def test_create_ads(self):
tag1 = Tag.objects.create(name="HTML 5", slug="html5")
tag2 = Tag.objects.create(name="jQuery", slug="jquery")
tag3 = Tag.objects.create(name="Python", slug="python")
self.assertEquals(0, Ad.objects.count())
ad1 = Ad()
ad1.company_name = "ACME International"
ad1.company_url = "http://acme.com/"
ad1.title = "Ninja Top Developer"
ad1.description = """
We are growing, we need more developers.
Apply if you are cool and sexy.
"""
ad1.apply_email = "*****@*****.**"
ad1.full_clean()
ad1.save()
self.assertEquals(1, Ad.objects.count())
ad1.tags.add(tag1)
self.assertEquals(1, tag1.ads.count())
tag2.ads.add(ad1)
self.assertEquals(2, ad1.tags.count())
self.assertEquals(0, tag3.ads.count())
def setUp(self): newspaper1 = Newspaper(name='Test Newspaper 1') newspaper1.save() ad = Ad() ad.name = 'Name 1' ad.content = 'Content 1' ad.save()
def setUp(self):
newspaper1 = Newspaper(name='Test Newspaper 1')
newspaper1.save()
ad = Ad()
ad.name = 'Name 1'
ad.content = 'Content 1'
ad.save()
def post(self, request, pk):
print("Add PK", pk)
t = get_object_or_404(Ad, id=pk)
ad = Ad(user=request.user, ad=t)
try:
ad.save() # In case of duplicate key
except IntegrityError as e:
pass
return HttpResponse()
def setUp(self): ad1 = Ad() ad1.name = 'Test Ad 1' ad1.content = 'Content 1' ad1.save() ad2 = Ad() ad2.name = 'Test Ad 2' ad2.content = 'Content 2' ad2.save()
def setUp(self):
ad1 = Ad()
ad1.name = 'Test Ad 1'
ad1.content = 'Content 1'
ad1.save()
ad2 = Ad()
ad2.name = 'Test Ad 2'
ad2.content = 'Content 2'
ad2.save()
def setUp(self): newspaper1 = Newspaper(name='Test Newspaper 1') newspaper1.save() newspaper2 = Newspaper(name='Test Newspaper 2') newspaper2.save() ad = Ad() ad.name = 'Name 1' ad.content = 'Content 1' ad.save() ad.newspapers.add(newspaper1, newspaper2) ad.save()
def setUp(self):
newspaper1 = Newspaper(name='Test Newspaper 1')
newspaper1.save()
newspaper2 = Newspaper(name='Test Newspaper 2')
newspaper2.save()
ad = Ad()
ad.name = 'Name 1'
ad.content = 'Content 1'
ad.save()
ad.newspapers.add(newspaper1, newspaper2)
ad.save()
def setUp(self): newspaper1 = Newspaper() newspaper1.name = 'Test Newspaper 1' newspaper1.save() newspaper2 = Newspaper() newspaper2.name = 'Test Newspaper 2' newspaper2.save() newspaper3 = Newspaper() newspaper3.name = 'Test Newspaper 3' newspaper3.save() ad1 = Ad() ad1.name = 'Test Ad 1' ad1.id = 1 ad1.content = 'Content 1' ad1.save() ad2 = Ad() ad2.name = 'Test Ad 2' ad2.id = 2 ad2.content = 'Content 2' ad2.save() ad2.newspapers.add(newspaper1, newspaper2) ad2.save() ad3 = Ad() ad2.id = 3 ad3.name = 'Test Ad 3' ad3.content = 'Content 3' ad3.save() self.browser = webdriver.Firefox() self.browser.implicitly_wait(3)
def setUp(self):
newspaper1 = Newspaper()
newspaper1.name = 'Test Newspaper 1'
newspaper1.save()
newspaper2 = Newspaper()
newspaper2.name = 'Test Newspaper 2'
newspaper2.save()
newspaper3 = Newspaper()
newspaper3.name = 'Test Newspaper 3'
newspaper3.save()
ad1 = Ad()
ad1.name = 'Test Ad 1'
ad1.id = 1
ad1.content = 'Content 1'
ad1.save()
ad2 = Ad()
ad2.name = 'Test Ad 2'
ad2.id = 2
ad2.content = 'Content 2'
ad2.save()
ad2.newspapers.add(newspaper1, newspaper2)
ad2.save()
ad3 = Ad()
ad2.id = 3
ad3.name = 'Test Ad 3'
ad3.content = 'Content 3'
ad3.save()
self.browser = webdriver.Firefox()
self.browser.implicitly_wait(3)
def setUp(self):
newspaper1 = Newspaper(name='Test Newspaper 1')
newspaper1.save()
newspaper2 = Newspaper(name='Test Newspaper 2')
