When forking, a test could be run as many times as its `TestDef` matched
`Fingerprint`s. When forking is disabled, sbt correctly runs the tests
only once.
The sbt Server is initialized with a callback onIncomingSocket. That
callback was created in CommandExchange and held references to a build
structure and a state. Neither the state nor structure would ever go out
of scope so they effectively leaked. It is possible for each
NetworkChannel to access a recent instance of state through the
CommandExchange.withState method. Using this, we can eliminate the
references to state and build structure in the onIncomingSocket
callback. In the sbt project, this reduced the memory utilization by
about 50mb on startup.
Intellij invokes sbt with "-Djline.terminal=jline.UnsupportedTerminal"
which Terminal rewrites to the value none. When that property is set, we
should be using a jline dumb terminal. While
https://github.com/sbt/sbt/pull/5788 did fix the import functionality,
jline 3 was still emitting some ansi characters to the intellij console.
When we feed a dumb terminal to the jline 3 line reader, the ansi
control characters go away.
On linux and mac, entering ctrl+c will automatically kill any forked
processes that were created by the sbt server because sigint is
automatically forwarded to the child process. This is not the case on
windows where it is necessary to forcibly kill these processes.
On windows with jline3, inputting ctrl+c in the sbt console just causes
the input stream to return -3 unlike mac and linux where ctrl+c always
signals.
Fixes https://github.com/sbt/sbt/issues/5791
The intellij import currentlly works by forking an sbt process and
writing command input through the process input stream. To make this
work, we need the SimpleTerminal (which is used when sbt is run with
-Dsbt.log.noformat=true) to be able to read input.
Attaching the input to the simple terminal caused watch tests to fail on
windows. This can be fixed by checking if the byte read from the input
stream is -1 and ignoring it if so.
The sbt.log.noformat parameter should be treated very similarly to
sbt.io.virtual. When it is true, we should just use the raw io streams
for the process. This came up because of
https://github.com/sbt/sbt/issues/5784 which reported that intellij
imports were not working and that ansi control characters were being
written to the output.
There can be race conditions where we try to interrupt and join a ui
thread before it becomes interruptible by blockign on a queue. To
workaround this, we can add the JoinThread class which adds an
extension method Thread.joinFor that takes a FiniteDuration parameter.
This variant of join will repeatedly interrupt and attempt to join the
thread for up to 10 milliseconds before retrying until the limit is
reached. If the limit is reached, we print a noisy error to the console.
I'm not 100% sure if we are leaking threads in the latest sbt version
but this gives me more piece of mind that either we are always
successfully joining the threads or we will be alerted if the joining
fails.
Moving forward, it would be great for sbt to use the latest code to run
the scripted tests. This could help detect regressions that aren't
noticed by manual dogfooding. We can do this by first publishing
1.4.0-SNAPSHOT binaries before running tests. I didn't want to do this
for all of the builds because it does make each build that publishes
binaries take about a minute longer. The last build in the matrix is
much faster than the other builds though so I made just that build use
the snapshot version. Since that build almost always finishes first,
this should have little effect on the actual time to run ci even though
the cpu time will increase a bit.
The CI output logs for the sbt build are truncated when using sbt
1.4.0-M2 because the virtual tty provided by the build agent doesn't
seem to work well with sbt's virtual io.
The travis tty does not work with virtual io. We should disable virtual
io by default when sbt detects that it's being run in CI but until we
release 1.4.0-M3 or 1.4.0-RC1, we need this flag to fix the travis
output.
Although log manager is in the internal package, akka uses it and it
seems worth preserving binary compatibility in their build, at least for
sbt 1.4.0-M2. Once the community build is passing, we can consider
reverting this.
```
[info] welcome to sbt 1.4.0-SNAPSHOT (AdoptOpenJDK Java 1.8.0_232)
[info] loading settings for project global-plugins from ...
[info] loading global plugins from ...
[info] loading project definition from /private/tmp/hello/project
[info] loading settings for project root from build.sbt ...
[info] set current project to hello (in build file:/private/tmp/hello/)
[info]
[info] Here are some highlights of this release:
[info] - Build server protocol (BSP) support
[info] - sbtn: a native thin client for sbt
[info] - VirtualFile + RemoteCache: caches build artifacts across different machines
[info] - Incremental build pipelining. Try it using `ThisBuild / usePipelining := true`.
[info] See http://eed3si9n.com/sbt-1.4.0-beta for full release notes.
[info] Hide the banner for this release by running `skipBanner`.
[info] sbt server started at local:///Users/eed3si9n/.sbt/1.0/server/478e6db75688771ddcf1/soc
```
While running ~scripted that multiple instance of the console terminal
were instantiated which caused problems with reading input. It turned
out that RunFromSourceMain was running in the same jvm process as sbt
and creating a new console terminal in a different classloader. This
both messed up the io of scripted tests when scriptedBufferLog was set
to false but it also made it so that I couldn't exit ~ with <enter>. To
workaround this, I deferred initializaiton of the console terminal to
Terminal.withStreams which is guarded by the sbt.io.virtual system
property.
In dogfooding sbt, I found that the WriteableInputStream used by the
console terminal initialized before it was needed. This would lead to
multiple instances of the WriteableInputStream being created, which
could lead to zombie threads reading from stdin. I'm not 100% sure what
the classloading scenario was that caused this to be a problem but in a
few days of using sbt after these changes, I haven't seem zombie
threads.
When exiting the thin client, an interrupted exception stack trace ends
up being printed because NonFatal doesn't include interrupted
exceptions.
Fixes https://github.com/sbt/sbt/issues/5759
Martin Duhem
eugene yokota
Ethan Atkins
Adrien Piquerez