def parse(conn_type='gsm'):
data = {}
domain = 'http://diesel.elcat.kg'
if conn_type == 'gsm':
url = '%s/index.php?showforum=192' % domain
elif conn_type == 'cdma':
url = '%s/index.php?showforum=194' % domain
elif conn_type == 'tdma':
url = '%s/index.php?showforum=193' % domain
source = get_html_from(url)
if source is None:
return
html = unicode(source, 'windows-1251').encode('utf-8')
html = html[html.rfind('<!-- END PINNED -->'):]
hrefs = []
for match in re.finditer('tid-link-', html):
sub = html[match.end():match.start()+30]
hrefs.append('%s/index.php?showtopic=%s' % (domain,
sub.split('" href=')[0]))
for step, href in enumerate(hrefs[:5]):
innersource = get_html_from(href)
if innersource is None:
continue
innerhtml = unicode(innersource, 'windows-1251').encode('utf-8')
match = innerhtml[innerhtml.find('style=\'word-wrap:break-word;\'><div><b>'):]
title = match[38:match.find('</div></td>')]
pattern = re.compile(ur'Продаю\s(\w+)\s(\w+)', re.UNICODE)
string_re = re.search(pattern, unicode(title.lower(), 'utf-8'))
if string_re is not None:
try:
str(string_re).decode('ascii')
except UnicodeEncodeError:
continue
brand = string_re.group(1)
model = string_re.group(2)
else:
continue
continue_loop = False
if brand == u'iphone':
if _is_model(string_re.group(), except_='iPhone'):
continue_loop = True
data['brand'] = TelephoneBrand.objects.get(name='Apple')
for iphone in IPHONE_MODELS:
if model == iphone.lower():
if data.get('model', None) is not None:
data['model'] = 'iPhone {0}'.format(iphone)
data['has_3g'] = 'yes'
data['has_wifi'] = 'yes'
else:
continue_loop = True
elif brand == u'samsung':
if _is_model(string_re.group(), except_='Samsung'):
continue_loop = True
data['brand'] = TelephoneBrand.objects.get(name='Samsung')
for samsung in SAMSUNG_MODELS:
if model == samsung.lower():
if data.get('model', None) is not None:
data['model'] = samsung
else:
continue_loop = True
elif brand == u'htc':
if _is_model(string_re.group(), except_='HTC'):
continue_loop = True
data['brand'] = TelephoneBrand.objects.get(name='HTC')
for htc in HTC_MODELS:
if model == htc.lower():
if data.get('model', None) is not None:
data['model'] = htc
else:
continue_loop = True
elif brand == u'blackberry':
if _is_model(string_re.group(), except_='BlackBerry'):
continue_loop = True
data['brand'] = TelephoneBrand.objects.get(name='BlackBerry')
for blackberry in BLACKBERRY_MODELS:
if model == blackberry.lower():
if data.get('model', None) is not None:
data['model'] = blackberry
else:
continue_loop = True
elif brand == u'motorola':
if _is_model(string_re.group(), except_='Motorola'):
continue_loop = True
data['brand'] = TelephoneBrand.objects.get(name='Motorola')
for motorola in MOTOROLA_MODELS:
if model == motorola.lower():
if data.get('model', None) is not None:
data['model'] = motorola
else:
continue_loop = True
elif brand == u'sony':
if _is_model(string_re.group(), except_='Sony'):
continue_loop = True
data['brand'] = TelephoneBrand.objects.get(name='Sony')
for sony in SONY_MODELS:
if model == sony.lower():
if data.get('model', None) is not None:
data['model'] = sony
else:
continue_loop = True
elif brand == u'nokia':
if _is_model(string_re.group(), except_='Nokia'):
continue_loop = True
data['brand'] = TelephoneBrand.objects.get(name='Nokia')
for nokia in NOKIA_MODELS:
if model == nokia.lower():
if data.get('model', None) is not None:
data['model'] = nokia
else:
continue_loop = True
elif brand == u'huawei':
if _is_model(string_re.group(), except_='Huawei'):
continue_loop = True
data['brand'] = TelephoneBrand.objects.get(name='Huawei')
for huawei in HUAWEI_MODELS:
if model == huawei.lower():
if data.get('model', None) is not None:
data['model'] = huawei
else:
continue_loop = True
if continue_loop or not data.get('brand', '') or \
not data.get('model', ''):
continue
if conn_type == 'gsm':
data['conn_type'] = 'gsm'
elif conn_type == 'cdma':
data['conn_type'] = 'cdma'
elif conn_type == 'tdma':
data['conn_type'] = 'tdma'
data['published'] = False
data['featured'] = False
data['seller'] = None
data['location'] = 'bishkek'
data['ip'] = '212.42.102.203'
price_pattern = re.compile(ur'(\d{1,3}?[\.,\s]?\d{3})\s?(сом|с\.)',
re.UNICODE)
price_re = re.search(price_pattern, unicode(title, 'utf-8'))
if price_re is None:
continue
price = ''
for s in price_re.group(1):
if s.isdigit():
price += s
try:
data['price'] = Decimal(price)
except TypeError:
continue
match = innerhtml[innerhtml.find('<div class="postcolor" id=\'post'):]
body = match[42:match.find(u'<!-- подпись -->'.encode('utf-8'))]
if u'новый' in unicode(title.lower(), 'utf-8') or \
u'новый' in unicode(body.lower(), 'utf-8'):
data['condition'] = 'new'
else:
data['condition'] = 'used'
phone_pattern = re.compile(r'([0]|\+996)\s?([357]\d{2})\s?(\d{2}\s?\d{2}\s?\d{2})\D',
re.UNICODE)
phone_re = re.search(phone_pattern, unicode(body, 'utf-8'))
if phone_re is None:
continue
try:
data['phone'] = clean_phone(str(phone_re.group()))
except:
continue
if data.get('phone', None) is None or str(data['price']) in data['phone']:
continue
try:
hash = u'%s|%s|%s|%s|%s' % (
data['brand'],
data['model'],
'bishkek',
data['phone'],
data['ip']
)
except KeyError:
continue
data['hash'] = hashlib.sha256(hash).hexdigest()
try:
Telephone.objects.get(hash=data['hash'])
except Telephone.DoesNotExist:
pass
else:
continue
instance = Telephone.objects.create(**data)
im_pattern = re.compile(r'monthly_\d{2}_\d{4}/post-\d+-\d+_thumb\.jpg',
re.UNICODE)
im_re = re.findall(im_pattern, unicode(innerhtml, 'utf-8'))
if not im_re:
continue
for id, src in enumerate(im_re):
im_url = 'http://diesel.elcat.kg/uploads/%s' % src.replace('_thumb.', '.')
uid = getpwnam('django').pw_uid
gid = grp.getgrnam('www-data').gr_gid
f = get_html_from(im_url)
if f is None:
return
fname = src.split('/post-')[-1].replace('_thumb.', '.')
path = image.encode_image_name(None, fname)
localpath = settings.MEDIA_ROOT + path
local_f = open(localpath, 'w')
local_f.write(f)
local_f.close()
im_size = Image.open(localpath).size
if im_size[0] < 370 and im_size[1] < 200:
if os.path.isfile(localpath):
os.remove(localpath)
else:
local_f = open(localpath, 'r')
if id == 0:
instance.image_1.save(fname, File(local_f))
local_f.close()
os.chown(instance.image_1.path, uid, gid)
elif id == 1:
instance.image_2.save(fname, File(local_f))
local_f.close()
os.chown(instance.image_2.path, uid, gid)
elif id == 2:
instance.image_3.save(fname, File(local_f))
local_f.close()
os.chown(instance.image_3.path, uid, gid)
elif id == 3:
instance.image_4.save(fname, File(local_f))
local_f.close()
os.chown(instance.image_4.path, uid, gid)
else:
local_f.close()
break
instance.save()
def parse():
data = {}
domain = 'http://diesel.elcat.kg'
url = '%s/index.php?showforum=225' % domain
source = get_html_from(url)
if source is None:
return
html = unicode(source, 'windows-1251').encode('utf-8')
html = html[html.rfind('<!-- END PINNED -->'):]
hrefs = []
for match in re.finditer('tid-link-', html):
sub = html[match.end():match.start()+30]
hrefs.append('%s/index.php?showtopic=%s' % (domain,
sub.split('" href=')[0]))
for step, href in enumerate(hrefs[:7]):
innersource = get_html_from(href)
if innersource is None:
continue
innerhtml = unicode(innersource, 'windows-1251').encode('utf-8')
match = innerhtml[innerhtml.find('style=\'word-wrap:break-word;\'><div><b>'):]
title = match[38:match.find('</div></td>')]
data['published'] = True
data['featured'] = False
data['seller'] = None
data['section'] = 'sale'
data['location'] = 'bishkek'
data['ip'] = '212.42.102.203'
data['rooms'] = get_rooms(title)
price_pattern = re.compile(r'(\d{1,3}[\.,\s]?\d{3}\s?[\u0024])|([\u0024]\s?\d{1,3}[\.,\s]?\d{3})',
re.UNICODE)
price_re = re.search(price_pattern, unicode(title, 'utf-8'))
if price_re is None:
continue
price = ''
for s in price_re.group():
if s.isdigit():
price += s
try:
data['price'] = currency.convert_currency(Decimal(price), 'KGS')
except:
continue
if u'куплю'.encode('utf-8') in title.lower() or not data['rooms'] or \
u'обмен'.encode('utf-8') in title.lower():
continue
if is_realtor(title):
data['is_realtor'] = 'yes'
else:
data['is_realtor'] = 'no'
match = innerhtml[innerhtml.find('<div class="postcolor" id=\'post'):]
body = match[42:match.find(u'<!-- подпись -->'.encode('utf-8'))]
if u'новый' in unicode(title.lower(), 'utf-8') or \
u'новый' in unicode(body.lower(), 'utf-8'):
data['condition'] = 'new'
else:
data['condition'] = 'used'
floor_pattern = re.compile(ur'(\d{1,2})((\s?/\s?|\sиз\s)\d{1,2})?\s?(этаж|эт\.)',
re.UNICODE)
floor_re = re.search(floor_pattern, unicode(body, 'utf-8'))
if floor_re is None:
floor_re = re.search(floor_pattern, unicode(title, 'utf-8'))
if floor_re is not None:
data['floor'] = floor_re.group(1)
area_pattern = re.compile(ur'(\d+[\.,]?\d+?)\s?(кв\.?[/\s]?м\.?|м2)',
re.UNICODE)
area_re = re.search(area_pattern, unicode(body, 'utf-8'))
if area_re is None:
area_re = re.search(area_pattern, unicode(title, 'utf-8'))
if area_re is not None:
data['area'] = area_re.group(1)
phone_pattern = re.compile(r'([0]|\+996)\s?([357]\d{2})\s?(\d{2}\s?\d{2}\s?\d{2})\D',
re.UNICODE)
phone_re = re.search(phone_pattern, unicode(body, 'utf-8'))
if phone_re is None:
continue
try:
data['phone'] = clean_phone(str(phone_re.group()))
except:
continue
if data.get('phone', None) is None or str(data['price']) in str(data['phone']):
continue
try:
hash = u'%s|%s|%s|%s' % (
data['rooms'],
'bishkek',
data['phone'],
data['ip']
)
except KeyError:
continue
data['hash'] = hashlib.sha256(hash).hexdigest()
try:
House.objects.get(hash=data['hash'])
except House.DoesNotExist:
pass
else:
continue
instance = House.objects.create(**data)
im_pattern = re.compile(r'monthly_\d{2}_\d{4}/post-\d+-\d+_thumb\.jpg',
re.UNICODE)
im_re = re.findall(im_pattern, unicode(innerhtml, 'utf-8'))
if not im_re:
continue
for id, src in enumerate(im_re):
im_url = 'http://diesel.elcat.kg/uploads/%s' % src.replace('_thumb.', '.')
