pdfbox-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Olaf Drümmer <olafl...@callassoftware.com>
Subject Re: PDF/X-1a?
Date Sat, 21 Jan 2017 10:10:48 GMT
Hi all,

first: everyone should try to be clear whether they are discussing PD/A-1 (archiving PDF)
or PDF/X-1a (using/exchanging PDF as print ready file in prepress) here.

Creating PDF:
- whenever creating a PDF from scratch, one has good control over what to put inside the PDF,
and how to do it. Thus, creating both PDF/A-1 or PDF/X-1a can be achieved with reasonable
effort. E.g. if RGB is not allowed at all in PDF/X-1a, use DeviceCMYK instead - if material
to be used (an image) is in RGB, a color conversion has to be done to the image before putting
it inside PDF. 

Converting existing PDF:
- here things can be arbitrarily complex and challenging. As one has no control over how the
PDF was created, anything could be inside it. Both PDF/A-1 and PDF/X-1a disallow lots of things.
For example, both disallow transparency. Turning a PDF with transparency into a PDF that still
looks (colors?) and behaves the same (text extraction?) but does not any longer use transparency
is far from being trivial.

While PDFBox will provide quite a nice and extensive toolbox for creating, reading and modifying
PDF, a lot of effort will have to go into implementation work in addition to and on top of
PDFBox for any decent conversion from arbitrary PDF to PDF/A-1 or PDF/X-1a. Think man years
if you do it yourself, or consider throwing in additional tools.


> On 21 Jan 2017, at 08:40, Harry Yoon <harryplus@outlook.com> wrote:
> Hi Tilman, Thanks for the answer.
> My knowledge of PDF is rather limited (I've been researching only for the last few days),
but yes, I understand that I have to create a standards-compliant doc using the APIs. It seems
that the requirements for PDF/X-1a are generally easy to satisfy, maybe except for CMYK colors.
I am not sure why it is harder than other requirements, but I hear that ghostscript tool ps2pdf,
for instance, cannot currently convert a ps/pdf to a fully compliant PDF/X-1a pdf because
of this CMYK requirement. I presume that PDFBox does not have such limitations?
> Just to follow up, can PDFBox be used to convert an existing pdf to PDF/X-1a conformant
pdf (in addition to generating a PDF/X-1a conformant pdf from scratch)? If so, is there a
facility in PDFBox to make this task easier? Or, is it just a manual process going through
the original pdf (after parsing) and re-generating output pdf (enforcing the requirements)?
> Thanks,
> ~Harry
> ________________________________
> From: Tilman Hausherr <THausherr@t-online.de>
> Sent: Saturday, January 21, 2017 7:06:19 AM
> To: users@pdfbox.apache.org
> Subject: Re: PDF/X-1a?
> Am 21.01.2017 um 07:38 schrieb Tilman Hausherr:
>> Am 21.01.2017 um 00:11 schrieb Harry Yoon:
>>> Hi,
>>> I'm new to PDFBox. I'm currently trying to figure out how to generate
>>> PDF/X-1a conformant PDF using PDFBox. I found some references related
>>> to PDF/A validation using PDFBox, but none pertinent to PDF/X.
>>> Is it possible to use PDFBox to generate PDF/X-1a compliant PDF docs?
>>> Or, at least to validate PDF files against PDF/X-1a standards?
>> Yes it is possible to use PDFBox to generate PDF/A-1a compliant docs.
>> Just read the standard ( this https://en.wikipedia.org/wiki/PDF/X
>> gives a preview) and respect the requirements. No, PDFBox doesn't have
>> a validator for that.
> To clarify this -
> PDFBox is rather low level, i.e. you have to know a bit about PDF
> itself. There isn't switch in PDFBox that sets PDF/A-1a compatibility.
> To see what I mean, have a look at the CreatePDFA.java example. This
> isn't what you need, but it shows what extra work is to be done for that
> standard, e.g. output intents and XMP.
> More on PDF/A here
> https://www.pdfa.org/pdfa-faq/
> Tilman
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@pdfbox.apache.org
> For additional commands, e-mail: users-help@pdfbox.apache.org

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

View raw message