July 2012 - tor-commits - lists.torproject.org

[metrics-tasks/master] Ignore a consensus without relays even sooner (#6232).
by karsten＠torproject.org 24 Jul '12

24 Jul '12

commit b8a3574dd5d5b96560d366b0082413484d7f05cc Author: Karsten Loesing <karsten.loesing(a)gmx.net> Date: Tue Jul 24 10:00:10 2012 +0200 Ignore a consensus without relays even sooner (#6232). --- task-6232/pyentropy.py | 6 +++--- 1 files changed, 3 insertions(+), 3 deletions(-) diff --git a/task-6232/pyentropy.py b/task-6232/pyentropy.py index 76a3d52..dd5e343 100644 --- a/task-6232/pyentropy.py +++ b/task-6232/pyentropy.py @@ -109,6 +109,9 @@ def run(file_name): except: pass + if len(routers) <= 0: + return + total_bw, total_exit_bw, total_guard_bw = 0, 0, 0 guards_no, exits_no = 0, 0 bw_countries, bw_as = {}, {} @@ -135,9 +138,6 @@ def run(file_name): else: bw_as[router.as_no] = router.bandwidth - if len(routers) <= 0: - return - entropy, entropy_exit, entropy_guard, entropy_country, entropy_as = 0.0, 0.0, 0.0, 0.0, 0.0 for router in routers: p = float(router.bandwidth) / float(total_bw)

1 0

[metrics-tasks/master] Skip consensuses that don't contain consensus weights (#6232).
by karsten＠torproject.org 24 Jul '12

24 Jul '12

commit f5a5bad9dac005e416e09a97e4512ac7bf0a667f Author: Karsten Loesing <karsten.loesing(a)gmx.net> Date: Tue Jul 24 10:57:10 2012 +0200 Skip consensuses that don't contain consensus weights (#6232). --- task-6232/pyentropy.py | 5 +++++ 1 files changed, 5 insertions(+), 0 deletions(-) diff --git a/task-6232/pyentropy.py b/task-6232/pyentropy.py index 8e28210..e6d3214 100644 --- a/task-6232/pyentropy.py +++ b/task-6232/pyentropy.py @@ -116,6 +116,8 @@ def run(file_name): guards_no, exits_no = 0, 0 bw_countries, bw_as = {}, {} for router in routers: + if not router.bandwidth: + continue total_bw += router.bandwidth if router.is_guard and router.is_exit: total_guard_bw += Wgd*router.bandwidth @@ -137,6 +139,9 @@ def run(file_name): else: bw_as[router.as_no] = router.bandwidth + if total_bw == 0: + return + entropy, entropy_exit, entropy_guard, entropy_country, entropy_as = 0.0, 0.0, 0.0, 0.0, 0.0 for router in routers: p = float(router.bandwidth) / float(total_bw)

1 0

[metrics-tasks/master] Treat unresolved AS numbers correctly (#6232).
by karsten＠torproject.org 24 Jul '12

24 Jul '12

commit 4275170f5cbcd7332df77f7ca7a36618c27c269c Author: Karsten Loesing <karsten.loesing(a)gmx.net> Date: Tue Jul 24 10:12:07 2012 +0200 Treat unresolved AS numbers correctly (#6232). So far, relays that couldn't be found in the AS database were ignored in the AS entropy calculation. That's not correct. We have to treat them as "other" AS, similar to how we handle relays for which we couldn't find the country code. Otherwise the probability of all ASes wouldn't sum up to 1. --- task-6232/pyentropy.py | 11 +++++------ 1 files changed, 5 insertions(+), 6 deletions(-) diff --git a/task-6232/pyentropy.py b/task-6232/pyentropy.py index dd5e343..8e28210 100644 --- a/task-6232/pyentropy.py +++ b/task-6232/pyentropy.py @@ -56,7 +56,7 @@ class Router: value = as_db.org_by_addr(str(ip)).split() return value[0] except: - return None + return "" def get_advertised_bw(self, hex_digest): try: @@ -132,11 +132,10 @@ def run(file_name): bw_countries[router.country] += router.bandwidth else: bw_countries[router.country] = router.bandwidth - if router.as_no: - if bw_as.has_key(router.as_no): - bw_as[router.as_no] += router.bandwidth - else: - bw_as[router.as_no] = router.bandwidth + if bw_as.has_key(router.as_no): + bw_as[router.as_no] += router.bandwidth + else: + bw_as[router.as_no] = router.bandwidth entropy, entropy_exit, entropy_guard, entropy_country, entropy_as = 0.0, 0.0, 0.0, 0.0, 0.0 for router in routers:

