7

We have a system that renders a report as a PDF, and displays it in the browser for the user. In the browser, the document always appears to display fine, but when printed on one machine, it sometimes changes some of the data in the report to seemingly random characters.

Here are some examples of the strings it inserts:

Ebuf;
Bvhvt ul1: -!3122
Ti jqqf e!Wjb;
Nfttf ohf s!Tf swjdf

Additionally, the inter-character spacing is weird. It sometimes writes characters overlapping each other.

I noticed some repetition in the garbled text, so I typed a few of them into Google, and surprisingly got a lot of hits. Here is the string I searched for: pdf cjmp ebuf nftf up!

The Google search summaries contain the garbled text. However, when I click on those links in Google, I get perfectly readable PDF files. It's as if Google's PDF crawler has the same bug.

Has anyone figured this out? Is this an Acrobat Reader bug?

2
  • 1
    Have you tried reading the PDFs in another viewer (such as sumatra)? or printing/creating them with a different method (Cutepdf, pdfcreator, openoffice, etc)?
    – MaQleod
    Aug 11, 2011 at 22:34
  • 1
    @MaQleod - it's not readily reproducible. Opening the same report again and printing it a second time seems to print it just fine. Also, on other computers it doesn't seem to happen. I don't have the option to open the report in a different viewer because the PDF is streamed from a report server and opened in Internet Explorer. Aug 12, 2011 at 15:47

7 Answers 7

4

I noticed that there were lots of good suggestions for how to fix this printing issue, but I thought it might be interesting and useful to post an explanation of the underlying cause.

It appears that these PDF misprints are due to a character encoding problem. All the ASCII characters are showing up as 1 higher than they should be (see ASCII character table here).

For example, the letter "D" is showing up as "E", and the letter "b" is showing up as "a". White "space" shows up as "!", and ":" shows up as ";". *Note that whitespace in the garbled PDF doesn't mean anything. "space" shows up as "!".

Here is the example string from the original post:

Ebuf;
Bvhvt ul1: -!3122
Ti jqqf e!Wjb;
Nfttf ohf s!Tf swjdf

Correcting for the "+1" encoding error, we get:

Date:
Augustk09, 2011 
Shipped Via:
Messenger Service

I suspect that the l on line 2 of the example was probably a !, which would be a "space" such that the correct message would read:

Date:
August 09, 2011 
Shipped Via:
Messenger Service
2

One fix which overcomes lots of Acrobat printing problems is to "Print as Image". As shown in the link, you can select that at print time, under the "advanced" button. Unfortunately it can also mean that the print jobs become significantly larger (this is printer and job dependent).

Another thing you can do is to try a different PDF reader (e.g. Foxit or CutePDF) but organisational readers may make this impossible.

2

Short Answer Uncheck "Rely on system fonts" in Acrobat Printer as follows: 1. Request to print the page (however you do that). 2. Select "Acrobat" as the printer (uses Acrobat Distiller) 3. Click on Acrobat Printer's "Properties" tab. 4. Go to bottom of dialog and UN-check "Rely on system fonts". 5. Click OK. Then "OK" again to print.

Long Answer I suffered the same garbled text problem today while trying to print an online receipt to the Acrobat printer (aka "distiller"). I have Acrobat XI Pro installed.

Previously, when I had the same problem, I used the "Print as Image" option which fixed it but at the large expense of creating a huge digital image file whereas a true Acrobat image is quite small in comparison.

Today, while searching for the "Print as Image" option, I noticed on the 1st Acrobat Printer Properties dialog window, there was a checked box near the bottom that was enabling "Rely on system fonts". That seems like a bad choice in any case because you never know if your system's fonts will exactly match the document fonts. When they don't, I assume there would have to be some kind of substitution logic being performed. If it cannot find a close visual relative, it probably uses something wild or just barfs into the file (which it does look like :-)

So, I unchecked that option and, voila, the page printed with all the text correct. I thought I had a side effect when the background behind the text was a light gray so I printed to Microsoft XPS which is a nice alternative to Acrobat. It also showed the gray background so I assume the web page that was offered that up when I clicked the web page's Print Page button and it did, indeed, have the same gray background. The fonts matched exactly as well so I think it's all correct.

So, problem solved. When I repeated this method, I noticed the option was re-enabled so you have to do it each time or find the option to permanently set it to disabled. For now, I will reserved judgement on whether to make it permanent and just uncheck it when Acrobat decides to barf.

Lastly, in one of my tests, I did come across an Acrobat text log file in which it said, Font not found", further validating my supposition as to what the problem was.

[unsolicited opinion] I, like so many others, can't wait for Adobe to go under or get some humility so we can buy and own our products rather being forced pay a mafia fee or lose use of our products. For now, I'll keep using the last purchasable versions until I am forced to look elsewhere. The competition is slowly closing in. [end of diatribe]

Hope that helps.

2

None of the other answers I've seen worked for me ("print as images", uncheck "rely on system fonts" [wasn't available]). In the end I had to switch to a different pdf viewer, and it printed my files just fine (the one I used is PDF-XChange PDF Viewer).

1

I also suffered from that problem from time to time. As comments indicate it is not easily reproducible.

However, I notice that, in my system, it almost always happened when a print job was given and I closed Adobe Reader before the print job actually started. Although figures and other geometrical aspects were kept OK, the text was gobbledygook (with boxes, copyright signs and extended ASCII characters, and so forth) in most parts of the document.

A solution, besides the one you pointed out, is to not close Adobe Reader, at least until the printing has finished.

0

The problem (and a solution) are described in this blog post.

TL;DR is “print as image.”

Specifically as outlined in the post:

Today, after reprinting the same PDF several times, and getting garbage out each time, I clicked “print”, and started looking around in the print window that opens for my HP 7260 printer. When I print PDFs, I get an “Advanced” button in the bottom left corner of the print window that I can click on. There are a bunch of “greyed-out” postscript options in this window, but there is also a check box that says “Print as image”. I decided to check this box and click “OK”, which closes the advanced printing options window, then I clicked “OK” to send the PDF to the printer. I think doing this causes the entire document to be sent to the printer as “raster graphics”, rather than as PCL containing a mix of graphics and ASCII text.

1
  • 4
    Link only answers are ghastly, abhorrent, and other kinds of nasty. Please, pretty please, quote the relevant parts of that link to avoid link rot (i.e. the site disappearing from the web) Nov 6, 2013 at 19:44
0

Wondering if it's the way certain typefaces are rendered (resized to fit the page for example) in the browser.

I also had issues with being able to print a auto generated document via the browser. The solution for me was to simply download it, then open the downloaded file in Acrobat. Printing directly from the application doesn't seem to interfere with how the typeface is being rendered.

A possible solution could be to make sure you've got the latest acrobat plugin for Chrome/Firefox, but I prefer downloading to relying on the browser at this point.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .