def ar_banks_iterator(
url_bank_list='http://www.bcra.gov.ar/Sistema_financiero/sisfin020101.asp',
url_bank_info='http://www.bcra.gob.ar/Sistema_financiero/'
'sisfin020101.asp?bco=%s',
country='Argentina'):
"""
Argentinian Banks list iterator.
>>> banks = ar_banks_iterator()
>>> banks.next().keys() == ['latitud', 'ghom', 'fax', 'code', 'office', \
'street2', 'site', 'number', 'phone', \
'street', 'address', 'active', 'gins', 'id', \
'longitud', 'city', 'name', 'zip', 'country', \
'state', 'email', 'vat']
True
"""
if BeautifulSoup is None:
return
page_list = urlopen(url_bank_list)
soup_list = BeautifulSoup(page_list)
for bank in soup_list('option'):
if 'value' in dict(bank.attrs):
id, name = bank['value'].strip(), bank.string.strip()
page_bank = urlopen(url_bank_info % id)
soup_bank = BeautifulSoup(page_bank)
data = {
'id': id,
'bcra_code': id,
'name': name,
'country': country,
'active': '1',
'number': '',
}
for line in soup_bank(id='texto_columna_2')[0]('table')[1]('tr'):
sline = line.td.renderContents()
for key in compiled_re.keys():
search = compiled_re[key].search(sline)
if search:
data[key] = unicode(search.group(1).strip(), encoding)
searchaddress = u"%(street)s, %(city)s, %(state)s, %(country)s" \
% data
geodata = unify_geo_data(strip_accents(searchaddress))
if geodata:
data.update(geodata)
for key in postprocessor_keys.keys():
if key in data:
data[key] = postprocessor_keys[key](key, data)
data[key] = data[key]
else:
_logger.warning("No geoposition %s." % (searchaddress))
yield data
return
def ar_banks_iterator(
url_bank_list='http://www.bcra.gov.ar/sisfin/sf010100.asp',
url_bank_info='http://www.bcra.gov.ar/sisfin/sf010100.asp?bco=%s',
country='Argentina'):
"""
Argentinian Banks list iterator.
>>> banks = ar_banks_iterator()
>>> banks.next().keys() == ['latitud', 'ghom', 'fax', 'code', 'office', \
'street2', 'site', 'number', 'phone', \
'street', 'address', 'active', 'gins', 'id', \
'longitud', 'city', 'name', 'zip', 'country', \
'state', 'email', 'vat']
True
"""
page_list = urlopen(url_bank_list)
soup_list = BeautifulSoup(page_list)
for bank in soup_list('option'):
if 'value' in dict(bank.attrs):
id, name = bank['value'].strip(), bank.string.strip()
page_bank = urlopen(url_bank_info % id)
soup_bank = BeautifulSoup(page_bank)
data = {
'id': id,
'name': name,
'country': country,
'active': '1',
}
for line in soup_bank('div')[5]('table')[0]('tr'):
sline = line.td.renderContents()
for key in compiled_re.keys():
search = compiled_re[key].search(sline)
if search:
data[key] = unicode(search.group(1).strip(), encoding)
searchaddress = u"%(street)s, %(city)s, %(state)s, %(country)s" % data
geodata = unify_geo_data(strip_accents(searchaddress))
data.update(geodata)
for key in postprocessor_keys.keys():
if key in data:
data[key] = postprocessor_keys[key](key, data)
data[key] = data[key].encode('utf-8')
yield data
def search_zip(street, number, city, state, country, unique=True):
"""
Return zipcode. Only works with argentina.