1 0

[metrics-tasks/master] Improve parser, style
by karsten＠torproject.org 24 Jul '12

24 Jul '12

commit 7e78fbbc80faf35a2bf770ce2ed766e13d639e60 Author: Sathyanarayanan Gunasekaran <gsathya.ceg(a)gmail.com> Date: Sun Jul 22 22:42:09 2012 +0530 Improve parser, style Parser doesn't need 'p' keyword anymore. --- task-6232/pyentropy.py | 21 +++++++++++++-------- 1 files changed, 13 insertions(+), 8 deletions(-) diff --git a/task-6232/pyentropy.py b/task-6232/pyentropy.py index 448cb44..6bd5132 100644 --- a/task-6232/pyentropy.py +++ b/task-6232/pyentropy.py @@ -87,6 +87,7 @@ class Router: def run(file_name): routers = [] + router = None Wed, Wee, Wgd, Wgg = 1, 1, 1, 1 # parse consensus with open(file_name, 'r') as f: @@ -94,11 +95,10 @@ def run(file_name): key = line.split()[0] values = line.split()[1:] if key =='r': + if router: + routers.append(router) router = Router() router.add(key, values) - elif key == 'p': - router.add(key, values) - routers.append(router) elif key == 'valid-after': valid_after = ' '.join(values) elif key == 'bandwidth-weights': @@ -190,11 +190,16 @@ def parse_args(): usage = "Usage - python pyentropy.py [options]" parser = OptionParser(usage) - parser.add_option("-g", "--geoip", dest="gi_db", default="GeoIP.dat", help="Input GeoIP database") - parser.add_option("-a", "--as", dest="as_db", default="GeoIPASNum.dat", help="Input AS GeoIP database") - parser.add_option("-s", "--server_desc", dest="server_desc", default="data/relay-descriptors/server-descriptors/", help="Server descriptors directory") - parser.add_option("-o", "--output", dest="output", default="entropy.csv", help="Output filename") - parser.add_option("-c", "--consensus", dest="consensus", default="in/consensus", help="Input consensus dir") + parser.add_option("-g", "--geoip", dest="gi_db", default="GeoIP.dat", + help="Input GeoIP database") + parser.add_option("-a", "--as", dest="as_db", default="GeoIPASNum.dat", + help="Input AS GeoIP database") + parser.add_option("-s", "--server_desc", dest="server_desc", + default="data/relay-descriptors/server-descriptors/", help="Server descriptors directory") + parser.add_option("-o", "--output", dest="output", default="entropy.csv", + help="Output filename") + parser.add_option("-c", "--consensus", dest="consensus", default="in/consensus", + help="Input consensus dir") (options, args) = parser.parse_args()

1 0

[metrics-tasks/master] Add argument parsing, delete GeoIP.dat
by karsten＠torproject.org 24 Jul '12

24 Jul '12

commit 14c41dc65d51b3723e38c2e01f9d529145cfbdc4 Author: Sathyanarayanan Gunasekaran <gsathya.ceg(a)gmail.com> Date: Thu Jul 19 22:39:07 2012 +0530 Add argument parsing, delete GeoIP.dat Following options were added - -o, --output = Output filename -c, --consensus = Input consensus dir -g, --geoip = GeoIP database Removed GeoIP.dat from the git repo. --- task-6232/GeoIP.dat | Bin 1343110 -> 0 bytes task-6232/pyentropy.py | 32 ++++++++++++++++++++------------ 2 files changed, 20 insertions(+), 12 deletions(-) diff --git a/task-6232/GeoIP.dat b/task-6232/GeoIP.dat deleted file mode 100644 index 00510d5..0000000 Binary files a/task-6232/GeoIP.dat and /dev/null differ diff --git a/task-6232/pyentropy.py b/task-6232/pyentropy.py index ae28f05..1a981fe 100644 --- a/task-6232/pyentropy.py +++ b/task-6232/pyentropy.py @@ -1,5 +1,5 @@ """ -Usage - python pyentropy.py <consensus-dir> <output-file> +Usage - python pyentropy.py -h Output - A CSV file of the format (without newlines): <valid-after>, <entropy for all nodes>, @@ -17,7 +17,7 @@ import sys import math import os import pygeoip -import getopt +from optparse import OptionParser KEYS = ['r','s','v','w','p','m'] @@ -119,17 +119,25 @@ def run(file_name): str(entropy_country), str(max_entropy_country)]) -def usage(): - print "Usage - python pyentropy.py <consensus-dir> <output-file>" +def parse_args(): + usage = "Usage - python pyentropy.py [options]" + parser = OptionParser(usage) + + parser.add_option("-g", "--geoip", dest="geoip", default="GeoIP.dat", help="Input GeoIP database") + parser.add_option("-o", "--output", dest="output", default="entropy.csv", help="Output filename") + parser.add_option("-c", "--consensus", dest="consensus", default="in/consensus", help="Input consensus dir") + + (options, args) = parser.parse_args() + + return options if __name__ == "__main__": - if len(sys.argv) != 3: - usage() - sys.exit() - - gi = pygeoip.GeoIP(os.path.join(os.path.dirname(__file__), 'GeoIP.dat')) - with open(sys.argv[2], 'w') as f: - for file_name in os.listdir(sys.argv[1]): - string = run(os.path.join(sys.argv[1], file_name)) + + options = parse_args() + gi = pygeoip.GeoIP(options.geoip) + + with open(options.output, 'w') as f: + for file_name in os.listdir(options.consensus): + string = run(os.path.join(options.consensus, file_name)) if string: f.write("%s\n" % (string))

1 0

[metrics-tasks/master] Use advertised bandwidths
by karsten＠torproject.org 24 Jul '12

24 Jul '12

commit cabd3c6397f3fdc7a50efb33b847d73d0e54b703 Author: Sathyanarayanan Gunasekaran <gsathya.ceg(a)gmail.com> Date: Sun Jul 22 14:49:14 2012 +0530 Use advertised bandwidths Parse the server descriptors and find the advertised bandwidths --- task-6232/pyentropy.py | 86 ++++++++++++++++++++++++++++++++++------------- 1 files changed, 62 insertions(+), 24 deletions(-) diff --git a/task-6232/pyentropy.py b/task-6232/pyentropy.py index 4328f98..448cb44 100644 --- a/task-6232/pyentropy.py +++ b/task-6232/pyentropy.py @@ -21,17 +21,21 @@ import os import pygeoip import StringIO import stem.descriptor -from stem.descriptor.server_descriptor import RelayDescriptor, BridgeDescriptor -from binascii import b2a_hex, a2b_base64, a2b_hex + from optparse import OptionParser +from binascii import b2a_hex, a2b_base64, a2b_hex +from stem.descriptor.server_descriptor import RelayDescriptor, BridgeDescriptor -KEYS = ['r','s','v','w','p','m'] +KEYS = ['r', 's', 'v', 'w','p', 'm'] class Router: def __init__(self): self.lines = [] self.nick = None + self.digest = None + self.hex_digest = None self.bandwidth = None + self.advertised_bw = None self.flags = None self.probability = None self.ip = None @@ -40,31 +44,50 @@ class Router: self.as_name = None self.is_exit = None self.is_guard = None - + def add(self, key, values): if key == 'r': self.nick = values[0] + self.digest = values[2] + self.hex_digest = b2a_hex(a2b_base64(self.digest+"=")) self.ip = values[5] self.country = gi_db.country_name_by_addr(self.ip) self.as_no, self.as_name = self.get_as_details() if key == 'w': - self.bandwidth = int(values[0].split('=')[1]) + self.advertised_bw = self.get_advertised_bw() + if self.advertised_bw: + self.bandwidth = self.advertised_bw + else: + self.bandwidth = int(values[0].split('=')[1]) if key == 's': self.flags = values if "Exit" in self.flags: self.is_exit = True if "Guard" in self.flags: self.is_guard = True - + def get_as_details(self): try: value = as_db.org_by_addr(str(self.ip)).split() return value[0], value[1] except: return None, None - + + def get_advertised_bw(self): + try: + with open(options.server_desc+self.hex_digest) as f: + data = f.read() + + desc_iter = stem.descriptor.server_descriptor.parse_file(StringIO.StringIO(data)) + desc_entries = list(desc_iter) + desc = desc_entries[0] + return min(desc.average_bandwidth, desc.burst_bandwidth, desc.observed_bandwidth) + except: + return None + def run(file_name): routers = [] + Wed, Wee, Wgd, Wgg = 1, 1, 1, 1 # parse consensus with open(file_name, 'r') as f: for line in f.readlines(): @@ -78,20 +101,28 @@ def run(file_name): routers.append(router) elif key == 'valid-after': valid_after = ' '.join(values) + elif key == 'bandwidth-weights': + Wed = float(values[6].split('=')[1]) / 10000 + Wee = float(values[7].split('=')[1]) / 10000 + Wgd = float(values[11].split('=')[1]) / 10000 + Wgg = float(values[12].split('=')[1]) / 10000 elif key in KEYS: router.add(key, values) - + total_bw, total_exit_bw, total_guard_bw = 0, 0, 0 - guards_no, exists_no = 0, 0 + guards_no, exits_no = 0, 0 bw_countries, bw_as = {}, {} for router in routers: total_bw += router.bandwidth - if router.is_guard: - total_guard_bw += router.bandwidth + if router.is_guard and router.is_exit: + total_guard_bw += Wgd*router.bandwidth + total_exit_bw += Wed*router.bandwidth + elif router.is_guard: + total_guard_bw += Wgg*router.bandwidth guards_no += 1 - if router.is_exit: - total_exit_bw += router.bandwidth - exists_no += 1 + elif router.is_exit: + total_exit_bw += Wee*router.bandwidth + exits_no += 1 if bw_countries.has_key(router.country): bw_countries[router.country] += router.bandwidth else: @@ -101,7 +132,7 @@ def run(file_name): bw_as[router.as_no] += router.bandwidth else: bw_as[router.as_no] = router.bandwidth - + if len(routers) <= 0: return @@ -110,12 +141,19 @@ def run(file_name): p = float(router.bandwidth) / float(total_bw) if p != 0: entropy += -(p * math.log(p, 2)) - if router.is_guard: - p = float(router.bandwidth) / float(total_guard_bw) + if router.is_guard and router.is_exit: + p = float(Wgd*router.bandwidth) / float(total_guard_bw) + if p != 0: + entropy_guard += -(p * math.log(p, 2)) + p = float(Wed*router.bandwidth) / float(total_exit_bw) + if p != 0: + entropy_exit += -(p * math.log(p, 2)) + elif router.is_guard: + p = float(Wgg*router.bandwidth) / float(total_guard_bw) if p != 0: entropy_guard += -(p * math.log(p, 2)) - if router.is_exit: - p = float(router.bandwidth) / float(total_exit_bw) + elif router.is_exit: + p = float(Wee*router.bandwidth) / float(total_exit_bw) if p != 0: entropy_exit += -(p * math.log(p, 2)) @@ -132,7 +170,7 @@ def run(file_name): # Entropy of uniform distribution of 'n' possible values: log(n) max_entropy = math.log(len(routers), 2) max_entropy_guard = math.log(guards_no, 2) - max_entropy_exit = math.log(exists_no, 2) + max_entropy_exit = math.log(exits_no, 2) max_entropy_country = math.log(len(bw_countries), 2) max_entropy_as = math.log(len(bw_as), 2) @@ -151,18 +189,18 @@ def run(file_name): def parse_args(): usage = "Usage - python pyentropy.py [options]" parser = OptionParser(usage) - + parser.add_option("-g", "--geoip", dest="gi_db", default="GeoIP.dat", help="Input GeoIP database") parser.add_option("-a", "--as", dest="as_db", default="GeoIPASNum.dat", help="Input AS GeoIP database") + parser.add_option("-s", "--server_desc", dest="server_desc", default="data/relay-descriptors/server-descriptors/", help="Server descriptors directory") parser.add_option("-o", "--output", dest="output", default="entropy.csv", help="Output filename") parser.add_option("-c", "--consensus", dest="consensus", default="in/consensus", help="Input consensus dir") - + (options, args) = parser.parse_args() - + return options if __name__ == "__main__": - options = parse_args() gi_db = pygeoip.GeoIP(options.gi_db) as_db = pygeoip.GeoIP(options.as_db)

1 0

[metrics-tasks/master] Find entropy based on AS No
by karsten＠torproject.org 24 Jul '12

24 Jul '12

commit 687d1e8b90e5a413aa56cec17abab097745bde8e Author: Sathyanarayanan Gunasekaran <gsathya.ceg(a)gmail.com> Date: Sat Jul 21 13:26:31 2012 +0530 Find entropy based on AS No Use -as to specify the AS db --- task-6232/pyentropy.py | 55 +++++++++++++++++++++++++++++++++++++---------- 1 files changed, 43 insertions(+), 12 deletions(-) diff --git a/task-6232/pyentropy.py b/task-6232/pyentropy.py index 1a981fe..fa50406 100644 --- a/task-6232/pyentropy.py +++ b/task-6232/pyentropy.py @@ -9,7 +9,9 @@ Output - A CSV file of the format (without newlines): <entropy for guard nodes>, <max entropy for guard nodes>, <entropy for countries>, - <max entropy for countries> + <max entropy for countries>, + <entropy for AS>, + <max entropy for AS> rsync -arz --delete metrics.torproject.org::metrics-recent/relay-descriptors/consensuses in """ @@ -17,6 +19,10 @@ import sys import math import os import pygeoip +import StringIO +import stem.descriptor +from stem.descriptor.server_descriptor import RelayDescriptor, BridgeDescriptor +from binascii import b2a_hex, a2b_base64, a2b_hex from optparse import OptionParser KEYS = ['r','s','v','w','p','m'] @@ -30,6 +36,8 @@ class Router: self.probability = None self.ip = None self.country = None + self.as_no = None + self.as_name = None self.is_exit = None self.is_guard = None @@ -37,7 +45,8 @@ class Router: if key == 'r': self.nick = values[0] self.ip = values[5] - self.country = gi.country_name_by_addr(self.ip) + self.country = gi_db.country_name_by_addr(self.ip) + self.as_no, self.as_name = self.get_as_details() if key == 'w': self.bandwidth = int(values[0].split('=')[1]) if key == 's': @@ -47,6 +56,13 @@ class Router: if "Guard" in self.flags: self.is_guard = True + def get_as_details(self): + try: + value = as_db.org_by_addr(str(self.ip)).split() + return value[0], value[1] + except: + return None, None + def run(file_name): routers = [] # parse consensus @@ -67,7 +83,7 @@ def run(file_name): totalBW, totalExitBW, totalGuardBW = 0, 0, 0 guards_n, exits_n = 0, 0 - bw_countries = {} + bw_countries, bw_as = {}, {} for router in routers: totalBW += router.bandwidth if router.is_guard: @@ -79,12 +95,17 @@ def run(file_name): if bw_countries.has_key(router.country): bw_countries[router.country] += router.bandwidth else: - bw_countries[router.country] = router.bandwidth + bw_countries[router.country] = router.bandwidth + if router.as_no: + if bw_as.has_key(router.as_no): + bw_as[router.as_no] += router.bandwidth + else: + bw_as[router.as_no] = router.bandwidth if len(routers) <= 0: return - - entropy, entropy_exit, entropy_guard, entropy_country = 0.0, 0.0, 0.0, 0.0 + + entropy, entropy_exit, entropy_guard, entropy_country, entropy_as = 0.0, 0.0, 0.0, 0.0, 0.0 for router in routers: p = float(router.bandwidth) / float(totalBW) if p != 0: @@ -97,18 +118,24 @@ def run(file_name): p = float(router.bandwidth) / float(totalExitBW) if p != 0: entropy_exit += -(p * math.log(p, 2)) - + for country in bw_countries.iterkeys(): p = float(bw_countries[country]) / float(totalBW) if p != 0: entropy_country += -(p * math.log(p, 2)) - + + for as_no in bw_as.iterkeys(): + p = float(bw_as[as_no]) / float(totalBW) + if p !=0: + entropy_as += -(p * math.log(p, 2)) + # Entropy of uniform distribution of 'n' possible values: log(n) max_entropy = math.log(len(routers), 2) max_entropy_guard = math.log(guards_n, 2) max_entropy_exit = math.log(exits_n, 2) max_entropy_country = math.log(len(bw_countries), 2) - + max_entropy_as = math.log(len(bw_as), 2) + return ",".join([valid_after, str(entropy), str(max_entropy), @@ -117,13 +144,16 @@ def run(file_name): str(entropy_guard), str(max_entropy_guard), str(entropy_country), - str(max_entropy_country)]) + str(max_entropy_country), + str(entropy_as), + str(max_entropy_as)]) def parse_args(): usage = "Usage - python pyentropy.py [options]" parser = OptionParser(usage) - parser.add_option("-g", "--geoip", dest="geoip", default="GeoIP.dat", help="Input GeoIP database") + parser.add_option("-g", "--geoip", dest="gi_db", default="GeoIP.dat", help="Input GeoIP database") + parser.add_option("-a", "--as", dest="as_db", default="GeoIPASNum.dat", help="Input AS GeoIP database") parser.add_option("-o", "--output", dest="output", default="entropy.csv", help="Output filename") parser.add_option("-c", "--consensus", dest="consensus", default="in/consensus", help="Input consensus dir") @@ -134,7 +164,8 @@ def parse_args(): if __name__ == "__main__": options = parse_args() - gi = pygeoip.GeoIP(options.geoip) + gi_db = pygeoip.GeoIP(options.gi_db) + as_db = pygeoip.GeoIP(options.as_db) with open(options.output, 'w') as f: for file_name in os.listdir(options.consensus):

1 0

[metrics-tasks/master] Refactor
by karsten＠torproject.org 24 Jul '12

24 Jul '12

commit 834f0e044d54c79970d3f3d37f232ddf7e05b33e Author: Sathyanarayanan Gunasekaran <gsathya.ceg(a)gmail.com> Date: Sat Jul 21 13:31:58 2012 +0530 Refactor Make it more pythonic. --- task-6232/pyentropy.py | 28 ++++++++++++++-------------- 1 files changed, 14 insertions(+), 14 deletions(-) diff --git a/task-6232/pyentropy.py b/task-6232/pyentropy.py index fa50406..4328f98 100644 --- a/task-6232/pyentropy.py +++ b/task-6232/pyentropy.py @@ -81,17 +81,17 @@ def run(file_name): elif key in KEYS: router.add(key, values) - totalBW, totalExitBW, totalGuardBW = 0, 0, 0 - guards_n, exits_n = 0, 0 + total_bw, total_exit_bw, total_guard_bw = 0, 0, 0 + guards_no, exists_no = 0, 0 bw_countries, bw_as = {}, {} for router in routers: - totalBW += router.bandwidth + total_bw += router.bandwidth if router.is_guard: - totalGuardBW += router.bandwidth - guards_n += 1 + total_guard_bw += router.bandwidth + guards_no += 1 if router.is_exit: - totalExitBW += router.bandwidth - exits_n += 1 + total_exit_bw += router.bandwidth + exists_no += 1 if bw_countries.has_key(router.country): bw_countries[router.country] += router.bandwidth else: @@ -107,32 +107,32 @@ def run(file_name): entropy, entropy_exit, entropy_guard, entropy_country, entropy_as = 0.0, 0.0, 0.0, 0.0, 0.0 for router in routers: - p = float(router.bandwidth) / float(totalBW) + p = float(router.bandwidth) / float(total_bw) if p != 0: entropy += -(p * math.log(p, 2)) if router.is_guard: - p = float(router.bandwidth) / float(totalGuardBW) + p = float(router.bandwidth) / float(total_guard_bw) if p != 0: entropy_guard += -(p * math.log(p, 2)) if router.is_exit: - p = float(router.bandwidth) / float(totalExitBW) + p = float(router.bandwidth) / float(total_exit_bw) if p != 0: entropy_exit += -(p * math.log(p, 2)) for country in bw_countries.iterkeys(): - p = float(bw_countries[country]) / float(totalBW) + p = float(bw_countries[country]) / float(total_bw) if p != 0: entropy_country += -(p * math.log(p, 2)) for as_no in bw_as.iterkeys(): - p = float(bw_as[as_no]) / float(totalBW) + p = float(bw_as[as_no]) / float(total_bw) if p !=0: entropy_as += -(p * math.log(p, 2)) # Entropy of uniform distribution of 'n' possible values: log(n) max_entropy = math.log(len(routers), 2) - max_entropy_guard = math.log(guards_n, 2) - max_entropy_exit = math.log(exits_n, 2) + max_entropy_guard = math.log(guards_no, 2) + max_entropy_exit = math.log(exists_no, 2) max_entropy_country = math.log(len(bw_countries), 2) max_entropy_as = math.log(len(bw_as), 2)

1 0

[metrics-tasks/master] Tweak consensus parsing (#6232).
by karsten＠torproject.org 24 Jul '12

