class Crawler(object):
def __init__(self):
"""Create an item representing the 'PeeringDB organization ID' class if
doesn't already exist. And fetch QIDs for organizations already in the
wikibase."""
sys.stderr.write('Initialization...\n')
# Helper for wiki access
self.wh = Wikihandy()
# Get the QID for the item representing the organization IDs
orgid_qid = self.wh.get_qid(
ORGID_LABEL,
create={ # Create it if it doesn't exist
'summary':
'add PeeringDB org IDs', # Commit message
'description':
'Identifier for an organization in the PeeringDB database'
})
# Load the QIDs for organizations already available in the wikibase
self.orgid2qid = self.wh.extid2qid(qid=orgid_qid)
# Added properties will have this reference information
today = self.wh.today()
self.reference = [(self.wh.get_pid('source'),
self.wh.get_qid('PeeringDB')),
(self.wh.get_pid('reference URL'), URL_PDB_ORGS),
(self.wh.get_pid('point in time'), today)]
def run(self):
"""Fetch organizations information from PeeringDB and push to wikibase"""
sys.stderr.write('Fetching PeeringDB data...\n')
req = requests.get(URL_PDB_ORGS)
if req.status_code != 200:
sys.exit('Error while fetching AS names')
organizations = json.loads(req.text)['data']
self.wh.login() # Login once for all threads
for i, res in enumerate(map(self.update_org, organizations)):
sys.stderr.write(f'\rProcessing... {i+1}/{len(organizations)}')
def update_org(self, organization):
"""Add the organization to wikibase if it's not there and update properties"""
# set property name
statements = [[
self.wh.get_pid('instance of'),
self.wh.get_qid('organization')
],
[
self.wh.get_pid('name'),
organization['name'].strip(), self.reference
]]
# set property website
if organization['website']:
statements.append([
self.wh.get_pid('website'), organization['website'],
self.reference
])
# set property country
if organization['country'] in iso3166.countries_by_alpha2:
country_qid = self.wh.get_qid(
iso3166.countries_by_alpha2[organization['country']].name)
if country_qid is not None:
statements.append(
[self.wh.get_pid('country'), country_qid, self.reference])
# Update name, website, and country for this organization
org_qid = self.org_qid(organization)
self.wh.upsert_statements('update peeringDB organization', org_qid,
statements)
return org_qid
def org_qid(self, organization):
"""Find the organization QID or add it to wikibase if it is not yet there.
Return the organization QID."""
# Check if the organization is in the wikibase
if str(organization['id']) not in self.orgid2qid:
# Set properties for this new organization
org_qualifier = [
(self.wh.get_pid('instance of'), self.wh.get_qid(ORGID_LABEL)),
]
statements = [[
self.wh.get_pid('external ID'),
str(organization['id']), [], org_qualifier
]]
# Add this organization to the wikibase
org_qid = self.wh.add_item('add new peeringDB organization',
label=organization['name'],
statements=statements)
# keep track of this QID
self.orgid2qid[str(organization['id'])] = org_qid
return self.orgid2qid[str(organization['id'])]
print('Adding items')
statements = defaultdict(list)
# wikidata = wikihandy.Wikihandy(wikidata_project='wikidata', lang='wikidata')
with open(BASIC_ITEMS_FNAME, 'r') as fp:
csvdata = csv.reader(decomment(fp), skipinitialspace=True)
for row in csvdata:
if not row:
continue
label, description, aliases, statements = [col.strip() for col in row]
print(label)
# Retrive statements from the csv file
# Assume all properties have the 'wikidata-item' datatype
claims = []
for statement in statements.split('|'):
try:
property, target = statement.split(':')
except ValueError:
# skip lines with no statement
continue
claims.append(
[wh.get_pid(property.strip()),
wh.get_qid(target), []])
wh.add_item("bootstrap", label, description, aliases, claims)
class Crawler(object):
def __init__(self):
"""Create an item representing the PeeringDB exchange point ID class if
doesn't already exist. And fetch QIDs for exchange points already in the
wikibase."""
# Helper for wiki access
self.wh = Wikihandy()
# Get the QID of the item representing PeeringDB IX IDs
ixid_qid = self.wh.get_qid(
IXID_LABEL,
create={ # Create it if it doesn't exist
'summary':
'add PeeringDB ix IDs', # Commit message
'description':
'Identifier for an exchange point in the PeeringDB database' # Description
})
# Load the QIDs for ix already available in the wikibase
self.ixid2qid = self.wh.extid2qid(qid=ixid_qid)
# Load the QIDs for peeringDB organizations
self.orgid2qid = self.wh.extid2qid(label=ORGID_LABEL)
# Added properties will have this reference information
self.today = self.wh.today()
self.reference = [(self.wh.get_pid('source'),
self.wh.get_qid('PeeringDB')),
(self.wh.get_pid('reference URL'), URL_PDB_IXS),
(self.wh.get_pid('point in time'), self.today)]
def run(self):
"""Fetch ixs information from PeeringDB and push to wikibase.
Using multiple threads for better performances."""
req = requests.get(URL_PDB_IXS)
if req.status_code != 200:
sys.exit('Error while fetching IXs data')
ixs = json.loads(req.text)['data']
self.wh.login() # Login once for all threads
for i, ix in enumerate(ixs):
# Get more info for this IX
req = requests.get(f'{URL_PDB_IXS}/{ix["id"]}')
if req.status_code != 200:
sys.exit('Error while fetching IXs data')
ix_info = json.loads(req.text)['data'][0]
# Update info in wiki
self.update_ix(ix_info)
sys.stderr.write(f'\rProcessing... {i+1}/{len(ixs)}')
def update_ix(self, ix):
"""Add the ix to wikibase if it's not already there and update its
properties."""
# set property name
statements = [[
self.wh.get_pid('instance of'),
self.wh.get_qid('Internet exchange point')
], [self.wh.get_pid('name'), ix['name'].strip(), self.reference]]
# link to corresponding organization
org_qid = self.orgid2qid.get(str(ix['org_id']))
if org_qid is not None:
statements.append(
[self.wh.get_pid('managed by'), org_qid, self.reference])
else:
print('Error this organization is not in wikibase: ', ix['org_id'])
# set property country
if ix['country']:
country_qid = self.wh.country2qid(ix['country'])
if country_qid is not None:
statements.append(
[self.wh.get_pid('country'), country_qid, self.reference])
# set property website
if ix['website']:
statements.append(
[self.wh.get_pid('website'), ix['website'], self.reference])
# set traffic webpage
if ix['url_stats']:
statements.append([
self.wh.get_pid('website'),
ix['url_stats'], # statement
self.reference, # reference
[
(self.wh.get_pid('instance of'),
self.wh.get_qid('traffic statistics')),
] # qualifier
])
ix_qid = self.ix_qid(ix)
# Update name, website, and organization for this IX
self.wh.upsert_statements('update peeringDB ixs', ix_qid, statements)
# update LAN corresponding to this IX
if 'ixlan_set' in ix:
for ixlan in ix['ixlan_set']:
pfx_url = f'{URL_PDB_LAN}/{ixlan["id"]}'
pfx_ref = [(self.wh.get_pid('source'),
self.wh.get_qid('PeeringDB')),
(self.wh.get_pid('reference URL'), pfx_url),
(self.wh.get_pid('point in time'), self.today)]
req = requests.get(pfx_url)
if req.status_code != 200:
sys.exit('Error while fetching IXs data')
lans = json.loads(req.text)['data']
for lan in lans:
for prefix in lan['ixpfx_set']:
pfx_qid = self.wh.prefix2qid(prefix['prefix'],
create=True)
pfx_stmts = [[
self.wh.get_pid('instance of'),
self.wh.get_qid('peering LAN'), pfx_ref
], [self.wh.get_pid('managed by'), ix_qid, pfx_ref]]
self.wh.upsert_statements('update peeringDB ixlan',
pfx_qid, pfx_stmts)
return ix_qid
def ix_qid(self, ix):
"""Find the ix QID for the given ix.
