Support MuPDF v1.18.0
This version fixes the following issues:
- #519 - method
Page.cleanContents()
should no longer destroy the PDF page's appearance. In earlier versions, this upstream bug occurred in rare cases. - #675 - unsuccessful storage allocations (e.g. for extremely large pixmaps), could occasionally lead to interpreter crashes. This should now always be prevented (fingers crossed).
- #668 - the specification of line dashes in PDF is now correctly documented.
- #669 - fixed a major cause of memory leakage in method
Document.insertPDF
.
The following new features or improvements are included:
- Text extraction method
Page.getText()
now also works for annotations:Annot.getText()
. - Text from within a rectangle can now be extracted directly via
Page.getTextbox(rect)
. This may obsolete extra scripts in many cases. - When applying redactions on PDF pages, the handling of images can now be fine-controlled via a new parameter.
- The DPI (resolution) of PNG images created from pixmaps is now automatically set from the
Pixmap.xres
andPixmap.yres
values.