Re: Interesting PDF on stackoverflow

[Tilman Hausherr]

Maybe this could be done with the ExtractTextByArea example. However 
IIRC the coordinates are awt-like (y 0 on top) coordinates, so the PDF 
coordinates should somehow be mapped to this.

Tilman
Am 21.07.2021 um 18:21 schrieb Tim Allison:
https://stackoverflow.com/questions/68402058/tika-isnt-reading-pdf-properly

Not sure there's much we should do on the Tika side.

How hard would it be to add an "extract only text that is on the page" feature?

Best,

        Tim