Unfortunately, we had to stop working on this project.
We actually got to a phase where we could modify the metadata of an uploaded ePUB, but, once we turned it back in eBook format, there was no way to run it.
The reason is that when a file is uploaded on HTTP protocol, the original file has to be copied in order to be sent. If you copy a ePUB, you can't run it for copyright reasons and protections on the ePUB algorithm.
It was possible, however, to generate a new ePUB using the information of the uploaded file. Doing this would make possible to run the new ePUB whilst still mantaining the original file on disk. The problem now lies that it would be possible for people to have an online ePUB copier, and so, piracy would be configured.
Piracy was always a concern when this project was concieved. We thought that, since we would be only modifying the book - not generating a new one - that would be no issues. We haven't taken in consideration that, if we wanted to approach as a online ePUB metadata editor, it would be necessary to generate a copy.
It's sad. But at least it was fun =).