def place_queen(N, queens_positions, current_row):
if current_row == N:
solution = str(queens_positions).strip('[]')
solution = solution.replace(' ', '')
if not solution_exists(N):
global first_time
first_time = True
queen_case = QueenCase(n=N, number_of_solutions=0)
session.add(queen_case)
session.commit()
if first_time:
query = session.query(QueenCase).filter_by(n=N)
for x in query:
queen_case_id = x.id
sol_obj = Solution(queen_case_id=queen_case_id, solution=solution)
session.add(sol_obj)
session.commit()
global sol
sol += 1
else:
for column in range(0, N):
if verify(queens_positions, current_row, column):
queens_positions[current_row] = column
place_queen(N, queens_positions, current_row + 1)
def calculate(N):
global sol, first_time
sol = 0
first_time = False
queens_positions = [-1] * N
place_queen(N, queens_positions, 0)
if first_time:
queen_case = session.query(QueenCase).filter_by(n=N).first()
queen_case.number_of_solutions = sol
session.commit()
elif not solution_exists(N):
queen_case = QueenCase(n=N, number_of_solutions=sol)
session.add(queen_case)
session.commit()
return sol
def getcountry(threeletter="PAK"):
print threeletter
baseurl = "http://api.worldbank.org/datafiles/%s_Country_MetaData_en_EXCEL.xls"
value = {'dsID': 'World Bank',
'region': threeletter,
'source': baseurl % threeletter,
'is_number': True}
fh = dl.grab(baseurl % threeletter, [404])
if not fh:
return
messy = messytables.excel.XLSTableSet(fh)
table = xypath.Table.from_messy(list(messy.tables)[0])
indicators = table.filter(is_in(indicator_list))
indname = indicators.shift(x=-1)
if not len(indname) == len(indicator_list):
print "missing indicators", [x.value for x in indname]
code = table.filter(equal_to('Indicator Code'))
years = code.fill(xypath.RIGHT)
junction = indname.junction(years)
for ind_cell, year_cell, value_cell in junction:
vdict = dict(value)
vdict['indID'] = ind_cell.value
vdict['period'] = year_cell.value
vdict['value'] = value_cell.value
indicator = {'indID': vdict['indID']}
nameunits = re.search('(.*)\((.*)\)', vdict['indID'])
if nameunits:
(indicator['name'], indicator['units']) = nameunits.groups()
else:
indicator['name'] = vdict['indID']
indicator['units'] = 'uno'
Indicator(**indicator).save()
v = Value(**vdict)
if not v.is_blank():
v.save()
print len(session.query(Value).filter(Value.dsID == 'World Bank').all())
session.commit()
for x in Entity.query:
c = x.children["owner"]
while c:
c.pop()
c = x.children["user"]
while c:
c.pop()
if not x.name in data:
session.delete(x)
for name, entity in owners.items():
nodes = names.get(name, [])
if nodes:
c = entity.children["owner"]
for node in nodes:
c.append(node)
else:
print "Node for '%s' not found" % name
for name, entities in users.items():
nodes = names.get(name, [])
if nodes:
for entity in entities:
c = entity.children["user"]
for node in nodes:
c.append(node)
else:
print "Node for '%s' not found" % name
session.commit()
def create_product(cls):
name = input("Ingrese el nombre del producto: ")
price = int(input("Ingrese el precio: "))
product = Products(name, price)
session.add(product)
session.commit()
def update_user(cls):
ob = session.query(Products).get(1)
ob.nombre = "Cambio!!"
session.commit()
def delete_user(cls):
ob = session.query(Products).get(2)
session.delete(ob)
session.commit()
code = table.filter(equal_to('Indicator Code'))
years = code.fill(xypath.RIGHT)
junction = indname.junction(years)
for ind_cell, year_cell, value_cell in junction:
vdict = dict(value)
vdict['indID'] = ind_cell.value
vdict['period'] = year_cell.value
vdict['value'] = value_cell.value
indicator = {'indID': vdict['indID']}
nameunits = re.search('(.*)\((.*)\)', vdict['indID'])
if nameunits:
(indicator['name'], indicator['units']) = nameunits.groups()
else:
indicator['name'] = vdict['indID']
indicator['units'] = 'uno'
Indicator(**indicator).save()
v = Value(**vdict)
if not v.is_blank():
v.save()
print len(session.query(Value).filter(Value.dsID == 'World Bank').all())
session.commit()
for country in getcountrylist():
try:
getcountry(country)
except Exception, e:
print country, e
raise
def getindicator(ind="100106", overridefunction=None):
if not overridefunction:
baseurl = 'http://hdrstats.undp.org/en/indicators/display_cf_xls_indicator.cfm?indicator_id=%s&lang=en' % ind
html = requests.get(baseurl).content
else:
html, baseurl = overridefunction()
value = {'dsID': 'HDRStats',
'indID': "HDR:"+ind,
'source': baseurl,
'is_number': True}
dataset = {'dsID': 'HDRStats',
'last_scraped': orm.now(),
'name': 'Human Development Indicators, UNDP'}
indicator = {'indID': "HDR:"+ind}
hdi_indicator = {'indID': 'HDR:HDI Rank',
'name': 'Human Development Index rank',
'units': ''}
Indicator(**hdi_indicator).save()
DataSet(**dataset).save()
print html
exit(3)
htmlio = StringIO.StringIO(html)
messy = messytables.html.HTMLTableSet(htmlio)
table = xypath.Table.from_messy(list(messy.tables)[0])
root = lxml.html.fromstring(html)
"get odd indicator / update time"
indicator_text = root.xpath("//h2/text()")[-1]
print indicator_text
try:
indicator_split, = re.findall("(.*)\(([^\(\)]+)\)", indicator_text)
except ValueError:
indicator_split = [indicator_text, ""]
indicator['name'], indicator['units'] = indicator_split
indicator['name'] = indicator['name'].strip()
access_text, = [x.tail.strip() for x in root.xpath("//br") if str(x.tail) != "None" and x.tail.strip()]
access_date_raw, = re.findall('Accessed:(.*)from', access_text)
dataset['last_updated'] = dateutil.parser.parse(access_date_raw).isoformat()
print dataset['last_updated'], indicator['name'], "*", indicator['units']
Indicator(**indicator).save()
country_cell = table.filter("Country").assert_one()
years = country_cell.fill(xypath.RIGHT).filter(lambda b: b.value != '')
countries = country_cell.fill(xypath.DOWN)
hdi_rank = table.filter("HDI Rank").assert_one()
max_year = max(year.value for year in years)
for i in countries.junction(hdi_rank):
newvalue = dict(value)
newvalue['indID'] = "HDR:HDI Rank"
newvalue['region'] = get_region(i[0])
newvalue['value'] = i[2].value.strip()
newvalue['period'] = 2012 # TODO Hard coded for now because year it pertains to is not clear
if newvalue['value'].strip() != '..':
Value(**newvalue).save()
for i in countries.junction(years):
newvalue = dict(value)
newvalue['region'] = get_region(i[0])
newvalue['value'] = i[2].value.strip()
newvalue['period'] =i[1].value.strip()
if newvalue['value'].strip() != '..':
Value(**newvalue).save()
print newvalue
session.commit()
def getindicator(ind="100106", overridefunction=None):
if not overridefunction:
baseurl = 'http://hdrstats.undp.org/en/indicators/display_cf_xls_indicator.cfm?indicator_id=%s&lang=en' % ind
html = requests.get(baseurl).content
else:
html, baseurl = overridefunction()
value = {
'dsID': 'HDRStats',
'indID': "HDR:" + ind,
'source': baseurl,
'is_number': True
}
dataset = {
'dsID': 'HDRStats',
'last_scraped': orm.now(),
'name': 'Human Development Indicators, UNDP'
}
indicator = {'indID': "HDR:" + ind}
hdi_indicator = {
'indID': 'HDR:HDI Rank',
'name': 'Human Development Index rank',
'units': ''
}
Indicator(**hdi_indicator).save()
DataSet(**dataset).save()
print html
exit(3)
htmlio = StringIO.StringIO(html)
messy = messytables.html.HTMLTableSet(htmlio)
table = xypath.Table.from_messy(list(messy.tables)[0])
root = lxml.html.fromstring(html)
"get odd indicator / update time"
indicator_text = root.xpath("//h2/text()")[-1]
print indicator_text
try:
indicator_split, = re.findall("(.*)\(([^\(\)]+)\)", indicator_text)
except ValueError:
indicator_split = [indicator_text, ""]
indicator['name'], indicator['units'] = indicator_split
indicator['name'] = indicator['name'].strip()
access_text, = [
x.tail.strip() for x in root.xpath("//br")
if str(x.tail) != "None" and x.tail.strip()
]
access_date_raw, = re.findall('Accessed:(.*)from', access_text)
dataset['last_updated'] = dateutil.parser.parse(
access_date_raw).isoformat()
print dataset['last_updated'], indicator['name'], "*", indicator['units']
Indicator(**indicator).save()
country_cell = table.filter("Country").assert_one()
years = country_cell.fill(xypath.RIGHT).filter(lambda b: b.value != '')
countries = country_cell.fill(xypath.DOWN)
hdi_rank = table.filter("HDI Rank").assert_one()
max_year = max(year.value for year in years)
for i in countries.junction(hdi_rank):
newvalue = dict(value)
newvalue['indID'] = "HDR:HDI Rank"
newvalue['region'] = get_region(i[0])
newvalue['value'] = i[2].value.strip()
newvalue[
'period'] = 2012 # TODO Hard coded for now because year it pertains to is not clear
if newvalue['value'].strip() != '..':
Value(**newvalue).save()
for i in countries.junction(years):
newvalue = dict(value)
newvalue['region'] = get_region(i[0])
newvalue['value'] = i[2].value.strip()
newvalue['period'] = i[1].value.strip()
if newvalue['value'].strip() != '..':
Value(**newvalue).save()
print newvalue
session.commit()