8381882: Convert test/jaxp/javax/xml/jaxp/unittest/stream tests to JUnit by david-beaumont · Pull Request #30643 · openjdk/jdk

david-beaumont · 2026-04-09T07:57:43Z

Convert and tidy-up unit tests in test/jaxp/javax/xml/jaxp/unittest/stream.

It is recommended to set the diff settings in github to hide whitespace (cog-wheel menu).

There are a lot of cases where manual catch blocks with explicit fail are removed to just allow the test to fail idiomatically by throwing the exception normally. This causes thousands of lines to be un-indented (with no other change).

About 200 explicit uses of fail() were either removed or converted to appropriate JUnit asserts.

There are also several non-trivial changes in this PR related to bad tests where, for example, catching and ignoring exceptions allowed the test to pass under all situations.

Progress

Change must be properly reviewed (1 review required, with at least 1 Reviewer)
Change must not contain extraneous whitespace
Commit message must refer to an issue

Issue

JDK-8381882: Convert test/jaxp/javax/xml/jaxp/unittest/stream tests to JUnit (Sub-task - P4)

Reviewing

Using git

Checkout this PR locally:
$ git fetch https://git.openjdk.org/jdk.git pull/30643/head:pull/30643
$ git checkout pull/30643

Update a local copy of the PR:
$ git checkout pull/30643
$ git pull https://git.openjdk.org/jdk.git pull/30643/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 30643

View PR using the GUI difftool:
$ git pr show -t 30643

Using diff file

Download this PR as a diff file:
https://git.openjdk.org/jdk/pull/30643.diff

Using Webrev

Link to Webrev Comment

bridgekeeper · 2026-04-09T07:59:40Z

👋 Welcome back dbeaumont! A progress list of the required criteria for merging this PR into master will be added to the body of your pull request. There are additional pull request commands available for use with this pull request.

openjdk · 2026-04-09T07:59:44Z

❗ This change is not yet ready to be integrated.
See the Progress checklist in the description for automated requirements.

openjdk · 2026-04-09T08:00:38Z

@david-beaumont The following label will be automatically applied to this pull request:

core-libs

When this pull request is ready to be reviewed, an "RFR" email will be sent to the corresponding mailing list. If you would like to change these labels, use the /label pull request command.

david-beaumont

Some "preloaded" comments explaining non-trivial changes.

david-beaumont · 2026-04-09T08:04:25Z

test/jaxp/javax/xml/jaxp/unittest/stream/Bug6489502.java

-            Assert.assertEquals(s1.getEventType(), XMLStreamConstants.START_DOCUMENT);
-            s1.next();
-            s1.next(); // advance to <TITLE>
-            Assert.assertTrue(s1.getLocalName().equals("TITLE"));


A lot of assertions have been rewritten to be idiomatic. Mostly this is handled via IntelliJ analysis, but not always.

david-beaumont · 2026-04-09T08:05:58Z

test/jaxp/javax/xml/jaxp/unittest/stream/Bug6509774.java

 public class Bug6509774 {

    @Test
-    public void test0() {


This test was obviously cut & pasted with alternating blank lines (making it much less readable), so I reformatted it to remove unwanted empty lines.

david-beaumont · 2026-04-09T08:06:24Z

test/jaxp/javax/xml/jaxp/unittest/stream/Bug6688002Test.java

        }
    }

-    public static/* synchronized */XMLStreamReader getReader(InputStream is) throws Exception {


No need for synchronization on factories.

Even worse, based on the @summary it sounds like the whole purpose of this test is to verify that these methods can be called concurrently, so the /* synchronized */ was quite misleading?

david-beaumont · 2026-04-09T08:09:33Z

test/jaxp/javax/xml/jaxp/unittest/stream/CoalesceTest/CoalesceTest.java

 */
 public class CoalesceTest {

-    String countryElementContent = "START India  CS}}}}}} India END";


As well as making these constants, I could switch to UPPER_SNAKE_CASE. Your (reviewer) call.

