and number of times in a priority list attrs and methods,
to be able to store the data.
Part of #28567.

Since all the other bandwidth values are in bytes and sbws should
work with bytes until it generates the final `bw`.

Closes: #29707. Bugfix v1.0.3.

used in the bandwidth file, so that it is less confusing and less
prone to errors.

to the state file.
To know the total number of consensus seen in the last days.

Part of #28566.

that have been seen for a relay and for the relay list.

Part of #28566.

for each relay and for the list of relays when fetching new
network statuses.
To count the number of times a relay was in a consensus and how
many consensuses have been seen.
In the test network router status entries don't have document
attribute.
Part of #28566

When the descriptor or network status of a relay was not obtained,
some attributes can't be obtained it can be return None instead
of failing the measurement.
Also, there's no need for hasattr when getattr can just return
None as default.

Closes #28870. Bugfix v0.4.0

They were used to resolve the IP of the destination and check
whether an exit policy allows to exit to that IP, but when the
destination is a CDN, the IP locally resolved would be different
to the IP resolved in the exit, and when the IP resolved to
IPv6, it was possible that the scanner didn't have IPv6.
The correct method to check whether an exit policy allows to exit
to an IP, would be to resolve the domain via Tor itself using
RESOLVE and ADDRMAP events with that exit.

In order to use it in torflow's scaling method

The flags are: not having a badflag and the policy allows to exit
to a port.

The flags are: not have the badflag and the policy allows to exit
to a port.

The flags are: not having a badexit and the policy allows to exit
to a port.

In order to do not relay on resolving a domain locally to check
the IP in the exit policy.

Even for a valid domain that usually resolves, sometimes the local DNS
resolver won't be able to resolve it (duh). Catch the exception that gets
thrown in that instance and fail gracefully by returning an empty list.

GH: closes #218

Matt Traudt authored Jun 22, 2018 and juga committed Jun 24, 2018

The chance of us needing a refresh at this point is TINY, but it's probably
better to return a list of relays stale by a millisecond than it is to cause
the user's program to crash.

Matt Traudt authored Jun 22, 2018 and juga committed Jun 24, 2018

I'm not sure if this is necessary, honestly. But it doesn't hurt.

Matt Traudt authored Jun 22, 2018 and juga committed Jun 24, 2018

GH: ref #205

Use the new can_exit_to member func of the Relay class in order to
remove the exits_can_exit_to member func of RelayList, which I never
really liked.

Relay.can_exit_to has the same DNS-based limitations as its older
brother.

Part of this takes care of... GH: closes #145

to make it less confussing.
Note that stem call it ed25519_master_key instead of
master_key_ed25519, but dir-spec.txt uses master-key-ed25519, which
is the term chosen here.

Matt Traudt authored Jun 07, 2018 and juga committed Jun 10, 2018

With unused support for ed25519 keys

* RelayList is now a list of RelayNS instead of RouterStatusEntryV3
  to have master_key_ed25519 as an attribute
* create pytest fixture to start Tor with tests
* create test to check master_key_ed25519

GH: closes #144