Hello Tilman,
Yes, I used the rotation magic as you advised and captured the angle of
rotation and applied the correct linear algebra formula;
Where I went astray was in the fact that I used the incorrect value from the
TextPosition. Having read a great many StackOverflow articles, I realized I
should use TextPosition.getTextMatrix().getTranslateX() and the corresponding
value for the y to return the coordinates of the start of the string in the
unrotated space.
A classic error. Hopefully posting this means that others don't fall into to
the same trap.
Have a great weekend.
On Wednesday, June 3, 2020, 7:31:54 PM GMT+1, Tilman Hausherr
<thaush...@t-online.de> wrote:
Hi,
The ExtractText tool (in the tools subproject) has a "magicRotation"
option, did you try that one?
Tilman
Am 03.06.2020 um 15:14 schrieb PDF Developer:
> Hello,
> I've experimented with the Extract Text Sample and have been successful in
> extracting text. As a requirement, I only extract text that is in a certain
> named separation and at (integer) rotation angle values.
>
> Now there is a requirement to get the start (x,y) position of the each of the
> rotated text strings. I tried capturing the TextPostion of the start of the
> line during the extraction process (writeString) and then applying a "reverse
> rotation" using the geometric x'=xcosθ-ysinθ etc. formula" but that didn't
> appear to give the correct results, as I am seeing negative coordinate
> values. I suspect that there is a matrix transformation that I am missing.
>
> Now I may have misunderstood the process here, so if someone could point me
> in the right direction then I would appreciate that.
>
> Thanks in advance.
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscr...@pdfbox.apache.org
For additional commands, e-mail: users-h...@pdfbox.apache.org
