meejah
0aaf00f803
get rid of prints
2021-04-14 15:05:44 -06:00
meejah
a89988af90
get rid of placeholder/test code; skip test_short for websockets
2021-04-14 15:05:44 -06:00
meejah
816e997b01
post-rebase fixups
2021-04-14 15:05:44 -06:00
meejah
f18edc89f9
refine
2021-04-14 15:05:44 -06:00
meejah
99c71112b6
a passing thing
2021-04-14 15:05:44 -06:00
meejah
dd1cc7d520
upgrade all tests; reactor_turn(); remove debug
2021-04-14 15:05:44 -06:00
meejah
002773d79f
WIP: first passing IOPump test
2021-04-14 15:05:44 -06:00
meejah
5210566150
websocket version of tests, with handshake
2021-04-14 15:05:44 -06:00
meejah
21af1f68a3
Transit is no longer a factory
2021-04-14 15:05:44 -06:00
meejah
5df5f86e42
not just localhost
2021-04-14 15:05:44 -06:00
meejah
1a461aa461
haxxor
2021-04-14 15:05:44 -06:00
meejah
34d039c38c
hack in prelim websocket support
2021-04-14 15:05:44 -06:00
meejah
c2147ee985
change from review: inline _got_handshake
2021-04-14 15:05:44 -06:00
meejah
c8fbc22120
dead code
2021-04-14 15:05:44 -06:00
meejah
b192b5ca71
dead code, correct input
2021-04-14 15:05:44 -06:00
meejah
34dd36158e
unused
2021-04-14 15:05:44 -06:00
meejah
2b2f06d984
unused
2021-04-14 15:05:44 -06:00
meejah
9cf42c560b
not sure we can hit this state at all
2021-04-14 15:05:44 -06:00
meejah
c09f15d866
re-instate log message
2021-04-14 15:05:44 -06:00
meejah
e0f5f556cc
does this ever get called?
2021-04-14 15:05:44 -06:00
meejah
b03801d155
guard
2021-04-14 15:05:44 -06:00
meejah
942f204140
log again
2021-04-14 15:05:44 -06:00
meejah
7e58767ac1
pyflakes
2021-04-14 15:05:44 -06:00
meejah
ca55509763
fix global stats-gathering / recording
2021-04-14 15:05:44 -06:00
meejah
60e70bac3c
cleanup / dead code
2021-04-14 15:05:44 -06:00
meejah
215a0f350b
restore 2 missing log-lines
2021-04-14 15:05:44 -06:00
meejah
03906ffe0d
pass actual database, not config
2021-04-14 15:05:44 -06:00
meejah
83de03c8c6
remove old test-code
2021-04-14 15:05:44 -06:00
meejah
9557bbf75a
we never remove backends
2021-04-14 15:05:44 -06:00
meejah
3ae3bb7443
cleanup, remove dead code
2021-04-14 15:05:44 -06:00
meejah
b7bcdfdca3
more stats / recording works
2021-04-14 15:05:44 -06:00
meejah
53864f57f0
use 'backends' for usage-recording
2021-04-14 15:05:44 -06:00
meejah
5ed572187b
unregister completely
2021-04-14 15:05:44 -06:00
meejah
40919b51be
count bytes missing
2021-04-14 15:05:44 -06:00
meejah
4669619f7e
skip usage-counting if we're jilted but other side is happy?
2021-04-14 15:05:44 -06:00
meejah
ff578fccf8
fix more tests (that examine internals)
2021-04-14 15:05:44 -06:00
meejah
7b91377e94
try to make 'redudant' mood work
2021-04-14 15:05:44 -06:00
meejah
734ed809c2
fix more tests
2021-04-14 15:05:44 -06:00
meejah
b51237d958
start of refactoring usage-recording: pass one test
2021-04-14 15:05:44 -06:00
meejah
0e64707459
count totals in state-machine
2021-04-14 15:05:44 -06:00
meejah
0e11f1b8f1
(wip) refactor to use Automat state-machine
2021-04-14 15:05:44 -06:00
meejah
555c23d4fe
first-cut of state-machine style code
2021-04-14 15:05:44 -06:00
meejah
00086a798d
flush cleanup
2021-04-12 08:50:22 -06:00
meejah
6efc274b81
get rid of double-flush() pairing with florian
2021-04-12 08:44:06 -06:00
meejah
591740ce5f
better name for interface
2021-04-10 18:42:12 -06:00
meejah
fc3507c1f6
flip around 'if' logic to simplify
2021-04-02 23:36:33 -06:00
meejah
f3c391e98b
more coverage
2021-04-02 23:15:48 -06:00
meejah
2903c7f2a0
re-org + comments
2021-04-02 23:05:52 -06:00
meejah
2d506de55a
upcall, 2.7-friendly
2021-04-02 20:33:57 -06:00
meejah
45c09fdd05
explicit interface, different naming
2021-04-02 20:10:01 -06:00
meejah
0434296415
still support py27
2021-04-02 19:57:22 -06:00
meejah
8447f88159
pyflakes
2021-04-02 16:11:18 -06:00
meejah
85f3f5b63c
'mock' location
2021-04-02 16:00:11 -06:00
meejah
5e21a3c35a
all tests pass
2021-04-02 16:00:11 -06:00
meejah
b9c2bbc524
refactor to use IOPump: one test passes
2021-04-02 16:00:11 -06:00
Brian Warner
ca309d5283
post-rebase: update newer tests
2020-05-23 17:50:26 -07:00
Joe Harrison
45824ca5d6
Use StringTransportWithDisconnection for transit server tests.
