This is a follow-up to #35049 with a few minor fixes related to using
the browser provided by `puppeteer` to run tests. Included fixes:
- Make the `webdriver-manager-update.js` really portable. (Previously,
it needed to be run from the directory that contained the
`node_modules/` directory. Now, it can be executed from a subdirectory
and will correctly resolve dependencies.)
- Use the `puppeteer`-based setup in AIO unit and e2e tests to ensure
that the downloaded ChromeDriver version matches the browser version
used in tests.
- Use the `puppeteer`-based setup in the `aio_monitoring_stable` CI job
(as happens with `aio_monitoring_next`).
- Use the [recommended way][1] of getting the browser port when using
`puppeteer` with `lighthouse` and avoid hard-coding the remote
debugging port (to be able to handle multiple instances running
concurrently).
[1]: https://github.com/GoogleChrome/lighthouse/blame/51df179a0/docs/puppeteer.md#L49
PR Close#35381
This means integration tests no longer need to depend on a $CI_CHROMEDRIVER_VERSION_ARG environment variable to specify which chromedriver version to download to match the locally installed chrome. This was bad DX and not having it specified was not reliable as webdriver-manager would not always download the chromedriver version to work with the locally installed chrome.
webdriver-manager update --gecko=false --standalone=false $CI_CHROMEDRIVER_VERSION_ARG is now replaced with node webdriver-manager-update.js in the root package.json, which checks which version of chrome puppeteer has come bundled with & downloads informs webdriver-manager to download the corresponding chrome driver version.
Integration tests now use "webdriver-manager": "file:../../node_modules/webdriver-manager" so they don't have to waste time calling webdriver-manager update in postinstall
"// resolutions": "Ensure a single version of webdriver-manager which comes from root node_modules that has already run webdriver-manager update",
"resolutions": {
"**/webdriver-manager": "file:../../node_modules/webdriver-manager"
}
This should speed up each integration postinstall by a few seconds.
Further, integration test package.json files link puppeteer via file:../../node_modules/puppeteer which is the ideal situation as the puppeteer post-install won't download chrome if it is already downloaded. In CI, since node_modules is cached it should not need to download Chrome either unless the node_modules cache is busted.
NB: each version of puppeteer comes bundles with a specific version of chrome. Root package.json & yarn.lock currently pull down puppeteer 2.1.0 which comes with chrome 80. See https://github.com/puppeteer/puppeteer#q-which-chromium-version-does-puppeteer-use for more info.
Only two references to CI_CHROMEDRIVER_VERSION_ARG left in integration tests at integration/bazel-schematics/test.sh which I'm not entirely sure how to get rid of it
Use a lightweight puppeteer=>chrome version mapping instead of launching chrome and calling browser.version()
Launching puppeteer headless chrome and calling browser.version() was a heavy-handed approach to determine the Chrome version. A small and easy to update mappings file is a better solution and it means that the `yarn install` step does not require chrome shared libs available on the system for its postinstall step
PR Close#35049
Previously we needed the `components-repo-ci` blocklist to disable
tests that were failing during the development of Ivy. Since we fixed
all those failing tests, and we don't want to regress, we can remove the
blocklist logic.
Resolves FW-1807
PR Close#35115
* Added a /tools/saucelabs/sauce-service.sh script that manages the sauce-connect as a service which is used by the karma-saucelabs.js wrapper to start the service.
* Added /tools/saucelabs/README.md that covers the details of SauceLabs karma testing with Bazel.
PR Close#34769
We rename the `material-unit-tests` job to `components-repo-unit-tests`
because the job runs all unit tests found in the Angular Components repository.
This includes the Angular CDK, Angular Material and more. Also the repository has
been renamed from `angular/material2` to `angular/components` in the past.
PR Close#34898
The components repository has a Yarn resolution to ensure that
dgeni-packages uses a specific TypeScript version. This resolution
causes the specified TS version to be considered as candidate for the
`@angular/bazel` peer dependency. Ultimately, Yarn decides to use
the TypeScript version from the resolution for `@angular/bazel`,
and builds will fail due to a version mismatch.
This is because `tsickle` will use the hoisted top-level TS
version (set to `3.7.4` ), while `@angular/bazel` uses the
version from the resolution (at the time of writing: v3.6.4)
PR Close#33717
Currently we only run Saucelabs on PRs using the legacy View Engine
build. Switching that build to Ivy is not trivial and there are various
options:
1. Updating the R3 switches to use POST_R3 by default. At first glance,
this doesn't look easy because the current ngtsc switch logic seems to
be unidirectional (only PRE_R3 to POST_R3).
2. Updating the legacy setup to run with Ivy. This sounds like the easiest
solution at first.. but it turns out to be way more complicated. Packages
would need to be built with ngtsc using legacy tools (i.e. first building
the compiler-cli; and then building packages) and View Engine only tests
would need to be determined and filtered out. Basically it will result in
re-auditing all test targets. This is contradictory to the fact that we have
this information in Bazel already.
3. Creating a new job that runs tests on Saucelabs with Bazel. We specify
fine-grained test targets that should run. This would be a good start
(e.g. acceptance tests) and also would mean that we do not continue maintaining
the legacy setup..
This commit implements the third option as it allows us to move forward
with the general Bazel migration. We don't want to spend too much time
on our legacy setup since it will be removed anyway in the future.
PR Close#34277
Before updating to remove the compile build variable, we must update
the components unit test integrations to a sha in the components
repo which no longer relies on the compile build variable.
PR Close#34280
We need to migrate to using angular_ivy_enabled value to determine whether to use
Ivy or ViewEngine for package building scripts and for size-tracking and
symbol-extract tooling.
PR Close#33983
The change type was only recorded for `aio/` and was not correct anyway.
For example:
- It considered `package.json` changes as `application` (even if only
`package.json` and `yarn.lock` had changed).
- It failed to account for changes in `@angular/*` dependencies, when
using the locally built Angular packages (instead reporting them as
`other`).
- It only looked at the last commit, so it failed to provide accurate
information for multi-commit builds (which are rare, but possible).
For the above reasons (and because there is no straight-forward way of
fixing it), this commit removes the change type from the uploaded data.
If necessary, it is still possible to find the type of changes from the
uploaded info (e.g. extract the associated commits and look at their
changes using git).
PR Close#33987
As part of the `payload-size` npm script in `aio/package.json` (which is
run on CI), the sizes of the angular.io app bundles are checked to
ensure they do not exceed certain limits and are also uploaded to
Firebase to be available for later analysis. The uploaded data include
the type of the changes (dependencies only, application only, or both).
The type of changes is inferred by looking at the files that have
changed inside the `aio/` directory.
When the `payload-size.sh` script was first introduced, the only files
that could affect bundle sizes were inside the `aio/` directory.
Therefore, the script would skip uploading the data and checking the
sizes if no changes were detected inside the `aio/` directory.
However, this assumption stopped being valid over time. For example:
- We started tracking/checking bundle sizes when building the angular.io
app with the locally built Angular packages (which live outside the
`aio/` directory.
- Due to CircleCI limitations, the `CI_COMMIT_RANGE` environment
variable (which is used for determining what files have been affected)
stopped reflecting the whole commit range of the build and only
included the last commit instead.
Based on the above, there were many cases were size data would not be
uploaded to Firebase, even when they may have been affected (because the
affecting changes were outside `aio/` - e.g. in framework packages).
This makes it harder to analyze size regressions, because important
data-points are missing.
Even worse, in these cases, the sizes were not even checked against the
specified limits, thus making it possible for size regressions to go
unnoticed (unless caught by other similar tests).
This commit fixes the `scripts/ci/payload-size.sh` script to always
track and check payload sizes for angular.io bundles.
NOTE: This change will result in more data being recorded (i.e.
recording data when it is not possible for the bundle sizes to
have been affected by the changes). This is still preferable to
failing to record and/or check when sizes could have been
affected.
PR Close#33987
The size diff threshold of 1% has proven to be too lenient for us
to catch size regressions in AIO. Since the AIO main bundle is
between 400-500 KB, a size regression must be between 4-5 KB before
it will cause the tests to fail. As a result, we may merge many
changes with smaller regressions of a few KB before the size test
eventually lets us know that the number has increased. The hope is
that lowering the threshold will help us catch the smaller
regressions during code review and prevent the size tests failing at
a random later time when someone catches the size "hot potato".
