TSEARCH2 Thesaurus limitations

Hi,
I'm new to Postgres and would appreciate some help
in understanding what the limitations of TSEARCH2 and
the Thesauri operation.
I'm trying to use the thesaurus as a geo-tagger/coder.
The first part of the problem is to create placename
list with additional information such as state, county
and country names.  But, the returned values are off.
There's less of a problem when the thesaurus is small
under 100 rows but I'm trying to upload 7 million rows.
I have not seen the latest TSEARCH2 code release so
I don't have a great deal of understanding of the inner
workings.
Is there specific code that I can hack which will remove
a fix limitation such as the number of tokens before the
indexer quits or is the index type insufficient for the scale
of data.
Thanks,
Ted