[
https://issues.apache.org/jira/browse/PDFBOX-66?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Jukka Zitting resolved PDFBOX-66.
---------------------------------
Resolution: Duplicate
> The patch in PDFBOX-374 basically does the same thing, but moves the logic to
> TextPosition.
OK, thanks. I'm resolving this issue as a duplicate of PDFBOX-374.
> TextPosition.width incorrect if text is rotated
> -----------------------------------------------
>
> Key: PDFBOX-66
> URL: https://issues.apache.org/jira/browse/PDFBOX-66
> Project: PDFBox
> Issue Type: Bug
>
> [imported from SourceForge]
> http://sourceforge.net/tracker/index.php?group_id=78314&atid=552832&aid=1228171
> Originally submitted by yves_bussard on 2005-06-27 04:38.
> Hi Ben,
> I noticed that if a text is rotated, then it's widht is
> not calculated correctly. This is because you only look
> at the page rotation rather than the transformation
> matrix. I.e. the page may not have a rotation but the
> text itself may be rotated (e.g. because the text
> transformation matrix contains a rotation).
> You do this just before calling showCharacter() at the
> end of the method showString(). I fixed this in my
> version by using the following method to retrieve the
> rotation:
> /**
> * Returns the current absolute rotation of the text.
> * @param ctm the current transformation matrix.
> * @param textTransformationMatrix the transformation
> matrix
> * of the text.
> * @return the current absolute rotation of the text. The
> * returned value is one of 0, 90, 180, 270
> */
> protected int getTextRotation (
> Matrix ctm, Matrix textTransformationMatrix
> ) {
> // compute the rotation
> Matrix m=(Matrix) textTransformationMatrix.clone();
> m.multiply(ctm);
> float a= m.getValue(0,0);
> float b= m.getValue(0,1);
> float c= m.getValue(1,0);
> float d= m.getValue(1,1);
> int rotation=0;
> if (Math.abs(a * d) > Math.abs(b * c)) {
> rotation= (d>0) ? 0 : 180;
> } else {
> rotation= (c<0) ? 90 : 270;
> }
> return rotation;
> }
> I'm using the rotation returned by this method instead
> of the pageRotation to calculate the
> totalStringDisplacement. Also, you may want to add the
> handling of a 180 rotation.
> I don't know if somehow we have to take the
> pageRotation in consideration or whether this is
> already included in the CTM.
> cheers
> Yves
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