david-beaumont · 2026-04-09T08:10:20Z

test/jaxp/javax/xml/jaxp/unittest/stream/CoalesceTest/CoalesceTest.java

-                    eventType = streamReader.next();
-                    if (eventType == XMLStreamConstants.CHARACTERS) {
-                        String text = streamReader.getText();
-                        if (!text.equals(descriptionElementContent)) {


None of the debug output is needed if assertEquals() is used.

david-beaumont · 2026-04-09T09:04:48Z

test/jaxp/javax/xml/jaxp/unittest/stream/XMLStreamWriterTest/NamespaceTest.java

    public void testDuplicateNamespaceURI() throws Exception {

        xmlStreamWriter.writeStartDocument();
-        xmlStreamWriter.writeStartElement(new String(""), "localName", new String("nsUri"));


It's not clear why the new String() was being used here. I could revert it if it's felt likely to be important.

david-beaumont · 2026-04-09T09:06:19Z

test/jaxp/javax/xml/jaxp/unittest/stream/XMLStreamWriterTest/NamespaceTest.java

 public class NamespaceTest {
-
-    /** debug output? */
-    private static final boolean DEBUG = true;


Debug output is extensive and unnecessary for unit tests. I've checked that no output was also affecting the test itself.

david-beaumont · 2026-04-09T09:07:07Z

test/jaxp/javax/xml/jaxp/unittest/stream/XMLStreamWriterTest/NamespaceTest.java

-        }
-
-        Assert.assertTrue(actualOutput.equals(EXPECTED_OUTPUT) || actualOutput.equals(EXPECTED_OUTPUT_2), "Expected: " + EXPECTED_OUTPUT + "\n" + "Actual: "
+        assertTrue(actualOutput.equals(EXPECTED_OUTPUT) || actualOutput.equals(EXPECTED_OUTPUT_2), "Expected: " + EXPECTED_OUTPUT + "\n" + "Actual: "


I've left this in, but realistically only one or these sub-clauses will ever be true.

david-beaumont · 2026-04-09T09:10:43Z

test/jaxp/javax/xml/jaxp/unittest/stream/XMLStreamWriterTest/NamespaceTest.java

A lot of removal of redundant catch clauses in this test... sorry!

david-beaumont · 2026-04-09T09:11:39Z

test/jaxp/javax/xml/jaxp/unittest/stream/XMLStreamWriterTest/NullUriDetectionTest.java

        XMLStreamWriter w = xof.createXMLStreamWriter(sw);
        w.writeStartDocument();
        w.writeStartElement("foo", "bar", "zot");
-        w.writeDefaultNamespace(null);


Not actually asserting anything.

mlbridge · 2026-04-09T09:56:41Z

Webrevs

01: Full - Incremental (9f54aaa6)
00: Full (5f6ef34c)

Marcono1234

Hopefully these comments are useful, if not please let me know. They are only suggestions; feel free to ignore them, I am not an OpenJDK member.

(This is the first half of the modified files; I will probably have a look at the other half afterwards.)

Marcono1234 · 2026-04-10T13:14:47Z

test/jaxp/javax/xml/jaxp/unittest/stream/AttributeLocalNameTest/AttributeLocalNameTest.java

- * @library /javax/xml/jaxp/libs /javax/xml/jaxp/unittest
- * @run testng/othervm stream.AttributeLocalNameTest.AttributeLocalNameTest
+ * @library /javax/xml/jaxp/unittest
+ * @run junit/othervm stream.AttributeLocalNameTest.AttributeLocalNameTest


In other JUnit PRs it was suggested to use ${test.main.class} for @run to avoid repetition, would that work here (for all files) as well?

Suggested change

* @run junit/othervm stream.AttributeLocalNameTest.AttributeLocalNameTest

* @run junit/othervm ${test.main.class}

Marcono1234 · 2026-04-10T13:19:47Z

test/jaxp/javax/xml/jaxp/unittest/stream/EntitiesTest/EntityTest.java

            if (!expectedLine.equals(actualLine)) {
                System.out.println("Entityreference expansion failed, line no: " + expectedOutput.getLineNumber());
                System.out.println("Expected: " + expectedLine);
                System.out.println("Actual  : " + actualLine);
                return false;
            }


Would it be more convenient to directly use assertEquals(expectedLine, actualLine, "in line " + expectedOutput.getLineNumber()) or similar here?