PR Close#33969
In efbbae5a4, the `publish_packages_as_artifacts` CircleCI job was
created to publish the build artifacts of PR builds on CI. In a8f4f14bd,
its scope was expanded to also publish build artifacts on non-PR builds.
The published artifacts names are constructed based on the PR number
(e.g. include `-pr12345-`), so on non-PR builds the names do not reflect
the source branch (instead, they include `-prfalse-`).
This commit fixes this by using the current branch name. For example,
artifact names for the `master` or `9.0.x` branch will include
`-master-` and `-9.0.x-` respectively (instead of `-prfalse-`).
(NOTE: For PRs, where branch name is `pull/12345`, the branch name is
transformed to `pr12345`.)
PR Close#33957
We have determined that the upload time cost of remote caching via http is too costly
to be effective for our distributed usage scenario. However this cost is still worth
it for us for Windows CI runs on a full cache hit move from a ~60 minute task to ~10
minutes.
It is worth noting that this remote http-caching is entirely separate from RBE caching
mechanisms.
PR Close#33907
Since `scripts/package-builder.js` gets run from the
`aio` folder sometimes, it is important to set the cwd
for the command that finds the bazel bin directory.
PR Close#33904
In #33823, `scripts/package-builds.sh` (which is used by both
`build-packages-dist.sh` and `build-ivy-npm-packages.sh`) was updated to
use `realpath`. It turns out that `realpath` does not exist on macOS, so
the build scripts do not work there.
In order to fix this (and also reduce the likelihood of introducing
similar issues in the future), this commit changes these bash scripts to
Node.js scripts (using [ShellJS](https://github.com/shelljs/shelljs) for
a cross-platform implementation of Unix shell commands where necessary).
PR Close#33854
In #33823, `scripts/package-builds.sh` (which is used by both
`build-packages-dist.sh` and `build-ivy-npm-packages.sh`) was updated to
use `realpath`. It turns out that `realpath` does not exist on macOS, so
the build scripts do not work there.
This commit fixes it by switching from `realpath $(pwd)/..` to
`dirname $(pwd)`, which does practically the same thing (in this
situation) and works on macOS as well.
PR Close#33843
- Move "Building packages" log above actual work (computing targets).
- Print corrent script name (`build-ivy-npm-packages.sh` would claim to
be `build-packages-dist.sh`).
- Make logs of different operations a little more consistent.
PR Close#33823
Since #33321, Angular packages have been persisted on each build as
CircleCI build artifacts (`.tgz` files), which can be used to install
dependencies on a project (for the purpose of testing or trying out a
change before a PR being merged and without having to build the packages
from source locally).
Previously, only packages published to npm under the `@angular` scope
were persisted as build artifacts.
This commit adds the `zone.js` package to the list of persisted
packages.
Fixes#33686
PR Close#33733
In #33046, internal uses of `zone.js` were switched to reference it
directly from source (built with Bazel) instead of npm. As a result, the
necessary scripts were updated to build `zone.js` as necessary. However,
some `integration/**/debug-test.sh` scripts were missed (apparently
because they are not used on CI, but only locally as helpers for
debugging the integration projects).
This commit updates the `scripts/build-packages-dist.sh` script to also
build `zone.js`, so that other scripts (such as the various
`debug-test.sh` scripts) can use it.
PR Close#33733
The Material units tests job has been disabled with the
typescript upgrade PR since the components repository was
still on an old TypeScript version (due to cyclic dependencies
between the framework and components repository).
Since the components repository has been updated to `9.0.0-next.15`
and now uses the compatible TypeScript version, we can re-enable
ds
Related to this change, we need to ignore the `package.json`
engines when installing the dependencies of the components repo. This
is because the components repo already updated to NodeJS v12 and Yarn
v1.19.1. This is not the case for the CI setup of framework. For now,
since we don't want to change the dev setup (as it slows down
development), we temporarily disable the engines.
Additionally, the material unit tests job now depends on the actual
release packages (not on the ngtsc compiled ones). This is because
the components repo setup relies on NGCC being run. This is also
helpful for validating ngcc against the framework packages.
PR Close#33530
Previously, when one wanted to try out the changes from a PR before it
was merged, they had to check out the PR locally and build the Angular
packages themselves (which is time-consuming and wasteful given that the
packages have already been built on CI).
