We were mistakenly not making a rejoin attempt on freezing connections
due to all closest connections going down. This fixes that, and tweaks
the test. I've still only done tens rather than hundreds of tests, but
I'm fairly confident that the conference test now consistently passes.
* add freezing and unfreezing of peers
* add rejoin packet
* revise handling of temporary invited connections
* rename "peer kill" packet to "peer leave" packet
* test rejoining in conference test
* use custom clock in conference test
This forces all the loop bodies to be executed at least once, which is
harmless since it just means one more tox event loop iteration. This
reduces the jitter we see in coverage measurements, which is partially
caused by loops sometimes being entered and sometimes not (because their
condition happens to randomly already be true).
* test names in conference_test
* raise error on attempt to invite friend to group before we are connected
* revise handling of temporary invited connections
We are now careful not to prematurely delete a connection to a peer
established during the invitation process; namely, before we have sufficient
other connections and have confirmed that we have an alternative route to the
peer.
* process out-of-order messages from a peer
* don't reset names when handling a Peer Response
Did my best to surmise the size requirements of
these integers, will do the rest of the tests soon. Also added a todo
and made an obsessive change to a for loop.
Also, renamed simple_conference_test to conference_simple_test so it's
sorted together with the other conference tests.
Next step is to use run_auto_test.h for the conference test.
Also reduce number of people in conference to 5, because on Circle CI the
test times out trying to connect more than 6 or 7 people. The persistent
conferences PR will improve this so we can set it much higher then.
Tests are not actually ran on appveyor for now, since they all fault for
some reason. For now, we just build them. Also, some tests are disabled
on msvc entirely, because they don't even compile. We'll need to look
into those, later. They are disabled using `MSVC_DONT_BUILD`.
- CFLAG gnu99 was changed to c99.
- CXXFLAG c++98 was changed to c++11.
- CFLAG -pedantic-errors was added so that non-ISO C now throws errors.
- _XOPEN_SOURCE feature test macro added and set to 600 to expose SUSv3
and c99 definitions in modules that required them.
- Fixed tests (and bootstrap daemon logging) that were failing due to
the altered build flags.
- Avoid string suffix misinterpretation; explicit narrowing conversion.
- Misc. additions to .gitignore to make sure build artifacts don't wind
up in version control.
Previously, all log messages generated by tox_new (which is quite a lot)
were dropped, because client code had no chance to register a logging
callback, yet. This change allows setting the log callback from the
beginning and removes the ability to unset it.
Since the log callback is forever special, since it can't be stateless,
we don't necessarily need to treat it uniformly (with `event`).
Compiling as C++ changes nothing semantically, but ensures that we don't
break C++ compatibility while also retaining C compatibility.
C++ compatibility is useful for tooling and additional diagnostics and
analyses.
In a next step, we will remove tests from each file to have a per-binary
split of tests. This will help identify which tests fail most often on
Travis CI.
In another future step, we will split the large one_test into several
auto tests, which will make testing quite a bit slower (adding about 10
seconds setup time to each), but hopefully a lot more stable ("Tox went
offline" should not happen as much anymore).
