Jakab, thanks for the tip, I will look into it when the time comes to refactor the current solution.

Unfortunately, it’s not an option at the moment due to the fact that I don’t have the resources to migrate every PDF generation to the new solution.

Does anyone have an idea how to solve this – is this a FOP bug or is there something wrong with my configuration?



From: Jakab Gipsz [mailto:szeak33@gmail.com]
Sent: 13. oktoober 2014. a. 10:16
To: fop-users@xmlgraphics.apache.org
Subject: Re: Encoding problem with one specific letter and postscript



You may can use the following strategy for generating watermarked PDF:
1. Generate Intermediate Format (IF) from yous fo/xslt.
2. Generate PDF from IF combined with another XSLT, which can put watermarks into the document.

I use this solution with good result.

You can find a good example for IF stampling in FOP 1.1 examples.

Bye, Szeak


2014-10-13 2:01 GMT+02:00 Lembit Gerz <Lembit.Gerz@nortal.com>:

I’ve included the 3 files including the fop-config.xml used to generate them in the attatchment. If for some reason this does not work, the same files can also be found here: https://www.dropbox.com/s/41x8so9swkuevjc/fopp.zip?dl=0



From: Glenn Adams [mailto:glenn@skynav.com]
Sent: 13. oktoober 2014. a. 0:08

To: FOP Users
Subject: Re: Encoding problem with one specific letter and postscript


You should provide the following:

  • maximally minimal input FO file (XML/XSLT input files are irrelevant)
  • the output PS file you obtain (when producing PS directly)
  • the output PDF file you obtain (when producing PDF directly)


On Sun, Oct 12, 2014 at 2:09 PM, Lembit Gerz <Lembit.Gerz@nortal.com> wrote:

No, if I generate the PDF directly using the same data and font file, then all the letters are displayed correctly and copying from the PDF is also possible.



From: Glenn Adams [mailto:glenn@skynav.com]
Sent: 12. oktoober 2014. a. 23:03

To: FOP Users
Subject: Re: Encoding problem with one specific letter and postscript


If you do try the same data using FOP generating PDF directly, then does the problem occur?


On Sun, Oct 12, 2014 at 1:42 PM, Lembit Gerz <Lembit.Gerz@nortal.com> wrote:

To add watermarks or other transformations to the document.

The current  setup is the following: generate postscript -> apply an awk skript to the ps, that for example adds a watermark -> convert the ps to pdf.

I know it might be a hacky solution, but unfortunately changing this setup is currently out of the question.


From: Glenn Adams [mailto:glenn@skynav.com]
Sent: 12. oktoober 2014. a. 22:31
To: FOP Users
Subject: Re: Encoding problem with one specific letter and postscript


Why aren't generating PDF directly from FOP?


On Sun, Oct 12, 2014 at 1:16 PM, Lembit Gerz <Lembit.Gerz@nortal.com> wrote:



I am using FOP 1.1 to generate postscript files with embedded fonts. Later I’m using ps2pdf to convert the postscript files to pdfs. The text includes Lithuanian letters.

However, after coverting them to pdf, two specific letters are displayed as squares, all the ohter Lituhuanian letters are displayed correctly. The problematic letters are the upper- and lowercase letters ė and Ė (e with dot, 0116 and 0117 in unicode). I can copy all the letters from the pdf (including the two problematic ones – when copying the square and pasting it somewhere, it displays the letter correctly).

I am using the standard Arial font from Windows fonts (arial.ttf).

This is my fop-config.xml:



        <renderer mime="application/postscript">



               <font embed-url="./arial.ttf" encoding-mode="single-byte">

                   <font-triplet name="Arial" style="normal" weight="normal"/>







When leaving out the encoding-mode=“single-byte“, the letters display correctly, but when copying from the pdf, I get gibberish.

When generating straight to pdf with FOP, everything is displayed correctly and copying is also possible.

I have tried other ps->pdf converters and they give the same result.

Using a metric file did not help.


The problem can be reproduced with the xml and xslt in the fop quick start guide (https://xmlgraphics.apache.org/fop/quickstartguide.html) with these modifications:

Set the name in name.xml to ABC14pąęčėųūĘĖŲČĄ.

Add the attribute font-familiy=“Arial“ to the fo:block in name2fo.xsl.

Use the above fop-config file and include the standard Arial font in FOP’s directory.

Run ’fop -xml name.xml -xsl name2fo.xsl -ps name.ps -c fop-config.xml’


Can anyone suggest, what could be the issue or how should I go about debugging this?

Thank You.








To unsubscribe, e-mail: fop-users-unsubscribe@xmlgraphics.apache.org
For additional commands, e-mail: fop-users-help@xmlgraphics.apache.org