srawlins

# This script is as dead simple as it can be. It emulates the behavior of the
# `.test` files at https://github.com/flutter/tests/blob/main/registry/
# but assumes that all of the mentioned repos are already cloned, and in the
# current directory. (For example, see in `analyze_assorted_layout_widgets`,
# the first step is `pushd flutter-assorted_layout_widgets` which is relative
# to the current directory.) Limiting the git work to `git fetch` and
# `git checkout` should keep the overhead low.:
#
# The hashes are hard-coded here, so you'll have to hand-update them with the
# registry mentioned above. If you have `flutter/tests` cloned, you can

srawlins

enum X { a, b }
const bool foo = X.a == X.b; // <-- won't compile
void main() {
  print(foo);
}

srawlins

import 'dart:math' as math;

// Poorly named wrapper for math.Random
class Random {
  math.Random inner;

  Random(this.inner);
}

void main() {

srawlins

void main() {
  var a = new AImpl();
  var b = new B(a);
  print("CALLING W/O NAMED ARG");
  b.a();
  print("CALLING W/ NAMED ARG");
  b.a(followLinks: false);
}

abstract class A {

srawlins

import 'dart:async';
import 'dart:collection';

void main() async {
  print('Function arguments that cannot be null');
  // dart:async
  //tryy('runZoned(null)', () => runZoned(null));
  //tryy('scheduleMicrotask(null)', () => scheduleMicrotask(null));
  await trya('Future.doWhile(null)', () => Future.doWhile(null));
  await trya('Future.forEach(_, null)', () => Future.forEach([1], null));

srawlins

# Ring the bell if any background window rang a bell
set -g bell-action any

# Default termtype. If the rcfile sets $TERM, that overrides this value.
set -g default-terminal screen-256color

# Keep your finger on ctrl, or don't
bind-key ^D detach-client

# Create splits and vertical splits

srawlins

Performance of new DEAD_CODE analyses

Before this change, the DEAD_CODE analysis only alerted on statements that follow:

• a return
• an unlabeled break
• an unlabeled continue

My change adds the following to that list:

srawlins

Performance of new UNUSED_SHOWN_NAME warning

Examining imports, and now the shown names on imports, is an expensive operation, as it potentially requires examining all of the elements in a library to discover if an import, or a shown name on an import, is used. So I wanted to test whether the new UNUSED_SHOWN_NAME warning significantly increases analysis time.

It obviously should increase analysis time by some amount, as it is doing extra work that wasn't being done before, but I wanted to measure this increase

srawlins

CommonMark-like spec text for tables

Motivation

Some very popular Markdown implementations include support for a table syntax (pandoc, PHP Markdown Extra, and GitHub-Flavored Markdown, to

srawlins

# Through vertical-arrows (first 10 ifs)

 %self      total      self      wait     child     calls  name
  7.60      1.284     1.284     0.000     0.000    25396   String#split
  5.60      0.946     0.946     0.000     0.000   117606   Array#flatten
  4.08      0.690     0.690     0.000     0.000   487582   Regexp#match
  4.03      1.053     0.680     0.000     0.372   393708   SCSSLint::ControlCommentProcessor#retrieve_comment_text
  3.53      0.597     0.597     0.000     0.000   522139   Kernel#is_a?
  3.42      0.860     0.578     0.000     0.282   396943   Enumerable#any?
  3.21      0.542     0.542     0.000     0.000   492675   Kernel#class