...
Replace the use of TCP in the test suite with Twisted's
StringTransport, specifically StringTransportWithDisconnection which
allows us to trigger a disconnect event on the server side during testing.
The `dataReceived` method on the server is now called directly, and any
effects will be realised immediately.
Responses are available to the test client using the `value()` method of
the transport objects, and the buffer can be cleared using `clear()`.
This allows all asynchronous behaviour to be removed from the transit
server test suite.
Furthermore, as we never have to wait for the server, tests no longer
hang if they fail: the errors are encountered immediately.
2020-05-23 17:18:47 -07:00
Brian Warner
ac7415a4d0
Merge branch 'pr13'
2020-05-21 23:08:06 -07:00
Brian Warner
1a7faf0654
tolerate clients who disconnect before sending a complete handshake
...
If the client connected and then immediately disconnected, or disconnected
before sending enough of a handshake to be classified as good or bad, we
tried to record the usage record as if the connection was complete ("happy"),
which triggered an assertion. This dumps an error to the log and failed to
record the usage record. In May 2020, this happened 55 times in 40 days.
2020-05-21 22:49:14 -07:00
Brian Warner
1242f36624
tolerate data arriving briefly after we hang up
...
If the file receiver hangs up on an established connection, we do a
`transport.loseConnection()` on the buddy (the file sender). But apparently
it takes a moment (perhaps a roundtrip through the kernel) for the incoming
data to stop, and that used to cause an error. In May 2020 this happened 11
times in 40 days. Now we just ignore this late data.
2020-05-21 22:48:07 -07:00
Joe Harrison
912cfa69b6
maintain python2.7 regex library compatibility
2020-03-08 21:13:49 +00:00
Joe Harrison
0dab7a4c63
restore impatient disconnect when message received after handshake but before sent_ok
2020-03-08 19:57:44 +00:00
Joe Harrison
8f89c8aaff
inline handshake parsing and explain MAX_LENGTH for transit server
2020-03-08 19:54:42 +00:00
Joe Harrison
4fdd89cb35
use line receiver to simplify handshake logic
2020-03-07 02:14:04 +00:00
Brian Warner
9758d83279
fix IPv4/IPv6 listening port
...
whoops, it's "::" (the unspecified address), not "::1" (the loopback address)
refs #12
2019-09-11 00:23:50 -07:00
Brian Warner
42a293213b
enable SO_KEEPALIVE on all connections
...
This timeout is notoriously long (about two hours), but it might eventually
prune stuck connections.
refs #9
2019-09-10 23:14:29 -07:00
Brian Warner
273c4d796c
change default port= to listen on both IPv4+IPv6
...
The default was "tcp:4001", which happens to expand into
"tcp:4001:interface=127.0.0.1", which limits the listening socket to IPv4
connections only.
Changing it to "tcp:4001:interface=\:\:1" means "listen on ::1", which is the
IPv6 name for the loopback interface, and gets us a socket that accepts both
IPv4 and IPv6 connections.
Note: this might cause incompatibilities with IPv4-only hosts, if they don't
accept the "::1" name. For these systems, run the server as `twist
transitrelay --port=tcp:4001" to revert to the old behavior.
closes #12
2019-09-10 23:09:04 -07:00
Brian Warner
6ceecb78bf
add TODO for one of the moods
2019-09-10 23:05:22 -07:00
Brian Warner
a9680918b9
test_stats: try more realistic fake time.time()
...
Maybe windows+py36 was failing because "456" is too small of a value, and the
OS time libraries can't figure out UTC for something back in 1970. Switch to
a fixed timestamp from just a few minutes ago, to see if it likes 2018
better.
