Docx4j xhtml to pdf retain styles/css

by **praveen065** » Mon Mar 31, 2014 9:04 pm

Hi Jason,
I am writing Java application with below functionality using docx4j

Input: XHTML div content + .css file
Ouput:PDF or Word File

I could able to convert the xhtml to pdf successfulle but could not retain the styles.
It is working only when styles sent inline

e.g.,
Working

Code: Select all: <div style="color:red;margin-left:20px;">Text 1</div>

Not working

Code: Select all: <div class="head-red">Text 1</div>

Is there anyway to read the .css file and apply the style to the Xhtml content.
Thanks is advance.

Regards,
Praveen_J

by **jason** » Mon Mar 31, 2014 9:34 pm

Hi Praveen,

Please see xhtml-import-f28/xhtmlimporter-and-headings-t1823.html#p6235

Does that help?

cheers .. Jason

by **praveen065** » Tue Apr 01, 2014 7:55 pm

Hi Jason,
I do not want to add styles manually like

Code: Select all: tt.addAttributeTransformation("class", "Heading1");

manually as I have hundreds of styles in my css file.
Could you help me in loading the styles from a .css file.

I tried

Code: Select all: XHTMLImporter.setRunFormatting(FormattingOption.CLASS_TO_STYLE_ONLY);

as well as adding style to XhtmlNamespaceHandler.css inside docx4j-ImportXHTML-nightly-20140121.jar
Still no effect in the generated pdf.

Actual Code:

Code: Select all: // Getting html part from request InputStream is = request.getInputStream(); ByteArrayOutputStream os = new ByteArrayOutputStream(); byte[] buf = new byte[32]; int r = 0; while (r >= 0) { r = is.read(buf); if (r >= 0) os.write(buf, 0, r); } String s = new String(os.toByteArray(), "UTF-8"); String decoded = URLDecoder.decode(s, "UTF-8"); //Creating document wordMLPackage = WordprocessingMLPackage.createPackage(); NumberingDefinitionsPart ndp = new NumberingDefinitionsPart(); wordMLPackage.getMainDocumentPart().addTargetPart(ndp); ndp.unmarshalDefaultNumbering(); XHTMLImporterImpl XHTMLImporter = new XHTMLImporterImpl(wordMLPackage); XHTMLImporter.setHyperlinkStyle("Hyperlink"); wordMLPackage.getMainDocumentPart().getContent().addAll(XHTMLImporter.convert(decoded, baseURL)); //Writing to pdf FOSettings foSettings = Docx4J.createFOSettings(); foSettings.setWmlPackage(wordMLPackage); // Docx4J.toFO(foSettings, os, Docx4J.FLAG_EXPORT_PREFER_XSL); Docx4J.toFO(foSettings, response.getOutputStream(), Docx4J.FLAG_NONE); response.setContentType("application/pdf");

Regards,
Praveen_J

by **jason** » Tue Apr 01, 2014 8:30 pm

Your code contains

Code: Select all: wordMLPackage.getMainDocumentPart().getContent().addAll(XHTMLImporter.convert(decoded, baseURL));

Have you set baseURL appropriately?

by **praveen065** » Tue Apr 01, 2014 8:35 pm

S dude..

String baseURL = "C:\\Users\\Praveen\\Desktop";

by **jason** » Tue Apr 01, 2014 9:49 pm

And what does your reference to the CSS file in your XHTML look like?

If that is in order, you might need to run your code in a debugger to see what is happening.

Have you got debug level logging enabled on org.docx4j.org.xhtmlrenderer ? That ought to give you some clues...

by **praveen065** » Tue Apr 01, 2014 10:35 pm

JaSON,
While debugging, I could see the styles loaded from docx4j-3.0.1.jar\org\docx4j\openpackaging\parts\WordprocessingML\styles.xml only

My Input to request would be a XHTML code

Code: Select all: <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"> <html xmlns="http://www.w3.org/1999/xhtml"> <head> <title>HTML Generator Sample Page</title> </head> <body> <div class="test-class">Test Text</div> </body> </html>

And I want to load/pick the class styles from a separate css file.

test.css

Code: Select all: test-class{ font-style: italic }

Could you share your mobile number to my mail id "praveen_spiceboy@yahoo.co.in" so that I could call and explain you the exact scenario whats happening.

Regards,
Praveen_J

by **jason** » Tue Apr 01, 2014 11:37 pm

You have 3 options(short of modifying the source code, which of course you can also do..)

Option 1: include a link to the external css file in the xhtml in the usual way (I had assumed so far that you were doing this!)

Option 2: set up you docx with the styles you want (with suitable names), and use FormattingOption.CLASS_TO_STYLE_ONLY

Option 3: add the style to XhtmlNamespaceHandler.css (that should've worked..)

by **praveen065** » Thu Apr 03, 2014 8:43 pm

Hi Jason,
Thanks for your suggestions.
I made it work by formimg the whole html page as given below.

Code: Select all: String header1 = "<html><head><style>"; String cssStyles = FileUtils.readFileToString( new File("C:\\Workspace_Rad\\PrintDocument\\WebContent\\css\\test.css")); String header2 = "</style></head><body>"; String divPart = request.getParameter("data");// Contains the only div part to print sent through ajax String footer = "</body></html>"; //Actual code given to docx4j to convert String toConvertPDF = header1 + cssStyles + header2 + divPart + footer;

But after converting, I could see many classes are not applied successfully.
Could you explain if there are any limitations using the XTHML Importer with respect to XHTML/CSS versions.
Please refer the attachment for the details.

Regards,
Praveen_J

by **jason** » Thu Apr 03, 2014 9:40 pm

As you'd be aware, you're actually doing 2 steps:

1. XHTML to docx
2. docx to PDF

To understand where any discrepancies are creeping in, you should analyse the 2 steps separately.

At present, we don't have comprehensive documentation as to what is and isn't supported.

Docx4j xhtml to pdf retain styles/css

Docx4j xhtml to pdf retain styles/css

Re: Docx4j xhtml to pdf retain styles/css

Re: Docx4j xhtml to pdf retain styles/css

Re: Docx4j xhtml to pdf retain styles/css

Re: Docx4j xhtml to pdf retain styles/css

Re: Docx4j xhtml to pdf retain styles/css

Re: Docx4j xhtml to pdf retain styles/css

Re: Docx4j xhtml to pdf retain styles/css

Re: Docx4j xhtml to pdf retain styles/css

Re: Docx4j xhtml to pdf retain styles/css

Who is online