Or maybe even do an assertEquals(expected.readAllLines(), actual.readAllLines()), unless that leads to too verbose assertions errors or behaves differently than LineNumberReader.

Marcono1234 · 2026-04-10T13:25:16Z

test/jaxp/javax/xml/jaxp/unittest/stream/EntitiesTest/EntityTest.java

        LineNumberReader expectedOutput = new LineNumberReader(expected);
        LineNumberReader actualOutput = new LineNumberReader(actual);

        while (expectedOutput.ready() && actualOutput.ready()) {


These ready() checks seem problematic:

if actualOutput.ready() is false, the test passes if there are less lines than expected

depending on the type of the given input stream, could ready() return false (even for expected) so no comparison is performed at all?

Maybe it would be better to just have a while (true) loop, which breaks once expectedOutput.readLine() == null && actualOutput.readLine() == null?

Marcono1234 · 2026-04-10T13:33:40Z

test/jaxp/javax/xml/jaxp/unittest/stream/EventsTest/Issue48Test.java

-    private void testListElems(List l, Class expType) {
-        Iterator it = l.iterator();
-        while (it.hasNext()) {
-            Object o = it.next();
-            Assert.assertNotNull(o);
-            Assert.assertTrue(expType.isAssignableFrom(o.getClass()));
-        }
-    }


Removing this made the test slightly less strict because this here also included a !list.contains(null) check which is now not performed anymore?

Marcono1234 · 2026-04-10T13:37:20Z

test/jaxp/javax/xml/jaxp/unittest/stream/EventsTest/Issue58Test.java

-            Assert.assertTrue(loc1.getLineNumber() == 1);
-            Location loc2 = evt.getLocation();
-            System.out.println("Location 2: " + loc2.getLineNumber() + "," + loc2.getColumnNumber());
-            evt = er.nextEvent(); // root


This nextEvent() // root call is gone now it seems. Maybe that was relevant before, to ensure loc1 and loc2 were not affected by it?

(When re-adding it, maybe add a comment to explain this?)

Marcono1234 · 2026-04-10T14:37:33Z

test/jaxp/javax/xml/jaxp/unittest/stream/XMLStreamFilterTest/Bug6481678.java

    @Test
-    public void testReadingNamespace() {
+    public void testReadingNamespace() throws Exception {
        is = new java.io.ByteArrayInputStream(getXML().getBytes());


Add import for java.io.ByteArrayInputStream? (applies to the other test methods here as well)

Marcono1234 · 2026-04-10T14:38:27Z

test/jaxp/javax/xml/jaxp/unittest/stream/XMLStreamFilterTest/Bug6481678.java

+            if (eventType == XMLStreamConstants.START_ELEMENT) {
+                if (sr.getLocalName().equals(rootElement)) {
+                    assertEquals(prefixApple, sr.getNamespacePrefix(0));
+                    assertEquals(namespaceURIApple, sr.getNamespaceURI(0));


For robustness set a foundElement = true and check assertTrue(foundElement) after the loop?

(applies to the other test methods here as well)

Marcono1234 · 2026-04-10T14:41:26Z

test/jaxp/javax/xml/jaxp/unittest/stream/XMLStreamFilterTest/HasNextTest.java

-
-        Assert.assertTrue((hasNext_1 == hasNext_2) && (hasNext_1 == hasNext_3),
-                "XMLStreamReader.hasNext() returns inconsistent values for each subsequent call: " + hasNext_1 + ", " + hasNext_2 + ", " + hasNext_3);
+        boolean hasNext = assertDoesNotThrow(r1::hasNext);


Add comment why assertDoesNotThrow is needed, that is, because callers call checkHasNext within a assertThrows(XMLStreamException, ...)?

Marcono1234 · 2026-04-10T14:56:45Z

test/jaxp/javax/xml/jaxp/unittest/stream/XMLStreamReaderTest/Bug6767322Test.java

-            Assert.fail("Exception occured: " + e.getMessage());
-        }
+        String version = r.getVersion();
+        System.out.println("Bug6767322.xml: " + version);


assertNotNull(version)? (or even assertEquals?)

