[
https://issues.apache.org/jira/browse/PDFBOX-4709?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17241887#comment-17241887
]
Tres Finocchiaro commented on PDFBOX-4709:
------------------------------------------
{quote}Surprisingly, this also happens in PDFBox 1.8, which uses font
operations (or so I thought) instead of vector graphics. (*update: no, it uses
graphics.drawGlyphVector()*)
{quote}
Whoops, I missed that updated note and sent our openjdk provider down a path of
studying 1.8.... I'm not sure if this information is helpful, but they've
gotten back to us with some interesting findings, quoting:
{quote}[...] the reason why the drawGlyphVector uses fill(g.getOutline(x, y))
method in
[PathGraphics.java#L650|https://github.com/openjdk/jdk/blob/461c5fc63708638c8f50aa89a298c2f45efd2a97/src/java.desktop/share/classes/sun/print/PathGraphics.java#L650]
because PDFBox sets a transform to the GlyphVector in[
PDSimpleFont.java#L349|https://github.com/apache/pdfbox/blob/41ae21bd4c3f304373d3b05f63af5325df248019/pdfbox/src/main/java/org/apache/pdfbox/pdmodel/font/PDSimpleFont.java#L349]and
it prevents calling drawString() method in [PathGraphics.java#L957
|https://github.com/openjdk/jdk/blob/461c5fc63708638c8f50aa89a298c2f45efd2a97/src/java.desktop/share/classes/sun/print/PathGraphics.java#L957]
...
The idea to make it work is to set the the glyph transform to the graphics
instead of the GlyphVector itself. Here is an example which I added to my
PDFBox 1.8.16 fork [pdfbox/commit
|https://github.com/AlexanderScherbatiy/pdfbox/commit/4877e6606d28a439bec52fdf028f439d340e2700]
I investigated it to understand what happens with a text printed by PDFBox 1.8.
As a result, the PDFBox 2.0. needs to be fixed not only by using
2d.drawGlyphVector() but also applying glyph vector transforms to graphics.
{quote}
[~tilman] is this information useful or helpful? If not, we can take this
advice downstream and try to modify the behavior there.
!4d6a9ed5-45ef-4742-a21f-0903a883e2e8.jpeg!
> PDFBox prints text poorly in comparison to Adobe, Chrome, other apps
> --------------------------------------------------------------------
>
> Key: PDFBOX-4709
> URL: https://issues.apache.org/jira/browse/PDFBOX-4709
> Project: PDFBox
> Issue Type: Bug
> Components: Rendering
> Affects Versions: 2.0.14
> Environment: Windows 10
> AdoptOpenJDK 11
> PDFBox 2.0.14
> Reporter: Lite M Finocchiaro
> Priority: Major
> Labels: Zebra
> Attachments: 4d6a9ed5-45ef-4742-a21f-0903a883e2e8.jpeg,
> Adoberesult.jpg, Drug-print.pdf, Drug-print.pdf, IMG_20191219_130048_2.jpg,
> PDFBoxVSgraphicsobj.jpg, PDFBoxresult.jpg, PrintedWithPDFBox.pdf, Screen Shot
> 2019-12-22 at 2.20.54 PM.png, Screen Shot 2019-12-22 at 2.21.00 PM.png,
> linux-thermal-test-graphics-frc-4pt.pdf, linux-thermal-test-graphics-frc.pdf,
> linux-thermal-test-pdfbox-4pt.pdf
>
>
> When printing a PDF from PDFBox to a Zebra GK420d thermal label printer, the
> text from the PDF is blurry and appears to have the top and bottom cut off
> compared to printing the same file through Adobe Acrobat.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
