Calibre is pyqt isn't it? There are python ports of lucene, have been for years:
http://lucene.apache.org/pylucene/ and Qt itself uses clucene internally (or used to, back when I last used it, which admittedly is 5-6 years and two major versions ago) so it was like a freebie anyway. There's no .net dependency specific to lucene itself, it was originally written in Java and it's been ported to practically everything (both in terms of operating system and programming language).
Ebook formats are also nothing special, most of them are a bundle of xml files wrapped in compression, and there's mountains of code already in calibre for taking apart, parsing and repackaging them, so that part isn't really a big deal I wouldn't think. There's a lot of things Calibre can do that MS doesn't
Indexing pdf's is probably harder, but there's utilities for that too. This is built into some bibliographic software like Zotero, but most of them simply run a commandline pdftotext and index the results. They aren't fancy at all, but they do the job.