Instead of chasing exports in each FileState library individually, we perform recursive computation, and remember computed export namespaces for visited branches. Cycles require special handling - we don't remember partial namespaces, we compute one full export namespace and so break the cycle; then the rest also can be computed in one pass. This makes Analysis Server a bit faster for some Quick Fixes, and what is more important, is a playground to test the idea of makeing the similar change to _Prelinker. We spend 8% of total analysis time chasing long chains of exports in Flutter. R=brianwilkerson@google.com, devoncarew@google.com Change-Id: Ia45cff9a7fb24aac2e72fe50752274b764462017 Reviewed-on: https://dart-review.googlesource.com/51442 Commit-Queue: Konstantin Shcheglov <scheglov@google.com> Reviewed-by: Brian Wilkerson <brianwilkerson@google.com>
Analyzer for Dart
This package provides a low-level library that performs static analysis of Dart code. It is useful for tool integration and embedding.
End-users should use the dartanalyzer command-line tool to analyze their Dart code.
Integrators that want to add Dart support to their editor should use the Dart Analysis Server. The Analysis Server API Specification is available. If you are adding Dart support to an editor or IDE, please let us know by emailing our list.
Configuring the analyzer
Both dartanalyzer and Dart Analysis Server can be configured with an
analysis_options.yaml file (using an .analysis_options file is deprecated).
This YAML file can control which files and paths are analyzed,
which lints are applied, and more.
If you are embedding the analyzer library in your project, you are responsible for finding the analysis options file, parsing it, and configuring the analyzer.
The analysis options file should live at the root of your project (for example,
next to your pubspec.yaml). Different embedders of analyzer, such as
dartanalyzer or Dart Analysis Server, may choose to find the file in various
different ways. Consult their documentation to learn more.
Here is an example file that instructs the analyzer to ignore two files:
analyzer:
exclude:
- test/_data/p4/lib/lib1.dart
- test/_data/p5/p5.dart
- test/_data/bad*.dart
- test/_brokendata/**
Note that you can use globs, as defined by the glob package.
Here is an example file that enables the analyzer's strong mode:
analyzer:
strong-mode: true
Here is an example file that enables two lint rules:
linter:
rules:
- camel_case_types
- empty_constructor_bodies
Check out all the available Dart lint rules.
You can combine the analyzer section and the linter section into a single
configuration. Here is an example:
analyzer:
exclude:
- test/_data/p4/lib/lib1.dart
linter:
rules:
- camel_case_types
Who uses this library?
Many tools embed this library, such as:
- dartfmt - a formatter for Dart code
- dartdoc - a documentation generator for Dart code
- Dart Analysis Server - a stateful server that supports IDEs and editors
Support
Post issues and feature requests at https://github.com/dart-lang/sdk/issues
Questions and discussions are welcome at the Dart Analyzer Discussion Group.
Background
The APIs in this package are, quite frankly, a mess at the moment. They were originally machine generated by a translator and were based on an earlier Java implementation. Several of the API's still look like their Java predecessors (or worse) rather than clean Dart API's.
In addition, there is currently no clean distinction between public and internal APIs. We plan to address this issue but doing so will, unfortunately, require a large number of breaking changes. We will try to minimize the pain this causes for our clients, but some pain is inevitable.
License
See the LICENSE file.