- Attach permission checks to the first argument (so the literal
command name) rather than the last argument. This fixes commands
showing up when they shouldn't.
- HelpingArgumentBuilder now inherits permissions of its leaf nodes.
This only really impacts the "track" subcommand.
- Don't autocomplete the computer selector for the "queue" subcommand.
As everyone has permission for this command, it's possible to find
all computer ids and labels in the world.
I'm in mixed minds about this, but don't think this is an exploit -
computer ids/labels are sent to in-range players so shouldn't be
considered secret - but worth patching none-the-less.
The main purpose of this is to backport the improved parse/runtime
errors to older versions. I think they're sufficiently useful that we
should try to make it as widely available as possible.
We've been running them for a week now on SC3 and the released version
and not seen any issues, so I think it's probably stable enough.
This is a pretty lazy commit: I ended up copying the whole ROM over and
then picking up a few other related changes along the way.
- Trim spaces from file paths (b8fce1eecc)
- Correctly format 12AM/PM with
%I (9f48395596)
- Fix http.request and htpt.websocketAsync not handling a few failure
edge-cases correctly (3b42f22a4f).
- Move the internal modules into the main package path, hidden under
cc.internal (34a31abd9c).
- Gather code coverage in Java instead of
Lua (28a55349a9).
- Make error messages in edit more
obvious (8cfbfe7ceb).
- Make mcfly's test methods global. This means we don't need to pass
stub everywhere (7335a892b5).
- Improve runtime and parse errors. This comes from numerous commits,
but chiefly a12b405acf, and
5502412181.
- Hide the internal redirect methods in
multishell (33b6f38339).
Note this does /not/ include the shebang changes (sorry Emma!). I've
tried to avoid adding any user-controllable features, mostly because I
don't know how to handle the versioning otherwise :).
- Switch to using OptionalInt/OptionalLong instead of @Nullable
Long/Integers. I know IntelliJ complains, but it avoids the risk of
implicit unboxing.
- Instead of mutating PartialOptions, we now define a merge() function
which returns the new options. This simplifies the logic in
AddressRule a whole bunch.
This is a noisier diff than I'd like as this is just a direct copy from
the multi-loader branch.
- Rename "ingame" package to "gametest"
- Don't chain GameTestSequence methods - it's actually much cleaner if
we just use Kotlin's implicit this syntax.
- Use our work in 71f81e1201 to write
computer tests using Kotlin instead of Lua. This means all the logic
is in one place, which is nice!
- Add a couple more tests for some of the more error-prone bits of
functionality.
This is an initial step before refactoring this into a separate module.
It's definitely not complete - there's a lot of work needed to remove
referneces to the main ComputerCraft class for instance - but is a
useful first step.
I still don't really understand why the ROOT locale is wrong here, but
there we go. We'll need to remember to uncomment the tests on the 1.18
branch!
Also add some code to map tests back to their definition side. Alas,
this only links to the file right now, not the correct line :/.
This offers very few advantages now, but helps support the following in
the future:
- Reuse test support code across multiple projects (useful for
multi-loader).
- Allow using test fixture code in testMod. We've got a version of our
gametest which use Kotlin instead of Lua for asserting computer
behaviour.
We can't use java-test-fixtures here for Forge reasons, so have to roll
our own version. Alas.
- Add an ILuaMachine implementation which runs Kotlin coroutines
instead. We can use this for testing asynchronous APIs. This also
replaces the FakeComputerManager.
- Move most things in the .support module to .test.core. We need to use
a separate package in order to cope with Java 9 modules (again,
thanks Forge).
This allows other mods to create wired-modem alike blocks, which expose
peripherals on the wired network, without having to reimplement the main
modem interface.
This is not currently documented, but a peripheral_hub should provide
the following methods:
- isPresentRemote
- getTypeRemote
- hasTypeRemote
- getMethodsRemote
- callRemote
- Add a new file_transfer event. This has the signature
"file_transfer", TransferredFiles.
TransferredFiles has a single method getFiles(), which returns a list
of all transferred files.
- Add a new "import" program which waits for a file_transfer event and
writes files to the current directory.
- If a file_transfer event is not handled (i.e. its getFiles() method
is not called) within 5 seconds on the client, we display a toast
informing the user on how to upload a file.
- Use <p> everywhere. This is uglier, but also technically more
correct. This requires a version bump to cct-javadoc, and will give
me a massive headache when merging.
- Link against the existing OpenJDK docs.
We now wait for workers to terminate when closing the computer thread.
I'll be honest, I'm not happy with this code. Multi-threading is really
hard to get right, and I can't say I'm convinced this is especially well
behaved. I did look at trying to model this in TLA+, but in the end
decided it wasn't worth it.
In the future we probably want to split ComputerExecutor into two
objects, where one is our entry in the ComputerThread queue (and so
holds timing information) while the other is responsible for actual
execution.
This converts ComputerThread from a singleton into a proper object,
which is setup when starting a computer, and tore down when the
ComputerContext is closed.
While this is mostly for conceptual elegance, it does offer some
concrete benefits:
- You can now adjust the thread count without restarting the whole
game (just leaving and rentering the world). Though, alas, no effect
on servers.
- We can run multiple ComputerThreads in parallel, which makes it much
easier to run tests in parallel. This allows us to remove our rather
silly IsolatedRunner test helper.
Computers now use a MainThreadScheduler to construct a
MainThreadScheduler.Executor, which is used to submit tasks. Our
previous (singleton) MainThread and MainThreadExecutor now implement
these interfaces.
The main purpose of this is to better manage the lifetime of the server
thread tasks. We've had at least one bug caused by us failing to reset
its state, so good to avoid those! This also allows us to use a fake
implementation in tests where we don't expect main thread tasks to run.
As we're now passing a bunch of arguments into our Computer, we bundle
the "global" ones into ComputerContext (which now also includes the Lua
machine factory!). This definitely isn't the nicest API, so we might
want to rethink this one day.
In some ways this isn't as nice as the Forge version (requires ATs,
doesn't check texture/model existence). However, it's more multi-loader
friendly and in some cases has much less boilerplate.
Blockstate JSON files are incredibly verbose, so we add a custom JSON
pretty printer which writes things in a slightly more compact manner.
This also changes how turtle upgrades are loaded - we now support
standard ResourceLocations (so computercraft:blocks/some_turtle_upgrade)
as well as ModelResourceLocations (computercraft:items/some_turtle_upgrade#inventory).
I don't think any resource packs need to touch our upgrades, but
apologies if this breaks anything.
