| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| ... | |
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Essentially, we fuse mixin and lazyvals into the fields phase.
With fields mixing in trait members into subclasses, we
have all info needed to compute bitmaps, and thus we can
synthesize the synchronisation logic as well.
By doing this before erasure we get better signatures,
and before specialized means specialized lazy vals work now.
Mixins is now almost reduced to its essence: implementing
super accessors and forwarders. It still synthesizes
accessors for param accessors and early init trait vals.
Concretely, trait lazy vals are mixed into subclasses
with the needed synchronization logic in place, as do
lazy vals in classes and methods. Similarly, modules
are initialized using double checked locking.
Since the code to initialize a module is short,
we do not emit compute methods for modules (anymore).
For simplicity, local lazy vals do not get a compute method either.
The strange corner case of constant-typed final lazy vals
is resolved in favor of laziness, by no longer assigning
a constant type to a lazy val (see widenIfNecessary in namers).
If you explicitly ask for something lazy, you get laziness;
with the constant-typedness implicit, it yields to the
conflicting `lazy` modifier because it is explicit.
Co-Authored-By: Lukas Rytz <lukas@lightbend.com>
Fixes scala/scala-dev#133
Inspired by dotc, desugar a local `lazy val x = rhs` into
```
val x$lzy = new scala.runtime.LazyInt()
def x(): Int = {
x$lzy.synchronized {
if (!x$lzy.initialized) {
x$lzy.initialized = true
x$lzy.value = rhs
}
x$lzy.value
}
}
```
Note that the 2.11 decoding (into a local variable and a bitmap) also
creates boxes for local lazy vals, in fact two for each lazy val:
```
def f = {
lazy val x = 0
x
}
```
desugars to
```
public int f() {
IntRef x$lzy = IntRef.zero();
VolatileByteRef bitmap$0 = VolatileByteRef.create((byte)0);
return this.x$1(x$lzy, bitmap$0);
}
private final int x$lzycompute$1(IntRef x$lzy$1, VolatileByteRef bitmap$0$1) {
C c = this;
synchronized (c) {
if ((byte)(bitmap$0$1.elem & 1) == 0) {
x$lzy$1.elem = 0;
bitmap$0$1.elem = (byte)(bitmap$0$1.elem | 1);
}
return x$lzy$1.elem;
}
}
private final int x$1(IntRef x$lzy$1, VolatileByteRef bitmap$0$1) {
return (byte)(bitmap$0$1.elem & 1) == 0 ?
this.x$lzycompute$1(x$lzy$1, bitmap$0$1) : x$lzy$1.elem;
}
```
An additional problem with the above encoding is that the `lzycompute`
method synchronizes on `this`. In connection with the new lambda
encoding that no longer generates anonymous classes, captured lazy vals
no longer synchronize on the lambda object.
The new encoding solves this problem (scala/scala-dev#133)
by synchronizing on the lazy holder.
Currently, we don't exploit the fact that the initialized field
is `@volatile`, because it's not clear the performance is needed
for local lazy vals (as they are not contended, and as soon as
the VM warms up, biased locking should deal with that)
Note, be very very careful when moving to double-checked locking,
as this needs a different variation than the one we use for
class-member lazy vals. A read of a volatile field of a class
does not necessarily impart any knowledge about a "subsequent" read
of another non-volatile field of the same object. A pair of
volatile reads and write can be used to implement a lock, but it's
not clear if the complexity is worth an unproven performance gain.
(Once the performance gain is proven, let's change the encoding.)
- don't explicitly init bitmap in bytecode
- must apply method to () explicitly after uncurry
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
- Modify `tools/scaladoc-diff` to use sbt instead of ant.
- Move `stability-test.sh` from `tools` to `scripts`. With the new
build process without separate `locker` and `strap` stages, it doesn’t
make sense to call this script without first setting up the proper
test environment in a CI build.
