def crawl_category(self, ctx='', **kwargs):
res = requests.get(HOST)
res.raise_for_status()
tree = lxml.html.fromstring(res.content)
dept_nodes = tree.cssselect('div#top-navigation ul.navigation li.menu-item a')
for dept_node in dept_nodes:
key = dept_node.text.strip()
if 'brand' in key.lower():
continue
combine_url = dept_node.get('href')
match = re.search(r'https?://.+', combine_url)
if not match:
combine_url = '%s%s' % (HOST, combine_url)
r = requests.get(combine_url)
r.raise_for_status()
t = lxml.html.fromstring(r.content)
pagesize_node = None
link_nodes = t.cssselect('div.atg_store_filter ul.atg_store_pager li')
for link_node in link_nodes:
if link_node.get('class') and 'nextLink' in link_node.get('class'):
break
pagesize_node = link_node
pagesize = int(pagesize_node.cssselect('a')[0].text.strip()) if pagesize_node else 1
is_new = False; is_updated = False
category = Category.objects(key=key).first()
if not category:
is_new = True
category = Category(key=key)
category.is_leaf = True
if combine_url and combine_url != category.combine_url:
category.combine_url = combine_url
is_updated = True
if pagesize and pagesize != category.pagesize:
category.pagesize = pagesize
is_updated = True
category.hit_time = datetime.utcnow()
category.save()
print category.key; print category.cats; print category.pagesize; print category.combine_url; print is_new; print is_updated; print;
common_saved.send(sender=ctx, obj_type='Category', key=category.key, url=category.combine_url, \
is_new=is_new, is_updated=((not is_new) and is_updated) )
def crawl_category(self, ctx='', **kwargs):
res = requests.get(HOST)
res.raise_for_status()
tree = lxml.html.fromstring(res.content)
primary_dept_nodes = tree.cssselect('nav#primary-nav > ul li.tab')
for primary_dept_node in primary_dept_nodes:
primary_dept = primary_dept_node.cssselect('h2 a span')[0].text
if primary_dept.lower() == 'brands':
continue
sub_dept = None
sub_dept_nodes = primary_dept_node.cssselect('div.nav-category div.nav-category-column ul li a')
for sub_dept_node in sub_dept_nodes:
sub_dept = sub_dept_node.text
combine_url = sub_dept_node.get('href')
key = combine_url.split('/')[-1]
match = re.search(r'https?://.+', combine_url)
if not match:
combine_url = '%s%s' % (HOST, combine_url)
is_new = False; is_updated = False
category = Category.objects(key=key).first()
if not category:
is_new = True
category = Category(key=key)
category.is_leaf = True
if primary_dept not in category.cats:
category.cats.append(primary_dept)
is_updated = True
if sub_dept not in category.cats:
category.cats.append(sub_dept)
is_updated = True
if combine_url and ((category.combine_url and combine_url.split('?')[0] != category.combine_url.split('?')[0]) \
or not category.combine_url):
category.combine_url = combine_url
is_updated = True
category.hit_time = datetime.utcnow()
category.save()
common_saved.send(sender=ctx, obj_type='Category', key=category.key, url=category.combine_url, \
is_new=is_new, is_updated=((not is_new) and is_updated) )
def crawl_category(self, ctx='', **kwargs):
res = requests.get(HOST)
res.raise_for_status()
tree = lxml.html.fromstring(res.content)
dept_nodes = tree.cssselect('div#nav div#moreDeptsWrap ul li a')
for dept_node in dept_nodes:
dept = dept_node.text.strip()
href = dept_node.get('href')
match = re.search(r'https?://.+', href)
if not match:
href = '%s%s' % (HOST, href)
res = requests.get(href)
res.raise_for_status()
tree = lxml.html.fromstring(res.content)
sub_dept_nodes = tree.cssselect('div#tcSideCol h5 a')
for sub_dept_node in sub_dept_nodes:
sub_dept = sub_dept_node.text.strip()
cats = [dept, sub_dept]
key = '_'.join(cats)
combine_url = sub_dept_node.get('href')
match = re.search(r'https?://.+', combine_url)
if not match:
combine_url = '%s%s' % (HOST, combine_url)
is_new = False; is_updated = False
category = Category.objects(key=key).first()
if not category:
is_new = True
category = Category(key=key)
category.is_leaf = True
if set(cats).difference(category.cats):
category.cats = list(set(cats) | set(category.cats))
is_updated = True
if combine_url and combine_url != category.combine_url:
category.combine_url = combine_url
is_updated = True
category.hit_time = datetime.utcnow()
category.save()
common_saved.send(sender=ctx, obj_type='Category', key=category.key, url=category.combine_url, \
is_new=is_new, is_updated=((not is_new) and is_updated) )
def crawl_category(self, ctx='', **kwargs):
res = requests.get(HOST)
res.raise_for_status()
tree = lxml.html.fromstring(res.content)
primary_cat_nodes = tree.cssselect('div.nav ul.menu li.menu-column')
for primary_cat_node in primary_cat_nodes:
primary_cat = primary_cat_node.cssselect('a.menu-column-link')[0].text.strip()
if primary_cat and 'designer' in primary_cat.lower():
continue
sub_cat_nodes = primary_cat_node.cssselect('ul.sub-menu li.sub-menu-column ul li a')
for sub_cat_node in sub_cat_nodes:
sub_cat = sub_cat_node.text.strip()
combine_url = sub_cat_node.get('href')
key = re.sub(HOST, '', combine_url)
key = re.sub('/', '_', key)
is_new = False; is_updated = False
category = Category.objects(key=key).first()
if not category:
is_new = True
category = Category(key=key)
category.is_leaf = True
if primary_cat not in category.cats:
category.cats.append(primary_cat)
is_updated = True
if combine_url and combine_url != category.combine_url:
category.combine_url = combine_url
is_updated = True
category.hit_time = datetime.utcnow()
category.save()
# print category.key; print category.cats; print category.combine_url; print is_new; print is_updated; print
common_saved.send(sender=ctx, obj_type='Category', key=category.key, url=category.combine_url, \
is_new=is_new, is_updated=((not is_new) and is_updated) )