Oops. On reflection many platforms (OSX, BSD, etc) lack proc contents so we
shouldn't attempt to use them when unavailable. Otherwise the connection panel
will be delayed. Caught thanks to a report on tor-relays@.

When we don't list geoip information there's no need to reserve space for it.

We padded the source and destination with different sizes to account for
listing a locale, but that padding was inverted for inbound connections.

If you have an exit policy that allows exiting then relay connections we could
conceivably exit to are labeled as 'EXIT'. That's wrong, and would
understandably spook folks.

On reflection it will be less confusing if inference is used as the primary
resolver in all scenarios.

In theory when the user has 'DisableDebuggerAttachment 0' in their torrc proc
permissions should allow standard connection resolvers to work. However, seems
that's not always the case. Using our new inference resolution method as a
final fallback.

Oops, just realized that we include the nyxrc but not the user's torrc (which
is a hell of a lot more important!).

Great catch from Hello71 that when a relay has IPv6 addresses the columns
overwrite each other. These addresses are long enough that they crowd out the
other fields, but them's the breaks. Shifting the other fields to make room.

For IPv4 alignment should be unchanged.

We catch the error, but didn't set is_successful causing a NameError.

Gah! Stupid me. Inverted a conditional, causing the accounting stats to not
show up until the interface is paused. Caught thanks to torix on tor-relays@...

  https://lists.torproject.org/pipermail/tor-relays/2017-November/013484.html

When the menu's open for a while we block the graph panel from redrawing
itself, causing events to accumilate. Then when the menu's closed we perform
a flurry of updates which makes the graph appear to catch up in fast forward.

Instead, if a panel is unable to draw itself after a second it gives up the
redraw attempt. This way when the menu's closed we perform just a single redraw
that catches us up.

Caught thanks to Guinness on tor-relays@.

Especially when first starting up nyx takes a couple seconds to shut down on my
relay. This is because the update method of our connection panel can take a
while if there's many relays. Aborting in the middle of its longest part if the
thread's terminated. This drops our shutdown time in practice to a few
milliseconds.

Oops, we displayed an unbounded decimal precision for the connection graph's
average count. Who cares what the sixth decimal point of precision is? :P

It's an average of a big whole number. As such just truncating to ints. Caught
thanks to Toralf.

I was unsure if FreeBSD had sqlite3 ports for all the python versions but a
reply on tor-relays@ seems to imply they do. Providing a command for whatever
interpreter version the user's running.

Turns out even when we suppress the exceptions it causes the interpreter to
print to stdout. Lets just avoid these calls entirely on FreeBSD. Dropping the
catch clauses since we were only doing this due to FreeBSD.

Caught thanks to Andrew.

Interesting catch from Toralf. I'm not entirely sure if this is it but if he
was using python3 it makes sense that keys() provides an iterator whereas
reversed() expects a collection. Not able to repro it though so not sure if
this is what's at play...

  Traceback (most recent call last):
    File "./run_nyx", line 14, in <module>
      nyx.main()
    File "/root/nyx/nyx/__init__.py", line 147, in main
      nyx.starter.main()
    File "/root/nyx/stem/util/conf.py", line 289, in wrapped
      return func(*args, config = config, **kwargs)
    File "/root/nyx/nyx/starter.py", line 94, in main
      nyx.curses.start(nyx.draw_loop, acs_support = config.get('acs_support', True), transparent_background = True, cursor = False)
    File "/root/nyx/nyx/curses.py", line 216, in start
      curses.wrapper(_wrapper)
    File "/usr/lib64/python3.4/curses/__init__.py", line 94, in wrapper
      return func(stdscr, *args, **kwds)
    File "/root/nyx/nyx/curses.py", line 214, in _wrapper
      function()
    File "/root/nyx/nyx/__init__.py", line 192, in draw_loop
      nyx.menu.show_menu()
    File "/root/nyx/nyx/menu.py", line 203, in show_menu
      menu = _make_menu()
    File "/root/nyx/nyx/menu.py", line 243, in _make_menu
      submenu = panel.submenu()
    File "/root/nyx/nyx/panel/log.py", line 273, in submenu
      [RadioMenuItem(opt, filter_group, opt) for opt in self._filter.latest_selections()],
    File "/root/nyx/nyx/log.py", line 432, in latest_selections
      return list(reversed(self._past_filters.keys()))
  TypeError: argument to reversed() must be a sequence

When displaying the menu we didn't redraw the background, causing old submenus
to remain visible while moving around.

Message thanks to Toralf.

Oops! Turns out this isn't available by default on FreeBSD or Gentoo. Providing
a nicer notice, encouraging folks to contact us so we can provide per-platform
suggestions for how to get it.

These are to sidestep issues on some platforms. Generally certainly fine if
they don't take effect. Caught thanks to tor-relays@ beta tester on FreeBSD...

  nyx: environment corrupt; missing value for
  Traceback (most recent call last):
    File "./run_nyx", line 14, in <module>
      nyx.main()
    File "/usr/home/ryan/nyx/nyx/__init__.py", line 147, in main
      nyx.starter.main()
    File "/usr/home/ryan/nyx/stem/util/conf.py", line 289, in wrapped
      return func(*args, config = config, **kwargs)
    File "/usr/home/ryan/nyx/nyx/starter.py", line 90, in main
      os.putenv('LANG', 'C')  # make subcommands (ps, netstat, etc) provide
  english results
  OSError: [Errno 14] Bad address

Hey, it's only a couple orders of magnitude.

... *sigh*.

Why did we divide by that? Weird. Because of it we always showed tor's cpu
usage as 0%.

We shouldn't silently swallow errors. Opting for notice runlevel so they're
visible to users by default so they can be reported.

The 'width' attribute already takes into account the initial 'x' offset, but in
the conditional we didn't account for that causing connection details to just
about never be shown.

Fixing a couple easy to make gotchas in our --log argument. It's especially
confusing becuase nyx expected 'NYX_WARNING' whereas it displayed 'NYX_WARN'.

Typo Roger spotted in the comments.

Lighten the work needed by the ConsensusTracker by having it process unparsed
consensus content. We already do this when starting up, so this actually
simplifies our code quite a bit.

If the dialog's title is shorter than the content (which happens when we lack
descriptors) the dialog crops the title.

It's rare, but once I spotted a "[0 duplicates hidden]" in my logs. Reason is
that the message deduplicated, but then the log got long enough that the
duplicate was dropped.

Now that stem can cache our address we can drop our our_address() helper.

Calling the endpoint of our own circuits 'exit' worries folks on occasion,
concerned that they might accidently be running an exit relay. Renaming
the termination point to 'end' to hopefully avoid those concerns...

  https://trac.torproject.org/projects/tor/ticket/12956

Replacing the textual listing with images to give the page a friendlier, more
approachable feel.

Rather than directing users toward the nyxrc.sample file providing them with a
webpage describing the options. Also making it more discoverable by putting it
in the 'more' menu.

Lots of breakages have slipped in while attempting to improve our performance.
Most were timing related due to our switch to static sleeps.

Oops! Forgot to add a couple small image resources.