- Replace the use of `build.number` in `bootstrap` with a new
`SHA-NIGHTLY` mode for `baseVersionSuffix`.
- Make `partest` call sbt instead of ant for initializing the classpath
and use the new classpath location (`quick` instead of `pack`).
|
| |\ \
| | |
| | | |
Switch Windows CI build to sbt (w/ some sbt build improvements) [ci: last-only]
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
- Use sbt in `integrate/windows`: This essentially combines the Unix CI
jobs `validate/publish-core` and `validate/test`, first publishing a
local release built with STARR and then building a new version with
that and running all tests on it.
- Unify repository handling across build scripts: A new function
`generateRepositoriesConfig` in `common`, based on the existing code
in `integrate/bootstrap`, writes the `repositories` file for sbt,
either with or without an extra bootstrap repository for resolving a
previously built version. It is used in all CI scripts to ensure that
artifacts are only resolved through the sanctioned proxies and
upstream repositories.
- The repository URL arguments in `setupPublishCore` and
`setupValidateTest` are now optional as well. These commands are used
without a URL from `integrate/windows`, which publishes to `local`
instead of a temporary remote repository.
- `testAll` is now a task instead of a command. It runs the same
sequence of sub-tasks as before but does not propagate failures
immediately. It always runs all subtasks and reports errors at the
end.
- The `generateBuildCharacterPropertiesFile` task now includes all
properties from `versions.properties` (whose values have potentially
been overwritten with `-D` options) in `buildcharacter.properties`.
|
| |/ /
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Use File.pathSeparator when processing classpath instead of just :
to don't end up with classpath like "C:\sth\a.jar:C:\sth\b.jar" what
was causing problems after split(":").
Display questions first and then wait for user input. I tested on 3
computes and without flush the printed questions were always displayed
just after user replied - never before.
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
All of the individual ant builds that occured during `bootstrap` are
replaced by equivalent sbt builds.
- Allow extra dashes in version suffix when using SPLIT
- Clean up ScriptCommands
- Building an extra `locker` for stability testing with ant was not
necessary but sbt also drops `strap`, so we need to build again
with `quick` to get the equivalent of `strap`. The script for checking
stability is invoked directly from the bootstrap script, not from sbt.
- `STARR` and `locker` build output is still logged to `logs/builds`,
the main build runs log directly to the main console with colored
output.
- Allow `—show-log` option on partest command line in sbt
- Normalize inferred LUB in `run/t7747-repl.scala`
- Add `normalize` feature from `ReplTest` to `InteractiveTest`
- Normalize inferred LUBs in `presentation/callcc-interpreter`
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This commit switches to using 2.12.0-M3-dc9effe as STARR,
so that we can switch to the new trait encoding where
each concrete trait member gets a a static member,
which has the actual implementation (as well as
serving as a target for for super calls using invokestatic),
and a default member (forwards to the static member).
Also bump partest to 1.0.17 -- the release that goes
with the in-sourcing of scalacheck.
Replace a few more -Yopt with -opt (for our new STARR)
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
- More memory for compiling partest tests
- Enable more detailed logging for partest failures:
Partest’s `AntRunner` has extra code for logging details of all failed
tests at the end of a run which is not present in `SbtRunner` but we
can set the `--show-log` and `--show-diff` options to get similar
output interspersed with the main test log.
- Update OSGi test dependencies for sbt build to same versions as in ant
(see 8247b8f077c96ba9a017b53de4d8062fe428c4a7 for the motivation for
this change).
- Disable MiMa checks (to be reactivated after M5).
|
| |\| |
|
| | | |
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
- Support directories in `-doc-external-doc`: It is documented as
accepting a “classpath_entry_path” for the keys but this only worked
for JARs and not for individual class files. When checking for
external-doc mappings for a Symbol, we now find the root directory
relative to a class file instead of using the full class file path.
The corresponding tests for SI-191 and SI8557 are also fixed to
support individual class files instead of JARs in partest. This is
required for the sbt build which runs partest on “quick” instead of
“pack”.