uid = getpwnam('django').pw_uid
gid = grp.getgrnam('www-data').gr_gid
f = get_html_from(im_url)
if f is None:
return
fname = src.split('/post-')[-1].replace('_thumb.', '.')
path = image.encode_image_name(None, fname)
localpath = settings.MEDIA_ROOT + path
local_f = open(localpath, 'w')
local_f.write(f)
local_f.close()
im_size = Image.open(localpath).size
if im_size[0] < 370 and im_size[1] < 200:
if os.path.isfile(localpath):
os.remove(localpath)
else:
local_f = open(localpath, 'r')
if id == 0:
instance.image_1.save(fname, File(local_f))
local_f.close()
os.chown(instance.image_1.path, uid, gid)
elif id == 1:
instance.image_2.save(fname, File(local_f))
local_f.close()
os.chown(instance.image_2.path, uid, gid)
elif id == 2:
instance.image_3.save(fname, File(local_f))
local_f.close()
os.chown(instance.image_3.path, uid, gid)
elif id == 3:
instance.image_4.save(fname, File(local_f))
local_f.close()
os.chown(instance.image_4.path, uid, gid)
else:
local_f.close()
break
instance.save()
def parse():
data = {}
domain = 'http://auto.doska.kg'
urls = ['%s/carsfind/page:%s/?order=date_new' % (domain, page) for page in range(1, 7)]
for url in urls:
soup = BeautifulSoup(get_html_from(url))
if soup is None:
return
href = str()
for id, table in enumerate(soup.find_all('table')):
if id == 3:
containers = table.find_all('tr')
for item in containers:
for id, td in enumerate(item.find_all('td')):
if id == 0:
if 'photononb' in td.a.img['src']:
return
href = domain + td.a['href']
if id == 1:
title = td.find('a', {'class': 'fs15'}).string
try:
title.decode('ascii')
except UnicodeEncodeError:
return
for brand in AutomobileBrand.objects.all():
if brand.name in title:
data['brand'] = brand
data['model'] = title.replace(brand.name, '').lstrip()
fields = td.find('div', {'class': 'fs12'}).contents[2]
try:
data['year'] = int(td.find('div', {'class': 'fs12'}).b.string[:4])
except:
return
if get_color(fields) is not None:
data['color'] = get_color(fields)
if get_body_type(fields) is not None:
data['body_type'] = get_body_type(fields)
if u'Механика' in fields.string:
data['transmission'] = 'mechanic'
elif u'Автомат' in fields.string:
data['transmission'] = 'automatic'
elif u'Типтроник' in fields.string:
data['transmission'] = 'tiptronic'
else:
return
if u'правый руль' in fields.string.lower():
data['steering_wheel'] = 'right'
else:
data['steering_wheel'] = 'left'
try:
sub = fields[:fields.find(u'л.')]
