PDF pages not readable by Adobe Acrobat/Reader after "Recompress Pages" SOLVED

[David.P]

Hello Forum and Tracker Support Team,

It seems to happen regularly that PDF files (that I have worked on for a longer period of time in PDF-XChange Editor) show sporadic erroneous pages in Adobe Reader and Adobe Acrobat. The corresponding PDF documents then no longer can be read correctly, and can't be saved anymore by Adobe products as a result. In PDF-XChange Editor, the files appear completely problem-free.

The issue seems to occur mainly in PDF files that have been assembled from various source documents, e.g. from scanned documents combined with Word and other documents printed to PDF. It seems to occur particularly frequently after using the Replace Pages feature of PDF-XChange Editor. Saving as Optimized, as PDF/A or as PDF/X usually does not fix such problems either.

Only saving the file as Sanitized PDF seems to fix all such problematic pages (except for pages with missing fonts). Sanitizing however is not a feasible solution because it deletes all comments and additionally seems to rasterize most pages.

I could provide a file compiled from about 30 such problematic pages, however, I can't post it on the forum for privacy reasons.

Please let me know if I can email this file to the Tracker Support e-mail address.

Thank you
David

[David.P]

PS: as discussed in the other thread with Victor:

https://forum.pdf-xchange.com/viewtopic.php?f=62&t=35707&p=148200#p148200

...I have sent the file containing the defective pages (defective only in Adobe products, that is) via e-mail.

The pages have become defective after working with the respective 5000 pages PDF file mentioned in the first post. That large PDF has been worked with very extensively over about 2 years, and has been saved hundreds of times thereby.

I have checked the file's version history, and the file has become defective on January 29 -- the day after I installed PDF-XChange Editor v.9.

So this time, it seems that the defects have been caused, or at least triggered, by PDF-XChange Editor -- right?

Best regards
David

[David.P]

Ok -- I believe that I have traced the problem down to some extent.

I'll send you the aforementioned, defective pages again, but this time in their previous, intact state as coming from the source file of January 28, when the pages were not yet defective (in Adobe products).

These pages apparently went defective because I had run "Recompress Pages" over the source file using the Editor v.9, with the following settings:



If this process is run over the pages in their intact state as just sent to you, then most of the pages go defective again, and cannot be displayed anymore in Adobe products.

In other words, the "Recompress Pages" operation currently seems to be causing certain types of pages to become unreadable in Adobe products.

I hope this helps you to figure out what could be the cause of this file corruption.

Best regards
David

[Lzcat - Tracker Supp]

Hi, David.P

Good news is that we found issue in Recompress image(s) and will fix it. Bad news is that there is no simple way to fix already broken files. In short - after recompressing inline images Adobe cannot handle them. If you know which image causing problems there is simple way to fix it - just edit it in one of external image editors (no need to change anything, open and save should be enough). But problem is that Editor cannot help to identify problem image (display both good and broken), when Adobe does not display page at all. So if your file have lot of images this solution is not good. For example in file "34 Pages ... pages are defective in Adobe products.pdf" images causing problems are last one in content, but in common case they can be any.

PS. Why all you files contains so much layers information? Even good file have more than 400KB of it, and as far as I can see it does not affect document. However in broken file this information use almost 8 MB, more than 2/3 of file size. What are you doing with files that such information grows?

image.png

image.png

Kind regards,
Lzcat - Tracker Supp

[David.P]

Hello Victor,

thank you for your news. I am glad to hear that you were able to identify the issue, and that it is about to be fixed.

Regarding the corrupted file, I'm currently trying to revert back to the last unbroken version, and to manually transfer any changes I made in the meantime.

Thanks also for the tip on how to repair those broken images.

Regarding your question about why these files contain so much layer information: I have no explanation for this, except that (as mentioned earlier) my source file has been edited thousands of times with PDF-XChange Editor over the course of two years. A few times, the file may also have been edited and saved with Adobe Acrobat.

I don't know if this explains why there is so much unneeded layer information present?

I also would love to know how to identify, and ideally delete, such unnecessary layer information. Are your two screenshots above related to this layer information?

Is it possible to batch-delete this unnecessary layer information from an entire large PDF file?

Best regards
David

[David.P]

PS:

I have now saved the 5000 page PDF file as "Optimized PDF", once using PDF-XChange Editor, once using Adobe Acrobat DC and once using Adobe Acrobat XI. In each case, I had all overhead optimized and everything deleted from the file except comments, bookmarks and links.

Below are the results:

Original file
282 MB


Optimized with PDF-XChange Editor v.9
226 MB


Optimized with Adobe Acrobat Pro DC
230 MB


Optimized with Adobe Acrobat Pro XI
229 MB

[Lzcat - Tracker Supp]

Hi, David.P

Regarding your question about why these files contain so much layer information: I have no explanation for this, except that (as mentioned earlier) my source file has been edited thousands of times with PDF-XChange Editor over the course of two years. A few times, the file may also have been edited and saved with Adobe Acrobat.

Editing file should not be a problem, unless you insert pages from documents with layers - in that case layers, related to inserted pages should be copied. Looks like you will need to monitor some your operations with file to see on which step layers information will grow extremely. As I can see you found that layers information is calculated in document overhead, so this may be a good indicator. But please not that document overhead include some other information too, so it can be only indicator. Normally if you insert pages from one document into another resulting document overhead should be less that sum of original overhead and overhead from inserted document.

As I told before in other thread Flatten All Optional Content command should remove all layers information (Flatten Layers in Adobe).
And yes, looks like Save As Optimized PDF will do this too (but it will affect other things, regarding to options).

Kind regards,
Lzcat - Tracker Supp

[David.P]

Thank you Victor, that was very helpful.

For completeness, this is what I get when I do a "Flatten All Optional Content" on that file:

All optional content flattened with PDF-XChange Editor
257 MB


For some reason however, every current version of that file, including all "Optimized" versions, still perform exceptionally bad in Adobe products. While there are no page errors anymore, Adobe Reader (latest version) sits there eating 25% CPU load when that file is only just opened, and not touched or moved any further.

While, using blazing-fast PDF-XChange Editor, this problem doesn't bother me too much at the moment, it still would be good to eventually find out the reason for this extremely bad performance of that file in Adobe.

It is to be noted that any version of that file that is older than January 28 (the day I installed PDF-XChange Editor v.9) performs without problems in Adobe. So it still seems to be likely that Editor v.9 is causing some (additional) problems when saving, as compared to v.8.

Edit: I just realized that older versions of that file, like of November 2020, behave equally bad in Adobe products. Thus, there is no additional, new issue with Editor v.9. Still, I'd be interested to find out why that file behaves so bad with Adobe.

There must be something wrong with the entire file, because when I save the first and second half of the file as two separate files, both still have the same bad performance in Adobe, even after saving them as "Optimized PDF".

Edit 2: one more find -- when I delete all ~5000 comments from the file, its performance in Adobe is flawless! So there seems to be a problem with the handling of (many) comments in the latest Adobe products. For example, it seems to take ages for Adobe Reader to recognize all comments, it reads about one comment per second (as can be seen in its comments pane) -- maybe because it tries to do something with those comments in the Adobe cloud.

One more reason why Adobe products are literally UNUSABLE for heavy PDF document lifting.

Edit 3: "No Comment":

Adobe Support Community: Acrobat Reader DC runs extremely slow when using Comments and Sticky Notes:
https://community.adobe.com/t5/acrobat-reader/acrobat-reader-dc-runs-extremely-slow-when-using-comments-and-sticky-notes/td-p/10361457?profile.language=de#bodyDisplay_7:~:text=I%20CANNOT%20BELIEVE%20THERE%20IS%20STILL%20NOT%20A%20FIX%20TO%20THIS

Acrobat Uservoice: Comment list takes too long to load:
https://acrobat.uservoice.com/forums/590923-acrobat-for-windows-and-mac/suggestions/19435291-comment-list-takes-too-long-to-load?page=1&per_page=40

I can't believe it how a company can get away with software that is so awfully bad. Another test file of mine containing 2500 comments has still not finished loading in Adobe Reader DC after 1 (one!) hour. It is now down to reading about three comments a minute -- and there are still a couple hundred comments left.

https://community.adobe.com/t5/acrobat/acrobat-dc-so-horrible-it-s-making-me-want-to-cry/m-p/7112272

"No comment"


Edit 4: I've marked this as solved. Thanks again to Victor for the helpful support.

Kind regards
David

[Tracker Supp-Stefan]