When we updated spotless's Java and Kotlin formatter, we changed the underlying
formatting rules.
However, due to spotless ratcheting, these formatting changes don't get applied unless a file
gets touched in a commit.
To avoid future PRs introducing lines of change that aren't related to the intention of the PR,
this is a one-time format of all files.
The current benchmark is not too useful now that we have a linear parser.
This changes the benchmarks to parse the standard library, and also the language snippet tests.
Co-authored-by: Islon Scherer <islonscherer@gmail.com>
Making these classes caused native-image to statically initialize
them at build time, which included CLI argument default values
(like working dir).
This turns them back into classes.
Co-authored-by: Islon Scherer <islonscherer@gmail.com>
If the object member is const, it only needs to be executed once, and all children in the prototype chain can use its cached value.
There is a possible other optimization here, not included in this PR: we can avoid adding these values to the child object's cachedMembers.
The Kotlin 2.0 version upgrade made us incompatible with Gradle 8.1.
Given that it is now over 1.5 years old, this drops support for it, and
raises the minimum version to 8.2.
Currently, we run against every release version, which is taking 20
minutes to run in CI, and this will grow as Gradle creates more
releases.
This limits the amount of Gradle versions being tested against;
theoretically, only testing against the minimum version and the maximum
version should ensure that we are compatible with every version in the
range.
Nested jars built by spring boot can possibly represent classpath
resource URIs as "jar:nested:".
This changes Pkl to by default trust them with the same level as
modulepath URIs.
Don't use the system Java on Windows builds, instead download them from Adoptium.
Also:
* Fail job if curl returns a 4xx status code
* Add java version to `GradleJob`
This fixes an issue where certain modules tasks fail due to the plugin
attempting to analyze their imports, but the arguments may not actually be
Pkl modules.
For example, the pkldoc task accepts entire packages in its "sourceMoules" property.
This changes the gather imports logic to only analyze file-based modules.
This is also a performance improvement; non file-based modules are unlikely to import
files due to insufficient trust levels.
Also: fix a bug when generating pkldoc on Windows
This fixes issues when a mapping/listing delegates to another mapping/listing.
The core issue is that `cachedValues` is not guaranteed to be the complete set
of the object's members; they may be stored on the delegate instead.
Therefore, it's not correct compute hash code and equality based on `cachedValues`.
Instead, it's better to use the `getCachedValue()` method, which has the correct
logic for looking up the existing cached value.
Any thrown Pkl Errors are colored in the simple test report!
Also:
* Refactor `TextFormatter` to be more generic; rename to `TextFormattingStringBuilder`
* Adjust test report slightly (no emojis, add more spacing).
* Introduce `ColorTheme` class.
* Make stack frame descriptors colored as "faint"
Also: this changes the summary so it summarizes _all_ modules, rather than a summary per module.
---------
Co-authored-by: Islon Scherer <islonscherer@gmail.com>
Co-authored-by: Philip K.F. Hölzenspies <holzensp@gmail.com>
There is currently a bug around resolving variables within the iterable of a for
generator or spread syntax (https://github.com/apple/pkl/issues/741)
Normally, mappings/listings are type-checked lazily. However, this results in the said
bug getting widened, for any object members declared in the iterable.
As a workaround for now, prevent the bug from being any worse by ensuring that these
object members are eagerly typechecked.
Changes:
* Move class `TestResults` to package `org.pkl.core`, because it is a public class (it's the result of `Evaluator#evaluateTest`)
* Change examples to treat individual examples as assertions in the same test. Previously, they were considered different tests with an incrementing number. This better aligns with how facts are treated.
* Change `TestResults` to be a record, and introduce builders.
* Remove "module" test result section (it is not really a section).
* Add javadoc to `TestResults`
* Formatting fix: prefix ✍️ emoji just like we do the ❌ and ✅ emojis
* Consider writing examples as failures, not successes. `pkl test` will exit with code 10 if the only failing tests are due to writing examples.
The addresses the following case:
local l1: Listing<String> = new { 1 }
l2: Listing<Int> = (l1) { 2 }
The member in `l1` should be checked against its owner; to check against `Listing<String>`.
This adds a new feature to build a dependency graph of Pkl programs, following the SPICE outlined in https://github.com/apple/pkl-evolution/pull/2.
It adds:
* CLI command `pkl analyze imports`
* Java API `org.pkl.core.Analyzer`
* Pkl stdlib module `pkl:analyze`
* pkl-gradle extension `analyze`
In addition, it also changes the Gradle plugin such that `transitiveModules` is by default computed from the import graph.
This adds support for adding the `const` modifier to object members.
Such object members are also required to have the `local` modifier applied.
This follows SPICE-0011.
This changes how the language performs typechecks for mappings and
listings.
Currently, Pkl will shallow-force any Mapping and Listing to check it
the type parameter (e.g. Listing<Person> means each element is checked
to be an instance of Person).
This changes the language to check each member's type when the member
is accessed.
This also adjust test runner to handle thrown errors from within tests.
With the change to make mapping/listing typechecks lazy, we can now
correctly handle thrown errors from within a single test case.
This adjusts the test runner to consider any thrown errors as a failure
for that specific test case.
* Add `getEffectiveOutputFiles` and `getEffectiveOutputDirs` to `EvalTask`, and mark them as output files/dirs so they are tracked by Gradle. This enables implicit dependency tracking between two tasks.
* Fix usage of `file()` notation in Gradle scripts when on Windows.
This addresses an issue where network requests may fail if cert revocation checks
error, which may occur due to availability issues, or due to lack of internet access.
