts

Steps to reproduce:

1. create new document (image policy does not matter)

2. insert a PDF file in the document and set it as PDF Passthrough

3. File > Export > PDF

4. PDF (for Print) > More > Rasterise: Everything

5. After exporting, the resulting PDF is blank.

If you add another object in the Publisher document (for example write some text with text tool), this other object is exported, but the inserted PDF is still missing (so in my example you would end with blank PDF containing only the text).

Link to comment
Share on other sites

  • Staff

The rasterise option is meant to basically make an image to place in a pdf. Having a PDF as passthrough wouldnt honour that - and reading the PDF to do the rasterisation wouldnt honour the passthrough either - So I think the best we could do in this case is have some sort of warning.

Link to comment
Share on other sites

Isn't Passthrough just the way how is the content of the PDF presented, or processed during export into object-based PDF? Technically nothing prevents Publisher to rasterize such PDFs, and actually it can do it already - if I select such inserted PDF, I can rasterise it using the context menu, and if I have on the page another objects alongside the inserted PDF, I can just select all, group it and then rasterise the group, and the result is what is supposed to be in PDF exported with Rasterise: Everything.

Also, there is no problem when exporting page with Passthrough PDFs to bitmap formats, and I guess the process is the same as when the page is being rasterised during export to PDF, so I don't really see the reason why it should not be possible into PDF...?

Link to comment
Share on other sites

@Tonda Forgive my ignorance: are you rasterising everything to prove a point? If not, why rasterise a pass-through PDF?

iMac Pro (2017) Processor: 3GHz 10-Core Intel Xeon W / Memory: 64 GB 2666 MHz DDR4 / Graphics: Radeon Pro Vega 64 16 GB | iPad Pro: 12.9 inch 2nd Gen

Link to comment
Share on other sites

@deeland Hello, there are actually two reasons for this situation:

1. As far as I know, passthough PDF mode is the only way how to get PDFs into Publisher reliably – and I am talking about huge amount of PDFs I am getting from various sources, who are not able to send anything else (usually it is some work that someone did for them in past and they ended only with the PDF), and I need to use these PDFs in further works (so I don't need to edit them, but only insert them as they are in other documents I work on). With the original PDF mode, I often ran into problems that Affinity wasn't able to interpret the content properly (so for example although the PDF had all fonts properly included and in past even commercial printer used this same PDF for printing the brochure without any problems, in Affinity some fonts are shown just as a character mess, or the pictures got white stripes over them).

2. Regarding the output - yes, of course, if the PDF goes into commercial printer, then no need for rasterising. But I often work on stuff like leaflets or PDF brochures, where the clients doesn't want the content to be in vectors (for example because they don't want the included logos to be extractable – and if you want to say that I can use PDF protection, then no – you can always extract the elements if you want, no matter how the PDF is protected). So exporting into rasterised PDF is quite common for me.

Link to comment
Share on other sites

@Tonda

Quote

1. As far as I know, passthough PDF mode is the only way how to get PDFs into Publisher reliably

From Affinity 1.9 promotional material:

"PDF passthrough (desktop only)
It’s now possible to flag your imported PDFs for PDF passthrough. This ensures that the file will be a perfect representation of the original PDF when exporting, regardless of whether you have the embedded fonts installed or not."

It’s designed to allow placed deliverables that match production criteria, not for helping in the hacking of an already delivered, PDF deliverable. As I said before: when these unfortunate and not ideal situations present themselves, opening the PDF in a program that can edit it correctly, would be the best option. Adobe Illustrator comes to mind, as does Affinity Designer. Of course, security can create issues and limitations, with the security put in place for a purpose. But as you say, perfectly crackable. As for typefaces and their font glyphs: these are always limited at time of PDF export to the bare-minimum for printer-publishers, in accordance with typeface licensing rules. The Passthrough method used by Affinity, should do just that: pass-through the original state of the placed PDF, at PDF export of the rest of your master Publisher document. Ergo: if you can’t hack the PDF the way you want it to be in Affinity Designer )or whatever program you choose), and have that become your placed artwork in your Publisher document, then one can’t expect Publisher to perform any magic tricks.