If this ix is not yet registered in the wikibase then add it.
Return the ix QID."""
# Check if the IX is in the wikibase
if str(ix['id']) not in self.ixid2qid:
# Set properties for this new ix
ix_qualifiers = [
(self.wh.get_pid('instance of'), self.wh.get_qid(IXID_LABEL)),
]
statements = [(self.wh.get_pid('instance of'),
self.wh.get_qid('Internet exchange point')),
(self.wh.get_pid('external ID'), str(ix['id']), [],
ix_qualifiers)]
# Add this ix to the wikibase
ix_qid = self.wh.add_item('add new peeringDB IX',
label=ix['name'],
description=ix['name_long'],
statements=statements)
# keep track of this QID
self.ixid2qid[str(ix['id'])] = ix_qid
return self.ixid2qid[str(ix['id'])]
class Crawler(object):
def __init__(self):
"""
"""
# Helper for wiki access
self.wh = Wikihandy(preload=True)
# Get the QID for RIPE Atlas
self.atlas_qid = self.wh.get_qid(
'RIPE Atlas',
create={ # Create it if it doesn't exist
'summary':
'add RIPE Atlas', # Commit message
'description':
'RIPE Atlas is a global, open, distributed Internet measurement platform, consisting of thousands of measurement devices that measure Internet connectivity in real time.', # Item description
'aliases':
'Atlas|atlas',
'statements':
[[self.wh.get_pid('managed by'),
self.wh.get_qid('RIPE NCC')]]
})
# Get the QID for Atlas Probe
self.atlas_probe_qid = self.wh.get_qid(
'Atlas probe',
create={ # Create it if it doesn't exist
'summary': 'add RIPE Atlas', # Commit message
'description':
'RIPE Atlas probes form the backbone of the RIPE Atlas infrastructure.', # Item description
'aliases': 'RIPE Atlas probe|atlas probe|RIPE atlas probe',
'statements': [[self.wh.get_pid('part of'), self.atlas_qid]]
})
# Get the QID for Atlas Anchor
self.atlas_anchor_qid = self.wh.get_qid(
'Atlas anchor',
create={ # Create it if it doesn't exist
'summary': 'add RIPE Atlas', # Commit message
'description':
'RIPE Atlas Anchors are located at hosts that can provide sufficient bandwidth to support a large number of incoming and outgoing measurements.', # Item description
'aliases': 'RIPE Atlas anchor|atlas anchor|RIPE atlas anchor',
'statements': [[self.wh.get_pid('part of'), self.atlas_qid]]
})
# Get the QID of the item representing PeeringDB IX IDs
self.probeid_qid = self.wh.get_qid(
PROBEID_LABEL,
create={ # Create it if it doesn't exist
'summary':
'add RIPE Atlas probes', # Commit message
'description':
'Identifier for a probe in the RIPE Atlas measurement platform' # Description
})
# Load the QIDs for probes already available in the wikibase
self.probeid2qid = self.wh.extid2qid(qid=self.probeid_qid)
# Added properties will have this additional information
today = self.wh.today()
self.reference = [(self.wh.get_pid('source'),
self.wh.get_qid('RIPE NCC')),
(self.wh.get_pid('reference URL'), URL),
(self.wh.get_pid('point in time'), today)]
self.v4_qualifiers = [(self.wh.get_pid('IP version'),
self.wh.get_qid('IPv4'))]
self.v6_qualifiers = [(self.wh.get_pid('IP version'),
self.wh.get_qid('IPv6'))]
def run(self):
"""Fetch probe information from Atlas API and push to wikibase. """
next_page = URL
while next_page is not None:
req = requests.get(next_page)
if req.status_code != 200:
sys.exit('Error while fetching the blocklist')
info = json.loads(req.text)
next_page = info['next']
for i, probe in enumerate(info['results']):
self.update_probe(probe)
sys.stderr.write(f'\rProcessed {i+1} probes')
sys.stderr.write(f'\n')
def update_probe(self, probe):
"""Add the probe to wikibase if it's not already there and update its
properties."""
