Print object fields via reflection when `toString()` is not implemented by asolntsev · Pull Request #3844 · mockito/mockito

asolntsev · 2026-07-12T14:48:32Z

Motivation

In my project, it often happens that value object doesn't have its own toString() method. And comes from a JAR that I cannot easily change.

For such objects, it's highly inconvenient to debug failing tests showing just MyClass@12345 in failure message.

@TimvdLippe @raphw

The problem

In verification failure messages, Mockito uses toString() method to print out the argument object. For objects not having their own toString(), the useless description like "MyClass@12345" was generated (the default Object.toString() implementation).

The solution

Now, ValuePrinter reflects over an object's declared fields when its class doesn't override toString(), thus producing readable description of the object. Too long field values are truncated.

Example of error message:

Argument(s) are different! Wanted:
mock.run(
    Credentials[username=john, password=secret]
);
-> at org.mockitousage.matchers.ReflectivePrintingTest.refEq_printsObjectFieldsUsingReflection(CustomToStringTest.java:53)
Actual invocations have different arguments:
mock.run(
    Credentials[username=usr, password=pwd]
);
-> at org.mockitousage.matchers.ReflectivePrintingTest.refEq_printsObjectFieldsUsingReflection(CustomToStringTest.java:51)

Where

/**
 * Class without own "toString()" method
 */
 class Credentials {
     private String username;
     private String password;
}

Checklist

Read the contributing guide
PR should be motivated, i.e. what does it fix, why, and if relevant how
If possible / relevant include an example in the description, that could help all readers
including project members to get a better picture of the change
Avoid other runtime dependencies
Meaningful commit history ; intention is important please rebase your commit history so that each
commit is meaningful and help the people that will explore a change in 2 years
The pull request follows coding style (run ./gradlew spotlessApply for auto-formatting)
Mention Fixes #<issue number> in the description if relevant
At least one commit should end with Fixes #<issue number> if relevant

codecov-commenter · 2026-07-12T14:52:32Z

Codecov Report

❌ Patch coverage is 84.37500% with 5 lines in your changes missing coverage. Please review.
✅ Project coverage is 86.92%. Comparing base (d8638e1) to head (f785fa9).
⚠️ Report is 1 commits behind head on main.

Files with missing lines	Patch %	Lines
...g/mockito/internal/matchers/text/ValuePrinter.java	83.87%	4 Missing and 1 partial ⚠️

Additional details and impacted files

@@             Coverage Diff              @@
##               main    #3844      +/-   ##
============================================
+ Coverage     86.84%   86.92%   +0.08%     
- Complexity     3044     3063      +19     
============================================
  Files           344      344              
  Lines          9172     9203      +31     
  Branches       1135     1138       +3     
============================================
+ Hits           7965     8000      +35     
+ Misses          917      916       -1     
+ Partials        290      287       -3

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

jselbo · 2026-07-13T14:48:35Z

+        }
+    }
+
+    private static String reflectiveToString(Object value) {


This is probably not nearly as robust as Apache commons ReflectionToStringBuilder but checking the source I don't think it would be trivial to pare down for this repo:
https://github.com/apache/commons-lang/blob/master/src/main/java/org/apache/commons/lang3/builder/ReflectionToStringBuilder.java

I think we don't need all the features of ReflectionToStringBuilder.
I would say, it's not "robust", it just allows configuration: "skip null values" etc. We don't need it.

jselbo · 2026-07-13T14:49:02Z

+    private static Iterable<Field> fields(Class<?> type) {
+        return Arrays.stream(type.getDeclaredFields())
+                .filter(field -> !Modifier.isStatic(field.getModifiers()))
+                .filter(field -> field.getName().indexOf('$') < 0)


use Field.isSynthetic ?

Good point, fixed.

Looks like you need to push a new commit with the change

Ooops :)
Pushed.

jselbo · 2026-07-13T14:50:24Z

+        return Arrays.stream(type.getDeclaredFields())
+                .filter(field -> !Modifier.isStatic(field.getModifiers()))
+                .filter(field -> field.getName().indexOf('$') < 0)
+                .sorted(comparing(Field::getName))


why would the order of type.getDeclaredFields() be non-deterministic? Not sure why sorting is necessary

Java reflection doesn't guarantee the order of fields/methods.