newspaper2.save()
newspaper3 = Newspaper(name='Test Newspaper 3')
newspaper3.save()
first_ad = Ad()
first_ad.name = 'Name 1'
first_ad.content = 'Content 1'
first_ad.save()
first_ad.newspapers.add(newspaper1, newspaper2)
first_ad.save()
second_ad = Ad()
second_ad.name = 'Name 2'
second_ad.content = 'Content 2'
second_ad.save()
second_ad.newspapers.add(newspaper2, newspaper3)
first_ad.save()
def setUp(self): newspaper1 = Newspaper(name='Test Newspaper 1') newspaper1.save() newspaper2 = Newspaper(name='Test Newspaper 2') newspaper2.save() newspaper3 = Newspaper(name='Test Newspaper 3') newspaper3.save() first_ad = Ad() first_ad.name = 'Name 1' first_ad.content = 'Content 1' first_ad.save() first_ad.newspapers.add(newspaper1, newspaper2) first_ad.save() second_ad = Ad() second_ad.name = 'Name 2' second_ad.content = 'Content 2' second_ad.save() second_ad.newspapers.add(newspaper2, newspaper3) first_ad.save()
def setUp(self): self.browser = webdriver.Firefox() self.browser.implicitly_wait(3) ad1 = Ad() ad1.name = 'Test Ad 1' ad1.content = 'Content 1' ad1.save() ad2 = Ad() ad2.name = 'Test Ad 2' ad2.content = 'Content 2' ad2.save() ad3 = Ad() ad3.name = 'Test Ad 3' ad3.content = 'Content 3' ad3.save()
def setUp(self):
self.browser = webdriver.Firefox()
self.browser.implicitly_wait(3)
ad1 = Ad()
ad1.name = 'Test Ad 1'
ad1.content = 'Content 1'
ad1.save()
ad2 = Ad()
ad2.name = 'Test Ad 2'
ad2.content = 'Content 2'
ad2.save()
ad3 = Ad()
ad3.name = 'Test Ad 3'
ad3.content = 'Content 3'
ad3.save()
def extract_data(ad_urls,uuid):
result = []
attribute_dict = { 'Date Listed' : 'date' ,
'Price' : 'price' ,
'Address' : 'address',
'Bathrooms (#)' : 'bathrooms' ,
'Furnished' : 'furnished' ,
'Pet Friendly' : 'pet_friendly'
}
for ad_url in ad_urls :
try:
items = dict() #empty dict used to create the db object
items['url'] = ad_url
#get ad data
ad_soup = bs(opener.open(ad_url).read())
map_link = ''
#title
title = ad_soup.find('h1',id='preview-local-title').getText().replace('google_ad_section_start','').replace('google_ad_section_end','')
items['title'] = title
#table data
for tr in ad_soup.find('table' , id='attributeTable').findAll('tr') :
for td in tr.findAll('td') :
key = td.getText()
if key in attribute_dict :
value = td.findNext('td').getText()
if attribute_dict.get(key) == 'date' :
items['pub_date'] = datetime.datetime.strptime(value,"%d-%b-%y")
if attribute_dict.get(key) == 'bathrooms' :
items['bathrooms'] = float(re.match('(\d+.?\d?) bathroom.*',value).group(1))
if attribute_dict.get(key) == 'price' :
if value.find('contact') > 0 :
items['rent'] = 0
else :
items['rent'] = int(float(value[1:].replace(',','')))
if attribute_dict.get(key) == 'address' :
items['address'] = value.replace('View map','')
if attribute_dict.get(key) == 'furnished' or attribute_dict.get(key) == 'pet_friendly' :
if value == 'No':
items[attribute_dict.get(key)] = False
else:
items[attribute_dict.get(key)]=True
# map coordinates
map_url = 'http://montreal.kijiji.ca' + ad_soup.find('a', attrs = { 'class' : 'viewmap-link' } ).get('href')
map_soup = bs(opener.open(map_url).read())
for noscript in map_soup.findAll('noscript') :
if noscript.find('img') :
map_link = noscript.find('img').get('src')
coords = urllib2.urlparse.parse_qs(urllib2.urlparse.urlparse(map_link).query)
lat_lng = coords['center'][0].split(',')
items['lat'] = float(lat_lng[0])
items['lng'] = float(lat_lng[1])
print items
