Java chrome dev tools add performance api #1
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Our `NetworkUtils` returns the first public address it finds in its list of NICs. This is normally fine, but there's no guarantee that the first address found is actually routeable or the default address of the machine. That means that anyone doing a look up of the address by host name may not be able to connect to the machine. Resolve this problem by attempting to ensure that the first NIC we return in our list of NICs is the default one.
… test When using the OSS protocol dialect, ChromeDriver does not handle unexpected alerts encountered when executing async JavaScript. If the alerts tests are enabled, many remaining tests fail because the alert is left active. The tests are still enabled, because when ChromeDriver uses the W3C protocol dialect, the alerts are properly handled. The addition of the attribute on the test class is to ensure that when running with the legacy OSS protocol dialect, that those cascading failures are eliminated. The attribute can be removed once ChromeDriver fully supports the W3C dialect and has made that dialect the default.s
… a directory. Fixes issue SeleniumHQ#7061
Signed-off-by: Alexei Barantsev <[email protected]>
Chrome WebDriver initializer modifies incoming options for driver. Ruby's "good practices" generally boil down to not mutating data you don't own.
…citly build server before using it
This sets the stage for us pulling the OSS dialect out into a legacy package without us breaking the java module system's rules on split packages.
It should be possible to start Grid TNG using a module-path.
This implementation does simple delegation to the OkHttp library, but is sympathetic to an implementation based on Java 11's WebSocket class.
This currently always assumes that it's possible to make a direct connection to the CDP host, which may not always be true.
It has been announced that OpenTracing and OpenCensus will be merging to create a unified API. https://medium.com/opentracing/merging-opentracing-and-opencensus-f0fe9c7ca6f0 This merged API has the goal of supporting both extant distributed tracing APIs, so rather than support both, we can standardise on one. I strongly suspect that the ultimate API will look a lot like OpenCensus, but right now OpenTracing is more widely adopted, and so that's the underlying API we shall use.
Introducing JavaScriptException, MoveTargetOutOfBoundsException, and WebDriverArgumentException. These all descend from WebDriverException, but having these distinct exception types allows users to better distinguish between differing error conditions.
No functional changes
No functional changes
Previously, the webserver target would clobber the webserver directory in the runfiles tree, and we wouldn't have access to webserver/keystore at runtime.
on startup. This is not ideal with bazel since we're still sorting
out build rules and the data dependencies haven't been nailed down.
As a quick fix to move forward, introduce a system property to
ignore the missing JS roots:
-Dorg.openqa.selenium.environment.webserver.ignoreMissingJsRoots=true
The server will still crash if it cannot find common/src/web, as
those dependencies are critical for the selenium test suite.
We don't really need to start its connection. Closes SeleniumHQ#7158
This allows to share a logic for setting default headers (e.g User-Agent) and any additional processing required to make HTTP requests. Closes SeleniumHQ#6662
The BenderProxy project is nothing more than a renamed and properly packaged version of the FryProxy fork we were using previously. Attempts to contact the owner of FryProxy through multiple channels concerning incorporating the dependency removal changes we require, so the fork has now been reversioned and published under another name. The package has been published to NuGet, as Bazel does not currently handle checked-in dependencies.
This target will run the tests in the Bazel build environment. At present, the tests run against Chrome. A future commit will enable executing tests against any of the supported browser configurations.
In case the user is trying to build the entire project using `./go build` or trying to build the standalone-server. The builds were failing with the below exception : BUILD FAILED: The rule //java/client/test/org/openqa/selenium/remote:common-tests could not be found. This was caused because the tasks in the Rakefile were still referring to "common-tests" task instead of the new "small-tests"
It doesn't play nicely with IEDriverServer which responds with text/html content-type, while client only supports application/json Basically reverts 8389311 while ensuring that default headers are used so that User-Agent is set properly
All of java/client/src should build with bazel. The java/client/test tree is a work in progress. Running tests is the current bottleneck as anything that tries to start a browser crashes due to the bazel sandbox: My latest results for `bazel test java/client/test/...`: //java/client/test/org/openqa/selenium/atoms:CompiledAtomsNotLeakingTest (cached) PASSED in 0.3s //java/client/test/org/openqa/selenium/atoms:InputAtomsTest (cached) PASSED in 0.4s //java/client/test/org/openqa/selenium/io:CircularOutputStreamTest (cached) PASSED in 0.7s //java/client/test/org/openqa/selenium/io:FileHandlerTest (cached) PASSED in 0.7s //java/client/test/org/openqa/selenium/io:TemporaryFilesystemTest (cached) PASSED in 0.7s //java/client/test/org/openqa/selenium/io:ZipTest (cached) PASSED in 0.7s //java/client/test/org/openqa/selenium/net:LinuxEphemeralPortRangeDetectorTest (cached) PASSED in 0.4s //java/client/test/org/openqa/selenium/net:NetworkUtilsTest (cached) PASSED in 0.4s //java/client/test/org/openqa/selenium/net:UrlCheckerTest (cached) PASSED in 1.1s //java/client/test/org/openqa/selenium/os:CommandLineTest (cached) PASSED in 3.0s //java/client/test/org/openqa/selenium/support:small-tests (cached) PASSED in 2.5s //java/client/test/org/openqa/selenium/testing:IgnoreComparatorUnitTest (cached) PASSED in 0.7s //java/client/test/org/openqa/selenium/support:large-tests FAILED in 12.2s
It was released from rb-3.x branch
The W3C WebDriver Specification does not recognize any property in the payload for the new session command other than "capabilities". The IE driver has long supported the presence of "desiredCapabilities" as a fallback for older, obsolete language bindings. Now that we are in the process of creating the Selenium 4.0 language bindings, this legacy compatibility should no longer be necessary. Also, this commit includes some long-overdue code formatting for the new session command handler. These changes are cosmetic only, and do not alter functionality.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Xin the preceding checkbox, I verify that I have signed the Contributor License Agreement