def get_ratio_of_discr(df: pd.DataFrame, admissibles, outcome, sensitive):
"""
Compute adjusted Ratio of Observational Discrimination
"""
dom_sensitive = sorted(get_domain(df, [sensitive]))
dom_outcome = sorted(get_domain(df, [outcome]))
if len(dom_sensitive) != 2 or len(dom_outcome) != 2:
return np.array([1.0])
s0 = dom_sensitive[0]
s1 = dom_sensitive[1]
o0 = dom_outcome[0]
o1 = dom_outcome[1]
cont_matrix_iter = contingency_matrices_iterator(df, outcome, sensitive,
admissibles)
rods = []
for cm in cont_matrix_iter:
if cm.shape == (2, 2):
cb = cm[s0][o1] * cm[s1][o0]
ad = cm[s0][o0] * cm[s1][o1]
if ad != 0:
rods.append(cb / ad)
else:
rods.append(1.0)
rods = np.array(rods)
return rods
def get_source_node(self):
if 'source_selector' in self.options:
if self.options['source_selector']:
nodes = self.doc.cssselect(self.options['source_selector'])
if len(nodes) == 1:
return nodes[0]
for node in nodes:
res = self.has_source(node)
if res is not None:
return res
body = self.doc.find('body')
if body is None:
return None
for node in body.iter():
res = self.has_source(node)
if res is not None:
return res
domain = get_domain(self.url)
for a in self.doc.iter('a'):
link = a.get('href')
if link and link.startswith('http') \
and get_domain(link) != domain:
text = self.get_block_text(a)
if len(text) > 2 \
and text.endswith(u'报') \
and not text.endswith(u'举报'):
return a
def get_source_node(self):
if 'source_selector' in self.options:
if self.options['source_selector']:
nodes = self.doc.cssselect(self.options['source_selector'])
if len(nodes) == 1:
return nodes[0]
for node in nodes:
res = self.has_source(node)
if res is not None:
return res
for node in self.doc.find('body').iter():
res = self.has_source(node)
if res is not None:
return res
domain = get_domain(self.url)
for a in self.doc.iter('a'):
link = a.get('href')
if link and link.startswith('http') \
and get_domain(link) != domain:
text = self.get_block_text(a)
if len(text) > 2 \
and text.endswith(u'报') \
and not text.endswith(u'举报'):
return a
def get_source_node(self):
if self.options.get('source_selector', ''):
nodes = self.article.select(self.options['source_selector'])
if len(nodes) == 1:
return nodes[0]
for node in nodes:
res = self.has_source(node)
if res is not None:
return res
for node in self.article.doc.find('body').iter():
res = self.has_source(node)
if res is not None:
return res
domain = get_domain(self.article.url)
for a in self.article.doc.iter('a'):
link = a.get('href')
if link and link.startswith('http') \
and get_domain(link) != domain:
text = self.article.get_block_text(a)
if len(text) >= 2:
if text.endswith(u'报') and not text.endswith(u'举报') \
or text[-2:] in CATES and len(text) == 4:
return a
def learn_page(request, theme_name=None, template='theme_page.html'):
theme = get_object_or_404(Theme, name=theme_name)
theme.layers = theme.layer_set.all().order_by('name')
context = {
'theme': theme,
'domain': get_domain(8000),
'domain8010': get_domain()
}
return render_to_response(template, RequestContext(request, context))
def data_needs(request, template='needs.html'):
themes = Theme.objects.all().order_by('display_name')
ordered_themes, theme_dict = add_ordered_needs_lists(themes)
context = {
'themes': themes,
'theme_dict': theme_dict,
'ordered_themes': ordered_themes,
'domain': get_domain(8000),
'domain8010': get_domain()
}
return render_to_response(template, RequestContext(request, context))
def data_needs(request, template="needs.html"):
themes = Theme.objects.all().order_by("display_name")
ordered_themes, theme_dict = add_ordered_needs_lists(themes)
context = {
"themes": themes,
"theme_dict": theme_dict,
"ordered_themes": ordered_themes,
"domain": get_domain(8000),
"domain8010": get_domain(),
}
return render_to_response(template, RequestContext(request, context))
def csw_listing(request, template='pycsw_catalog_view.html'):
if logger:
logger.info("Start csw_listing")
csw_recs = pycsw_records.objects.using('pycsw_test').all().order_by('organization')
html_id = 0
for rec in csw_recs:
rec.html_id = html_id
html_id += 1
context = {'records': csw_recs, 'domain': get_domain(8000), 'domain8010': get_domain()}
if logger:
logger.info("End csw_listing")
return render_to_response(template, RequestContext(request, context))
def topic_page(request, topic_name=None, template='topic_page.html'):
topic = get_object_or_404(Topic, name=topic_name)
views = MapView.objects.filter(topic=topic).order_by('ordering')
viewsList = simplejson.dumps([view.name for view in views])
layers = topic.layers.all().order_by('name')
context = {
'topic': topic,
'views': views,
'views_list': viewsList,
'initial_view': views[0].name,
'layers': layers,
'domain': get_domain(8000),
'domain8010': get_domain()
}
return render_to_response(template, RequestContext(request, context))
def get_har(
self,
remove_domain_request=True,
domains_to_remove={
'facebook.com', 'facebook.it', 'youtube.it', 'youtube.com',
'twitter.it', 'twitter.com'
},
file_type_to_remove={'jpg', 'png', 'jpeg'}):
result = list()
if self.logging and self.logs:
domain = None
if remove_domain_request:
domain = utils.get_domain(self.current_url)
for log in self.logs:
message = json.load(StringIO(log['message']))['message']
if 'method' in message:
method = message['method']
if method and method == 'Network.responseReceived':
url = message['params']['response']['url']
if utils.is_valid_url(url):
to_insert = (domain and not utils.is_domain_link(
url, domain)) or domain is None
to_insert = to_insert and utils.get_filetype_from_url(
url) not in file_type_to_remove
if to_insert:
for d in domains_to_remove:
if utils.is_domain_link(url, d):
to_insert = False
break
if to_insert:
result.append(url)
result = list(set(result))
#print('har len: ' + str(len(result)))
return result
def __init__(self, link, base_url):
self.text = self.get_text(link)
self.class_ = self.get_class(link)
self.href = self.get_href(link, base_url)
self.domain = get_domain(self.href)
self.parent = link.parent
self.base_url = base_url
def add(self, cate):
url = cate['url']
domain = get_domain(url)
subdomains = get_subdomains(url)
paths = get_path(url).split('/')
query = urlparse.urlparse(url).query
if domain not in self.root:
self.root[domain] = {'sub':{}, 'path':{}}
node = self.root[domain]
if len(subdomains) > 1 or len(subdomains) == 1 and subdomains[0] != 'www':
for sub in subdomains:
if sub not in node['sub']:
node['sub'][sub] = {'sub':{}, 'path':{}}
node = node['sub'][sub]
for path in paths:
if path not in node['path']:
node['path'][path] = {'path':{}}
node = node['path'][path]
if query:
node['path']['query___' + query] = {'path':{}}
node = node['path']['query___' + query]
node['cate'] = cate
def bookmark_link(self):
if not self.bookmark and self.is_sublayer and self.parent.bookmark:
return self.parent.bookmark.replace('<layer_id>', str(self.id))
if not self.bookmark:
domain = get_domain(8000)
return '%s/planner/#%s' % (domain, self.slug)
return self.bookmark
def bookmark_link(self):
if not self.bookmark and self.is_sublayer and self.parent.bookmark:
return self.parent.bookmark.replace('<layer_id>', str(self.id))
if not self.bookmark:
domain = get_domain(8000)
return '%s/planner/#%s' %(domain, self.slug)
return self.bookmark
def deploy_prometheus_route():
'''Deploy Prometheus Route'''
topic = 'Prometheus Operator Route'
src_file = os.path.join(os.getcwd(),\
"deploy/monitoring/prometheus/assisted-installer-ocp-prometheus-route.yaml")
dst_file = os.path.join(os.getcwd(),\
"build/assisted-installer-ocp-prometheus-route.yaml")
try:
# I have permissions
ingress_domain = utils.get_domain()
except:
# I have not permissions, yes it's ugly...
# This ingress should be there because of UI deployment
json_path_ingress = '{.spec.rules[0].host}'
cmd = "{} get ingress assisted-installer -o jsonpath='{}'".format(
CMD_BIN, json_path_ingress)
assisted_installer_ingress_domain = utils.check_output(cmd)
if assisted_installer_ingress_domain.split(
".")[0] != 'assisted-installer':
print("Error recovering the ingress route")
sys.exit(1)
ingress_domain = assisted_installer_ingress_domain.split(".",
maxsplit=1)[1]
with open(src_file, "r") as src:
with open(dst_file, "w+") as dst:
data = src.read()
data = data.replace("INGRESS_DOMAIN", ingress_domain)
print("Deploying {}: {}".format(topic, dst_file))
dst.write(data)
utils.apply(dst_file)
def top_things(db_file):
urls = {}
people = {}
graph = {}
db = DB()
if not db.open("{0}".format(db_file), DB.OREADER | DB.OCREATE):
print "Could not open database. (Top things)"
cur = db.cursor()
cur.jump_back()
while True:
rec = cur.get(False)
if not rec:
break
loaded_rec = loads(rec[1])
split = get_domain(loaded_rec)
if urls.get(split, False) == False:
urls[split] = 1
else:
urls[split] = urls[split] + 1
person = loaded_rec['person']
if people.get(person, False) == False:
people[person] = 1
else:
people[person] = people[person] + 1
if split is not None and split is not "" and \
person is not None and person is not "":
# Build a crazy relational graph out of my nosql data
if graph.get(split, False) == False:
graph[split] = {
"is_person": False,
"data": [person],
"linked_to_count": 1
}
elif person not in graph[split]:
graph[split]["data"].append(person)
graph[split][
"linked_to_count"] = graph[split]["linked_to_count"] + 1
if graph.get(person, False) == False:
graph[person] = {"is_person": True, "data": [split]}
elif split not in graph[person]:
graph[person]["data"].append(split)
cur.step_back()
cur.disable()
db.close()
def get_one(x):
return x[1]
return (sorted(urls.items(), key=get_one, reverse=True),
sorted(people.items(), key=get_one, reverse=True), graph)
def main():
deploy_options = handle_arguments()
# TODO: delete once rename everything to assisted-installer
if deploy_options.target == "oc-ingress":
service_host = "assisted-installer.{}".format(
utils.get_domain(deploy_options.domain))
service_port = "80"
else:
service_host = utils.get_service_host(
SERVICE, deploy_options.target, namespace=deploy_options.namespace)
service_port = utils.get_service_port(
SERVICE, deploy_options.target, namespace=deploy_options.namespace)
with open(SRC_FILE, "r") as src:
with open(DST_FILE, "w+") as dst:
data = src.read()
data = data.replace("REPLACE_URL", '"{}"'.format(service_host))
data = data.replace("REPLACE_PORT", '"{}"'.format(service_port))
data = data.replace("REPLACE_DOMAINS",
'"{}"'.format(deploy_options.base_dns_domains))
data = data.replace('REPLACE_NAMESPACE', deploy_options.namespace)
print("Deploying {}".format(DST_FILE))
versions = {
"IMAGE_BUILDER": "installer-image-build",
"AGENT_DOCKER_IMAGE": "agent",
"KUBECONFIG_GENERATE_IMAGE":
"ignition-manifests-and-kubeconfig-generate",
"INSTALLER_IMAGE": "assisted-installer",
"CONTROLLER_IMAGE": "assisted-installer-controller",
"CONNECTIVITY_CHECK_IMAGE": "connectivity_check",
"INVENTORY_IMAGE": "inventory"
}
for env_var_name, image_short_name in versions.items():
image_fqdn = deployment_options.get_image_override(
deploy_options, image_short_name, env_var_name)
versions[env_var_name] = image_fqdn
# Edge case for controller image override
if os.environ.get("INSTALLER_IMAGE"
) and not os.environ.get("CONTROLLER_IMAGE"):
versions[
"CONTROLLER_IMAGE"] = deployment_options.IMAGE_FQDN_TEMPLATE.format(
"assisted-installer-controller",
deployment_options.get_tag(
versions["INSTALLER_IMAGE"]))
versions["SELF_VERSION"] = deployment_options.get_image_override(
deploy_options, "assisted-service", "SERVICE")
deploy_tag = get_deployment_tag(deploy_options)
if deploy_tag:
versions["RELEASE_TAG"] = deploy_tag
y = yaml.load(data)
y['data'].update(versions)
data = yaml.dump(y)
dst.write(data)
utils.apply(DST_FILE)
def get_allowed_from(self, child_urls):
"""
:param child_urls: List of child urls to check robots.txt on
:return: A list of allowed child urls to crawl
"""
allowed = []
domains = list(set('{0}'.format(get_domain(url)) for url in child_urls))
domain_to_children = {domain: filter(lambda u: get_domain(u) == domain, child_urls) for domain in domains}
for domain in domain_to_children:
try:
rules = self.robots.fetch(domain)
for url in domain_to_children[domain]:
if rules.allowed(url, self._agent):
allowed.append(url)
except:
allowed.extend(domain_to_children[domain])
return allowed
def introspect(domain):
filter_func = lambda x: get_domain(loads(x[1])).lower() in domain.lower()
pages, requested_page = get_effective_page(request.args.get("page", 0),
filter_func)
items = get_items(filter_func, g.db_file, requested_page)
return render_template("index.html", items=items, pages=pages,
requested_page=requested_page, current_page=request.args.get('page', 0))
def is_image_link(url):
if url.split('.')[-1] in img_extensions:
return True
domain = get_domain(url).split('.')
for sharer in img_sharers:
if sharer in domain:
return True
return False
def add_learn_links(themes):
context = []
domain = get_domain()
for theme in themes:
link = '%s/portal/learn/%s' %(domain, linkify(theme.name))
#print link
context.append({'theme': theme, 'learn_link': link})
return context
def get_error_rate(self, response):
self.out_domains.add(get_domain(response.request.url))
self.crawler.stats.inc_value("no_requests")
if not self.domain.check_request_url(response.request.url):
self.crawler.stats.inc_value('no_new_posts')
self.sum_download_time += response.meta['request_time']
urls = [response.urljoin(url.strip()) for url in response.xpath("//a/@href").getall() if fix_url(url)]
for url in urls:
yield Request(url=url, callback=self.get_error_rate, errback=self.check_error_back_rate)
def __init__(self, name, url):
self.name = name
self.url = url.replace(' ','')
self.html = self.get_html()
self.domain = get_domain(self.url) # e.g. stackoverflow.com
self.base_url = self.get_base_url() # e.g. http://stackoverflow.com/
self.all_links = self.get_all_links()
self.some_valid_links = self.get_some_valid_links()
self.valid_links = self.get_all_valid_links()
def top_things(db_file):
urls = {}
people = {}
graph = {}
db = DB()
if not db.open("{0}".format(db_file), DB.OREADER | DB.OCREATE):
print "Could not open database. (Top things)"
cur = db.cursor()
cur.jump_back()
while True:
rec = cur.get(False)
if not rec:
break
loaded_rec = loads(rec[1])
split = get_domain(loaded_rec)
if urls.get(split, False) == False:
urls[split] = 1
else:
urls[split] = urls[split] + 1
person = loaded_rec['person']
if people.get(person, False) == False:
people[person] = 1
else:
people[person] = people[person] + 1
if split is not None and split is not "" and \
person is not None and person is not "":
# Build a crazy relational graph out of my nosql data
if graph.get(split, False) == False:
graph[split] = {"is_person": False, "data": [person], "linked_to_count": 1}
elif person not in graph[split]:
graph[split]["data"].append(person)
graph[split]["linked_to_count"] = graph[split]["linked_to_count"] + 1
if graph.get(person, False) == False:
graph[person] = {"is_person": True, "data": [split]}
elif split not in graph[person]:
graph[person]["data"].append(split)
cur.step_back()
cur.disable()
db.close()
def get_one(x):
return x[1]
return (sorted(urls.items(), key=get_one, reverse=True),
sorted(people.items(), key=get_one, reverse=True),
graph)
def __init__(self,
url_list,
save_base_dir,
header,
encoding=None,
grab_out_site_link=False,
to_single_page=False,
full_site=False,
ref_model=False,
framework=None):
"""
:param url_list:
:param save_base_dir:
:param header:
:param encoding:
:param grab_out_site_link:
:param to_single_page:
:param full_site:
:param ref_model:
"""
self.parent_save_dir = save_base_dir
self.date_str = get_date()
self.zip_save_base_abs_dir = f"{self.parent_save_dir}/{config.template_archive_dir}/{self.date_str}/" # zip /xx/xx/archive/2019-00-01/
self.download_temp_abs_dir = f"{self.parent_save_dir}/{config.template_temp_dir}/{self.date_str}/"
self.domain = get_domain(url_list[0])
self.tpl_dl_dir, self.js_dir, self.img_dir, self.css_dir, self.other_dir = self.__prepare_dirs(
)
self.dl_urls = {} # 去重使用,存储 url=>磁盘绝对路径
self.error_grab_resource = {} # 记录 http url => disk url ，最后生成一个报告打包
self.header = header
self.charset = encoding
self.is_grab_outer_link = grab_out_site_link
self.is_ref_model = ref_model
if self.is_ref_model:
self.is_grab_outer_link = False # 盗链模式下，一定不抓外部的资源,内部资源也会被改写绝对路径
self.is_to_single_page = to_single_page # 是否把图片，css, js等压缩到一个页面里
self.single_page = [] #report生成时使用
self.is_full_site = full_site #是否是整站
self.html_link_queue = Queue() # html 页面的队列
for u in url_list:
self.html_link_queue.put(u)
self.downloaded_html_url = [
] # 已经下载过的html，保存(disk_path, file_name, url, ), 最后用这个重新修正链接
self.download_queue = Queue(
) # 数据格式json {'cmd':quit/download, "url":'http://baidu.com', "save_path":'/full/path/file.ext', 'type':'bin/text'}
self.download_finished = False # url消耗完毕不代表网络请求都返回了
self.task_finished = False # 全部网络都返回， eventloop结束
self.zip_result_file = None
self.file_name_dup_checker = {
} # file_name => url 。用于检查生成的文件名字是否有重复的，如果重复了就要重新生成了
self.framework_support = framework
self.thread = threading.Thread(target=self.__download_thread)
self.thread.start()
def __init__(self, input, **options):
self.input = input
self.url = options.get('url', '')
self.debug = options.get('debug', False)
self.title = options.get('title', '^^')
self.pages = options.get('pages', None)
self.texts = options.get('texts', None)
self.domain = get_domain(self.url)
self.options = options
self.doc = clean_html(input, return_doc=True)
self.text = self.doc.text_content()
self.len = word_count(self.text) if self.text else 0
def __init__(self, input, **options):
self.input = input
self.url = options.get('url', '')
self.debug = options.get('debug', False)
self.title = options.get('title', '^^')
self.pages = options.get('pages', None)
self.texts = options.get('texts', None)
self.domain = get_domain(self.url)
self.options = options
self.doc = clean_html(input, return_doc=True)
self.text = self.doc.text_content()
self.len = word_count(self.text) if self.text else 0
def get_cookie_syncs_for_multiple_sites(self,
sites,
cookie_length=8,
filepath=''):
"""Get cookie syncing data for multiple sites, and write results to disk.
Cookies must be at least cookie_length characters long to be considered.
"""
sites = self._filter_site_list(sites)
cookie_sync_data = defaultdict(defaultdict)
for site in sites:
cookie_sync_data[site] = self.get_cookie_syncs_by_site(
site, cookie_length=cookie_length)
# Write complete output as csv
with open(os.path.join(filepath, 'full_cookie_syncs.csv'), 'w') as f:
writer = csv.writer(f)
writer.writerow(
['site', 'sending_domain', 'receiving_url', 'cookie_value'])
for site in cookie_sync_data:
for receiving_url in cookie_sync_data[site]:
for sending_url, cookie_value in cookie_sync_data[site][
receiving_url]:
writer.writerow(
[site, sending_url, receiving_url, cookie_value])
# Write partial output as CSV, only identifying sending domain and receiving domain
# (rather than the full receiving URL)
cooks_just_domains = defaultdict(defaultdict)
for site in cookie_sync_data:
cooks_just_domains[site] = defaultdict(set)
for receiving_url in cookie_sync_data[site]:
for sending_domain, value in cookie_sync_data[site][
receiving_url]:
cooks_just_domains[site][utils.get_domain(
receiving_url)].add(sending_domain)
with open(os.path.join(filepath, 'condensed_cookie_syncs.csv'),
'w') as f:
writer = csv.writer(f)
writer.writerow(['site', 'sending_domain', 'receiving_domain'])
for site in cooks_just_domains:
for receiving_domain in cooks_just_domains[site]:
if len(cooks_just_domains[site][receiving_domain]
) > 1 and 'NOT_FOUND' in cooks_just_domains[site][
receiving_domain]:
cooks_just_domains[site][receiving_domain].discard(
'NOT_FOUND')
for sending_domain in cooks_just_domains[site][
receiving_domain]:
writer.writerow(
[site, sending_domain, receiving_domain])
def __init__(self, function):
super(LowAnalyzer, self).__init__()
self.function = function
# Dominio
self.domain = get_domain(self.function)
# Raíces
self.roots = get_roots(self.function)
# Signo
self.negative, self.positive = get_sign(self.function)
def render_template(self,
template_path,
template_context,
to_string=False):
"""
Render a Template to output
"""
from modules.events.internal import api as events_api
# Debug - Show what non-js search engines see
template_context['no_client'] = bool(
self.request.get('no_client', False))
