Also, defer to test package for throttling (this will require a test
package update as well).
Also, add a lot more instrumentation to --verbose mode for tests, and
fix minor trivial things here and there, and add error handling in
more places.
Also, refactor how coverage works to be simpler and not use statics.
Previously, it was possible for the test harness to bail
and the test runner to complete before the platform plugin
triggered the collection of coverage data. This fixes the
race condition such that the pending coverage collection
task is recorded immediately after starting the process.
This removes direct file access from within flutter_tools
in favor of using `package:file` via a `FileSystem` that's
accessed via the `ApplicationContext`.
This lays the groundwork for us to be able to easily swap
out the underlying file system when running Flutter tools,
which will be used to provide a record/replay file system,
analogous to what we have for process invocations.
With this change, they're run via instance methods on an object
obtained through the context. This will allow us to substitute
that object in tests with replay/record versions to allow us to
mock out the os-layer in tests.
* Remove the workaround that pinned args to v0.13.6
This reverts most of the changes in commit 6331b6c8b5
* throw exception if exit code is not an integer
* rework command infrastructure to throw ToolExit when non-zero exitCode
* convert commands to return Future<Null>
* cleanup remaining commands to use throwToolExit for non-zero exit code
* remove isUnusual exception message
* add type annotations for updated args package
* refactor _run to runCmd
* replace requiresProjectRoot getter with call to commandValidator
* replace requiresDevice getter with call to findTargetDevice
* trace command requires a debug connection, not a device
* inline androidOnly getter
* rename command methods to verifyTheRunCmd and runCmd
* move common verification into BuildSubCommand
* rename deviceForCommand to device
* rename methods to verifyThenRunCommand and runCommand
This prevents multiple simultaneous runs of the analyzer from stomping
over each other (e.g. multiple runs of 'update-packages'). Certain
long-lived commands (like analyze, run, logs) are exempted once they've
done enough work to be safe from most stomping action.
This still doesn't make us entirely safe from craziness, e.g. if you're
half way through an 'update-packages' run and you call 'git pull', who
knows what state you'll end up in. But there's only so much one can do.
Fixes https://github.com/flutter/flutter/issues/2762
Also, make it clear the screen between results so it's more obvious
what's going on when you have new results (especially when you have
fixed everything).
* refactor the --resident run option into a separate file
* update daemon to run --resident apps
* re-plumbing daemon start
* send app logs
* update tests
* review changes
* fix test runner
* remove PackageMap.createGlobalInstance; rely on the ctor
* review comments
* run pub for analyze and test
* don't run pub when doing repo analysis
* move logic for when to run pub into the commands
* re-write ternary expression
Well, all the easy ones, anyway.
For some reason `// ignore:` isn't working for me so I've disabled
lints that need that. Also disabled those that require a ton of work
(which I'm doing, but not in this PR, to keep it reviewable).
This adds:
- avoid_init_to_null
- library_names
- package_api_docs
- package_names
- package_prefixed_library_names
- prefer_is_not_empty
- sort_constructors_first
- sort_unnamed_constructors_first
- unnecessary_getters_setters
1) Moved basic utility code into base/ directory to make it clear which code
doesn't depend on Flutter-specific knowldge.
2) Move the CommandRunner subclasses into a runner/ directory because these
aren't commands themselves.
Other changes in this patch:
- Make the 'flutter' tool say "Updating flutter tool..." when it calls
pub get, to avoid confusion about what the pub get output is about.
- Make the bash flutter tool call pub get when the revision has
changed. (This was already happening on Windows.)
- Fix a raft of bugs found by the analyzer.
- Fix some style nits in various bits of code that happened to be near
things the analyzer noticed.
- Remove the logic in "flutter test" that would run "pub get", since
upon further reflexion it was determined it didn't work anyway.
We'll probably have to add better diagnostics here and say to run the
updater script.
- Remove the native velocity tracker script, since it was testing code
that has since been removed.
Notes on ignored warnings:
- We ignore warnings in any packages that are not in the Flutter repo or
in the author's current directory.
- We ignore various irrelevant Strong Mode warnings. We still enable
strong mode because even though it's not really relevant to our needs,
it does (more or less accidentally) catch a few things that are
helpful to us.
- We allow CONSTANTS_LIKE_THIS, since we get some of those from other
platforms that we are copying for sanity and consistency.
- We allow one-member abstract classes since we have a number of them
where it's perfectly reasonable.
- We unfortunately still ignore warnings in mojom.dart autogenerated
files. We should really fix those but that's a separate patch.
- We verify the actual source file when we see the 'Name non-constant
identifiers using lowerCamelCase.' lint, to allow one-letter variables
that use capital letters (e.g. for physics expressions) and to allow
multiple-underscore variable names.
- We ignore all errors on lines that contain the following magic
incantation and a "#" character:
// analyzer doesn't like constructor tear-offs
- For all remaining errors, if the line contains a comment of the form
// analyzer says "..."
...then we ignore any errors that have that "..." string in them.
This script runs the Flutter unit tests. By default, the script assumes you
have compiled a SkyShell in an "engine/src" that's a peer to the "flutter"
directory.