if sub:
if len(sub[sub.rfind(',')+1:].strip()) <= 10:
data['engine_capacity'] = sub[sub.rfind(',')+1:].strip()
except:
pass
if get_drive_type(fields) is not None:
data['drive_type'] = get_drive_type(fields)
contents = u''.join([unicode(td) for td in item.contents])
price = str()
for s in contents[contents.find('$')+1:contents.find('$')+7]:
if s.isdigit():
price += s
data['price'] = currency.convert_currency(Decimal(price), 'KGS')
innersoup = BeautifulSoup(get_html_from(href))
data['phone'] = clean_phone(innersoup.find('span', {'class': 'text_c_blue'}).string)
if data.get('phone', None) is None:
continue
data['ip'] = '212.42.117.66'
try:
hash = u'%s|%s|%s|%s' % (
data['brand'].name,
data['model'],
data['phone'],
data['ip']
)
data['hash'] = hashlib.sha256(hash).hexdigest()
except:
continue
try:
Automobile.objects.get(hash=data['hash'])
except Automobile.DoesNotExist:
pass
else:
continue
data['published'] = True
data['featured'] = False
data['seller'] = None
data['condition'] = 'used'
data['location'] = 'bishkek'
auto = Automobile.objects.create(**data)
id = 0
for link in innersoup.find_all('a'):
if link.has_key('rel'):
im_url = link['href']
uid = getpwnam('django').pw_uid
gid = grp.getgrnam('www-data').gr_gid
f = get_html_from(im_url)
if f is None:
return
fname = im_url.replace('http://auto.static.akipress.org/', '')[im_url.find('/')-1:]
path = image.encode_image_name(None, fname)
localpath = settings.MEDIA_ROOT + path
local_f = open(localpath, 'w')
local_f.write(f)
local_f.close()
im_size = Image.open(localpath).size
if im_size[0] < 370 and im_size[1] < 200:
if os.path.isfile(localpath):
os.remove(localpath)
else:
local_f = open(localpath, 'r')
if id == 0:
auto.image_1.save(fname, File(local_f))
local_f.close()
os.chown(auto.image_1.path, uid, gid)
elif id == 1:
auto.image_2.save(fname, File(local_f))
local_f.close()
os.chown(auto.image_2.path, uid, gid)
elif id == 2:
auto.image_3.save(fname, File(local_f))
local_f.close()
os.chown(auto.image_3.path, uid, gid)
elif id == 3:
auto.image_4.save(fname, File(local_f))
local_f.close()
os.chown(auto.image_4.path, uid, gid)
else:
local_f.close()
break
id += 1
auto.save()
def parse():
data = {}
domain = 'http://www.avtogid.kg'
urls = ['%s/search/search?cat=1&page=%s' % (domain, page) for page in range(1, 7)]
for url in urls:
soup = BeautifulSoup(get_html_from(url))
if soup is None:
return
containers = soup.find_all('div', {'class': 'sub-selected'})
for item in containers:
table_tag = item.find('div', {'class': 'table'})
title_tag = item.find('div', {'class': 'res-car-name'})
title = title_tag.h3.a.string
href = domain + title_tag.h3.a['href']
try:
title.decode('ascii')
except UnicodeEncodeError:
continue
for brand in AutomobileBrand.objects.all():
if brand.name in title:
data['brand'] = brand
data['model'] = title.replace(brand.name, '').lstrip()
price = str()
for s in title_tag.h4.string:
if s.isdigit():
price += s
data['price'] = currency.convert_currency(Decimal(price), 'KGS')
for id, td in enumerate(table_tag.table.tbody.find_all('td')):
if id == 0:
data['year'] = int(td.string[:4])
if id == 1:
data['engine_capacity'] = td.string
if id == 2:
if get_color(td) is not None:
data['color'] = get_color(td)
if id == 3:
if td.string == u'Автомат':
data['transmission'] = 'automatic'
elif td.string == u'Типтроник':
data['transmission'] = 'tiptronic'
elif td.string == u'Механическая':
data['transmission'] = 'mechanic'
else:
continue
innersoup = BeautifulSoup(get_html_from(href))
if innersoup is None:
return
desc_strs = innersoup.find('div', {'class': 'description'}).p.strings
try:
data['desc'] = u'. '.join(desc_strs)
except AttributeError:
pass
else:
if len(data['desc']) < 5:
