Question

我有两个scala.xml.Elem个对象（实际的，预期的）。我使用的是JUnit 4，但也包含了XMLUnit 1.3。

有没有简单的方法来比较两个对象的相等性，忽略XML中的属性顺序和无关紧要的空格？

我尝试XMLUnit.assertXMLEqual()，但它抱怨类型为scala.xml.Elem。

我知道我可以使用equals或==，但我希望断言在不相等时打印两个值。如果我使用assertTrue(actual.equals(expected))，并且它们不相等，则唯一的输出将是“断言失败”。

Answer 1

如果要与忽略空格的XML Elem对象进行比较，可以使用scala.xml.Utility.trim方法从中删除空格。

scala> val a = <foo>bar</foo>
a: scala.xml.Elem = <foo>bar</foo>

scala> val b = <foo>   bar   </foo>
b: scala.xml.Elem = <foo>   bar   </foo>

scala> a == b
res8: Boolean = false

scala> import scala.xml.Utility.trim
import scala.xml.Utility.trim

scala> trim(a) == trim(b)
res9: Boolean = true

如果使用XML文字，Scala不关心属性的顺序：

scala> val a = <foo first="1" second="2" />
a: scala.xml.Elem = <foo first="1" second="2"></foo>

scala> val b = <foo second="1" first="1"  />
b: scala.xml.Elem = <foo first="1" second="1"></foo>

scala> a == b
res22: Boolean = true

我建议ScalaTest进行单元测试，你有ShouldMatchers：

// Scala repl started with scalatest-1.2.jar in the classpath

scala> val a = <foo>bar</foo>
a: scala.xml.Elem = <foo>bar</foo>

scala> val b = <foo>bar</foo>
b: scala.xml.Elem = <foo>bar</foo>

scala> a should equal(b)

scala> val b = <foo>bar2</foo>
b: scala.xml.Elem = <foo>bar2</foo>

scala> a should equal(b)
org.scalatest.TestFailedException: <foo>bar</foo> did not equal <foo>bar2</foo>
    at org.scalatest.matchers.Matchers$class.newTestFailedException(Matchers.scala:148)
    at org.scalatest.matchers.ShouldMatchers$.newTestFailedException(ShouldMatchers.scala:2329)
    at org.scalatest.matchers.ShouldMatchers$ShouldMethodHelper$.shouldMatcher(ShouldMatchers.scala:871)
    at org.scalatest.matchers.ShouldMatchers$SeqShouldWrapper.should(ShouldMatchers.scala:1724)
    at .<init>(<console>:15)
    at .<clinit>(<console>)
    at RequestResult$.<init>(<console>:9)
    at RequestResult$.<clinit>(<console>)
    at RequestResult$scala_repl_result(<console>)
    at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
    at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
    at sun.reflect.Delega...

Answer 2

使用允许传递自定义消息的assertTrue版本

public static void assertTrue(java.lang.String message,
                              boolean condition)

和（例如）diff生成具有不等于的下降节点的字符串

scala> val xml1 = <person><name>john</name><lastname>smith</lastname></person>
xml1: scala.xml.Elem = <person><name>john</name><lastname>smith</lastname></person>

scala> val xml2 = <person><name>jane</name><lastname>smith</lastname></person>
xml2: scala.xml.Elem = <person><name>jane</name><lastname>smith</lastname></person>

scala> assert(xml1 == xml2, xml1.child diff xml2.child mkString(", "))
java.lang.AssertionError: assertion failed: <name>john</name>
        at scala.Predef$.assert(Predef.scala:91)
        at .<init>(<console>:8)
        at .<clinit>(<console>)

Answer 3

早期的答案对我有帮助，尽管我发现有时候我想检查更大的XML块，并且显示两块XML的故障比较有点难以理解。此方法将首先尝试递归到子元素以比较这些元素，因此如果深层嵌套元素不正确，它将显示更简洁的错误。根据您的XML，这可能无法为您提供足够的上下文来确定其实际失败的位置，但我发现它很有用。

/** Check that the XMLs are the same, ignoring empty text nodes (whitespace). */
private def assertEqual(actual: xml.Node, expected: xml.Node) {

    def recurse(actual: xml.Node, expected: xml.Node) {
        // depth-first checks, to get specific failures
        for ((actualChild, expectedChild) <- actual.child zip expected.child) {
            recurse(actualChild, expectedChild)
        }
        actual should be (expected)
    }

    recurse(scala.xml.Utility.trim(actual), scala.xml.Utility.trim(expected))

}

Answer 4

我修改了@Nick的代码以与JDom2一起使用。在他的代码中，由于zip的工作原理，如果expectedXML的尾随元素不在actualXML中，则测试通过。我修复了该错误，并使尾随元素的比较为可选：

trait XMLTest extends XMLSupport {
  /** Verify that the XMLs are the same, regardless of attribute or element ordering and ignoring whitespace. */
  def assertEqual(actual: Element, expected: Element, ignoreTrailingElements: Boolean=false): Assertion = {
    // depth-first comparison
    def recurse(actual: Element, expected: Element): Assertion = {
      import scala.collection.JavaConverters._
      val actualChildren: Seq[Element] = actual.getChildren.asScala.sortBy(_.getName)
      val expectedChildren: Seq[Element] = expected.getChildren.asScala.sortBy(_.getName)
      (actualChildren zip expectedChildren) foreach { case (actualChild, expectedChild) =>
        recurse(actualChild, expectedChild)
      }
      actual.getName shouldEqual expected.getName
      actual.getTextNormalize shouldEqual expected.getTextNormalize
      actual.getAttributes.asScala.map(_.toString).sorted shouldEqual expected.getAttributes.asScala.map(_.toString).sorted
      if (!ignoreTrailingElements && actualChildren.size < expectedChildren.size) {
        val diff = expectedChildren.drop(actualChildren.size)
        fail("Extra XML children found: " + prettyPrint(diff))
      } else succeed
    }

    recurse(actual, expected)
  }
}

我写了这个特征以混合到测试代码中：

trait XMLSupport {
  import org.jdom2.output.{Format, XMLOutputter}

  def prettyPrint(doc: Document): String = {
    val xmlOutput = new XMLOutputter()
    xmlOutput.setFormat(Format.getPrettyFormat)
    xmlOutput.outputString(doc)
  }

  def prettyPrint(elements: Seq[Element]): String = {
    import scala.collection.JavaConverters._

    val xmlOutput = new XMLOutputter()
    xmlOutput.setFormat(Format.getPrettyFormat)
    xmlOutput.outputString(elements.asJava)
  }
}

我以这种方式调用测试：

class XmlTest extends WordSpec with MustMatchers {
  // test code here
  assertEqual(actualXML.getRootElement, expectedXML.getRootElement, ignoreTrailingElements=true)
}

比较单元测试中的scala.xml.Elem对象

4 个答案: