* Upgrade error-prone to 3.3.4
This would fix the failure with openjdk 11.0.9 in
3.3.3.
Fixed new antipatterns raised by the new version:
- Replaced unnecessary lambdas with methods.
- Switched wait/sleep calls to equivalent methods using java.time types
- Types inheriting Object.toString() should not be assigned to string
parameter in logging statements.
* CertificateChecker with checks for expiration and key length
* Add validity length check
* Get rid of hard-coded constants and DSA checks
* add files that for some reason weren't included in last commit
* Rename violations and other fixes
* Add displayMessage to CertificateViolation enum
* Switch violations from an enum to a class
* small changes
* Get rid of ECDSA checks
* add checks for old validity length
* Change error message for validity length
For some inexplicable reasons I have to move the javax.mail package one
spot up to avoid its classes being shadowed by those provided in the
appengine package...
* Enable Java 11 features
As of this commit Java 11 must be used to build. The generated bytecode
is still at Java 8 due to App Engine task queue limit.
Also fixed a bug where the included google-java-format jar file is not
used, requiring the user to install it separately.
See: https://cloud.google.com/appengine/docs/standard/java/taskqueue/push
* Get rid of all remaining JUnit 4 usages except in prober & proxy subprojects
Caveat: Test suites aren't yet implemented in JUnit 5 so we still use the ones
from JUnit 5 in the core subproject.
* Fix some build errors
1. The Gradle apt plugin is no longer needed to process annotations.
2. Without the apt plugin, Gralde puts the source files generated by
annotation processors in build/generated/sources/annotationProcessor.
3. Change the location of custom generated files to be consistent.
4. Fix a javadoc formatting error.
* Add JUnit Params and start using it
* Convert rest of RDE tests
* Don't check headers for generated tests
* Expand visibility to fix build breakage
* Bump JUnit versions to 5.6.2
SystemPropertyRule in some cases should be applied last:
when multiple rules exist and and modified property is checked
in cleanups.
ConsoleOteSetupActionTest and ConsoleRegistrarCreatorActionTest
are two such classes, and can be flaky in JUnit 4. This PR
migrates them to JUnit5 and applies ordering to extensions in
them.
Added a mockito dependency, and upgraded mockito-core to 3.3.3.
Meaningful changes: SystemPropertyRule.java and
ConsoleOteSetupActionTest.java, and
ConsoleRegistrarCreatorActionTest.java
* Add testcontainers' Junit5 support dependency
Also updated guava, dagger, hibernate, postgresql, and cloud socket factory
to latest version.
Migrated PersistenceModuleTest as an example.
Real changes:
- dependencies.gradle
- core/build.gradle
- PersistenceModuleTest.java
* Add Test suite support for JUnit 5 classes
Added Gradle dependencies and updated lockfiles.
Updated SqlInegrationTestSuite to use new annotations.
Migrated one member class in SqlIntegrationTestSuite (CursorDaoTest)
to JUnit 5, and verified that the new Suite runner can handle a
mixture of JUnit 4 and 5 tests in one suite.
Note that Gradle tests that run TestSuites must choose JUnit 4.
Updated core/build.gradle and integration/build.gradle.
* Fix broken builds when Maven Central is used
Gradle 6.2.1 apparently introduces a behavior change wrt boolean
expression: empty string used to eval to false, but now evals to
true.
Pre Gradle 6.2.1, root project's Gradle properties apparently were
not set to buildSrc. Now they are passed on to buildSrc -- mavenUrl
in buildSrc changes from null to "".
Both changes break the project when mavenUrl and/or pluginsUrl are
not set on command line.
Also added junit.jupiter-api as testCompile dependencies to projects.
This is a directly used dependency, whose absence causes a Lint
warning.
* Start using JUnit 5
This converts a single test class over to JUnit 5 (YamlUtilsTest). The main
differences you'll notice are that @RunWith isn't needed anymore, test classes
and test methods can now be package-private, and the @Test annotation comes from
the org.junit.jupiter.api package instead of org.junit. There's a lot more
differences between 4 and 5 than this that we'll need to keep in mind when
converting more test classes; for some more details, see:
https://www.baeldung.com/junit-5-migration
In order to allow JUnit 4 and 5 test classes to coexist, I've had to add two new
dependencies, org.junit.jupiter:junit-jupiter-engine and
org.junit.vintage:junit-vintage-engine, which exist in addition to junit:junit
for now. Eventually, once we've completed migrating over all JUnit 4 test
classes, then we can remove junit and junit-vintage-engine and just be left with
junit-jupiter-engine.
* Delete no longer needed lockfiles
* Merge branch 'master' into first-junit5
* Use CA cert in tests
Our tests were using a self-signed cert created by a netty utility.
