Dilator.stop() now shuts everything down, and returns a Deferred when it all
stops moving. This needed some Manager state machine changes (to notify
Dilator when it enters the STOPPED state). This also revealed problems in the
delivery of connector_connection_made() (which was misnamed) and
connector_connection_lost() (which wasn't being called at all).
This is only used by tests so far (and will simplify the integration test
that hasn't landed yet), but is not yet wired up to Boss, so there's no way
for applications to enable it yet.
I mistakenly believed that Noise handshakes are simultaneous. In fact, the
Responder waits until it sees the Initiator's handshake before sending its
own. I had to update the Connection state machines to work this way (the
Record machine now has set_role_leader and set_role_follower), and update the
tests to match.
For debugging I added a `_role` property to Record, but it should probably be
removed.
only install 'noiseprotocol' (which is necessary for dilation to work) if the
"dilate" feature is requested (e.g. `pip install magic-wormhole[dilate]`)
test_ipaddrs.py had an invalid regexp, caught by the latest version of flake8
The new flake8 complains about both W503 (line break inside a conditional
before a binary operator) and W504 (line break *after* the operator). I think
break-before is the new preferred style, but for now I'm just going to ignore
them both and leave the code alone.
This establishes what our mailbox protocol does to encrypt the
individual (post-SPAKE2) messages, which combines NaCl SecretBox and our own
key-derivation choices. I'd like to move off of NaCl/libsodium and I think
some RFC7539-implementing library might be compatible, and with these test
vectors I can check that. I also want to copy these tests into the Rust port.
See https://bugs.python.org/issue26175 . tempfile.SpooledTemporaryFile
doesn't fully implement the IOBase abstract class, which breaks because
py3.7.0's new zipfile module tries to delegate .seekable down to the wrapped
file and causes an AttributeError.
refs #306
The flake8 config excludes E741, which would complain about using
'l' (lower-case ell) as a variable name. We use this for the Lister object in
one test that uses single-character variable names for all the machines ('b'
for Boss, 'm' for Mailbox, etc). That comment was added before excluding
E741. If we ever restore that warning, we might want to rename the variable.
We care about how long it takes to import all the wormhole-specific things,
to investigate user-perceived latency from the time the command is launched
to the time they can actually interact with it. So we need to record
`time.time()` before doing the rest of the imports, even though pep8 says all
imports should be done before any non-importing statements.
tested with/on:
- ubuntu linux 18.04 amd64
- pyinstaller 3.3.1 (pip install pyinstaller)
- python 3.6.5
There is a good chance it also works on FreeBSD, maybe also on macOS.
The change in __main__.py was required because otherwise it complains about
__main__ not being a package when trying the dot-relative import.
Tests can pass an EventualQueue into wormhole.create(), to override the
default. This lets the tests flush the queue without using a haphazard
real-time delay.
closes#23
(in fact, we added multiple-Deferreds-per-API a while ago, but this does it
in a much cleaner fashion, and with the safety of an eventual-send)
This factors out the various "give me a Deferred for an value that may or may
not eventually be successfully generated" routines in _DeferredWormhole. It
uses an eventual-send to fire the Deferreds to avoid plan-coordination
hazards when the attached callbacks then call back into the Wormhole object
before the rest of the state transition has finished.
We defer starting a new timer until we've completely emptied the queue, since
we know we'll get to any new events added inside one of our callbacks. The
old design in Foolscap (which copied the list, cleared the original, then
fired everything in the copy) didn't look at these new events. OTOH, this
pop(0)-until-empty approach makes it easier to get into an infinite loop (any
callback which queues a new callback will get priority over anything else).
But the code is simpler.
I've seen intermittent failures in
test_cli.PregeneratedCode.test_text_subprocess where the host was slow (or
overloaded) enough that the "Waiting for sender.." pacifier message was
displayed, which flunks the test because we're looking for a specific output
string. We patch this 1-second timer in the non-subprocess tests, but you
can't patch across a process boundary.
This patch adds an undocumented environment variable that lets you override
the timer values. The test then sets it to something large.
For future consideration: another approach would be to change the test to
tolerate the extra message. This would be trickier to validate, though.
This will be necessary to avoid reentrancy hazards, in case the application
code makes other wormhole API calls from the callbacks for
get_unverified_key(), get_verifier(), or get_message().
We were missing two (the calls to choose_nameplate() and choose_words() that
happen after the input() function has finished, but while we're still inside
the thread that makes it safe for input() to block). This almost certainly
caused the crash seen in issue #280.
Update the tests to match: CodeInputter.finish must now be called with
deferToThread from inside tests, or the internal blockingCallFromThread must
be stubbed out.
This causes two threads to use the reactor at the same time, with horrible
results. The _rlcompleter code currently violates this requirement, causing
occasional failures if the messages arrive in just the wrong way (refs #280).
Also log events at the beginning and end of Input.choose_nameplate and
Input.choose_words, since those are the two big locally-driven (UI) triggers
that cause multiple messages to be fired and lots of work to happen.
lgtm.com noticed some unreachable code paths, and it turns out that nothing
in the rest of the code base could ever raise WormholeClosedError (I guess it
was leftover from before the big API refactoring). Both sender and receiver
are simpler without the unnecessary checks and state variables.
This comments out some "if 0: debug()" stuff I keep around to investigate
problems, since lgtm thinks of it as accidentally-unreachable code.
I also deleted a server usage command
entirely (src/wormhole/server/cmd_usage.py show_usage) which was disabled
while I rewrote that schema: the new plan is to move the server into a new
repository altogether, and use a completely different approach to the
usage database.
A slow Travis-CI host caused one the 1.0s KEY_TIMER to fire by accident,
making the test fail because it wasn't expecting to see the "please be
patient" message. Fixed this by increasing the timeout to a very large value
when we aren't explicitly testing it.
A tiny update to show the command to execute along with the code required on
the same line, rather than split across two lines. This small change helps
when sending the information to others using copy and paste.
Fixes#266
-=david=-
