Maybe you could write own rules for ImportTXT plugin? (Just an idea)
I didn't know this plugin, maybe it will provide a quite easy way to get the result.
The output format is json (a zip file containing an html file and a json file inside a folder, to be precise), and contains a lot of "rubbish". I think that it is necessary to study the json structure and how ImportTXT works too.
Maybe an easier solution would be to use the online php parser I mentioned above, but I don't think that this third passage is a good idea.
I'm not sure I've the ability to filter the json properly, maybe some other user with more skill could be interested to work on the import filter?