This commit persists all Angular packages on each build as `.tgz` files,
which can be used to install dependencies on an project (supported by
both [npm][1] and [yarn][2]). In addition to individual `.tgz` files for
each package, a `.tgz` file including all packages is also stored, which
can be used to test the packages locally by overwriting the ones in the
`node_modules/` directory of a project.
CircleCI [build artifacts][3] an be used for longer-term storage of the
outputs of a build and are designed to be useful around the time of the
build, which suits our needs.
[1]: https://docs.npmjs.com/cli/install.html
[2]: https://yarnpkg.com/lang/en/docs/cli/add
[3]: https://circleci.com/docs/2.0/artifacts
PR Close#33321
The `payload-size.sh` script is mainly used on CI to calculate, check
and potentially save (on non-PR builds) the sizes of the bundles for
various apps (including angular.io). If everything goes well (i.e. the
checks pass, meaning that the sizes did not increase above the specified
threshold) nothing is shown in the CI logs.
In some cases, it is useful to be able to see what the sizes were in a
specific build; e.g. for debugging purposes or when investigating a
gradual increase that happened over time. (Some of this info is
available on https://size.angular.io/, but not all.)
Previously, the only way to find out what the sizes were for a specific
build was to checkout the corresponding commit locally and build the
target app, which in turn requires building all Angular packages and can
take some time. Given that the sizes are already calculated on CI, this
was a waste.
This commit makes it easy to find out the bundle sizes for a specific
build/commit by always printing out the calculated sizes (thus making
them show up in the CI logs).
PR Close#33099
Updates the `material-unit-test` job to run tests against
the latest commit of the Angular Components repository.
The components repository updated to rules_nodejs#0.38.2 before Angular Bazel did.
To do this, the `@angular/bazel` v0.38.2 compatibility changes were patched on postinstall.
This now conflicts because we install a `@angular/bazel` version in the `material-unit-tests` job
that already includes these compatibility changes. This would result in the patch being a noop
for which the `patch` command throws an error.
We can remove this once components can install a released version of `@angular/bazel` that is
compatible with `rules_nodejs#0.38.2`.
PR Close#33073
The components repository updated to rules_nodejs#0.38.0 before Angular Bazel did.
To do this, the `@angular/bazel` v0.38.0 compatibility changes were patched on postinstall.
This now conflicts because we install a `@angular/bazel` version in the `material-unit-tests` job
that already includes these compatibility changes. This would result in the patch being a noop
for which the `patch` command throws an error.
We can remove this once components can install a released version of `@angular/bazel` that is
compatible with `rules_nodejs#0.38.0`.
PR Close#32889
Ensures that the `material-unit-tests` job does not accidentally build
subpackages of the `src/dev-app` in the components repo. This is now
an issue because the components repository updated their dev-app
to use Bazel with more individual subpackages. These subpackages
are not excluded by the `--deleted_packages` flag.
For best practice, we should use the exclusion target pattern
instead of the undocumented `--deleted_packages` flag anyway.
PR Close#32485
The commit range that is associated with a CI build is used for a couple
of things (mostly related to payload-size tracking):
- Determine whether a size change was caused by application code or
dependencies (or both).
- Add the messages of the commits associated with the build (and thus
the payload-size change).
NOTE: The commit range is only used on push builds.
Previously, the commit range was computed based on the
`CIRCLE_COMPARE_URL` environment variable. With [CircleCI Pipelines][1]
enabled, `CIRCLE_COMPARE_URL` is no longer available and the commit
range cannot be reliably detected.
This commit switches `CI_COMMIT_RANGE` to only include the last commit.
This can be less accurate in some rare cases, but is true in the
majority of cases (on push builds). Additionally, it stores the CircleCI
build URL in the database along with the payload data, so the relevant
info can be retrieved when needed.
[1]: https://circleci.com/docs/2.0/build-processing
PR Close#32537
This commit changes the Angular compiler (ivy-only) to generate `$localize`
tagged strings for component templates that use `i18n` attributes.
BREAKING CHANGE
Since `$localize` is a global function, it must be included in any applications
that use i18n. This is achieved by importing the `@angular/localize` package
into an appropriate bundle, where it will be executed before the renderer
needs to call `$localize`. For CLI based projects, this is best done in
the `polyfills.ts` file.