Publisher is giving me a hard time right for very different reasons, but I know it’s limitations (and similar programs), regardless.

Quote

…where the clients doesn't want the content to be in vectors (for example because they don't want the included logos to be extractable…

Following my point of generating a Designer file that replaces the hacked PDF: you could then separate the logo elements (or other objects), and rasterise them here, instead.

Sounds like you have a lot on your plate and a very demanding workflow with unexpected issues around every corner. Very stressful. Back in the day, PitStop was used in PrePress for essentially ripping apart and editing elements at the PrePress stage, but it wasn't a miracle-maker.

Anyway, I think your answer is to adjust your workflow, so you have full control of the hacked PDF, before pushing through to Publisher. This would also allow you to control colour profiles. It might spare a few premature grey hairs, no? Regardless, good luck to you!

iMac Pro (2017) Processor: 3GHz 10-Core Intel Xeon W / Memory: 64 GB 2666 MHz DDR4 / Graphics: Radeon Pro Vega 64 16 GB | iPad Pro: 12.9 inch 2nd Gen

Link to comment
Share on other sites

@deeland Sorry, but I think there is some misinterpretation going on. I don’t need to hack any PDFs. Let me describe the whole situation more thoroughly:

I need to be able to insert into Publisher the PDF documents, that I receive from clients or others. This is quite common thing, especially if you work on stuff like advertisement magazines or conference brochures, where you create the material itself, but lots of the content comes already done by the companies who advertise there (their ads etc.) and you just insert these PDFs into designated areas. So for example in the conference brochure, each participating company may have 1 page, where first half is their logo and text description with some table, and second half their advertisement – and that advertisement is usually provided by them. So the work on the page looks like this: you insert their logo, write or paste the text and table content, and then insert PDF with their advertisement. You don’t need to do any edits on the PDF, because it is supposed to be in the brochure exactly as they sent it to you.



Until version 1.9, if you received any PDF with embedded fonts that weren’t installed on your computer, this was impossible to do only in Publisher, because Publisher reinterprets PDF during inserting, so the fonts get substituted or in worse case the texts end being complete mess. Many other DTP applications solve this by letting convert the PDF to paths during the inserting, but unfortunately Publisher does not have this functionality, and if I recall correctly, long time ago someone from staff mentioned here that it will never be part of Affinity because of some technical or licensing reasons.

 So my solution was to convert the PDF to paths in callas pdfToolbox first, and then insert the PDF in paths into Publisher, which works fine, but quite undermines the approach of Publisher being self-sufficient tool for DTP, not to mention that pdfToolbox is quite expensive, so getting a DTP program and then being forced to buy another software, which cost roughly 10times of the DTP program, just to being able to insert PDF without problems seems quite ridiculous for me. 

So I was really happy to see that there is now PDF passthrough, with which if you insert a PDF in Publisher, it lets the PDF stay in its original form – so it completely solves my problem for getting PDFs into Publisher reliably.

But then I have discovered the bug which is this thread about. If you export such document with inserted passthroughed PDFs into PDF with rasterisation on, the PDFs are missing in the export. I perfectly understand the sentence from manual you have quoted, because it means what I am writing. Also, there is no problem in this inserted PDF being rasterised in Publisher, because you can do it right now - just cmd-A on the page, group everything and rasterise, and you get the whole page rasterised, including the PDFs. If you then export it as PDF, this is what I need, just done using many steps, that would not be necessary if the export worked properly.

So I think there is just some problem in what Publisher does when exporting the PDF - probably the code of export wasn’t adjusted for the new situation, when there may be some PDFs in passthrough mode on the page, so right now the Publisher just doesn’t have proper instructions what to do with these, so they are missing. Of course I don’t know the code, but no matter of the current state of the code, it can be easily implemented by adding something like “if Rasterise: Everything is selected, group and render all objects on all pages and then do the export and then return the file to original state”, because all this works if you do it manually.

