Hi,
No, an implicit save could bring new problems if the file is also
encrypted (if it was originally). A recent SO question shows that when
saving, the file is no longer usable in memory. So at this time the only
thing to do would be to change the documentation.
Re the other problem: no I don't know. Please create an issue with the
relevant files.
Tilman
Am 07.08.2020 um 18:07 schrieb Rainer Romatka:
Hi Tilman,
Thank you so much. This works.
I added the following lines to the end of the method that creates the overlay
document:
try (ByteOutputStream devNull = new ByteOutputStream())
{
watermarkDoc.save(devNull);
}
and that solved the problem with the embedding of the font.
Would it be an idea to add these lines (prophylactically) to the code in the
Overlay class? Ideally it would happen only if the PDDocument hadn’t been saved
yet but I don’t know if it’s possible to check for that. If there is interest
in this approach I’d be happen to submit a patch that adds this.
- Rainer
PS: I encountered a different problem while troubleshooting and trying to fix
this one: I originally used PdfBox 2.0.14 and upgraded to 2.0.20 while
troubleshooting. After that I had a problem with an jpeg image (a logo) that is
part of the overlay. It shows up on some pages but not on others in most
viewers. When opening the file Acrobat it pops up a dialog saying “There was an
error while reading a stream”. This went away after reverting too 2.0.14. Is
this a known problem?
On Aug 6, 2020, at 8:45 PM, Tilman Hausherr <thaush...@t-online.de> wrote:
Hi,
This is related to a known problem, that the font subsetting is done when
saving. So try saving to a dummy, e.g. to a ByteArrayOutputStream.
I would like to know whether that works, or whether you have to always reload,
so that the documentation can be improved.
Tilman
Am 07.08.2020 um 00:34 schrieb Support:
I am using the org.apache.pdfbox.multipdf.Overlay class to add a watermark to an
existing PDF file. The watermark contains the name of the customer and therefore can
contain a wide range of characters. To accommodate that I am embedding the
NotoSansCJKtc-Regular.ttf font in the overlay document. (If I write the overlay
document to a file the font is embedded as expected). However, when I use
Overlay.overlayDocuments(Map<Integer,PDDocument>) the resulting file does not
embed the font.
How the resulting PDF file is displayed depends on the PDF viewer and on
whether the font happens to be installed on the system. Sometimes helpful
warnings are produced (Acrobat if the font is not installed), sometimes the
overlay text is displayed with what appears to random characters and random
font (Google Chrome) and sometimes everything is shown beautifully.
Can anyone direct me to how I can get the true type font embedded in the
PDDocument created by overlayDocuments? What part am I missing? I will gladly
provide some sample code / documents if there is obvious answer.
Thanks,
- Rainer
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscr...@pdfbox.apache.org
For additional commands, e-mail: users-h...@pdfbox.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscr...@pdfbox.apache.org
For additional commands, e-mail: users-h...@pdfbox.apache.org
