The metadata.error change was brought back in
https://github.com/microsoft/playwright/pull/29271and it broke java port
as we could have error and result set simulteniously. This PR moves the
logic to the trace recorder instead and keeps the protocol contract
clear that either error or result is present, but not both.
Otherwise, we forever block SIGTERM and SIGHUP by registering a handler
that does not do anything (due to no browsers to close) and prevents
default handler that exits from running.
Fixes#28091.
This moves the fix in #27095 from `modernize` to `appendEvent`. The
reason is that `trace V4` is used both for older traces that do not have
`consoleMessage.args` and the new ones with `args`. Since we do not call
`modernize` for traces of the same version, the original fix does not
help in this case.
Fixes#27144.
The bare-bones `debian` distribution we use for testing doesn't have
`ps`. This patch switches to reading `/proc` file system directly on
Linux instead of relying on `ps`.
Performance measurements for the 20000 active processes on Debian Docker
container, tested on my M1 Max Mac:
- the `ps -eo pid,ppid,pgid` call + parsing takes 293ms
- the manual synchronous `/proc` traversal + parsing takes 326ms
So this is ~10% perf penalty.
Drive-by: rename `pgid` into `pgrp` so that it stands out from `pid`
(process ID) and `ppid` (parent process ID).
This patch:
- changes the `childProcess` fixture to reliably SIGKILL all descendants
(children and grand-children, regardless of their process group).
This is achieved using the `ps` command to build the process tree, and
then send
`SIGKILL` to the descendant process groups.
- changes the `runCLI` fixture to **not** auto-close codegen by default;
the `childProcess` fixture will clean up all processes. This makes
sure that all `runCLI.waitFor()` commands actually wait until the
necessary
output.
- for a handful of tests that do actually want to auto-close codegen,
introduce an optional `autoCloseWhen` flag for the `runCLI` fixture
that makes sure to close the codegen once a certain output was reached.
This reverts commit 303c5998f8.
Reason for revert: I tried enabling `ssim-cie94` by default on
ionic-framework test suite, and it proves to be overly strict for their
usecase.
The option defines a comparator to be used to compare images.
Possible values are `"pixelmatch"` and `"ssim-cie94"`.
Note: This reverts commit 8167f8bf54.
This removes everything related to docker integration experiments that
we conducted over the last 6 months.
I'll send a follow-up with an alternative suggestion that was demo'ed on
a team meeting in the end of December.
This patch:
- updates documentation to lead users from `TestConfig.snapshotDir` and
`testInfo.snapshotSuffix` to `TestConfig.snapshotPathTemplate` as a
better and more flexible alternative.
- drops `{snapshotSuffix}` from documentation
- stops using `snapshotSuffix = ''` in our own tests and switches us
to the `snapshotPathTemplate`.
- adds `{testName}` token.
This patch implements a new mode of network tethering for Playwright
server & its clients.
With this patch:
- playwright server could be launched with the
`--browser-proxy-mode=tether` flag to engage in the new mode
- a new type of client, "Network Tethering Client" can connect to the
server to provide network traffic to the browsers
- all clients that connect to the server with the `x-playwright-proxy:
*` header will get traffic from the "Network Tethering Client"
This patch also adds an environment variable
`PW_OWNED_BY_TETHER_CLIENT`. With this env, playwright server will
auto-close when the network tethering client disconnects. It will also
auto-close if the network client does not connect to the server in the
first 10 seconds of the server existence. This way we can ensure that
`npx playwright docker start` blocks terminal & controls the lifetime of
the started container.
This makes `HttpServer` accept `preferredPort` option that will first
try to listen on that port, and if that port is already in use, listen
on some available port instead.
Fixes#17201.
This moves some expect() matchers tests from test runner tests to page tests,
because these are implemented through a library call anyway.
Makes tests more readbable, faster and easier to test specific details.