Link to comment
Share on other sites

2 hours ago, Tonda said:

So I think there is just some problem in what Publisher does when exporting the PDF - probably the code of export wasn’t adjusted for the new situation, when there may be some PDFs in passthrough mode on the page, so right now the Publisher just doesn’t have proper instructions what to do with these, so they are missing. Of course I don’t know the code, but no matter of the current state of the code, it can be easily implemented by adding something like “if Rasterise: Everything is selected, group and render all objects on all pages and then do the export and then return the file to original state”, because all this works if you do it manually.

First, it's not clear to me why you need to select "Rasterize Everything" during the export.

But it sounds to me like the developers decided that since they could not do both Passthrough and Rasterize they decided to believe the Passthrough request. Perhaps, instead, they could count this as an error and refuse to do the export. But since Rasterize will require Interpreting the PDF, just select Interpret instead of Passthrough if you're going to use Rasterize Everything.

Basically, Passthrough and Rasterize Everything are an incompatible combination.

-- Walt
Designer, Photo, and Publisher V1 and V2 at latest retail and beta releases
PC:
    Desktop:  Windows 11 Pro, version 23H2, 64GB memory, AMD Ryzen 9 5900 12-Core @ 3.00 GHz, NVIDIA GeForce RTX 3090 

    Laptop:  Windows 11 Pro, version 23H2, 32GB memory, Intel Core i7-10750H @ 2.60GHz, Intel UHD Graphics Comet Lake GT2 and NVIDIA GeForce RTX 3070 Laptop GPU.
iPad:  iPad Pro M1, 12.9": iPadOS 17.4.1, Apple Pencil 2, Magic Keyboard 
Mac:  2023 M2 MacBook Air 15", 16GB memory, macOS Sonoma 14.4.1

Link to comment
Share on other sites

@Tonda technically, programs like PitStop or Callas pdfToolbox, do a “backend number” on a PDF. A PDF was always intended as a way of securing a document. Embedded glyphs of a font could be shipped to prepress without the entire typeface being required. Then the poor folks at prepress always had issues of some sort, so these programs allowed you to “hack” the PDF. If it walks like a duck, and it quacks is like a duck, well, there it is: you know the rest. So let’s not get stuck on the word “hack”, here.

Quote

So the work on the page looks like this: you insert their logo, write or paste the text and table content, and then insert PDF with their advertisement. You don’t need to do any edits on the PDF, because it is supposed to be in the brochure exactly as they sent it to you.



Understood. Standard workflow. So you have their logo as a vector file or, a tiny GIF in Word (just kidding, or am I?) separate to the PDF, plus the final PDF advert is supplied to spec, and as you say, no need to edit it.

Quote

Many other DTP applications solve this by letting convert the PDF to paths during the inserting, but unfortunately Publisher does not have this functionality

Well, kind of. The PDF still had to be made “correctly” with the typeface glyphs subsetted correctly. This is why pdfToolBox and other prepress tools were developed. But I get what you mean. Supplied PDFs can always have nasty surprises.

Quote

…the approach of Publisher being self-sufficient tool for DTP, not to mention that pdfToolbox is quite expensive, so getting a DTP program and then being forced to buy another software, which cost roughly 10times of the DTP program, just to being able to insert PDF without problems seems quite ridiculous for me.

Well, we just discussed this very thing. Maybe Publisher, and in turn, Affinity, was allowing you to do what you want, due to some oversight by Affinity, in older versions? No idea. Could it be that the PDF sources you are now being supplied are substandard? Can you use an old PDF from your older, successful files still in Publisher? Just trying to understand what has happened here. Sounds like you had it good for a while, but Publisher may have tightened up, or was forced to, on its commitment to PDF security or typeface licensing protection? If Publisher continues to do the work of prepress tools like pdfToolBox or PitStop, that could be an amazing advantage for you, but outside the intended scope of the program. Which in turn could be a legal minefield.

Quote

Of course I don’t know the code, but no matter of the current state of the code, it can be easily implemented by adding something like “if Rasterise: Everything is selected, group and render all objects on all pages and then do the export and then return the file to original state”, because all this works if you do it manually.

That last sentence “… because all this works if you do it manually.” Intrigues me. What do you mean?

To be clear, though. I’m not putting up a fight for Affinity here: I still have many issues not resolved myself. I am trying to help you out, here. Is it a “bug” or is it a limitation, introduced intentionally? That’s what I’m trying to understand. Maybe you now have to purchase a prepress “hacking” tool like PDF ToolBox? You said it yourself, that’s what you needed before discovering Publisher, be it in its earlier versions. But do tell me if your older and trusted PDF source files do or don’t work in your workflow with the new Publisher version using passthrough. Of course, I could be missing the point entirely: in which case, let’s walk away. I’m not paid by Affinity to engage with you. I just like to try and solve problems. Fair enough?

iMac Pro (2017) Processor: 3GHz 10-Core Intel Xeon W / Memory: 64 GB 2666 MHz DDR4 / Graphics: Radeon Pro Vega 64 16 GB | iPad Pro: 12.9 inch 2nd Gen

Link to comment
Share on other sites

@walt.farrell 

Quote

Basically, Passthrough and Rasterize Everything are an incompatible combination.

Well, that’s a very good point! Well said!

iMac Pro (2017) Processor: 3GHz 10-Core Intel Xeon W / Memory: 64 GB 2666 MHz DDR4 / Graphics: Radeon Pro Vega 64 16 GB | iPad Pro: 12.9 inch 2nd Gen

Link to comment
Share on other sites

@walt.farrell @deeland

1 hour ago, walt.farrell said:

First, it's not clear to me why you need to select "Rasterize Everything" during the export.

I personally have 3 reasons why I need to use often "Rasterize Everything", but I am sure there are others:

- speed - if you have file with many objects, from some point the exported PDF is easier to view if the content is being rasterized, rather than in being in vectors (this is important especially if you are dealing with people who have slow computers)

- size - for PDFs that are meant for example for online viewing or for proofing only, and there is no need to keep the vector quality, it is often much more space efficient to export is as rasterized PDF

- security - some people require the content not being extractable from the resulting PDF, which rasterized content fullfils best from all possible options

 

1 hour ago, walt.farrell said:

But it sounds to me like the developers decided that since they could not do both Passthrough and Rasterize they decided to believe the Passthrough request. Perhaps, instead, they could count this as an error and refuse to do the export. But since Rasterize will require Interpreting the PDF, just select Interpret instead of Passthrough if you're going to use Rasterize Everything.

Basically, Passthrough and Rasterize Everything are an incompatible combination.

1 hour ago, deeland said:

Well, that’s a very good point! Well said!

1 hour ago, deeland said:

That last sentence “… because all this works if you do it manually.” Intrigues me. What do you mean?

I don't understand why you are both convincing me that it is not possible to rasterize placed PDF (when it is being set to Passthrough mode), when I already twice mentioned that it works. If you don't believe me, try it yourself:

1. create new document (image policy does not matter)

2. place a PDF file in the document and set the mode of the object to PDF Passthrough: Passthrough

3. use text tool and write whatever you want somewhere on the page

4. use the vector tool and draw whatever you want somewhere on the page

5. Go to menu Select and choose "Select All"

6. Go to menu Layer and choose "Group"

7. Go to menu Layer and select "Rasterize"

Do you still want to keep arguing that when you set the inserted PDF to Passthrough, it is not possible to rasterize it alongside the other objects on the page?

Link to comment
Share on other sites

@Pauls

23 hours ago, Pauls said:

The rasterise option is meant to basically make an image to place in a pdf. Having a PDF as passthrough wouldnt honour that - and reading the PDF to do the rasterisation wouldnt honour the passthrough either - So I think the best we could do in this case is have some sort of warning.

I am not sure if we talk about the same goal. My goal is to export what I see in the document into regular rasterized PDF. Passthrough is for me just the way to force the Publisher to interpret the source PDFs without altering their content (so fonts are not being snown in substitutes etc.). And during the export with "Rasterize: Everything" checked, all I expect from Publisher is to rasterize everything I see on the pages and save it into PDF (which I supposed is what "Rasterize: Everything" does).

Link to comment
Share on other sites

1 hour ago, Tonda said:

I don't understand why you are both convincing me that it is not possible to rasterize placed PDF (when it is being set to Passthrough mode), when I already twice mentioned that it works. If you don't believe me, try it yourself:

1. create new document (image policy does not matter)

2. place a PDF file in the document and set the mode of the object to PDF Passthrough: Passthrough

3. use text tool and write whatever you want somewhere on the page

4. use the vector tool and draw whatever you want somewhere on the page

5. Go to menu Select and choose "Select All"

6. Go to menu Layer and choose "Group"

7. Go to menu Layer and select "Rasterize"

Do you still want to keep arguing that when you set the inserted PDF to Passthrough, it is not possible to rasterize it alongside the other objects on the page?

Of course it's possible to rasterize it.

But once it is rasterized, it is no longer a Passthrough PDF. It is a pixel layer.

And, to rasterize it, it had to be read and interpreted.

So, there's certainly an inconsistency:

  • The application is willing to Rasterize a Passthrough PDF if you select it and manually choose Rasterize, even though that requires interpretation and converts the PDF into a pixel layer.
  • The application is not willing to Rasterize a Passthrough PDF if you select Rasterizate Everything during Export.

Of those, I'm not sure which is the bug, or whether it's just an inconsistency.

-- Walt
Designer, Photo, and Publisher V1 and V2 at latest retail and beta releases
PC:
    Desktop:  Windows 11 Pro, version 23H2, 64GB memory, AMD Ryzen 9 5900 12-Core @ 3.00 GHz, NVIDIA GeForce RTX 3090 

    Laptop:  Windows 11 Pro, version 23H2, 32GB memory, Intel Core i7-10750H @ 2.60GHz, Intel UHD Graphics Comet Lake GT2 and NVIDIA GeForce RTX 3070 Laptop GPU.
iPad:  iPad Pro M1, 12.9": iPadOS 17.4.1, Apple Pencil 2, Magic Keyboard 
Mac:  2023 M2 MacBook Air 15", 16GB memory, macOS Sonoma 14.4.1

Link to comment
Share on other sites

  • 2 weeks later...
On 3/4/2021 at 10:50 AM, Tonda said:

Passthrough is for me just the way to force the Publisher to interpret the source PDFs without altering their content (so fonts are not being snown in substitutes etc.).

I'm a little late as I am catching up after a couple weeks. I did not see this last part answered, so maybe it will still be of help to you.

What the others are telling you about the inherent incompatibility between passthrough and rasterize everything is true. I think the key difficulty that is making this confusing to you is your understanding of what "passthrough" really is, and that is evident by the portion of your most recent post that I have quoted above: you are choosing passthrough as a way "to force Publisher to interpret the source PDFs without altering their content." In fact, passthrough does no interpreting at all during export: it just embeds the existing PDF as a simple stream of bytes without trying to understand them. On the other hand, rasterizing the PDF (which would also include rendering it to the screen) requires Affinity to interpret the meaning of those bytes in order to render them either to screen or to a raster file.

If a "passthrough" PDF is rendered, it is no longer passthrough: it must be interpreted to be rendered, and to the extent that Publisher fails at faithful interpretation (hence the need for passthrough), it would fail in exactly the same way whether rendering to screen or to a fixed pixel image.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...

Important Information

Terms of Use | Privacy Policy | Guidelines | We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.