# TODO add status, geometry (geo-location) and IPs?
# Properties for this probe
statements = []
if probe['is_anchor']:
statements.append(
[self.wh.get_pid('instance of'), self.atlas_probe_qid])
statements.append(
[self.wh.get_pid('instance of'), self.atlas_anchor_qid])
if probe['asn_v4']:
as_qid = self.wh.asn2qid(probe['asn_v4'])
if as_qid:
statements.append([
self.wh.get_pid('part of'), as_qid, self.reference,
self.v4_qualifiers
])
if probe['asn_v6']:
as_qid = self.wh.asn2qid(probe['asn_v6'])
if as_qid:
statements.append([
self.wh.get_pid('part of'), as_qid, self.reference,
self.v6_qualifiers
])
if probe['prefix_v4']:
prefix_qid = self.wh.prefix2qid(probe['prefix_v4'])
if prefix_qid:
statements.append(
[self.wh.get_pid('part of'), prefix_qid, self.reference])
if probe['prefix_v6']:
prefix_qid = self.wh.prefix2qid(probe['prefix_v6'])
if prefix_qid:
statements.append(
[self.wh.get_pid('part of'), prefix_qid, self.reference])
if probe['country_code']:
statements.append([
self.wh.get_pid('country'),
self.wh.country2qid(probe['country_code']), self.reference
])
if probe['first_connected']:
statements.append([
self.wh.get_pid('start time'),
self.wh.to_wbtime(probe['first_connected']), self.reference
])
if 'name' in probe['status']:
# Get the QIDs for probes status
status_qid = self.wh.get_qid(
f'RIPE Atlas probe status: {probe["status"]["name"]}',
create={ # Create it if it doesn't exist
'summary': 'add RIPE Atlas probe status', # Commit message
})
if probe['status_since']:
statements.append([
self.wh.get_pid('status'), status_qid, self.reference,
[(self.wh.get_pid('start time'),
self.wh.to_wbtime(probe['status_since']))]
])
# set end time if the probe is abandonned
if probe['status']['name'] == 'Abandoned' and probe['status_since']:
statements.append([
self.wh.get_pid('end time'),
self.wh.to_wbtime(probe['status_since'])
])
# Add probe tags
for tag in probe['tags']:
statements.append([
self.wh.get_pid('tag'),
self.wh.get_qid(tag['name'],
create={
'summary': 'Add RIPE Atlas tag',
})
])
# Commit to wikibase
# Get the probe QID (create if probe is not yet registered) and commit changes
probe_qid = self.probe_qid(probe)
self.wh.upsert_statements('update from RIPE Atlas probes', probe_qid,
statements)
def probe_qid(self, probe):
"""Find the ix QID for the given probe ID.
If this probe is not yet registered in the wikibase then add it.
Return the probe QID."""
id = str(probe['id'])
# Check if the IX is in the wikibase
if id not in self.probeid2qid:
# Set properties for this new probe
probeid_qualifiers = [
(self.wh.get_pid('instance of'), self.probeid_qid),
]
statements = [
(self.wh.get_pid('instance of'), self.atlas_probe_qid),
(self.wh.get_pid('external ID'), id, [], probeid_qualifiers)
]
# Add this probe to the wikibase
probe_qid = self.wh.add_item('add new RIPE Atlas probe',
label=f'RIPE Atlas probe #{id}',
description=probe['description'],
statements=statements)
# keep track of this QID
self.probeid2qid[id] = probe_qid
return self.probeid2qid[id]