Hacktoberfest 2026:メンテナが10月に向けて印を付けた、オープンで初心者向けの issue。 Hacktoberfest の issue を見る

2.8.1: ExecJS.compile consistently exits 1 in one environment and 0 elsewhere.

オープン
#105 コメント 4 件 リアクション 2 件 担当者 0 名 GitHub で見る

まだ誰も着手していません。

評価

難易度
4/5
見積もり時間
3〜5日
初心者へのやさしさ
35/100
issue の種類
バグ
明瞭さ
おおむね明確
活発さ
停滞
技術スタック
node.js, ruby
領域
tooling

調査の方向性

Start with the reproduction script and the ExecJS entry points in lib/execjs/module.rb, runtime.rb, and external_runtime.rb, then compare the behavior introduced by a6abf130c721ca6ec05cfc7ed79cf2e87a7dcc35 with 2.8.0. Reproduce the exit-status difference using Node 12.21.0 in the affected RHEL environment and verify that compilation succeeds with exit code 0 without regressing the hanging-process fix.

索引モデルが issue の本文から書いたものです。

説明

I have a few environments I'm testing execjs's compile in combination with uglifier and have a strange problem that happens consistently in one environment but not elsewhere. I have minimized the problem to this commit: a6abf130c721ca6ec05cfc7ed79cf2e87a7dcc35 released in 2.8.1 and can reliably recreate it with this script:

require 'bundler/inline'

gemfile do
  source 'http://rubygems.org'
  gem 'execjs',   "=2.8.1"
  gem 'uglifier', "=3.0.4"
end

require 'uglifier'
puts "Uglifier: #{Uglifier::VERSION}"
puts "ExecJS: #{ExecJS::VERSION}"

puts "Running test..."
ExecJS.compile(File.read(Uglifier::SourceMapPath)); nil
puts $?

In the bad environment, I'm seeing that this exits with 'ok' but with exit code 1:

Uglifier: 3.0.4
ExecJS: 2.8.1
Running test...
Traceback (most recent call last):

        7: from XXX/gems/execjs-2.8.1/lib/execjs/module.rb:27:in `compile'
        6: from XXX/gems/execjs-2.8.1/lib/execjs/runtime.rb:72:in `compile'
        5: from XXX/gems/execjs-2.8.1/lib/execjs/runtime.rb:72:in `new'
        4: from XXX/gems/execjs-2.8.1/lib/execjs/external_runtime.rb:14:in `initialize'
        3: from XXX/gems/execjs-2.8.1/lib/execjs/external_runtime.rb:39:in `exec'
        2: from XXX/gems/execjs-2.8.1/lib/execjs/external_runtime.rb:219:in `exec_runtime'
        1: from (execjs):1

ExecJS::RuntimeError (["ok"])
pid 134 exit 1

When I try to recreate this issue in other environments with 2.8.1, it exits correctly with exit code 0.

I determined it was this change: a6abf130c721ca6ec05cfc7ed79cf2e87a7dcc35 by testing this bad environment with execjs 2.8.0 and 2.7.0 without changing anything else and they both exit with exit code 0 as expected:

Uglifier: 3.0.4
ExecJS: 2.7.0
Running test...
pid 364 exit 0
Uglifier: 3.0.4
ExecJS: 2.8.0
Running test...
pid 249 exit 0

I'm using node 12.21.0 and it works on OSX and RHEL 8.4 and debian. The bad environment is also RHEL 8.4 based with a very similar setup as the working one.

It looks like we're explicitly calling process.exit() to avoid it hanging: c5fd11d7b0338ae7779f69a293964f3440ee3787. The docs say:

Rather than calling process.exit() directly, the code should set the process.exitCode and allow the process to exit naturally by avoiding scheduling any additional work for the event loop

The commit for 2.8.1 is such a minor change. I don't know if the process isn't gracefully terminating or some other low level error is happening but I cannot get more information to determine why this exits 1, which in turn makes execjs believe it failed. I've tried getting more output from node but the following options didn't give me anything extra:

--report-uncaught-exception
--report-on-fatalerror

Do you have any ideas on things I can try? Thanks!

主要言語
Ruby
スター
578
フォーク
286
PR マージ指標
30日以内にマージされた PR はありません

環境構築

はじめの一歩

  1. issue を最後まで読み、次にプロジェクトのコントリビューションガイドを読みます。
  2. 着手することを issue にコメントします — 二人が同じ作業をするのを防げます。
  3. リポジトリをフォークし、ブランチを切って変更します。
  4. issue 番号を参照したプルリクエストを送ります。

rails/execjs のほかの issue

rails/execjs の issue をすべて見る

似ている issue

Ruby の issue をもっと見る

新しい issue をメールで受け取る

初心者向けの GitHub issue を短くまとめたダイジェスト。