- Fix version and repository handling for bootstrapping. The bootstrap
`scalaInstance` can now be resolved from any repository added to the
project (not just the bootstrap repositories) by using a different
workaround for https://github.com/sbt/sbt/issues/1872.
- Workaround for https://github.com/sbt/sbt/issues/2640 (putting the
wrong `scalaInstance` on partest’s classpath). The required
`ScalaInstance` constructor is deprecated, so we have to disable
deprecation warnings and fatal warnings until there is a better fix.
- Add MiMa to the sbt build (port of the old `test.bc` ant task). The
sbt-mima plugin doesn’t have all the features we need, so we do it
manually in a similar way to what the plugin does. Checks are done
in both directions for the `library` and `compiler` projects. The
base version has to be set in `build.sbt`. When set to `None`, MiMa
checks are skipped. MiMa checks are run sequentially to avoid spurious
errors (see https://github.com/typesafehub/migration-manager/issues/115).
- Port the OSGi tests to the sbt build. The set of JARs that gets copied
into build/osgi as bundles is a bit different from the ant build. We
omit the source JARs but add additional modules that are part of the
Scala distribution, which seems more correct.
- Get rid up `pull-binary-libs.sh` for the sbt build. Add artifacts are
resolved from the special bootstrap repository through Ivy. The
special `code.jar` and `instrumented.jar` artifacts are copied to the
location where partest expects them (because these paths are hardcoded
in partest). Other extra JARs for partest in `test/files/lib` are
referenced directly from the Ivy cache.
- Move common settings that should be available with unqualified names
in local `.sbt` files and on the command line into an auto-plugin.
- Add an `antStyle` setting to sbt to allow users to easily enable
ant-style incremental compilation instead of sbt’s standard name
hashing with `set antStyle := true`.
- Disable verbose `info`-level logging during sbt startup for both,
`validate/test` and `validate/publish-core` jobs. Update logging is
no longer disabled when running locally (where it is useful and
doesn’t generate excessive output).
- Pass optimization flags for scalac down to partest, using the new
partest version 1.0.15\6.
- Call the new sbt-based PR validation from `scripts/jobs/validate/test`.
- Disable the tests `run/t7843-jsr223-service` and `run/t7933` from
https://github.com/scala/scala/pull/4959 for now. We need to set up
a new test project (either partest or junit) that can run them on a
packaged version of Scala, or possibly move them into a separate
project that would naturally run from a packaged Scala as part of the
community build.
|
| |\| |
|
| | | |
|
| |\ \
| | |
| | | |
SI-2712 Add support for partial unification of type constructors
|
| | | | |
|
| |/ /
| |
| |
| | |
Keep -Yopt-inline-heuristics and -Yopt-trace unchanged
|
| |\|
| |
| |
| |
| |
| |
| |
| |
| | |
merge/2.11.x-to-2.12.x-20160517
Conflicts:
build.sbt
test/files/run/repl-javap-app.check
test/files/run/repl-javap-app.scala
|
| | |\
| | |
| | | |
Eliminate major sources of daily noise in SBT build.
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
- Intercept incorrect "binary conflict" warning issued by SBT.
Fixes https://github.com/scala/scala-dev/issues/100
- Bump to a new version of pantsbuild/jarjar to fix an
incompatibility with Java 8 parameter names in class
files, which we run into on the 2.12.x branch. See:
https://github.com/pantsbuild/jarjar/pull/19
- Disable info level logging for dependency resolve/download.
|
| |\| | |
|
| | |\ \
| | |/
| |/| |
Publish usable Scala distributions to Ivy repositories
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
- sbt requires a `default` configuration in the Scala distribution but
doesn’t publish such a configuration to Ivy repositories by default.
