def _parse_dirreq_line(keyword, recognized_counts_attr,
unrecognized_counts_attr, descriptor, entries):
value = _value(keyword, entries)
recognized_counts = {}
unrecognized_counts = {}
is_response_stats = keyword in ('dirreq-v2-resp', 'dirreq-v3-resp')
key_set = DirResponse if is_response_stats else DirStat
key_type = 'STATUS' if is_response_stats else 'STAT'
for status, count in _mappings_for(keyword, value, divider=','):
if not count.isdigit():
raise ValueError(
'%s lines should contain %s=COUNT mappings: %s %s' %
(keyword, key_type, keyword, value))
if status in key_set:
recognized_counts[status] = int(count)
else:
unrecognized_counts[status] = int(count)
setattr(descriptor, recognized_counts_attr, recognized_counts)
setattr(descriptor, unrecognized_counts_attr, unrecognized_counts)
def _parse_hs_stats(keyword, stat_attribute, extra_attribute, descriptor,
entries):
# "<keyword>" num key=val key=val...
value, stat, extra = _value(keyword, entries), None, {}
if value is None:
pass # not in the descriptor
elif value == '':
raise ValueError("'%s' line was blank" % keyword)
else:
if ' ' in value:
stat_value, remainder = value.split(' ', 1)
else:
stat_value, remainder = value, None
try:
stat = int(stat_value)
except ValueError:
raise ValueError("'%s' stat was non-numeric (%s): %s %s" %
(keyword, stat_value, keyword, value))
for key, val in _mappings_for(keyword, remainder):
extra[key] = val
setattr(descriptor, stat_attribute, stat)
setattr(descriptor, extra_attribute, extra)
def _parse_body(descriptor: 'stem.descriptor.Descriptor',
entries: ENTRY_TYPE) -> None:
# In version 1.0.0 the body is everything after the first line. Otherwise
# it's everything after the header's divider.
content = io.BytesIO(descriptor.get_bytes())
if descriptor.version == '1.0.0':
content.readline() # skip the first line
else:
while content.readline().strip() not in ('', HEADER_DIV,
HEADER_DIV_ALT):
pass # skip the header
measurements = {}
for line_bytes in content.readlines():
line = stem.util.str_tools._to_unicode(line_bytes.strip())
attr = dict(_mappings_for('measurement', line))
fingerprint = attr.get('node_id', '').lstrip(
'$') # bwauths prefix fingerprints with '$'
if not fingerprint:
raise ValueError("Every meaurement must include 'node_id': %s" %
stem.util.str_tools._to_unicode(line))
elif fingerprint in measurements:
raise ValueError(
'Relay %s is listed multiple times. It should only be present once.'
% fingerprint)
measurements[fingerprint] = attr
descriptor.measurements = measurements
def _parse_bridge_ip_transports_line(descriptor, entries):
value, ip_transports = _value('bridge-ip-transports', entries), {}
for protocol, count in _mappings_for('bridge-ip-transports', value, divider = ','):
if not count.isdigit():
raise stem.ProtocolError('Transport count was non-numeric (%s): bridge-ip-transports %s' % (count, value))
ip_transports[protocol] = int(count)
descriptor.ip_transports = ip_transports
def _parse_bridge_ip_versions_line(descriptor: 'stem.descriptor.Descriptor', entries: ENTRY_TYPE) -> None:
value, ip_versions = _value('bridge-ip-versions', entries), {}
for protocol, count in _mappings_for('bridge-ip-versions', value, divider = ','):
if not count.isdigit():
raise stem.ProtocolError('IP protocol count was non-numeric (%s): bridge-ip-versions %s' % (count, value))
ip_versions[protocol] = int(count)
descriptor.ip_versions = ip_versions
def _parse_port_count_line(keyword, attribute, descriptor, entries):
# "<keyword>" port=N,port=N,...
value, port_mappings = _value(keyword, entries), {}
for port, stat in _mappings_for(keyword, value, divider = ','):
if (port != 'other' and not stem.util.connection.is_valid_port(port)) or not stat.isdigit():
raise ValueError('Entries in %s line should only be PORT=N entries: %s %s' % (keyword, keyword, value))
port = int(port) if port.isdigit() else port
port_mappings[port] = int(stat)
setattr(descriptor, attribute, port_mappings)
def _parse_padding_counts_line(descriptor, entries):
# "padding-counts" YYYY-MM-DD HH:MM:SS (NSEC s) key=val key=val...
value = _value('padding-counts', entries)
timestamp, interval, remainder = _parse_timestamp_and_interval('padding-counts', value)
counts = {}
for k, v in _mappings_for('padding-counts', remainder, require_value = True):
counts[k] = int(v) if v.isdigit() else v
setattr(descriptor, 'padding_counts_end', timestamp)
setattr(descriptor, 'padding_counts_interval', interval)
setattr(descriptor, 'padding_counts', counts)
def _parse_port_count_line(keyword: str, attribute: str, descriptor: 'stem.descriptor.Descriptor', entries: ENTRY_TYPE) -> None:
# "<keyword>" port=N,port=N,...
value, port_mappings = _value(keyword, entries), {}
for port, stat in _mappings_for(keyword, value, divider = ','):
if (port != 'other' and not stem.util.connection.is_valid_port(port)) or not stat.isdigit():
raise ValueError('Entries in %s line should only be PORT=N entries: %s %s' % (keyword, keyword, value))
port = int(port) if port.isdigit() else port # type: ignore # this can be an int or 'other'
port_mappings[port] = int(stat)
setattr(descriptor, attribute, port_mappings)
def _parse_geoip_to_count_line(keyword, attribute, descriptor, entries):
# "<keyword>" CC=N,CC=N,...
#
# The maxmind geoip (https://www.maxmind.com/app/iso3166) has numeric
# locale codes for some special values, for instance...
# A1,"Anonymous Proxy"
# A2,"Satellite Provider"
# ??,"Unknown"
value, locale_usage = _value(keyword, entries), {}
for locale, count in _mappings_for(keyword, value, divider = ','):
if not _locale_re.match(locale) or not count.isdigit():
raise ValueError('Entries in %s line should only be CC=N entries: %s %s' % (keyword, keyword, value))
locale_usage[locale] = int(count)
setattr(descriptor, attribute, locale_usage)