I have published two books so far using the most obscene of methods; I parse my own HTML into XeTeX using a mix of bash and Python. Parsing HTML is supposedly

I have published two books so far using the most obscene of methods; I parse my own HTML into XeTeX using a mix of bash and Python. Parsing HTML is supposedly a big no-no but what’s the problem when it’s your own code? Anyway, it seems like most comment here are trying to reinvent the wheel when any variant of LaTeX is a great solution; there are never any compromises to be made with TeX.