data['desc'] = ''
tel_strs = innersoup.find('div', {'class': 'tel-soc'}).strings
for id, string in enumerate(tel_strs):
if id == 2:
if string:
data['phone'] = clean_phone(string)
else:
continue
if data.get('phone', None) is None:
continue
data['ip'] = '212.42.117.66'
try:
hash = u'%s|%s|%s|%s' % (
data['brand'].name,
data['model'],
data['phone'],
data['ip']
)
except KeyError:
continue
data['hash'] = hashlib.sha256(hash).hexdigest()
try:
Automobile.objects.get(hash=data['hash'])
except Automobile.DoesNotExist:
pass
else:
continue
data['published'] = True
data['featured'] = False
data['seller'] = None
data['condition'] = 'used'
data['location'] = 'bishkek'
auto = Automobile.objects.create(**data)
try:
im_url = innersoup.find('a', {'class': 'main-image'})['href']
except TypeError:
pass
else:
uid = getpwnam('django').pw_uid
gid = grp.getgrnam('www-data').gr_gid
f = get_html_from(domain + im_url)
if f is None:
return
fname = im_url.replace('/upload/images/', '')
path = image.encode_image_name(None, fname)
localpath = settings.MEDIA_ROOT + path
local_f = open(localpath, 'w')
local_f.write(f)
local_f.close()
im_size = Image.open(localpath).size
if im_size[0] < 370 and im_size[1] < 200:
if os.path.isfile(localpath):
os.remove(localpath)
else:
local_f = open(localpath, 'r')
auto.image_1.save(fname, File(local_f))
local_f.close()
os.chown(auto.image_1.path, uid, gid)
try:
thumbs = innersoup.find_all('a', {'class': 'thumbnail'})
except TypeError:
pass
else:
for id, im in enumerate(thumbs):
f = get_html_from(domain + im['href'])
if f is None:
return
fname = im['href'].replace('/upload/images/', '')
path = image.encode_image_name(None, fname)
localpath = settings.MEDIA_ROOT + path
local_f = open(localpath, 'w')
local_f.write(f)
local_f.close()
im_size = Image.open(localpath).size
if im_size[0] < 370 and im_size[1] < 200:
if os.path.isfile(localpath):
os.remove(localpath)
else:
local_f = open(localpath, 'r')
if id == 0:
auto.image_2.save(fname, File(local_f))
local_f.close()
os.chown(auto.image_2.path, uid, gid)
elif id == 1:
auto.image_3.save(fname, File(local_f))
local_f.close()
os.chown(auto.image_3.path, uid, gid)
elif id == 2:
auto.image_4.save(fname, File(local_f))
local_f.close()
os.chown(auto.image_4.path, uid, gid)
else:
local_f.close()
break
auto.save()
def parse():
data = {}
model = None
domain = 'http://www.domik.kg'
urls = ['%s/offers/sell/page_%s/' % (domain, page) for page in range(1, 7)]
for url in urls:
soup = BeautifulSoup(get_html_from(url))
if soup is None:
return
containers = soup.find_all('div', {'class': 'pix_r'})
for item in containers:
title = unicode()
title_tags = item.find('h1', {'class': 'i_sell'}).find_all('a')
for id, tag in enumerate(title_tags):
if id == 0:
href = tag['href']
title = tag.string.strip().replace('\n', '').replace('\t', '')
break
rooms = str()
for s in title:
if s.isdigit():
rooms += s
try:
data['rooms'] = int(rooms)
except ValueError:
continue
for string in item.strings:
if u'Бишкек' in string:
data['location'] = 'bishkek'
elif u'Иссык-Кульская область' in string:
data['location'] = 'issykkul'
elif u'Чуйская область' in string:
data['location'] = 'chui'
elif u'Джалал-Абадская область' in string:
data['location'] = 'jalalabad'
elif u'Баткенская область' in string:
data['location'] = 'batken'
elif u'Ошская область' in string:
data['location'] = 'osh'
elif u'Таласская область' in string:
data['location'] = 'talas'
else:
continue
price = str()
try:
for s in item.find('div', {'class': 'price_item'}).string:
if s.isdigit():
price += s
except TypeError:
continue
else:
if price:
data['price'] = currency.convert_currency(Decimal(price), 'KGS')
else:
continue