>>> search_zip("rivadavia", "9800", "buenos aires",
"capital federal", "argentina")
u'C1407DZT'
>>> search_zip("jose clemente paz", "1200",
"jose clemente paz", "buenos aires", "argentina")
u'B1665BBB'
>>> search_zip("general paz", "5445",
"general san martin", "buenos aires", "argentina")
u'1650'
"""
street = strip_accents(street.lower())
city = strip_accents(city.lower())
state = strip_accents(state.lower())
country = strip_accents(country.lower())
codpos = None
if country in ['argentina', 'ar']:
# Elimina titulos de calles y avidas
street = re.sub('\s*av\s+', '', street)
re_cpa = re.compile('>(\w{8})<')
if state in ['capital federal']:
codloca = ['5001']
codpos = ['']
else:
url = "http://www3.correoargentino.com.ar/scriptsN/cpa/' \
'cpa_loca.idc?codprov=%s&pnl=%s"
inpage = urlopen(url % (codprov_dict[country][state], _st(city)))
soup = BeautifulSoup(inpage)
options = soup.findAll('option')
if len(options) == 0:
raise RuntimeError(
'No locations for "%s"' %
','.join([street, number, city, state, country]))
loca = map(
lambda opt: re.search('(.*)\s*\(\d+\)', opt.string.lower()).
groups()[0].strip(), options)
codloca = map(lambda opt: opt['value'], options)
codpos = map(
lambda opt: re.search('\((\d+)\)', opt.string).groups()[0],
options)
for i in xrange(len(codloca)):
url = 'http://www3.correoargentino.com.ar/scriptsN/cpa/' \
'cpa_calle.idc?codloca=%s&pnc=%s&alt=%s'
inpage = urlopen(url % (codloca[i], _st(street), number))
soup = BeautifulSoup(inpage)
output = soup.body.div.table.tr.td.renderContents()
match = re_cpa.search(output)
if match:
codpos[i] = match.group(1)
if len(codloca) > 1 and unique:
i = mostequivalent(loca, city)
return codpos[i]
else:
return unicode(codpos[0])
else:
raise NotImplementedError
def search_zip(street, number, city, state, country, unique=True):
"""
Return zipcode. Only works with argentina.
>>> search_zip("rivadavia", "9800", "buenos aires",
"capital federal", "argentina")
u'C1407DZT'
>>> search_zip("jose clemente paz", "1200",
"jose clemente paz", "buenos aires", "argentina")
u'B1665BBB'
>>> search_zip("general paz", "5445",
"general san martin", "buenos aires", "argentina")
u'1650'
"""
street = strip_accents(street.lower())
city = strip_accents(city.lower())
state = strip_accents(state.lower())
country = strip_accents(country.lower())
codpos = None
if country in ['argentina', 'ar']:
# Elimina titulos de calles y avidas
street = re.sub('\s*av\s+', '', street)
re_cpa = re.compile('>(\w{8})<')
if state in ['capital federal']:
codloca = ['5001']
codpos = ['']
else:
url = "http://www3.correoargentino.com.ar/scriptsN/cpa/' \
'cpa_loca.idc?codprov=%s&pnl=%s"
inpage = urlopen(url % (codprov_dict[country][state], _st(city)))
soup = BeautifulSoup(inpage)
options = soup.findAll('option')
if len(options) == 0:
raise RuntimeError('No locations for "%s"' % ','.join(
[street, number, city, state, country]))
loca = map(
lambda opt: re.search('(.*)\s*\(\d+\)',
opt.string.lower()).groups()[0].strip(),
options)
codloca = map(
lambda opt: opt['value'],
options)
codpos = map(
lambda opt: re.search('\((\d+)\)',
opt.string).groups()[0],
options)
for i in xrange(len(codloca)):
url = 'http://www3.correoargentino.com.ar/scriptsN/cpa/' \
'cpa_calle.idc?codloca=%s&pnc=%s&alt=%s'
inpage = urlopen(url % (codloca[i], _st(street), number))
soup = BeautifulSoup(inpage)
output = soup.body.div.table.tr.td.renderContents()
match = re_cpa.search(output)
if match:
codpos[i] = match.group(1)
if len(codloca) > 1 and unique:
i = mostequivalent(loca, city)
return codpos[i]
else:
return unicode(codpos[0])
else:
raise NotImplementedError