Marcono1234 · 2026-04-10T14:58:24Z

test/jaxp/javax/xml/jaxp/unittest/stream/XMLStreamReaderTest/Bug6847819Test.java

-            inputFactory.createXMLStreamReader(new StringReader(xml));
-            Assert.fail("Expected an exception for " + msg);
-        } catch (XMLStreamException ex) { // good
-            System.out.println("Expected failure: '" + ex.getMessage() + "' " + "(matching message: '" + msg + "')");


Should check exception message? While it was not done previously, this previous println suggests assertTrue(message.contains("illegal declaration"), "message: " + message)?

Marcono1234 · 2026-04-10T15:10:43Z

test/jaxp/javax/xml/jaxp/unittest/stream/XMLStreamReaderTest/Issue47Test.java

-        }
+        XMLInputFactory xif = XMLInputFactory.newInstance();
+        XMLStreamReader r = xif.createXMLStreamReader(new StringReader(xml));
+        assertTrue(!r.standaloneSet() && !r.isStandalone());


Split into two asserts (assertFalse)? (applies also to the other test methods)

Marcono1234 · 2026-04-10T15:13:48Z

test/jaxp/javax/xml/jaxp/unittest/stream/XMLStreamReaderTest/IssueTracker70.java

                String v;
                v = xsr.getAttributeValue(nameSpace, attrName);


Move assignment to declaration?

String v = xsr.getAttributeValue(nameSpace, attrName);

Marcono1234 · 2026-04-10T15:23:38Z

test/jaxp/javax/xml/jaxp/unittest/stream/XMLStreamReaderTest/NamespaceTest.java

+    public void testRootElementNamespace() throws Exception {
+        XMLInputFactory xif = XMLInputFactory.newInstance();
+        xif.setProperty(XMLInputFactory.IS_NAMESPACE_AWARE, Boolean.TRUE);
+        InputStream is = new java.io.ByteArrayInputStream(getXML().getBytes());


Add import for java.io.ByteArrayInputStream? (also for the other test methods below)

Marcono1234 · 2026-04-10T15:29:23Z

test/jaxp/javax/xml/jaxp/unittest/stream/XMLStreamReaderTest/SupportDTDTest.java

-        reset();
+    public void supportDTD(boolean supportDTD, boolean replaceEntity, int inputType) throws Exception {
        print("\n");
        print((supportDTD ? "SupportDTD=true" : "SupportDTD=false") + ", " + (replaceEntity ? "replaceEntity=true" : "replaceEntity=false"));


Can simplify this?

Suggested change

print((supportDTD ? "SupportDTD=true" : "SupportDTD=false") + ", " + (replaceEntity ? "replaceEntity=true" : "replaceEntity=false"));

print("SupportDTD=" + supportDTD + ", replaceEntity=" + replaceEntity);

Marcono1234 · 2026-04-10T15:35:10Z

test/jaxp/javax/xml/jaxp/unittest/stream/XMLStreamWriterTest/Bug6600882Test.java

+        XMLStreamWriter w = of.createXMLStreamWriter(new ByteArrayOutputStream());
+        XMLStreamWriter w1 = of.createXMLStreamWriter(new ByteArrayOutputStream());
+        assertTrue(w.equals(w) && w.hashCode() == w.hashCode());
+        assertNotEquals(w1, w);


For consistency explicitly check assertFalse(w1.equals(w)) here, since this test is about the equals implementation?

JUnit is not intended for checking equals implementations, and might take shortcuts.

Marcono1234 · 2026-04-10T19:05:10Z

test/jaxp/javax/xml/jaxp/unittest/stream/StreamReaderDelegateTest.java

+                        assertNull(delegate.getAttributeNamespace(0));
+                        assertEquals("http://ns1.java.com", delegate.getAttributeNamespace(1));
+                        assertEquals("ns1", delegate.getAttributePrefix(1));
+                        assertEquals(delegate.getAttributeName(1).toString(), "{" + delegate.getAttributeNamespace(1) + "}" + delegate.getAttributeLocalName(1));


Switch arguments here? Is delegate.getAttributeName(1).toString() rather the 'actual' value?

