Question

我试图覆盖默认的XMLUnit行为，只报告两个输入之间的第一个差异，其中包含发现的所有差异的（文本）报告。

到目前为止我已经完成了这个：

private static void reportXhtmlDifferences(String expected, String actual) {
  Diff ds = DiffBuilder.compare(Input.fromString(expected))
    .withTest(Input.fromString(actual))
    .checkForSimilar()
    .normalizeWhitespace()
    .ignoreComments()
    .withDocumentBuilderFactory(dbf).build();

  DefaultComparisonFormatter formatter = new DefaultComparisonFormatter();
  if (ds.hasDifferences()) {
    StringBuffer expectedBuffer = new StringBuffer();
    StringBuffer actualBuffer = new StringBuffer();
    for (Difference d: ds.getDifferences()) {
      expectedBuffer.append(formatter.getDetails(d.getComparison().getControlDetails(), null, true));
      expectedBuffer.append("\n----------\n");

      actualBuffer.append(formatter.getDetails(d.getComparison().getTestDetails(), null, true));
      actualBuffer.append("\n----------\n");
    }
    throw new ComparisonFailure("There are HTML differences", expectedBuffer.toString(), actualBuffer.toString());
  }
}

但我不喜欢：

必须在客户端代码中迭代Differences。
进入DefaultComparisonFormatter的内部并使用getDetails ComparisonType调用null。
用线破折号查明差异。

也许这只是来自一种不合理的坏直觉，但我想知道是否有人对这个用例有一些意见。

Answer 1

XMLUnit建议仅打印出差异，请参阅“旧XMLUnit 1.x的DetailedDiff”部分：https://github.com/xmlunit/user-guide/wiki/Migrating-from-XMLUnit-1.x-to-2.x

您的代码如下：

private static void reportXhtmlDifferences(String expected, String actual) {
  Diff ds = DiffBuilder.compare(Input.fromString(expected))
    .withTest(Input.fromString(actual))
    .checkForSimilar()
    .normalizeWhitespace()
    .ignoreComments()
    .withDocumentBuilderFactory(dbf).build();

  if (ds.hasDifferences()) {
    StringBuffer buffer = new StringBuffer();
    for (Difference d: ds.getDifferences()) {
      buffer.append(d.toString());
    }
    throw new RuntimeException("There are HTML differences\n" + buffer.toString());
  }
}

打印XMLUnit中所有差异的惯用方法是什么？

1 个答案: