Re: [dita-ot] There are no Polish characters when publishing to PDF

[Radu Coravu]

Hi,

You just need to use for generating the PDF a font which supports Polish characters.

There is a font mappings XML file in the DITA OT:

http://www.oxygenxml.com/doc/ug-oxygen/#topics/DITA-font-mappings-details.html

Regards,

Radu

On Fri, Mar 20, 2015 at 3:35 PM, Grzegorz Krystaszek <grzek...@gmail.com> wrote:

Hello.
Is there a way to output PDF Polish handle special characters?
I know it is not officially supported PL language, but is there a way around?
An interesting situation is that when you export to PDF are PL knew but only on the left side in Bookmarks.
An example of a PDF file attached.

--
You received this message because you are subscribed to the Google Groups "DITA-OT Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dita-ot-user...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[ch...@kravogel.com]

Is there anything else I have to do?
In fo/cfg/fo I have updated the font-mappings.xml and added the font Aller to the list of default

    <logical-font name="Sans">
      <physical-font char-set="default">
        <font-face>Arial Unicode MS, Helvetica, Aller</font-face>
      </physical-font>

But when I render the PDF, it is still using Arial and not Aller and the Polish characters are still not rendered.

What else is required?

Hope you can help.

Chris

[Radu Coravu]

Hi Chris,

When you have an enumeration of font families in the font-face element, the PDF processor will prefer to use them from left to right.

So if the first font says that it can render a character, it will be used for that. Otherwise the second font will be asked if it can render the character and so on.

So you should probably set Aller as the first font.

After the PDF publishing is over you can also look in the console output received from ANT, it should tell at the end what fonts it uses and if certain characters/glyphs cannot be rendered with it.

One more thing, if you also have the Polish characters both in the titles and in the content, you should also take care of the physical fonts used for the "Serif" alias.

Regards,

Radu

--

[ch...@kravogel.com]

Hi Radu,
after updating the font-mapping.xml I found out that I also have to change the settings of the xep.xml file of renderX.
But that seems to be the half way only.

After this update I get the right font in the table of contents of the PDF and in the table of contents within the file.

But the content itself has still Helvetica.

Any ideas what the reason for that behaviour could be?

Best regards

Chris

[Radu Coravu]

Hi Chris,

I did not know you were using RenderX XEP to obtain the PDF.

You should set before publishing the parameters "clean.temp" to "no" and "retain.topic.fo" to "yes".

After publishing you should find in the temporary files folder defined for the transformation a file called "topic.fo". Try to open it and search to see exactly what font-families are used for what content. If the proper font families are used, then you probably need to have an extra configuration done to XEP.

As a test you could also try to publish directly using the Apache FOP, just to see if the same fonts are used in the same parts of the PDF file.

Regards,

Radu

--