That makes sense. If you look at KindleUnpack debug output run an azw3 (KF8) that uses an svg with image inside an xhtml file, you will see that Amazon has pulled out the svg into its own section. It always does this. This creates an artificial page break. The KindleUnpack code has to find the extracted svg and put it back inside the xhtml file when trying to unpack to epub from an azw3.
|