This is not a problem when publishing to a Maven repository because
Maven doesn’t use the concept of configurations and Ivy creates a
standard set (including `default`) when resolving artifacts from Maven
repositories, but it prevents the use of any Scala distribution
published with `publishLocal`.
The underlying issue is that sbt requires `default` instead of
`default(compile)`. We work around this limitation by publishing a
dummy `default` configuration.
- sbt has hardcoded dependencies on the `scala-library` artifact of type
`jar`. In the tradition of `sbt-osgi` we used type `bundle` when
publishing via Ivy but this makes the artifacts unusable from sbt. We
now publish the OSGi bundles directly as type `jar` (which is
compatible with how they appear in Ivy after resolving from a Maven
repository).
- We have to be more aggressive about not publishing certain
subprojects, otherwise `ivy.xml` files could still be published even
when using `publishArtifacts := false`.
- `removePomDependencies` now also modifies `ivy.xml` in addition to
the Maven POM so that bogus dependencies do not leak into the Ivy
descriptors.
Fixes https://github.com/scala/scala-dev/issues/130
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Implicit conversions are now in package convert as ImplicitConversions,
ImplicitConversionsToScala and ImplicitConversionsToJava.
Deprecated WrapAsJava, WrapAsScala and the values in package object.
Improve documentation.
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | | |
The backend replaces .box / .unbox methods by corresponding invocations
to BoxesRunTime, but not for Unit.
This commit restores the body of `Unit.box` and `Unit.unbox`.
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
The sbt command `generateSources` now generates both the AnyVal
sources and the tuple/function sources (previously done by `genprod`).
Source generation is part of the sbt build, the `scala.tools.cmd.gen`
package is removed from `scala-compiler`. This simplifies bootstrapping.
Generated sources are still in the same location and checked into git.
The shell scripts `tools/codegen` and `tools/codegen-anyvals` are
also removed. The ant build and the build scripts do not call these.
Regenerating sources is a manual step at the moment.
|
| |\ \ \
| | | |
| | | | |
Seal collection classes that were annotated with deprecatedInheritance in 2.11.0
|
| | | | | |
|
| |\ \ \ \
| | |/ /
| |/| | |
|
| | | |/
| |/| |
|
| | |/
|/| |
|
| |\| |
|
| | |\
| | |
| | | |
upgrade to sbt 0.13.11
|
| | | |
| | |
| | |
| | | |
yum! delicious dogfood!
|
| |\| |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
merge/2.11.x-to-2.12.x-20160225
Conflicts:
scripts/jobs/integrate/bootstrap
src/build/maven/scala-actors-pom.xml
test/files/pos/t3420.flags
Conflicts were trivial to resolve.
|
| | |/ |
|
| | |
| |
| |
| |
| |
| | |
Added a deprecation warning for `-optimize`.
Later we'll also graduate `-Yopt` to `-opt`, probably for 2.12.0-M5.
|
| |\| |
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
`scala -deprecation` (without a trailing script argument)
wasn't allowed. Now it is.
I also supported trailing whitespace on all commands.
Also fixed: a bug with completion of `scalac ./san<TAB>`.
It was completing as though the `./` had not been typed,
which threw the suggestion off by a few characters.
|
| |\|
| |
| |
| |
| |
| |
| |
| |
| | |
Conflicts:
src/library/scala/collection/Iterator.scala
|
`-- trivial conflicts only. Parens were added to the
next() calls in 2.12.x, while in the meantime
`{Concat,Join}Iterator` were optimized in 2.11.x
|
| | |
| |
| |
| |
| | |
Also refactor the partest parser to use the improved tab
completion for test paths.
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
We can complete partest options (I've excluded some that
aren't relevant in SBT), as well as test file names.
If `--srcpath scaladoc` is included, completion of test
paths will be based on `test/scaladoc` rather than
the default `test/files`. Note that the `--srcpath` option
is currently broken via scala partest interface, this
change to scala-partest is needed to make it work:
https://github.com/scala/scala-partest/pull/49
I've also hijacked the `--grep` option with logic
in the SBT command itself, rather than passing this
to `partest`. Just like `./bin/partest-ack`, this looks
for either test file names or regex matches within
the contents of test, check, or flag files.
