I am an english-speaking software developer who has recently added support for the rendering of right-to-left languages in the software I maintain. This involved using ICU to find visual runs of text and convert them from right-to-left to left-to-right to be rendered using Cairo.
I arbitrarily chose Hebrew and Arabic to test the new features and noticed that the final form characters were not rendered by Cairo.
For example the test script contains:
which is rendered as:
with the HEBREW LETTER FINAL NUN character missing.
Is that acceptable?