From Class.getDeclaredFields javadoc:

The elements in the returned array are not sorted and are not in any particular order.

jselbo · 2026-07-13T14:53:32Z

+        }
+    }
+
+    private static String truncate(String text) {


what if the different value gets truncated? This wouldn't be very useful then. Ideally this would have some behavior to report only the differing fields, or ensure the difference isn't truncated

While it's theoretically possible, it hopefully doesn't happen too often.
Anyway, it will be better than now. :)

Yes, I totally agree that ideally, Mockito should report the differences in the fields. For too long string, probably report the exact position of diff in the string. Maybe in another PR?..

jselbo · 2026-07-13T14:54:21Z

This seems reasonable at first glance, and low risk if it's only scoped to refEq. Still, I left some comments

asolntsev · 2026-07-13T15:14:51Z

@jselbo It's scoped not only to refEq, but also to eq and same. BUT it's scoped for objects that don't have own toString method. So anyway, it's much better than previously :)

jselbo · 2026-07-14T16:55:33Z

 checksums/
 /.tool-versions
+
+/.claude/settings.local.json


please remove

No problems, I've removed the file.

P.S. Though, it's usually a common practice to ignore this file.

jselbo · 2026-07-14T16:59:03Z

+        assertThatThrownBy(() -> verify(mock).run(expected))
+                .isInstanceOf(AssertionError.class)
+                .hasMessageContaining("Argument(s) are different! Wanted:")
+                .hasMessageContaining("Credentials[password=secret, username=john]")


I wonder if there is a privacy/security concern here. Suppose test data has real sensitive information (like reading some secret keys -- probably bad practice but I'm sure it happens). Now Mockito will print the private fields which would otherwise not be exposed if the class doesn't implement toString(). This could be a bad surprise for someone.

Valid concern.
But I would say it should not be a problem in tests.

Mockito is mostly used in unit-tests (and sometimes integration tests). Usually people use test data, not real sensitive data in tests.

And if they use some real passwords in tests, they should be ready for the risk that these passwords will appear in logs, test reports or CI notifications.

Usually Jenkins, GitHub Actions etc. do mask the secure parameters like password.

In verification failure messages, Mockito uses `toString()` method to print out the argument object. For objects not having their own `toString()`, the useless description like "MyClass@12345" was generated (the default `Object.toString()` implementation). Now, ValuePrinter reflects over an object's declared fields when its class doesn't override `toString()`, thus producing readable description of the object. Too long field values are truncated.

jselbo · 2026-07-15T14:18:10Z

Ok I'm comfortable merging this. Thanks for this improvement!

asolntsev force-pushed the feature/print-object-fields-with-reflection branch 3 times, most recently from 04792af to 8f62553 Compare July 12, 2026 21:17

jselbo reviewed Jul 13, 2026

View reviewed changes

jselbo reviewed Jul 14, 2026

View reviewed changes

asolntsev force-pushed the feature/print-object-fields-with-reflection branch from b5b83a9 to f785fa9 Compare July 15, 2026 08:16

asolntsev requested a review from jselbo July 15, 2026 12:51

jselbo approved these changes Jul 15, 2026

View reviewed changes

jselbo merged commit ee8a330 into mockito:main Jul 15, 2026
20 checks passed

asolntsev deleted the feature/print-object-fields-with-reflection branch July 16, 2026 15:44

This was referenced Sep 26, 2026

Bump org.mockito:mockito-junit-jupiter from 5.23.0 to 5.24.0 JanKocabek/JavaSwingTetris#57

Closed

Bump org.mockito:mockito-core from 5.23.0 to 5.24.0 JanKocabek/JavaSwingTetris#58

Closed

Uh oh!

Conversation

asolntsev commented Jul 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Motivation

The problem

The solution

Checklist

Uh oh!

codecov-commenter commented Jul 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

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!

jselbo commented Jul 13, 2026

Uh oh!

asolntsev commented Jul 13, 2026

Uh oh!

Choose a reason for hiding this comment

Uh oh!

asolntsev Jul 14, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

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!

jselbo commented Jul 15, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

asolntsev commented Jul 12, 2026 •

edited

Loading

codecov-commenter commented Jul 12, 2026 •

edited

Loading

asolntsev Jul 14, 2026 •

edited

Loading