24 Jul '12

commit ff0fee27015f8f5b1d589c259525d6678d99b22f Author: Karsten Loesing <karsten.loesing(a)gmx.net> Date: Tue Jul 24 09:33:04 2012 +0200 Tweak consensus parsing (#6232). - Remove unnecessary function calls for 'v', 'p', and 'm' lines which we don't parse anyway. - Fix a minor bug where we would not append the last router to the list, because there's no "r" line after that one. --- task-6232/pyentropy.py | 24 ++++++++++++------------ 1 files changed, 12 insertions(+), 12 deletions(-) diff --git a/task-6232/pyentropy.py b/task-6232/pyentropy.py index e782974..8af09c6 100644 --- a/task-6232/pyentropy.py +++ b/task-6232/pyentropy.py @@ -26,8 +26,6 @@ from optparse import OptionParser from binascii import b2a_hex, a2b_base64, a2b_hex from stem.descriptor.server_descriptor import RelayDescriptor, BridgeDescriptor -KEYS = ['r', 's', 'v', 'w','p', 'm'] - class Router: def __init__(self): self.lines = [] @@ -45,27 +43,28 @@ class Router: self.is_exit = None self.is_guard = None - def add(self, key, values): - if key == 'r': + def add_router_info(self, values): self.nick = values[0] self.digest = values[2] self.hex_digest = b2a_hex(a2b_base64(self.digest+"=")) self.ip = values[5] self.country = gi_db.country_name_by_addr(self.ip) self.as_no, self.as_name = self.get_as_details() - if key == 'w': + + def add_weights(self, values): self.advertised_bw = self.get_advertised_bw() if self.advertised_bw: self.bandwidth = self.advertised_bw else: self.bandwidth = int(values[0].split('=')[1]) - if key == 's': + + def add_flags(self, values): self.flags = values if "Exit" in self.flags: self.is_exit = True if "Guard" in self.flags: self.is_guard = True - + def get_as_details(self): try: value = as_db.org_by_addr(str(self.ip)).split() @@ -105,10 +104,13 @@ def run(file_name): key = line.split()[0] values = line.split()[1:] if key =='r': - if router: - routers.append(router) router = Router() - router.add(key, values) + routers.append(router) + router.add_router_info(values) + elif key == 's': + router.add_flags(values) + elif key == 'w': + router.add_weights(values) elif key == 'valid-after': valid_after = ' '.join(values) elif key == 'bandwidth-weights': @@ -120,8 +122,6 @@ def run(file_name): Wgg = data['Wgg'] except: pass - elif key in KEYS: - router.add(key, values) total_bw, total_exit_bw, total_guard_bw = 0, 0, 0 guards_no, exits_no = 0, 0

1 0

[metrics-tasks/master] Parse bandweight-weights better
by karsten＠torproject.org 24 Jul '12

24 Jul '12

commit 960327bfd58eb506aea3df9d34f97a4c14469e4c Author: Sathyanarayanan Gunasekaran <gsathya.ceg(a)gmail.com> Date: Tue Jul 24 01:01:17 2012 +0530 Parse bandweight-weights better Instead of depending on position, depend on keys --- task-6232/pyentropy.py | 22 ++++++++++++++++++---- 1 files changed, 18 insertions(+), 4 deletions(-) diff --git a/task-6232/pyentropy.py b/task-6232/pyentropy.py index 6bd5132..e782974 100644 --- a/task-6232/pyentropy.py +++ b/task-6232/pyentropy.py @@ -85,6 +85,16 @@ class Router: except: return None +def parse_bw_weights(values): + data = {} + try: + for value in values: + key, value = value.split("=") + data[key] = float(value) / 10000 + return data + except: + return None + def run(file_name): routers = [] router = None @@ -102,10 +112,14 @@ def run(file_name): elif key == 'valid-after': valid_after = ' '.join(values) elif key == 'bandwidth-weights': - Wed = float(values[6].split('=')[1]) / 10000 - Wee = float(values[7].split('=')[1]) / 10000 - Wgd = float(values[11].split('=')[1]) / 10000 - Wgg = float(values[12].split('=')[1]) / 10000 + data = parse_bw_weights(values) + try: + Wed = data['Wed'] + Wee = data['Wee'] + Wgd = data['Wgd'] + Wgg = data['Wgg'] + except: + pass elif key in KEYS: router.add(key, values)

1 0