This is a parser and address normalizer for the bulk data from OpenStates

A large portion of this code is devoted to taking unstructred address strings and turning them into structured data with the help of http://federalgovernmentzipcodes.us/.

(Eventually I'd like to use TIGER data to help with the parsing and normalization.)

Please see free-zipcode-database-parser for the table required.

My most current address normalization and the errors encountered can be found at http://jimkeener.com/state-legislators.tar.bz2.

Running

To run, a database user and name must be supplied. See the example config file for the format if you choose to use a config file. The command line options override the config file.

$ python3 create-csv.py -h
Usage: create-csv.py [options]

Options:
  -h, --help            show this help message and exit
  -u USER, --user=USER  User to log into the database as
  -D DB_NAME, --db=DB_NAME
                        Database to log into
  -d DB_HOST, --host=DB_HOST
                        Database host to log into (Default: localhost)
  -v VERBOSITY, --verbosity=VERBOSITY
                        How much should I complain? OFF, CRITICAL, ERROR,
                        WARNING, INFO or DEBUG (Default: DEBUG)
  -c CONFIG, --config=CONFIG
                        Location of a config file
  -o OUTFILE, --outfile=OUTFILE
                        Name of the output file (Default: ./legislators.csv)
  -l LOGFILE, --logfile=LOGFILE
                        Name of the log file (Default: -)
  -i INDIR, --indir=INDIR
                        Location of the input directory (Default:
                        ./openstates.org/legislators)