def check_generate_pages_html(self, current_page, total_pages, window=7, check_middle=False):
extra_links = 4 # first, prev, next, last
search = "'>\"/><img src=x onerror=alert(1)>"
html_str = utils.generate_pages(current_page, total_pages, search=search)
self.assertNotIn(search, html_str, "The raw search string shouldn't appear in the output")
self.assertIn('search=%27%3E%22%2F%3E%3Cimg+src%3Dx+onerror%3Dalert%281%29%3E', html_str)
self.assertTrue(callable(html_str.__html__), "Should return something that is HTML-escaping aware")
dom = BeautifulSoup(html_str, 'html.parser')
self.assertIsNotNone(dom)
ulist = dom.ul
ulist_items = ulist.find_all('li')
self.assertEqual(min(window, total_pages) + extra_links, len(ulist_items))
page_items = ulist_items[2:-2]
mid = int(len(page_items) / 2)
for i, item in enumerate(page_items):
a_node = item.a
href_link = a_node['href']
node_text = a_node.string
if node_text == str(current_page + 1):
if check_middle:
self.assertEqual(mid, i)
self.assertEqual('javascript:void(0)', href_link)
self.assertIn('active', item['class'])
else:
self.assertRegex(href_link, r'^\?', 'Link is page-relative')
query = parse_qs(href_link[1:])
self.assertListEqual(query['page'], [str(int(node_text) - 1)])
self.assertListEqual(query['search'], [search])
def check_generate_pages_html(self,
current_page,
total_pages,
window=7,
check_middle=False):
extra_links = 4 # first, prev, next, last
search = "'>\"/><img src=x onerror=alert(1)>"
html_str = utils.generate_pages(current_page,
total_pages,
search=search)
assert search not in html_str, "The raw search string shouldn't appear in the output"
assert 'search=%27%3E%22%2F%3E%3Cimg+src%3Dx+onerror%3Dalert%281%29%3E' in html_str
assert callable(
html_str.__html__
), "Should return something that is HTML-escaping aware"
dom = BeautifulSoup(html_str, 'html.parser')
assert dom is not None
ulist = dom.ul
ulist_items = ulist.find_all('li')
assert min(window, total_pages) + extra_links == len(ulist_items)
page_items = ulist_items[2:-2]
mid = int(len(page_items) / 2)
for i, item in enumerate(page_items):
a_node = item.a
href_link = a_node['href']
node_text = a_node.string
if node_text == str(current_page + 1):
if check_middle:
assert mid == i
assert 'javascript:void(0)' == href_link
assert 'active' in item['class']
else:
assert re.search(r'^\?', href_link), 'Link is page-relative'
query = parse_qs(href_link[1:])
assert query['page'] == [str(int(node_text) - 1)]
assert query['search'] == [search]
def check_generate_pages_html(self,
current_page,
total_pages,
window=7,
check_middle=False):
extra_links = 4 # first, prev, next, last
html_str = utils.generate_pages(current_page, total_pages)
# dom parser has issues with special « and »
html_str = html_str.replace('«', '')
html_str = html_str.replace('»', '')
dom = minidom.parseString(html_str)
self.assertIsNotNone(dom)
ulist = dom.getElementsByTagName('ul')[0]
ulist_items = ulist.getElementsByTagName('li')
self.assertEqual(
min(window, total_pages) + extra_links, len(ulist_items))
def get_text(nodelist):
rc = []
for node in nodelist:
if node.nodeType == node.TEXT_NODE:
rc.append(node.data)
return ''.join(rc)
page_items = ulist_items[2:-2]
mid = int(len(page_items) / 2)
for i, item in enumerate(page_items):
a_node = item.getElementsByTagName('a')[0]
href_link = a_node.getAttribute('href')
node_text = get_text(a_node.childNodes)
if node_text == str(current_page + 1):
if check_middle:
self.assertEqual(mid, i)
self.assertEqual('javascript:void(0)',
a_node.getAttribute('href'))
self.assertIn('active', item.getAttribute('class'))
else:
link_str = '?page=' + str(int(node_text) - 1)
self.assertEqual(link_str, href_link)
def check_generate_pages_html(self, current_page, total_pages,
window=7, check_middle=False):
extra_links = 4 # first, prev, next, last
search = "'>\"/><img src=x onerror=alert(1)>"
html_str = utils.generate_pages(current_page, total_pages,
search=search)
self.assertNotIn(search, html_str,
"The raw search string shouldn't appear in the output")
self.assertIn('search=%27%3E%22%2F%3E%3Cimg+src%3Dx+onerror%3Dalert%281%29%3E',
html_str)
self.assertTrue(
callable(html_str.__html__),
"Should return something that is HTML-escaping aware"
)
dom = BeautifulSoup(html_str, 'html.parser')
self.assertIsNotNone(dom)
ulist = dom.ul