# TODO: This needs to abstract the jinja env out further...
from main import JINJA_ENVIRONMENT
template_context['settings_dict'] = {}
template_context['settings_dict']['is_appspot'] = is_appspot()
template_context['settings_dict']['domain'] = get_domain()
# Tack on the google analytics profiles
if is_appspot(): # TODO: Remove this before being prod ready
template_context['settings_dict'][
'ga_profile_id'] = 'UA-54271335-1'
else:
template_context['settings_dict'][
'ga_profile_id'] = 'UA-54271335-2'
# Temporary Serverside rendering handler - this should be done via jina extensions likely
template_context[
'upcoming_event_resources'] = events_api.get_upcoming_event_resources(
)
template_context[
'ongoing_event_resources'] = events_api.get_ongoing_event_resources(
)
# TODO: This should come from some sort of middleware likely
template_context['settings_dict']['is_authenticated'] = bool(
users.get_current_user())
template_context['settings'] = json.dumps(
template_context['settings_dict'])
template = JINJA_ENVIRONMENT.get_template(template_path)
rendered_content = template.render(template_context)
if to_string:
return rendered_content
self.response.write(rendered_content)
def tiles_page(request, slug=None, template='tiles_page.html'):
layer = get_object_or_404(Layer, slug_name=slug)
orig_url = layer.url
arctile_url = orig_url.replace('{z}',
'{level}').replace('{x}', '{col}').replace(
'{y}', '{row}')
arcrest_url = orig_url.replace('/export', '')
context = {
'layer': layer,
'arctile_url': arctile_url,
'arcrest_url': arcrest_url,
'domain': get_domain(8000)
}
return render_to_response(template, RequestContext(request, context))
def fetch_from(self, urls):
"""
:param urls: A list of urls to fetch sitemaps of
:return: A list of urls that was found within each sitemap of given urls
"""
unique_domains = list(set(get_domain(u) for u in urls))
sitemaps = self._try_fetch_sitemaps(unique_domains)
results = []
for url in sitemaps:
sitemaps_content = self.requests_getter.get_content_from(sitemaps[url])
for content in sitemaps_content:
locations = self.sitemap_url_extractor.extract_from(content)
locations = filter(lambda u: not u.endswith('.xml'), locations)
results.extend(locations)
return results
def article():
url = request.args.get('url')
article = mongo.article.find_one({'_id': url})
if not article:
try:
html = get_or_cache(url)
article = html2article(html, url, selector=True, merge=True)
if article and not article['src_name']:
article['src_name'] = get_domain(url)
tpl = url2tpl(url)
urls = html2urls(html, url)
texts = dict(
map(lambda x: (x[0], max(x[1], key=lambda y: len(y))),
urls.iteritems()))
tmp = dict(map(lambda x: (x, url2tpl(x)), texts.iterkeys()))
urls = {}
for u, t in tmp.iteritems():
if u != url and t == tpl:
urls[u] = texts[u]
if len(urls) >= 10:
break
if article:
article['urls'] = urls
article['_id'] = url
article['view'] = 1
article['last'] = time.time()
copy = article.copy()
copy['urls'] = json.dumps(copy['urls'])
mongo.article.save(copy)
except:
pass
else:
article['urls'] = json.loads(article['urls'])
mongo.article.update({'_id': url},
{'$set': {
'view': article['view'] + 1
}})
if article:
article['pubtime'] = article['pubtime'][:10]
return render_template('extract/article.html', article=article, url=url)
def deploy_grafana_route():
# Deploy Grafana Route
topic = 'Grafana Route'
src_file = os.path.join(
os.getcwd(),
"deploy/monitoring/grafana/assisted-installer-ocp-grafana-route.yaml")
dst_file = os.path.join(os.getcwd(),
"build/assisted-installer-ocp-grafana-route.yaml")
ingress_domain = utils.get_domain()
with open(src_file, "r") as src:
with open(dst_file, "w+") as dst:
data = src.read()
data = data.replace("INGRESS_DOMAIN", ingress_domain)
print("Deploying {}: {}".format(topic, dst_file))
dst.write(data)
utils.apply(dst_file)
def sortFrontier(frontier, domainurl_count):
inlinks_count = [len(i.inlinks) for i in frontier]
inlinks_score = normalize(inlinks_count)
domain_count = [domainurl_count[get_domain(i.url)] for i in frontier]
domain_score = normalize(domain_count)
keyword_score = match_keywords(frontier)
for idx,obj in enumerate(frontier):
final_score = inlinks_score[idx] + domain_score[idx] + keyword_score[idx]
obj.score = final_score
sortedList = sorted(frontier, key=lambda obj: obj.score, reverse=True)
logging.info("Returning sorted list of objects (top 30){}".format([(obj.url, obj.score) for obj in sortedList[:30]]))
return sortedList
def domain_out_domains():
with open('domain_out_domains.jsonl', mode='w') as f:
for file_name in os.listdir('url_outlinks'):
with open(f"url_outlinks/{file_name}") as fp:
domains = json.load(fp)
for domain, url_outlinks in domains.items():
out_domains_set = set()
for outlinks in url_outlinks.values():
for outlink in outlinks:
out_domains_set.add(get_domain(outlink))
f.write(
f"{json.dumps({'domain': domain, 'out_domains': list(out_domains_set)})}\n"
)
fp.close()
f.flush()
f.close()
def fetch_stories(self, correlation_id=-1):
"""Fetches new stories from the datasource. Uses the last story external id to
fetch only new stories."""
try:
url = "http://%s/twitter_sensor/?user=%s&password=%s" % (get_domain(), self.user.user_name, self.user.user_password)
tweets = urllib.urlopen(url).read()
tweets = json.loads(tweets)
print tweets
for key in tweets:
try :
authors = []
authors.append(tweets[key])
self.add_read_story(key, authors)
self.add_user(tweets[key])
except:
log_event("fetch_stories_failed", "AgentCell", self.id, "Adding fetched story %s failed, for %s" % (key, self.user), correlation_id)
except:
log_event("fetch_stories_failed", "AgentCell", self.id, "Failed to fetch stories for %s" % self.user, correlation_id)
def article():
url = request.args.get('url')
article = mongo.article.find_one({'_id':url})
if not article:
try:
html = get_or_cache(url)
article = html2article(html, url, selector=True, merge=True)
if article and not article['src_name']:
article['src_name'] = get_domain(url)
tpl = url2tpl(url)
urls = html2urls(html, url)
texts = dict(map(lambda x: (x[0], max(x[1], key=lambda y:len(y))), urls.iteritems()))
tmp = dict(map(lambda x: (x, url2tpl(x)), texts.iterkeys()))
urls = {}
for u, t in tmp.iteritems():
if u != url and t == tpl:
urls[u] = texts[u]
if len(urls) >= 10:
break
if article:
article['urls'] = urls
article['_id'] = url
article['view'] = 1
article['last'] = time.time()
copy = article.copy()
copy['urls'] = json.dumps(copy['urls'])
mongo.article.save(copy)
except:
pass
else:
article['urls'] = json.loads(article['urls'])
mongo.article.update({'_id':url}, {'$set':{'view':article['view'] + 1}})
if article:
article['pubtime'] = article['pubtime'][:10]
return render_template('extract/article.html', article=article, url=url)
def crawl(self, url, max_page_depth=5, max_external_sites_page_depth=4, request_rate_limit=4):
"""
Will crawl a given url up to max_page_depth and max_external_sites_page_depth on a max rate of
request_rate_limit.
