I have created a couple of tools:
- osmreader:it can be used from the command line. It can read big .osm.bz2 files (I am testing it with spain.osm.bz2). It can be used as library as well. Making very easy to read the information from those files. It is actually parsing XML (libxml2 via lxml).
- osm2hdf5: it converts .osm.bz2 to an .h5 file. It takes around 25min to convert spain. After that it is crazy fast to access the information. I am using it to create some validation tools.
I have just posted the information here:
They might be useful for some of you.