import-census2010

data_dir10 = '../data/2010/tx2010.pl/'

FILES10 = {
    'cols': '../data/2010/2010_PLSummaryFile_FieldNames.xlsx',
    'geoheader': data_dir10+'txgeo2010.pl',
    'seq01': data_dir10+'tx000012010.pl',
    'seq02': data_dir10+'tx000022010.pl',
}

def return_dtypes_dict(df):
    return dict(zip(df['Name'], df['dtype']))
    
####################################
## READ IN THE GEOHEADER
####################################

geo_defs10 = pd.read_excel(FILES10['cols'],sheet_name='geo')

geo_cols10 = geo_defs10['Name'].tolist()
geo_dtypes10 = return_dtypes_dict(geo_defs10)
geo_widths10 = geo_defs10['Size'].astype(int).tolist()

gh_df10 = pd.read_fwf(
    FILES10['geoheader'],
    encoding='latin1',
    header=None,
    names=geo_cols10,
    widths=geo_widths10,
    dtype=geo_dtypes10
)

#creating this slice so we can join on the human-readable geo name
gh_df10_place_hr = gh_df10[['LOGRECNO','NAME','STATE','PLACE']]
gh_df10_cnty_hr = gh_df10[['LOGRECNO','NAME','STATE','COUNTY']]


####################################
## READ IN P1
####################################

p1_defs10 = pd.read_excel(FILES10['cols'],sheet_name='p1')

p1_cols10 = p1_defs10['Name'].tolist()
p1_dtypes10 = return_dtypes_dict(p1_defs10)

p1_df10 = pd.read_csv(
    FILES10['seq01'],
    header=None,
    names=p1_cols10,
    dtype=p1_dtypes10
)


####################################
## READ IN P2
####################################

p2_defs10 = pd.read_excel(FILES10['cols'],sheet_name='p2')

p2_cols10 = p2_defs10['Name'].tolist()
p2_dtypes10 = return_dtypes_dict(p2_defs10)

p2_df10 = pd.read_csv(
    FILES10['seq02'],
    header=None,
    names=p2_cols10,
    dtype=p2_dtypes10
)