:param url: The to-be crawled url
:param max_page_depth: Max internal (same-domain) depth
:param max_external_sites_page_depth: Max external (different-domain) depth
:param request_rate_limit: Up to n requests at once
:return: List of Url objects (See schemas/url.py)
"""
self._url_scanner.set_request_limit(request_rate_limit)
self._max_page_depth = max_page_depth
self._max_external_sites_page_depth = max_external_sites_page_depth
self._domain = get_domain(url)
self._internal_urls_to_scan.append(url)
self._crawl_internal_urls()
self._crawl_external_urls()
return self._get_crawled_urls()
def main():
# TODO: delete once rename everything to assisted-installer
if args.target == "oc-ingress":
service_host = "assisted-installer.{}".format(
utils.get_domain(args.domain))
service_port = "80"
else:
service_host = utils.get_service_host(SERVICE, args.target)
service_port = utils.get_service_port(SERVICE, args.target)
with open(SRC_FILE, "r") as src:
with open(DST_FILE, "w+") as dst:
data = src.read()
data = data.replace("REPLACE_URL", '"{}"'.format(service_host))
data = data.replace("REPLACE_PORT", '"{}"'.format(service_port))
print("Deploying {}".format(DST_FILE))
if args.deploy_tag is not "":
versions = {
"IMAGE_BUILDER": "quay.io/ocpmetal/installer-image-build:",
"AGENT_DOCKER_IMAGE": "quay.io/ocpmetal/agent:",
"KUBECONFIG_GENERATE_IMAGE":
"quay.io/ocpmetal/ignition-manifests-and-kubeconfig-generate:",
"INSTALLER_IMAGE": "quay.io/ocpmetal/assisted-installer:",
"CONNECTIVITY_CHECK_IMAGE":
"quay.io/ocpmetal/connectivity_check:",
"INVENTORY_IMAGE": "quay.io/ocpmetal/inventory:",
"HARDWARE_INFO_IMAGE": "quay.io/ocpmetal/hardware_info:",
"SELF_VERSION": "quay.io/ocpmetal/installer-image-build:"
}
versions = {
k: v + args.deploy_tag
for k, v in versions.items()
}
y = yaml.load(data)
y['data'].update(versions)
data = yaml.dump(y)
else:
y = yaml.load(data)
y['data'].update({"SELF_VERSION": os.environ.get("SERVICE")})
data = yaml.dump(y)
dst.write(data)
utils.apply(DST_FILE)
def read_seeds(seedfile):
logging.info("Reading the seeds")
frontier_map = {} # url: url object
frontier = {} # waveno : list of url objects
domain_urlcount = {}
wave_no = 1
with open(seedfile, "r") as f:
for line in f:
line = line.strip()
url = clean_url(line)
if validators.url(url) is True:
obj = URL(url, wave_no)
frontier_map[url] = obj
if wave_no not in frontier:
frontier[wave_no] = []
frontier[wave_no].append(obj)
domain_urlcount = checkDomain(get_domain(obj.url), domain_urlcount)
return frontier_map, frontier, domain_urlcount
def deploy_grafana_route():
'''Deploy Grafana Route'''
topic = 'Grafana Route'
src_file = os.path.join(os.getcwd(),\
'deploy/monitoring/grafana/assisted-installer-ocp-grafana-route.yaml')
dst_file = os.path.join(os.getcwd(),\
'build', deploy_options.namespace, 'assisted-installer-ocp-grafana-route.yaml')
try:
# I have permissions
ingress_domain = utils.get_domain(target=deploy_options.target,
namespace=deploy_options.namespace,
profile=deploy_options.profile)
except:
# I have not permissions, yes it's ugly...
# This ingress should be there because of UI deployment
json_path_ingress = '{.spec.rules[0].host}'
cmd = "{} -n {} get ingress assisted-installer -o jsonpath='{}'".format(
CMD_BIN, deploy_options.namespace, json_path_ingress)
assisted_installer_ingress_domain = utils.check_output(cmd)
if assisted_installer_ingress_domain.split(
".")[0] != 'assisted-installer':
print("Error recovering the ingress route")
sys.exit(1)
ingress_domain = assisted_installer_ingress_domain.split(".",
maxsplit=1)[1]
with open(src_file, "r") as src:
with open(dst_file, "w+") as dst:
data = src.read()
data = data.replace("INGRESS_DOMAIN", ingress_domain)
data = data.replace('REPLACE_NAMESPACE',
f'"{deploy_options.namespace}"')
print("Deploying {}: {}".format(topic, dst_file))
dst.write(data)
utils.apply(target=deploy_options.target,
namespace=deploy_options.namespace,
profile=deploy_options.profile,
file=dst_file)
def extract_essence(self, correlation_id):
"""
Analyze the story text, to extract the essence from it. For the essence, look for a matching StoryEssence cell.
If found, link the story cell to the StoryEssence cell. Else create a new StoryEssence cell & link the story to it.