I tried for some time to make the tab completion of
thousands of filenames more user friendly, but wasn't
able to get something working. Ideally, it should
only suggest to `test/files/{pos, neg, ...}` on the
first <TAB>, and then offer files on another TAB.
Files should also be offered if a full directory
has been entered. Hopefully a SBT parser guru will
step in and add some polish here.
|
| | | |
|
| |\|
| |
| |
| |
| |
| | |
Conflicts:
build.sbt
scripts/jobs/integrate/bootstrap
|
| | |\
| | |
| | | |
sbt-based PR validation job
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
# Improve version handling in the sbt build:
The new settings `baseVersion` and `baseVersionSuffix` make it easier
to set version numbers for different kinds of builds in a consistent
way without e.g. having to first get a git SHA outside of sbt.
The new task `generateBuildCharacterPropertiesFile` writes the file
`buildcharacter.properties` to the root dir. The format is compatible
with the same file produced by the ANT build but it only contains a
subset of the properties, in particular the Maven version, which is
needed in publishing scripts and passed around between different
Jenkins jobs as `jenkins.properties`.
SHAs in version numbers are consistently shortened to 7 digits (as used
by git and github). Previously we used 7 digits in Maven snapshot
version numbers but 10 digits in canonical and OSGi version numbers.
# Add Jenkins script support to the sbt build:
The new command `setupPublishCore` takes the PR validation snapshot
repository as an argument and changes the required settings for the
`publish-core` build (use SHA-SNAPSHOT versioning, compile with
optimization enabled, do not publish scaladoc sets)
For example, the following command can be used to generate
`buildcharacter.properties` with the version numbers needed for PR
validation builds:
sbt setupPublishCore dummy generateBuildCharacterPropertiesFile
The sbt build will now automatically detect and use a “~/.credentials”
file with the credentials for publishing to a remote repository.
# Call sbt from `publish-core`:
The correct`$SBT_CMD` is set directly in `bootstrap` and used by
`publish-core` to first generate `buildcharacter.properties` and then
build and publish. Parsing the git revision, computing a version
number and getting binary dependencies are no longer required in the
script. This is all done in the sbt build.
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | | |
We now normalize EOLs in `ScalaTool` to LF when reading the templates,
and write them as CRLF or LF depending on the target platform. This
allows the script templates to be stored locally with either LF or CRLF
line endings (which can both happen, depending on the platform and git
configuration).
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
The generated scripts for "quick" are not portable anyway (neither in
the sbt build nor in the ant build) because they contain absolute
paths. When using the sh versions on Windows in MSYS or Cygwin, the
classpath must be in Windows format because "java" expects it that way.
Note that none of this applies to "pack" (and thus to distribution
builds) where the classpath is constructed dynamically by the launcher
scripts itself.
|
| | |/ |
|
| |\|
| |
| |
| |
| |
| |
| |
| |
| | |
Conflicts:
src/compiler/scala/tools/nsc/backend/opt/ConstantOptimization.scala
src/compiler/scala/tools/nsc/transform/Constructors.scala
src/compiler/scala/tools/nsc/typechecker/Contexts.scala
src/scaladoc/scala/tools/nsc/doc/html/page/Template.scala
src/scaladoc/scala/tools/nsc/doc/html/resource/lib/jquery.layout.js
|
| | |
| |
| |
| |
| |
| |
| |
| |
| | |
- Create the directory for the output file when building an OSGi
bundle. Otherwise a publishing command would fail if dist/mkPack had
not been run before.
- Depend only on “packagedArtifact” in “mkPack”, not on “packageBin”.
The latter is sbt’s standard packaging task but some of the required
dependencies need to be packaged as OSGi bundles.
|