That cert is not CA, and cannot be used with newer JDK versions
(starting from u242 with Ubuntu openjdk).
Now we generate a unsafe cert for tests. The new cert,
SelfSignedCaCertificate, needs to be made test-only.
To do so we need to refactor both prober and proxy.
* Upgradle JUnit to 4.13
Removed third_party/junit folder and all usage of the
JunitBackPort class. As a result, third_party is no
longer a Gradle subproject.
Minor code changes were needed to work around an
error-prone pattern: multiple statement in assertThrows'
runnable lambda.
Also third_party/activation and third_party/jsch. These
dependencies are loaded from remote maven repo. The local
copies are not in use.
* Consolidate certificate supplier module
Both the proxy and the proxy needs certificate suppliers. The PR
consolidates the module that providings those bindings to a shared
module and switched the proxy to use that module. The prober currently
uses P12 file to store its certificates. I am debating keeping that
supplier ro converting them to PEM files for simplicity.
* Rename mode enum values to be more descriptive
* Update annotation names to be more descriptive
* Allow project dependency to use runtimeClasspath
Project dependency should use runtimeClasspath. However, if
left unspecified, it uses 'default', which is the same as
the legacy 'runtime' configuration. (runtimeOnly dependencies
are left out).
Since runtimeClasspath cannot be referenced directly, we use
a custom config (deploy_jar) as a proxy.
By excluding testjars (leaked into 'compile' by third-party
dependencies) from runtimeClasspath, we prevent them from
getting into release artifacts.
Two meaningful changes in appengine_war.gradle and java_common.gradle
TESTED=Diffed contents of services/{module}/build/exploded-*
Only three jars are removed: hamcrest-core, junit, and
mockito-core.
Client SSL handler already performs the necessary validation. Only tests are
added.
Server SSL handler does not currently check for the validity period of
the client certificate as the insecure trust manager is used. This PR
added the check but does not actually terminate the connection yet. It
will log the expired certificates so that we can contact the registrars
to update them.
Once we are certain that all certificates are updated, we can turn off
dryrun mode.
We should also consider checking if the certificate has too long a
validity period as it defeats the purpose of using regularly updated
certificates to deprecate insecure cipher suites.
The prober now uses the common SSL initializer in the networking
subproject.
Also changed both initializers to take an ImmutableList of certificates
other than an array of those, for better immutability.
I have no idea where these lockfile changes are coming from. They seem
to be pure noise as far as code review is concerned.
* Break circular dependency between core and util
Created a new :common project and moved a minimum
number of classes to break the circular dependency
between the two projects. This gets rid of the
gradle lint dependency warnings.
Also separated api classes and testing helpers into
separate source sets in :common so that testing
classes may be restricted to test configurations.
* Add a converter for CurrencyUnits stored in the database
This uses the well-known String representation for currency units. It also
provides a base class for other converters that will be persisting the
toString() representation.
* Add DB and formatting changes
* Add tests, make minor fixes
* Instrument proxy frontend latency metric
Backend latency only captures the time spent waiting for the GAE backend
to respond to a request, which is not representitive of what clients
experience. For instance it does not take into account the time the
proxy spends on processing the requests and the time it takes to send a
response to the client.
This PR adds a metric for frontend latency, which is the time spent on
answering a client request. This should serve as a better proxy for the
latency clients observe.
* Rename AbstractMetrics to BaseMetrics
Export of cl/270900150.
To refer to a KMS key or key ring, we should use the stable `.self_link`. Using `.id` instead provides an unstable identifier which may change (and it will change in the upcoming update of the google provider to 2.9.1). A change in the identifier will cause Terraform to destroy and recreate the key. Destroying the key means all data associated with it is lost; the key cannot be recreated.
This CL replaces `.id` with `.self_link`, so all of those problems will not happen. In addition, `prevent_destroy` protects the key against delete-and-recreate in general.
* Upgrade to Truth 1.0
Refactored fail(...) to assertWithMessage().fail().
Upgraded com.google.monitoring-client family of dependencies to 1.0.6
Also fixed bad use of io.StringIO (on binary buffer) recently introduced to
google-java-format-diff.py.
* Fix dependency-locking config
Reenable dependency locking after a bug errorneouly turned it off.
Removed the guava-related workaround that forcefully resolve to
the -jre distribution.
Enabled locking for buildSrc by updating its property file.
Updated all lock files.
https://github.com/google/nomulus/pull/129 migrated `GoogleCredential`
to `GoogleCredentialsBundle` and introduced a subtle bug. I don't fully
understand why but there are times when the access token is null but
`credentials.refresh()` is not called, resulting in NullPointerException
when `credentials.getAccessToken().getTokenValue()` is called.
Since the new GoogleCredentials class supports `shouldRefresh()`, we now
just rely on it to make sure that we always get a value access token.