"""
try:
print "extract_essence called for story '%s'" % self.core
client = Client()
response = client.get('http://%s/text_analyzer/extract_essence/' % get_domain(), {'text': self.core}).content
print "got extract essence response: ", response
if response != "":
try :
self.add_essence(response)
except:
print sys.exc_info()
print "essence=", response
log_event("extract_essence_failed", STORY_CELL, self.id, "Adding essence '%s' extracted from story '%s' failed" % (response, self.core), correlation_id)
# all went all, update the flag
self.is_essence_extracted = True
self.save()
except:
print "Failed to extract essence", sys.exc_info()
log_event("extract_essence_failed", STORY_CELL, self.id, "Failed to extract essence from story '%s'" % self.core, correlation_id)
def fetch_stories(self, correlation_id=-1):
"""Fetches new stories from the datasource. Uses the last story external id to
fetch only new stories."""
try:
#url = "http://%s/twitter_sensor/?user=%s&password=%s" % (get_domain(), self.user.user_name, self.user.user_password)
#tweets = urllib.urlopen(url).read()
client = Client()
tweets = client.get('http://%s/twitter_sensor/' % get_domain(), {'user': self.user.user_name, 'password': self.user.user_password}).content
tweets = json.loads(tweets)
print tweets
for key in tweets:
try :
authors = []
for story in StoryCell.objects.all():
if story.core == key:
return
authors.append(tweets[key][0])
self.add_read_story(key, authors)
self.add_user(tweets[key][0])
except:
log_event("fetch_stories_failed", AGENT_CELL, self.id, "Adding fetched story %s failed, for %s" % (key, self.user), correlation_id)
except:
print "Failed to fetch stories", sys.exc_info()
log_event("fetch_stories_failed", AGENT_CELL, self.id, "Failed to fetch stories for %s" % self.user, correlation_id)
def tiles_link(self):
if self.is_shareable and self.layer_type in ['XYZ', 'ArcRest', 'WMS']:
domain = get_domain(8000)
return '%s/explore/%s' %(domain, self.slug)
return None
def extract_concepts(self, correlation_id):
"""
Analyze the story text, to extract named entities. For each named entity, look for a matching Concept cell.
If found, link the concept cell to the story. Else create a new concept cell & link the story to it.
"""
try:
client = Client()
response = client.get('http://%s/text_analyzer/extract_named_entities/' % get_domain(), {'text': self.core}).content
if response != "[]":
named_entities = json.loads(response)
for ne in named_entities:
try :
self.add_concept(ne)
except:
print sys.exc_info()
log_event("extract_concepts_failed", STORY_CELL, self.id, "Adding concept '%s' extracted from story '%s' failed" % (ne[0], self.core), correlation_id)
# all went all, update the flag
self.is_concepts_extracted = True
self.save()
except:
print "Failed to extract concepts", sys.exc_info()
log_event("extract_concepts_failed", STORY_CELL, self.id, "Failed to extract concepts from story '%s'" % self.core, correlation_id)
def filter_func(x):
return get_domain(loads(x[1])).lower() in domain.lower()
# Create A set of results based upon this result set - iterator??
ctx['posts'] = entities
rss_content = self.render_template('./templates/newsfeeds/rss.html', ctx, to_string=True)
# Set Cache
ubercache.cache_set(cache_key, rss_content, category='written')
self.response.headers['Content-Type'] = 'application/xml'
self.response.write(rss_content)
return
# Rest Controllers
resource_url = 'http://' + get_domain() + '/api/posts/%s'
category_resource_url = 'http://' + get_domain() + '/api/post_categories/%s'
CATEGORY_REST_RULES = [
ResourceIdField(output_only=True),
ResourceUrlField(category_resource_url, output_only=True),
SlugField(BlogCategory.slug, required=True),
RestField(BlogCategory.title, required=True),
]
REST_RULES = [
ResourceIdField(output_only=True),
ResourceUrlField(resource_url, output_only=True),
SlugField(BlogPost.slug, required=True),
def data_catalog_bs3(request, template='bs3_catalog.html'):
themes = Theme.objects.all().order_by('display_name')
themes_with_links = add_learn_links(themes)
add_ordered_layers_lists(themes_with_links)
context = {'themes': themes_with_links, 'domain': get_domain(8000), 'domain8010': get_domain()}
return render_to_response(template, RequestContext(request, context))
def data_needs(request, template='needs.html'):
themes = Theme.objects.all().order_by('display_name')
ordered_themes, theme_dict = add_ordered_needs_lists(themes)
context = {'themes': themes, 'theme_dict': theme_dict, 'ordered_themes': ordered_themes, 'domain': get_domain(8000), 'domain8010': get_domain()}
return render_to_response(template, RequestContext(request, context))
def get_user_stats(username, db_file):
item = {
"username": username,
"aliases": [],
"total_posts": 0,
"domains": {},
"first_post_date": None,
"first_post_date_unix": None,
"most_recent_post": None,
"most_recent_post_unix": 0,
"average_posts_per_hour": 0.0,
"average_posts_per_day": 0.0,
"average_posts_per_week": 0.0
}
db = DB()
if not db.open("{0}".format(db_file), DB.OREADER | DB.OCREATE):
print "Could not open database."