Marcono1234 · 2026-04-10T19:13:55Z

test/jaxp/javax/xml/jaxp/unittest/stream/StreamReaderDelegateTest.java

-                        Assert.assertTrue(delegate.getAttributeCount() == 5);
-                        Assert.assertTrue(delegate.getAttributeType(1) == "CDATA");
+                        assertEquals(5, delegate.getAttributeCount());
+                        assertSame("CDATA", delegate.getAttributeType(1));


Maybe assertSame("CDATA", ...) is actually too strict here; does it really have to be the same exact string object?

Though it matches the previous assertion behavior.

Marcono1234 · 2026-04-10T19:16:08Z

test/jaxp/javax/xml/jaxp/unittest/stream/StreamReaderDelegateTest.java

            while (delegate.hasNext()) {
                if (delegate.getEventType() == XMLStreamConstants.START_ELEMENT) {
                    if (delegate.getLocalName().equals("name") || delegate.getLocalName().equals("price")) {
                        System.out.println(delegate.getElementText());


Replace println with assertNotNull, if that was the intention? (or add it in addition?)

Marcono1234 · 2026-04-10T19:23:43Z

test/jaxp/javax/xml/jaxp/unittest/util/BaseStAXUT.java

+            assertEquals("Expected getText() and getTextCharacters() to return same value for event of type (" + tokenTypeDesc(sr.getEventType()) + ")",
                    text, text2);


Message comes last, currently it is the 'expected' argument.

But how has this assertion passed then before? Is it even reachable?

Marcono1234 · 2026-04-10T19:28:04Z

test/jaxp/javax/xml/jaxp/unittest/util/BaseStAXUT.java

            byte[] data = content.getBytes("UTF-8");
            return constructStreamReader(f, data);


Avoid try-catch by using StandardCharsets?

Marcono1234 · 2026-04-10T19:51:42Z

It is recommended to set the diff settings in github to hide whitespace (cog-wheel menu).

A tip also to other reviewers: When you open this PR in github.dev (docs) (which opens when you press the . key while viewing a PR), the diff is often better, though the comment UI might not be as nice.

Convert to JUnit

3937114

openjdk bot added the core-libs core-libs-dev@openjdk.org label Apr 9, 2026

pre-review tidying

5f6ef34

david-beaumont commented Apr 9, 2026

View reviewed changes

david-beaumont marked this pull request as ready for review April 9, 2026 09:51

openjdk bot added the rfr Pull request is ready for review label Apr 9, 2026

one more assertion simplification

9f54aaa

Marcono1234 reviewed Apr 10, 2026

View reviewed changes

	* @run junit/othervm stream.AttributeLocalNameTest.AttributeLocalNameTest
	* @run junit/othervm ${test.main.class}

	print((supportDTD ? "SupportDTD=true" : "SupportDTD=false") + ", " + (replaceEntity ? "replaceEntity=true" : "replaceEntity=false"));
	print("SupportDTD=" + supportDTD + ", replaceEntity=" + replaceEntity);

		assertEquals("Expected getText() and getTextCharacters() to return same value for event of type (" + tokenTypeDesc(sr.getEventType()) + ")",
		text, text2);

		byte[] data = content.getBytes("UTF-8");
		return constructStreamReader(f, data);

Conversation

david-beaumont commented Apr 9, 2026 • edited by openjdk bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Progress

Issue

Reviewing

Uh oh!

bridgekeeper bot commented Apr 9, 2026

Uh oh!

openjdk bot commented Apr 9, 2026

Uh oh!

openjdk bot commented Apr 9, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

david-beaumont left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

mlbridge bot commented Apr 9, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Webrevs

Uh oh!

Marcono1234 left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

david-beaumont commented Apr 9, 2026 •

edited by openjdk bot

Loading

openjdk bot commented Apr 9, 2026 •

edited

Loading

mlbridge bot commented Apr 9, 2026 •

edited

Loading