```ts
import '@angular/localize';
```
For non-CLI applications this could be added as a script to the index.html
file or another suitable script file.
PR Close#31609
Whenever someone tries to commit (by running `git commit` directly or
indirectly), a `commit-msg` git hook is run to validate the commit
message. If the validation fails, an error message is printed and the
commit is aborted.
Occasionally, people may have written a non-trivial commit message which
could turn out to be invalid (due a small typo for example). In that
case, it is frustrating to "lose" the whole message and have to write it
all over again (from memory). This is frustrating and has happened to me
enough times to finally seek a solution.
Fortunately, it turns out that git stores the last commit message in
`.git/COMMIT_EDITMSG`, so it is easy to get it back (as long as you know
where to look for it). This commit mentions this info in the validation
error to help people that might not know about it.
(This issue is probably mostly relevant for people using git from the
command-line and not through a UI, but it won't hurt in either case.)
PR Close#32420
Ivy snapshots no longer contain metadata.json which breaks the detection logic for
Angular Package Format (APF) in npm/yarn install.
This PR adds ANGULAR_PACAKGE as a magic file to denote APF.
See https://github.com/bazelbuild/rules_nodejs/issues/927
PR Close#31900
Initially the blocklist has been removed because there were
no remaining disabled tests that failed. Also the blocklist
logic didn't work anymore because the `material-unit-tests` CI
job now runs against `angular/components#master` with Bazel.
388578fec9 tried to revert the removal
of the blocklist in favor of a new upcoming breaking change with
HammerJS, but the revert doesn't help since the blocklist still
doesn't work with Bazel.
In order to make the blocklist work with the unit tests running
with Bazel, a PR has been submitted on the components repository.
See: https://github.com/angular/components/pull/16833.
This commit updates the blocklist logic on the framework side to
work with the new logic on the components repo side.
PR Close#32239
Initially when the `material-unit-tests` job got wired up,
Ivy was not really backwards-compatible and a few bugs caused
test failures when running the Angular Material test suites w/ Ivy.
These bugs got fixed progressively and eventually the test
blocklist became empty. At this point we don't want to regress
in the future and the blocklist should never have new items.
Additionally since we switched the unit-tests job to run against
Angular Material `master` with Bazel, the blocklist is no
longer respected. Therefore we can safely remove the blocklist.
PR Close#32138
No longer locks the Material unit tests job to a specific branch, but rather allows
locking to a specific commit from a given branch. This allows us to use the
"master" branch from the `components` repository.
PR Close#31569
Previously, the ivy-2019 branch of the Material (aka components) repo was
used, which contains some changes that were necessary to work with Ivy.
These changes are not longer necessary, as Material's master branch is
fully working with Ivy today. To be up-to-date with recent Material
development and its support for more recent dependencies, e.g. TypeScript,
it is desirable for us to be on a newer version of Material.
This commit moves the Material tests away from the ivy-2019 branch, to a
recent commit on master. We are not targeting the master branch itself,
as that would introduce a moving target into Angular's CI checks, which
is undesirable.
Lastly, the usage of gulp to run Material's tests is changed into using
Bazel, as Material itself is now also built with Bazel.
PR Close#31569
Payload size tracking was temporarily disabled in #31057, due to
`CIRCLE_COMPARE_URL` stopping being available. It turned out this was
related to turning on the new [Pipelines][1] feature, which was required
for testing Windows on CircleCI.
Since then, we have turned `Pipelines` off and got `CIRCLE_COMPARE_URL`
back (e.g. see [build 362971][2]). According to CircleCI, failing to
populate `CIRCLE_COMPARE_URL` with `Pipelines` on is a bug and they are
working on fixing it.
[1]: https://circleci.com/docs/2.0/build-processing/
[2]: https://circleci.com/gh/angular/angular/362971Fixes#31121
PR Close#31138
The `CIRCLE_COMPARE_URL` is not available in builds any more since we
enabled `Pipelines` on CircleCI. We have contected CircleCI, but until
this is solved we cannot get the commit range and thus disabling
uploading of payload size data to avoid broken builds.
PR Close#31057
George Kalpakas
Greg Magolan
Paul Gschwendtner
Igor Minar
atscott
Joey Perrott
Kara Erickson
Pete Bacon Darwin
Keen Yee Liau
JoostK