ad = Ad(**items)
ad.save()
result.append(ad)
except Exception as e:
print "[FAILED]" , ad_url
print e
pass #skip to the next one
cache_obj = cache.get(uuid)
new_obj = { 'total' : cache_obj['total'] , 'found' : cache_obj['found'] + 1 }
cache.set(uuid,new_obj)
return result
def import_ads(file_id):
imported_file = ImportFile.objects.get(id=file_id)
clear_prev_ads(imported_file)
with PersistentDict(imported_file.settings.path, 'w', format='json') as params:
#with closing(shelve.open(file_path)) as params:
ids = []
count = len(params['lines'])
i = 0
for line in params['lines']:
ad = {}
for field in params['matching']:
if params['matching'][field] != 'Constant' and 'Constant' not in params['matching'][field]:
poss = params['matching'][field] if type(params['matching'][field]) is list else [params['matching'][field]]
value = ""
for pos in poss:
pos = int(pos)
if len(line) < pos or not line[pos].strip():
continue
value += ", " + line[pos] if value else line[pos]
if not value:
continue
if field in CompareFieldsForm.compared_fields:
curr_comp_field = CompareFieldsForm.compared_fields[field]
compared_obj = curr_comp_field['model'].objects.get(pk=value)
if field == 'category':
if compared_obj.sub_category is not None:
ad['category'] = compared_obj.sub_category.category
ad['sub_category'] = compared_obj.sub_category
else:
ad['category'] = compared_obj.category
elif field == 'district':
if compared_obj.district:
ad['district'] = compared_obj.district
ad['town'] = compared_obj.district.town
else:
ad['town'] = compared_obj.town
else:
ad[field] = compared_obj.value
elif field == 'pub_date':
pub_date = dateutil.parser.parse(value)
if str(pub_date.time()) == "00:00:00":
pub_date = datetime.combine(pub_date.date(), datetime.now().time())
ad[field] = ad['order_date'] = pub_date
else:
ad[field] = Ad.filter_field(value, field)
elif 'constants' in params and field in params['constants']: # constant
value = Ad.filter_field(str(params['constants'][field]), field)
if field in CompareFieldsForm.compared_fields:
if field == 'category':
cat_values = value.split('-')
if len(cat_values) > 1:
sub_cat = SubCategory.objects.get(pk=cat_values[1])
ad['category'] = sub_cat.category
ad['sub_category'] = sub_cat
else:
cat = Category.objects.get(pk=cat_values[0])
ad['category'] = cat
elif 'foreign' in CompareFieldsForm.compared_fields[field]:
ad[field] = CompareFieldsForm.compared_fields[field]['foreign'].objects.get(pk=value)
else:
for key, val in CompareFieldsForm.compared_fields[field]['choices']:
if str(key) == value:
ad[field] = val
else:
ad[field] = value
empty = set(MatchFieldsForm.required) - set(ad.keys())
if not empty:
ad_obj = Ad(**ad)
ad_obj.imported = imported_file
ad_obj.save()
find_similar.delay(ad_obj.id)
if ad_obj.url:
parsed_uri = urlparse(ad_obj.url)
if parsed_uri.netloc in Crawler.DOMAINS:
import_attachments.delay(ad_obj.id)
ids.append(ad_obj.id)
i += 1
celery.current_task.update_state(state='PROGRESS', meta={'current': i, 'total': count})
params['ids'] = ids
return ids
def get_ads(cls):
errors = 5
match = {'cena': 'price', 'datap': 'pub_date', 'dom_etag': 'floor_max', 'etag': 'floor',
'kol_komn': 'rooms_count', 'pl1': 'area_living', 'plk': 'area_kitchen', 'plosh': 'area',
'pl_land': 'area_land', 'textob': 'desc'}
now = datetime.now().strftime("%Y-%m-%d")
yesterday = (date.today() - timedelta(1)).strftime("%Y-%m-%d")
data = {
're_base_name': 'kvart',
're_base_section': 'green',
'rem': 0,
'subq_start': 0,
'subq_lines': 200,
'order_by[data]': 'desc',
'order_by[datap]': 'desc',
'data_start': yesterday,
'data_end': now,
}
statistic = {}
log = ""
try:
for table in ['arenda', 'kvart', 'komn', 'domm', 'negil', 'client']: # all tales
data['re_base_name'] = table
print('Load from ' + table)
statistic[table] = {}
for private_type in ['green']: #['green', 'red']: # all bases
statistic[table][private_type] = 0
print('table type ' + private_type)
data['re_base_section'] = private_type
current = 0
total = 200
while current < total: # all pages
print('select 200 starts from ' + str(current) + ', total is ' + str(total))
data['subq_start'] = current
str_data = urllib.urlencode(data)
content = cls.get('http://estate-in-kharkov.com/ps/re_base/ajax/real-estate-database.php', str_data)
current += 200
if 're_base_query_count' in content:
total = int(content['re_base_query_count'])
if 'items' not in content and errors:
print("hasn't content , errors left %d" % errors)
errors -= 1
print(content)
time.sleep((6-errors)*10)
current -= 200
continue
assert 'items' in content, 'unexpected content: ' + str(content)
if content['items'] is False:
break
for item in content['items'].values():
aid = item['kod']
try:
existed = Ad.objects.get(gkey=aid)
print(str(existed) + ' already imported!')
continue
except Ad.DoesNotExist:
pass
ad_item = {'gkey': aid}
for field in item:
if item[field] and field in match:
ad_item[match[field]] = item[field]
ad = cls.parse_categories(ad_item, item, table)
# continue
if 'sub_category' in ad:
print(ad['sub_category'])
if 'rooms_count' in ad:
print('rooms: ' + str(ad['rooms_count']))
if item['textob']:
ad['title'] = truncatesmart(item['textob'], 45)
if not ad['title']:
ad['title'] = item['textob'][:45]
ad['offering'] = table != 'client'
ad['private'] = private_type == 'green'
ad['phone'] = ""
for i in range(1, 5):
name = 'tel'+str(i)
if name in item and item[name].strip():
if i > 1:
ad['phone'] += ', '
ad['phone'] += item[name].strip()
if item['metro']:
try:
metro = Metro.objects.get(pk=item['metro'])
ad['desc'] += ' ' + metro.name
except Metro.DoesNotExist:
pass
if item['ulica']:
try:
ad['address'] = Street.objects.get(pk=item['ulica'])
except Street.DoesNotExist:
pass
if item['raj']:
try:
ad['district'] = District.objects.get(pk=item['raj'])
except District.DoesNotExist:
pass
if item['nasp']:
item['nasp'] = 1 if item['nasp'] == '293' else item['nasp']
try:
ad['town'] = Town.objects.get(pk=item['nasp'])
except Town.DoesNotExist:
continue
if ad['pub_date']:
pub_date = dateutil.parser.parse(ad['pub_date'])
if str(pub_date.time()) == "00:00:00":
pub_date = datetime.combine(pub_date.date(), datetime.now().time())
pub_date_utc = pub_date.replace(tzinfo=timezone.get_current_timezone())
ad['pub_date'] = ad['order_date'] = pub_date_utc
if item['fotosite'] and item['fotosite'].find('&have_images'):
ad['url'] = item['fotosite'].replace('&have_images', '')
ad_obj = Ad(**ad)
ad_obj.save()
statistic[table][private_type] += 1
find_similar.delay(ad_obj.id)
if ad_obj.url:
parsed_uri = urlparse(ad_obj.url)
if parsed_uri.netloc in Crawler.DOMAINS:
import_attachments.delay(ad_obj.id)
print(ad_obj)
time.sleep(1)
except:
log += "Error: " + traceback.format_exc() + "\n"
raise
finally:
for name, stats in statistic.items():
log += name + ': '
if 'green' in stats:
log += str(stats['green'])
log += " / "
if 'red' in stats:
log += str(stats['red'])
log += "\n"
send_log(log)