ulist_items = ulist.find_all('li')
self.assertEqual(min(window, total_pages) + extra_links, len(ulist_items))
page_items = ulist_items[2:-2]
mid = int(len(page_items) / 2)
for i, item in enumerate(page_items):
a_node = item.a
href_link = a_node['href']
node_text = a_node.string
if node_text == str(current_page + 1):
if check_middle:
self.assertEqual(mid, i)
self.assertEqual('javascript:void(0)', href_link)
self.assertIn('active', item['class'])
else:
self.assertRegex(href_link, r'^\?', 'Link is page-relative')
query = parse_qs(href_link[1:])
self.assertListEqual(query['page'], [str(int(node_text) - 1)])
self.assertListEqual(query['search'], [search])
def check_generate_pages_html(self, current_page, total_pages,
window=7, check_middle=False):
extra_links = 4 # first, prev, next, last
html_str = utils.generate_pages(current_page, total_pages)
# dom parser has issues with special « and »
html_str = html_str.replace('«', '')
html_str = html_str.replace('»', '')
dom = minidom.parseString(html_str)
self.assertIsNotNone(dom)
ulist = dom.getElementsByTagName('ul')[0]
ulist_items = ulist.getElementsByTagName('li')
self.assertEqual(min(window, total_pages) + extra_links, len(ulist_items))
def get_text(nodelist):
rc = []
for node in nodelist:
if node.nodeType == node.TEXT_NODE:
rc.append(node.data)
return ''.join(rc)
page_items = ulist_items[2:-2]
mid = int(len(page_items) / 2)
for i, item in enumerate(page_items):
a_node = item.getElementsByTagName('a')[0]
href_link = a_node.getAttribute('href')
node_text = get_text(a_node.childNodes)
if node_text == str(current_page + 1):
if check_middle:
self.assertEqual(mid, i)
self.assertEqual('javascript:void(0)', a_node.getAttribute('href'))
self.assertIn('active', item.getAttribute('class'))
else:
link_str = '?page=' + str(int(node_text) - 1)
self.assertEqual(link_str, href_link)
def index(self, session=None):
DM = models.DagModel
# restrict the dags shown if filter_by_owner and current user is not superuser
do_filter = FILTER_BY_OWNER
owner_mode = conf.get('webserver', 'OWNER_MODE').strip().lower()
hide_paused_dags_by_default = conf.getboolean(
'webserver', 'hide_paused_dags_by_default')
show_paused_arg = request.args.get('showPaused', 'None')
def get_int_arg(value, default=0):
try:
return int(value)
except ValueError:
return default
arg_current_page = request.args.get('page', '0')
arg_search_query = request.args.get('search', None)
dags_per_page = PAGE_SIZE
current_page = get_int_arg(arg_current_page, default=0)
if show_paused_arg.strip().lower() == 'false':
hide_paused = True
elif show_paused_arg.strip().lower() == 'true':
hide_paused = False
else:
hide_paused = hide_paused_dags_by_default
# read orm_dags from the db
query = session.query(DM)
query = query.filter(~DM.is_subdag, DM.is_active)
# optionally filter out "paused" dags
if hide_paused:
query = query.filter(~DM.is_paused)
if arg_search_query:
query = query.filter(
DM.dag_id.ilike('%' + arg_search_query + '%')
| DM.owners.ilike('%' + arg_search_query + '%'))
query = query.order_by(DM.dag_id)
start = current_page * dags_per_page
end = start + dags_per_page
dags = query.offset(start).limit(dags_per_page).all()
import_errors = session.query(errors.ImportError).all()
for ie in import_errors:
flash("Broken DAG: [{ie.filename}] {ie.stacktrace}".format(ie=ie),
"error")
from airflow.plugins_manager import import_errors as plugin_import_errors
for filename, stacktrace in plugin_import_errors.items():
flash(
"Broken plugin: [{filename}] {stacktrace}".format(
stacktrace=stacktrace, filename=filename), "error")
num_of_all_dags = query.count()
num_of_pages = int(math.ceil(num_of_all_dags / float(dags_per_page)))
auto_complete_data = set()
for row in query.with_entities(DM.dag_id, DM.owners):
auto_complete_data.add(row.dag_id)
auto_complete_data.add(row.owners)
return self.render(
'airflow/dags.html',
dags=dags,
hide_paused=hide_paused,
current_page=current_page,
search_query=arg_search_query if arg_search_query else '',
page_size=dags_per_page,
num_of_pages=num_of_pages,
num_dag_from=min(start + 1, num_of_all_dags),
num_dag_to=min(end, num_of_all_dags),
num_of_all_dags=num_of_all_dags,
paging=wwwutils.generate_pages(current_page,
num_of_pages,
search=arg_search_query,
showPaused=not hide_paused),
auto_complete_data=auto_complete_data)
