Docx to PDF conversion: PDF/A support

[rama]

Hi All,

For Docx to PDF conversion XDocReport uses either iText or Apache FOP. Both of these underlying frameworks offer the possibility to set PDF/A Conformance level but unfortunately i could not

find any way to set PDF/A Conformance level using XDocReport. 

Is it possible to set PDF/A Conformance level using XDocReport or not? This is the decision point in my project whether to use XDocReport or not.

Please help me :)

[Angelo zerr]

Hi rama,

At first PDF converter with Apache FOP is just a POC for XDocReport, we have gaven this implementation(FOP uses a lof of Memory and it's not performant because it requires FO and code is awfull because it is based on XSLT).

I don't know "PDF/A Conformance level", but if you know how to set this configuration with iText we could modify the converter.

Regards Angelo

2013/11/7 rama <inform...@gmail.com>

--
You received this message because you are subscribed to the Google Groups "xdocreport" group.
To unsubscribe from this group and stop receiving emails from it, send an email to xdocreport+...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[rama]

Hi Angelo,

see PdfWriter.setPDFXConformance(PdfWriter.PDFA1A/PdfWriter.PDFA1B)

http://www.coderanch.com/how-to/javadoc/itext-2.1.7/com/lowagie/text/pdf/PdfWriter.html

Thanks.

[Angelo zerr]

Hi rama,

Many thank's for your explanation.

I have commited on git https://code.google.com/p/xdocreport/source/detail?r=c96c3a9178bc9e5ab387569db466854e57b87fef

Now you have IPdfWriterConfiguration#configure( PdfWriter writer ) interface that you can implement to customize the PdfWriter as you wish.

Here a sample to use it : 

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------

        org.apache.poi.xwpf.converter.pdf.PdfOptions options = new PdfOptions();

        options.setConfiguration( new IPdfWriterConfiguration()

        {

            

            @Override

            public void configure( PdfWriter writer )

            {

                writer.setPDFXConformance( PdfWriter.PDFA1A );

            }

        } );

        PdfConverter.getInstance().convert( document, out, options );

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Tell me if this idea please you, after that I will update the wiki with this new feature.

Regards Angelo

2013/11/7 rama <inform...@gmail.com>

Hi Angelo,

--

[rama]

Thanks. I dont use Converter directly but through IXDocReport.  For me it is important to utilize the Templating engine during the conversion process and from examples i can see only way to use templating engine through IXDocReport. But IXDocReport.convert method uses fr.opensagres.xdocreport.converter.Options as parameter. This Options class is different from PdfOptions or its parent class. Is it possible to provide such a configuration there? Or tell me how can i use templating engine using Converters directly?

On Thursday, November 7, 2013 11:57:32 AM UTC+1, rama wrote:

[Angelo zerr]

If you wish to use org.apache.poi.xwpf.converter.pdf.PdfOptions, you must use fr.opensagres.xdocreport.converter.Options#subOptions like this : 

 org.apache.poi.xwpf.converter.pdf.PdfOptions pdfOptions = ...

Options.getTo(ConverterTypeTo.PDF).via(ConverterTypeVia.XWPF).subOptions(pdfOptions);

IXDocReport report = ...

report.convert(context, options, out);

Regards Angelo

2013/11/8 rama <inform...@gmail.com>

--

[rama]

Perfect. Until when can i expect this fix. I would test this with a snapshot version. Is there any possibility to use maven dependency to a specific snapshot version?

On Thursday, November 7, 2013 11:57:32 AM UTC+1, rama wrote:

[Angelo zerr]

Yes sure, please read https://code.google.com/p/xdocreport/wiki/Download#Maven_repository_artifacts

2013/11/8 rama <inform...@gmail.com>

--

[rama]

Thanks, actually i have to use XDocReporrt for docx --> pdf & xhtml conversion. I know Docx4J is slower & memory hungry than others but on my test system Docx4J gives better results in terms of formatting. 

In iText conversion to pdf: one page docx was converted to 2 pages pdf. tabs were not correctly formatted. footer seems correct, haven't tried with header yet! Is there any limitation on using the image in Header section???

I found your page https://code.google.com/p/xdocreport/wiki/XWPFConverterPDFViaIText very helpful.

In Docx4J conversion to pdf: pdf seems much better but only image was misplaced. dont know why???

In DocX4J conversion to xhtml: result is fine BUT many HTML closing tags were missing. Is there any option to do HTML cleaning during conversion process?

Do you know where can i find the Docx4J formatting limitations (pdf & xhtml)?

On Thursday, November 7, 2013 11:57:32 AM UTC+1, rama wrote:

[Angelo zerr]

> but on my test system Docx4J gives better results in terms of formatting. 

Please create several issues with simple docx in order to we try to improve our iText converter.

For docx4j question, please post your question on docx4j forum.

Regards Angelo

2013/11/8 rama <inform...@gmail.com>

Thanks, actually i have to use XDocReporrt for docx --> pdf & xhtml conversion. I know Docx4J is slower & memory hungry than others but on my test system Docx4J gives better results in terms of formatting. 

--