innersoup = BeautifulSoup(get_html_from(url))
if innersoup is None:
return
try:
main_tags = innersoup.find('td', {'id': 'fotoz'})
addr_strs = main_tags.find_next_sibling('td').h3.find_next_sibling('div').strings
addr = unicode()
for id, string in enumerate(addr_strs):
if id > 1:
addr += u'%s. ' % string
except:
continue
if len(addr.strip()):
data['address'] = addr.strip()
else:
continue
desc = innersoup.find('div', {'class': 'fullcont'}).string
try:
data['desc'] = desc.strip()
except AttributeError:
data['desc'] = ''
else:
if len(data['desc']) < 5:
data['desc'] = ''
data['phone'] = clean_phone(innersoup.find('div', {'class': 'fullcont'}).find_next_sibling('div', {'class': 'private'}).string)
if data.get('phone', None) is None:
continue
data['ip'] = '176.126.165.34'
if len(data['phone']) > 15:
continue
try:
hash = u'%s|%s|%s|%s' % (
data['rooms'],
data['location'],
data['phone'],
data['ip']
)
except KeyError:
continue
data['hash'] = hashlib.sha256(hash).hexdigest()
data_strs = main_tags.find_next_sibling('td').h3.find_next_sibling('div').find_next_sibling('div').strings
floor = str()
for id, string in enumerate(data_strs):
if id == 2:
for s in string:
if s.isdigit():
floor += s
if id == 5:
try:
data['area'] = int(string)
except ValueError:
pass
try:
int(floor)
except ValueError:
continue
else:
data['floor'] = floor
data['is_realtor'] = 'yes'
for tag in innersoup.find_all('div', {'class': 'private'}):
if u'(собственник)' in tag.string:
data['is_realtor'] = 'no'
data['published'] = True
data['featured'] = False
data['seller'] = None
data['condition'] = 'used'
data['section'] = 'sale'
if u'квартиру' in title:
try:
Apartment.objects.get(hash=data['hash'])
except Apartment.DoesNotExist:
pass
else:
continue
instance = Apartment.objects.create(**data)
elif u'дом' in title:
try:
House.objects.get(hash=data['hash'])
except House.DoesNotExist:
pass
else:
continue
instance = House.objects.create(**data)
else:
continue
try:
im_url = innersoup.find('img', {'id': 'foto_prev'})['src'].replace('_m.', '.')
except TypeError:
pass
else:
uid = getpwnam('django').pw_uid
gid = grp.getgrnam('www-data').gr_gid
f = get_html_from(im_url)
if f is None:
return
fname = im_url.split('/')[-1]
path = image.encode_image_name(None, fname)
localpath = settings.MEDIA_ROOT + path
local_f = open(localpath, 'w')
local_f.write(f)
local_f.close()
im_size = Image.open(localpath).size
if im_size[0] < 370 and im_size[1] < 200:
if os.path.isfile(localpath):
os.remove(localpath)
else:
local_f = open(localpath, 'r')
instance.image_1.save(fname, File(local_f))
local_f.close()
os.chown(instance.image_1.path, uid, gid)
try:
thumb_tags = innersoup.find_all('a', {'class': 'fotoz_link'})
except TypeError:
pass
else:
for id, tag in enumerate(thumb_tags):
if id == 0:
continue
if isinstance(tag['rel'], list):
thumb_url = tag['rel'][0].replace('_m.', '.')
else:
thumb_url = tag['rel'].replace('_m.', '.')
f = get_html_from(thumb_url)
if f is None:
return
fname = thumb_url.split('/')[-1]
path = image.encode_image_name(None, fname)
localpath = settings.MEDIA_ROOT + path
local_f = open(localpath, 'w')
local_f.write(f)
local_f.close()
im_size = Image.open(localpath).size
if im_size[0] < 370 and im_size[1] < 200:
if os.path.isfile(localpath):
os.remove(localpath)
else:
local_f = open(localpath, 'r')
if id == 1:
instance.image_2.save(fname, File(local_f))
local_f.close()
os.chown(instance.image_2.path, uid, gid)
elif id == 2:
instance.image_3.save(fname, File(local_f))
local_f.close()
os.chown(instance.image_3.path, uid, gid)
elif id == 3:
instance.image_4.save(fname, File(local_f))
local_f.close()
os.chown(instance.image_4.path, uid, gid)
else:
local_f.close()
break
instance.save()
