Пример #1
0
def convert_file(filename):
    doc = etree.parse(filename)
    engine = util.make_engine()
    base = {'source_file': filename, 'source_id': 0}
    for i, commitment in enumerate(doc.findall('//commitment')):
        base.update({'source_line': commitment.sourceline,
                     'source_contract_id': i})
        write = lambda r: write_row(engine, base, r)
        convert_commitment(commitment, write)
Пример #2
0
def merge():
    engine = util.make_engine()
    table = sl.get_table(engine, 'fts')
    for row in sl.distinct(engine, table, 'beneficiary', 'country_code'):
        canonical, uri, score = lookup(row.get('beneficiary'), row.get('country_code'), engine)
        row['beneficiary_canonical'] = canonical
        row['beneficiary_uri'] = uri
        row['beneficiary_score'] = score
        sl.upsert(engine, table, row, ['beneficiary', 'country'])
Пример #3
0
def merge():
    read_countries()
    engine = util.make_engine()
    table = sl.get_table(engine, 'fts')
    for row in sl.distinct(engine, table, 'country'):
        country = row.get('country')
        data = match(country)
        row['country_code'] = data.get('iso_3166-1_2')
        row['country_common'] = data.get('common')
        sl.upsert(engine, table, row, ['country'])
Пример #4
0
def merge():
    geocoder = shapegeocode.geocoder(
        'nuts2-shapefile/data/NUTS_RG_10M_2006.shp',
        filter=lambda r: r['STAT_LEVL_'] == 3)
    regions = load_region_hierarchy()
    engine = util.make_engine()
    table = sl.get_table(engine, 'fts')
    for row in sl.distinct(engine, table, *KEYS):
        loc = geocode(row)
        if loc is None:
            continue
        row.update(loc)
        reg = find_region(geocoder, regions, row)
        row.update(reg)
        log.info("Geocoded: %s/%s - %s",
            row['lat'], row['lon'], row.get('nuts3_label'))
        sl.upsert(engine, table, row, KEYS)
Пример #5
0
def merge(codes):
    engine = util.make_engine()
    table = sl.get_table(engine, 'fts')
    for level in LEVELS:
        src_col = 'budget_code' if level == 'item' else level
        for data in sl.distinct(engine, table, src_col):
            value = data[src_col]
            if level == 'item' and len(value) < 11:
                continue
            if value not in codes:
                print value
                continue
            code_data = codes.get(value)
            data['%s_name' % level] = value
            data['%s_label' % level] = code_data['label']
            data['%s_description' % level] = code_data['description']
            data['%s_legal_basis' % level] = code_data['legal_basis']
            sl.upsert(engine, table, data, [src_col])