2018-02-19 13:23:40 -08:00
Brian Warner
071471a353
test_stats: mock time.time() with a float, to appease windows
...
For some reason, the appveyor (windows) tests fail on py36 with an OSError
during an internal (twisted.python.log) call to datetime.fromtimestamp().
It's possible that my mock times are too old (I'm using made-up values like
"456"), but maybe it's just that real time.time() returns a float instead of
an integer.
2018-02-19 13:12:14 -08:00
Brian Warner
5a762b16ad
call increase_rlimits() when creating the service
2018-02-19 12:28:17 -08:00
Brian Warner
f7b8c5b19a
copy increase_rlimits.py from magic-wormhole-mailbox-server
...
This code was originally in magic-wormhole, embedded in the server startup
code. I carved it out to a separate file (with tests) in mailbox-server, but
it should be used here too.
2018-02-19 12:26:10 -08:00
Brian Warner
098e315624
remove a few lines by using defaultdict()
2018-02-12 17:34:11 -08:00
Brian Warner
73d233a7be
comment about updating 'current' row immediately, or waiting for timer
2018-02-12 17:29:27 -08:00
Brian Warner
10e754fc9a
update v1 schema comments with new "redundant" mood
2018-02-12 17:04:24 -08:00
Brian Warner
4e9b6c53a9
rework mood tracking, full tests of usage events
2018-02-12 16:57:39 -08:00
Brian Warner
547ead75ba
test_transit_server: factor out wait()
2018-02-12 16:55:37 -08:00
Brian Warner
b3100344eb
_version.py: hush lgtm.com alert
...
This needs to be upstreamed into Versioneer, as it will disappear the next
time we upgrade.
2018-01-03 00:26:29 +01:00
Brian Warner
ecf3edb3f7
transit_server.py: remove unnecessary 'pass'
...
caught by lgtm.com analysis
2018-01-03 00:26:03 +01:00
Brian Warner
05ffcb2d55
test_transit_server: improve coverage
2017-11-09 16:39:44 -08:00
Brian Warner
c270ad6e0b
DB: small cleanups, improve test coverage
2017-11-09 16:39:25 -08:00
Brian Warner
54be4cd8d3
database: add open_existing_db() for migration tool
2017-11-07 21:54:59 -06:00
Brian Warner
69d66dd4c1
database: add create-only function, for migration tool
2017-11-07 21:54:53 -06:00
Brian Warner
c6e3347e68
copy test_database from magic-wormhole package
...
disable the upgrader test for now, since we only have the one schema version
2017-11-07 21:44:57 -06:00
Brian Warner
0b93725709
more test coverage
2017-11-07 21:21:04 -06:00
Brian Warner
fa31777db0
increase test coverage in the bad handshake path
2017-11-07 21:09:12 -06:00
Brian Warner
65b2192e89
replace --log-stdout with --log-fd=
2017-11-07 20:35:29 -06:00
Brian Warner
ff48518e37
handle --blur-usage= correctly, log it
2017-11-07 20:14:54 -06:00
Brian Warner
9ccb0424b0
move docs out of usage and into markdown file in docs/
2017-11-07 19:50:32 -06:00
Brian Warner
83e1c8acfe
make it all basically work, improve test coverage
2017-11-07 19:39:01 -06:00
Brian Warner
a898a65b09
start on usage-db
2017-11-04 12:54:49 -07:00
Brian Warner
d36e0c44bd
server_tap.py: minor docs
2017-10-26 11:36:37 +08:00
Brian Warner
849a90f0fa
test_stats: more test coverage
2017-09-13 18:26:54 -07:00
Brian Warner
a1333e51cd
fix unicode/json on py3
2017-09-13 18:18:29 -07:00
Brian Warner
2e0c97cea8
test_stats: start improving coverage
2017-09-13 18:14:41 -07:00
Brian Warner
b15e955fec
get tests to pass
2017-09-13 17:51:55 -07:00
Brian Warner
26efac7da9
transit_server.py: rewrite stats
2017-09-13 17:24:37 -07:00
Brian Warner
ea9e24913c
more WIP
2017-09-13 12:29:51 -07:00
Brian Warner
c287175d38
remove unused files
...
This server won't use a database, just a logfile and statsfile. And it
doesn't have a CLI command, just a twist/twistd plugin.
2017-09-13 00:39:10 -07:00
Brian Warner
0dc3fd5e7e
write the twist/twistd plugins and setup code
2017-09-13 00:37:29 -07:00
