disable print causes nullpointer

[mahmutcanprehcm]

i have an html page which can be dynamically grow. i want to get one pdf page.

This code causes NullPointerException:

         ITextRenderer iTextRenderer = new ITextRenderer();
         iTextRenderer.setDocumentFromString(readHtmlFileAsString("foo.html"));
         SharedContext sharedContext = iTextRenderer.getSharedContext();
         sharedContext.setPrint(false);
         iTextRenderer.layout();

ITextRendere.java:

        public void createPDF(OutputStream os, boolean finish, int initialPageNo) throws DocumentException {
        List<PageBox> pages = _root.getLayer().getPages();
        RenderingContext c = newRenderingContext();
        c.setInitialPageNo(initialPageNo);
        PageBox firstPage = pages.get(0);
        ....

it this a bug, or wrong usage?

[asolntsev]

@mahmutcanprehcm Please show the NullPointerException itself with full stacktrace.

[mahmutcanprehcm]

java.lang.NullPointerException
	at org.xhtmlrenderer.layout.Layer.layoutPages(Layer.java:1066)
	at org.xhtmlrenderer.pdf.ITextRenderer.layout(ITextRenderer.java:232)
	at de.printer.service.pdf.DefaultTemplateBuilderService.xhtmlToPdf(DefaultTemplateBuilderService.java:72)
	at de.printer.service.pdf.DefaultTemplateBuilderService.dtoToByte(DefaultTemplateBuilderService.java:33)

[mahmutcanprehcm]

#Created by Apache Maven 3.3.9
version=9.1.11
groupId=org.xhtmlrenderer
artifactId=flying-saucer-core

[asolntsev]

@mahmutcanprehcm But version 9.1.11 is very old. Please try with the latest version 9.9.1.

[mahmutcanprehcm]

the same with 9.9.1

java.lang.IndexOutOfBoundsException: Index 0 out of bounds for length 0
	at java.base/jdk.internal.util.Preconditions.outOfBounds(Preconditions.java:100)
	at java.base/jdk.internal.util.Preconditions.outOfBoundsCheckIndex(Preconditions.java:106)
	at java.base/jdk.internal.util.Preconditions.checkIndex(Preconditions.java:302)
	at java.base/java.util.Objects.checkIndex(Objects.java:385)
	at java.base/java.util.ArrayList.get(ArrayList.java:427)
	at org.xhtmlrenderer.pdf.ITextRenderer.createPDF(ITextRenderer.java:372)
	at org.xhtmlrenderer.pdf.ITextRenderer.createPDF(ITextRenderer.java:328)
	at com.solutionfactory.HtmlToPdfOpenPDF.main(HtmlToPdfOpenPDF.java:32)

"The problem occurs when you call setPrint(false) on SharedContext."

ITextRenderer.java
Line 372

  RenderingContext c = newRenderingContext();
        c.setInitialPageNo(initialPageNo);
        PageBox firstPage = pages.get(0);
        com.lowagie.text.Rectangle firstPageSize = new com.lowagie.text.Rectangle(0, 0, firstPage.getWidth(c) / _dotsPerPoint,
                firstPage.getHeight(c) / _dotsPerPoint);

[asolntsev]

@mahmutcanprehcm I would say, it's rather a wrong usage.

Why do you call setPrint(false)?

When you convert a web page to PDF, it's usually intended for printing. It means setPrint(true) would be reasonable, and it is the default value.

[mahmutcanprehcm]

I need a single page. It is an invoice that goes to a thermal printer and if the customer wants to download a PDF, he should not receive several pages.

[asolntsev]

Ok, but how setPrint(false) is related to single/multiple pages?
It dictates "media: print" or "media: screen". It doesn't affect number of pages.

[mahmutcanprehcm]

OK. Unfortunately I couldn't find any documentation about this method.

How can I then ensure that everything only ends up on one page?

[pbrant]

I'm not sure I entirely understand your use case, but one way to effectively disable pagination would be to set a page height of e.g. 500 cm.

…

On Wed, Sep 4, 2024, 10:11 mahmutcanprehcm ***@***.***> wrote: OK. Unfortunately I couldn't find any documentation about this method. How can I then ensure that everything only ends up on one page? — Reply to this email directly, view it on GitHub <#375 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAAF42F4XTEURSDIWJE2ZKTZU26DBAVCNFSM6AAAAABNMN4NTGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGMRYGIYDANZZGQ> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[mahmutcanprehcm]

this produces always a large pdf file

[pbrant]

(Urk, sorry the late response here.)

Thanks, that makes sense. The behavior you're looking for (PDF output on a single page with page extents trimmed to the content extent) isn't something that's come up before. FS doesn't support it directly.

One fairly low effort way you could hack around this would be to use ITextRenderer#findPagePositionsByID(). It takes a regex that returns the content area of block-level elements with an id attribute set that match the pattern. With an empty <div id="content-end"></div> element you could use that to figure out where the content stops on the huge page. You could then use that to render the receipt a second time with the corrected page size.

This would have the negative that you'd be laying out the receipt twice (but only rendering the PDF once). FS is fast enough though that that may not matter.

If that is a problem, I can't think of an alternative beyond doing surgery on ITextRenderer itself to provide the option to adjust page sizes post layout.

Edit: Another option would be to use iText or PDFBox to update the crop box of the page after the fact (vs. doing a second layout). I'm not entirely sure which would be faster though.