cur = db.cursor()
cur.jump()
while True:
rec = cur.get(False)
if not rec:
break
loaded_rec = loads(rec[1])
if loaded_rec['person'] != username:
cur.step()
continue
# Looks like this is a post by the user we're looking for
split = get_domain(loaded_rec)
if item['domains'].get(split, False) == False:
item['domains'][split] = 1
else:
item['domains'][split] = item['domains'][split] + 1
if item['first_post_date_unix'] is None:
item['first_post_date_unix'] = loaded_rec['created_at']
if item['most_recent_post_unix'] < loaded_rec['created_at']:
item['most_recent_post_unix'] = loaded_rec['created_at']
item['total_posts'] = item['total_posts'] + 1
cur.step()
cur.disable()
db.close()
# Clean up everything
first_time = None
if item['first_post_date_unix'] is not None:
unix = float(item['first_post_date_unix'])
first_time = datetime.fromtimestamp(unix)
item['first_post_date'] = first_time.isoformat()
recent_time = None
if item['most_recent_post_unix'] is not None:
unix = float(item['most_recent_post_unix'])
recent_time = datetime.fromtimestamp(unix)
item['most_recent_post'] = recent_time.isoformat()
if first_time and recent_time:
delta = recent_time - first_time
item['user_age_days'] = delta.days
item['user_age_seconds'] = delta.total_seconds()
item['average_posts_per_hour'] = item['total_posts'] / (
delta.total_seconds() / 60.0)
item['average_posts_per_day'] = item['total_posts'] / (
delta.total_seconds() / 60.0 / 24.0)
item['average_posts_per_week'] = item['total_posts'] / (
delta.total_seconds() / 60.0 / 24.0 / 7.0)
return item
def get_user_stats(username, db_file):
item = {
"username": username,
"aliases": [],
"total_posts": 0,
"domains": {},
"first_post_date": None,
"first_post_date_unix": None,
"most_recent_post": None,
"most_recent_post_unix": 0,
"average_posts_per_hour": 0.0,
"average_posts_per_day": 0.0,
"average_posts_per_week": 0.0
}
db = DB()
if not db.open("{0}".format(db_file), DB.OREADER | DB.OCREATE):
print "Could not open database."
cur = db.cursor()
cur.jump()
while True:
rec = cur.get(False)
if not rec:
break
loaded_rec = loads(rec[1])
if loaded_rec['person'] != username:
cur.step()
continue
# Looks like this is a post by the user we're looking for
split = get_domain(loaded_rec)
if item['domains'].get(split, False) == False:
item['domains'][split] = 1
else:
item['domains'][split] = item['domains'][split] + 1
if item['first_post_date_unix'] is None:
item['first_post_date_unix'] = loaded_rec['created_at']
if item['most_recent_post_unix'] < loaded_rec['created_at']:
item['most_recent_post_unix'] = loaded_rec['created_at']
item['total_posts'] = item['total_posts'] + 1
cur.step()
cur.disable()
db.close()
# Clean up everything
first_time = None
if item['first_post_date_unix'] is not None:
unix = float(item['first_post_date_unix'])
first_time = datetime.fromtimestamp(unix)
item['first_post_date'] = first_time.isoformat()
recent_time = None
if item['most_recent_post_unix'] is not None:
unix = float(item['most_recent_post_unix'])
recent_time = datetime.fromtimestamp(unix)
item['most_recent_post'] = recent_time.isoformat()
if first_time and recent_time:
delta = recent_time - first_time
item['user_age_days'] = delta.days
item['user_age_seconds'] = delta.total_seconds()
item['average_posts_per_hour'] = item['total_posts'] / (delta.total_seconds() / 60.0)
item['average_posts_per_day'] = item['total_posts'] / (delta.total_seconds() / 60.0 / 24.0)
item['average_posts_per_week'] = item['total_posts'] / (delta.total_seconds() / 60.0 / 24.0 / 7.0)
return item
def description_link(self):
theme_name = self.themes.all()[0].name
domain = get_domain(8000)
return '%s/learn/%s#%s' %(domain, theme_name, self.slug)
from rest.resource import RestField, SlugField, ResourceIdField, ResourceUrlField
from rest.resource import BooleanField, ResourceField
from rest.params import coerce_to_datetime
from rest.utils import get_key_from_resource_id
from files.rest_helpers import REST_RESOURCE_RULES as FILE_REST_RULES
from modules.events.internal import api as events_api
from modules.events.internal.models import Event
from modules.events.constants import CATEGORY, PRIMARY_IMAGE_PROP
from utils import ubercache
from cal.rest_helpers import EventDateField
from utils import get_domain
resource_url = 'http://' + get_domain() + '/api/events/%s' # TODO: HRM?
# verbosity vs. input vs. output
REST_RULES = [
ResourceIdField(output_only=True),
ResourceUrlField(resource_url, output_only=True),
SlugField(Event.slug, required=True),
RestField(Event.name, required=True),
RestField(Event.url, required=False),
EventDateField(Event.event_dates, required=True),
RestField(Event.content),
RestField(Event.summary),
BooleanField(Event.featured),
RestField(Event.primary_image_resource_id, required=False),
ResourceField(PRIMARY_IMAGE_PROP,
def learn_page(request, theme_name=None, template='learn_page.html'):
topics = Topic.objects.filter(active=True).order_by('ordering')
context = {'topics': topics, 'domain': get_domain(8000), 'domain8010': get_domain()}
return render_to_response(template, RequestContext(request, context))
def topic_page(request, topic_name=None, template='topic_page.html'):
topic = get_object_or_404(Topic, name = topic_name)
views = MapView.objects.filter(topic=topic).order_by('ordering')
viewsList = simplejson.dumps([view.name for view in views])
layers = topic.layers.all().order_by('name')
context = {'topic': topic, 'views': [views[0]], 'views_list': [viewsList[0]], 'initial_view': views[0].name, 'layers': layers, 'domain': get_domain(8000), 'domain8010': get_domain()}
return render_to_response(template, RequestContext(request, context))
def is_filtered_jid(self, user_jid):
if (self._whitelist and user_jid not in self._whitelist
and get_domain(user_jid) not in self._whitelist):
return True
def learn_link(self):
domain = get_domain(8000)
return '%s/learn/%s' %(domain, self.name)
def get_absolute_url(self):
return "http://%s/cells/view/story/%d" % (get_domain(), self.id)