Revocation checking can still be enabled by setting JVM property com.sun.net.ssl.checkRevocation if on the JVM.
Also:
* Load built-in certs from resources, and move them to pkl-commons-cli
* Fix an issue where HttpInitException is not caught when loading a module
This adds release notes for 0.26.
Also:
* Adjust IJ settings to ignore certain http: strings
* Add note about http introduced in 0.26 in message passing api
Co-authored-by: Philip K.F. Hölzenspies <holzensp@gmail.com>
Rationale: "proxy" can mean very different things (e.g. java.lang.reflect.Proxy in Java).
This makes the flag name more specific.
CLI:
* `--proxy` -> `--http-proxy`
* `--no-proxy` -> `--http-no-proxy`
Gradle:
* `proxyAddress` -> `httpProxy`
* `noProxy` -> `httpNoProxy`
This fixes an issue where an error coming from loading a project file is shown as a PklBugException.
There were two problems here:
1. proxyAddress needs to be a lazy value, because it can try to load a PklProject
2. accessing proxyAddress can throw a PklException, so it needs to be within the try/catch
Instead of bundling Pkl's built-in CA certificates as a class path resource and loading them at runtime,
pass them to the native image compiler as the default SSL context's trust store.
This results in faster SSL initialization and is more consistent with how default certificates
are handled when running on the JVM.
Further related improvements:
- Remove HttpClientBuilder methods `addDefaultCliCertificates` and `addBuiltInCertificates`.
- Remove pkl-certs subproject and the optional dependencies on it.
- Move `PklCARoots.pem` to `pkl-cli/src/certs`.
- Fix certificate related error messages that were missing an argument.
- Prevent PklBugException if initialization of `CliBaseOptions.httpClient` fails.
- Add ability to set CA certificates as a byte array
- Add CA certificates option to message passing API
Now that Pkl syntax highlighting is available on GitHub,
we can remove the Groovy association with these files.
This also adds changes to mark `*.pcf` and `PklProject` files as Pkl.
This adds a language change that requires references from typealiases
to the enclosing module to be `const`.
This is required because typealiases are not late-bound.
Rationale is laid out in SPICE-0007.
Also:
* Update documentation to reflect new rules.
* Fix `Project.pkl`; mark method `const` to not break said rule.
Fix an issue where the CI job attempts to install the wrong executable
when on macOS/aarch64.
Also, adjusts native test definitions in pkl-server to match pkl-core.
This adds changes to support loading project dependencies in non-file based projects.
The design for this feature can be found in SPICE-0005: https://github.com/apple/pkl-evolution/pull/6
Changes:
* Consider all imports prefixed with `@` as dependency notation.
* Bugfix: fix resolution of glob expressions in a local dependency.
* Adjust pkl.Project:
- Allow local dependencies from a scheme-local paths.
- Disallow certain evaluator settings if not loaded as a file-based module.
* Breaking API change: `ProjectDependenciesManager` constructor now requires `ModuleResolver` and `SecurityManager`.
* Don't expose JDK internal classes; instead solve msgpack issue with `--initialize-at-run-time`.
* Use quick build mode for non-release builds: 40% faster compilation, 20% smaller executable.
* Remove options that were commented out.
* Also run ServerTest against native executable
This adds support for Windows.
The in-language path separator is still `/`, to ensure Pkl programs are cross-platform.
Log lines are written using CRLF endings on Windows.
Modules that are combined with `--module-output-separator` uses LF endings to ensure
consistent rendering across platforms.
`jpkl` does not work on Windows as a direct executable.
However, it can work with `java -jar jpkl`.
Additional details:
* Adjust git settings for Windows
* Add native executable for pkl cli
* Add jdk17 windows Gradle check in CI
* Adjust CI test reports to be staged within Gradle rather than by shell script.
* Fix: encode more characters that are not safe Windows paths
* Skip running tests involving symbolic links on Windows (these require administrator privileges to run).
* Introduce custom implementation of `IoUtils.relativize`
* Allow Gradle to initialize ExecutableJar `Property` values
* Add Gradle flag to enable remote JVM debugging
Co-authored-by: Philip K.F. Hölzenspies <holzensp@gmail.com>
GenericUrl is a catch-all that uses URL.openConnection().
Since we now have special handling of HTTP urls, it makes more sense to
put it in its own module key.
This changes the file paths to use characters that are safe for Windows.
Channges the output of the following:
* Package cache directory
* Generated pkl-doc files
* Kotlin generated code
Unsafe characters are encoded as (<hex>).
For example, the colon character `:` is encoded as `(3a)`.
Additionally, this changes the cache directory prefix (package-1 to
package-2).
Follows the design of https://github.com/apple/pkl-evolution/pull/3
This is technically a breaking change, but follows the intended use-case
and documentation of settings.
* Require that the global settings file at `~/.pkl/settings.pkl`
amends stdlib module `pkl:settings`, or otherwise sets `output.value`
to an instance of `pkl.Settings`.
* Simplify the object mapping of `PklSettings.java`
* Remove unnecessary strictfp modifier
* Add annotations to address Truffle DSL warnings (@Idempotent, @Exclusive)
* Adjust build logic to allow building cross-arch on macOS
* Add warning suppression for specialization limit (left this one as a TODO)
* Class-based converters are covariant, not invariant.
* The most specific class-based converter wins
* The first declared path-based converter that matches wins
Fixes an issue where source links are incorrectly URI encoded; i.e. `https%3A//github.com` instead of `https://github.com`.
This was causing the browser to resolve these as relative to the enclosing page.
Daniel Chao