Class IterableOfProtosSubject<M extends Message>

  • Type Parameters:
    M - the type of the messages in the Iterable

    public class IterableOfProtosSubject<M extends Message>
    extends IterableSubject
    Truth subject for the iterables of protocol buffers.

    ProtoTruth.assertThat(actual).containsExactly(expected) performs the same assertion as Truth.assertThat(actual).containsExactly(expected). By default, the assertions are strict with respect to repeated field order, missing fields, etc. This behavior can be changed with the configuration methods on this subject, e.g. ProtoTruth.assertThat(actual).ignoringRepeatedFieldOrder().containsExactlyEntriesIn(expected).

    By default, floating-point fields are compared using exact equality, which is probably not what you want if the values are the results of some arithmetic. To check for approximate equality, use usingDoubleTolerance(double), usingFloatTolerance(float), and their per-field equivalents.

    Equality tests, and other methods, may yield slightly different behavior for versions 2 and 3 of Protocol Buffers. If testing protos of multiple versions, make sure you understand the behaviors of default and unknown fields so you don't under or over test.

    • Constructor Detail

      • IterableOfProtosSubject

        protected IterableOfProtosSubject​(FailureMetadata failureMetadata,
                                          @Nullable Iterable<M> messages)
    • Method Detail

      • actualCustomStringRepresentation

        protected String actualCustomStringRepresentation()
        Description copied from class: Subject
        Supplies the direct string representation of the actual value to other methods which may prefix or otherwise position it in an error message. This should only be overridden to provide an improved string representation of the value under test, as it would appear in any given error message, and should not be used for additional prefixing.

        Subjects should override this with care.

        By default, this returns String.ValueOf(getActualValue()).

        Overrides:
        actualCustomStringRepresentation in class IterableSubject
      • displayingDiffsPairedBy

        public IterableOfProtosUsingCorrespondence<M> displayingDiffsPairedBy​(Function<? super M,​?> keyFunction)
        Specifies a way to pair up unexpected and missing elements in the message when an assertion fails. For example:
        
         assertThat(actualFoos)
             .ignoringRepeatedFieldOrder()
             .ignoringFields(Foo.BAR_FIELD_NUMBER)
             .displayingDiffsPairedBy(Foo::getId)
             .containsExactlyElementsIn(expectedFoos);
         

        On assertions where it makes sense to do so, the elements are paired as follows: they are keyed by keyFunction, and if an unexpected element and a missing element have the same non-null key then the they are paired up. (Elements with null keys are not paired.) The failure message will show paired elements together, and a diff will be shown.

        The expected elements given in the assertion should be uniquely keyed by keyFunction. If multiple missing elements have the same key then the pairing will be skipped.

        Useful key functions will have the property that key equality is less strict than the already specified equality rules; i.e. given actual and expected values with keys actualKey and expectedKey, if actual and expected compare equal given the rest of the directives such as ignoringRepeatedFieldOrder and ignoringFields, then it is guaranteed that actualKey is equal to expectedKey, but there are cases where actualKey is equal to expectedKey but the direct comparison fails.

        Note that calling this method makes no difference to whether a test passes or fails, it just improves the message if it fails.

      • ignoringFieldAbsence

        public IterableOfProtosFluentAssertion<M> ignoringFieldAbsence()
        Specifies that the 'has' bit of individual fields should be ignored when comparing for equality.

        For version 2 Protocol Buffers, this setting determines whether two protos with the same value for a field compare equal if one explicitly sets the value, and the other merely implicitly uses the schema-defined default. This setting also determines whether unknown fields should be considered in the comparison. By ignoringFieldAbsence(), unknown fields are ignored, and value-equal fields as specified above are considered equal.

        For version 3 Protocol Buffers, this setting does not affect primitive fields, because their default value is indistinguishable from unset.

      • ignoringFieldAbsenceOfFields

        public IterableOfProtosFluentAssertion<M> ignoringFieldAbsenceOfFields​(int firstFieldNumber,
                                                                               int... rest)
        Specifies that the 'has' bit of these explicitly specified top-level field numbers should be ignored when comparing for equality. Sub-fields must be specified explicitly (via Descriptors.FieldDescriptor) if they are to be ignored as well.

        Use ignoringFieldAbsence() instead to ignore the 'has' bit for all fields.

        See Also:
        for details
      • ignoringRepeatedFieldOrder

        public IterableOfProtosFluentAssertion<M> ignoringRepeatedFieldOrder()
        Specifies that the ordering of repeated fields, at all levels, should be ignored when comparing for equality.

        This setting applies to all repeated fields recursively, but it does not ignore structure. For example, with ignoringRepeatedFieldOrder(), a repeated int32 field bar, set inside a repeated message field foo, the following protos will all compare equal:

        
         message1: {
           foo: {
             bar: 1
             bar: 2
           }
           foo: {
             bar: 3
             bar: 4
           }
         }
        
         message2: {
           foo: {
             bar: 2
             bar: 1
           }
           foo: {
             bar: 4
             bar: 3
           }
         }
        
         message3: {
           foo: {
             bar: 4
             bar: 3
           }
           foo: {
             bar: 2
             bar: 1
           }
         }
         

        However, the following message will compare equal to none of these:

        
         message4: {
           foo: {
             bar: 1
             bar: 3
           }
           foo: {
             bar: 2
             bar: 4
           }
         }
         

        This setting does not apply to map fields, for which field order is always ignored. The serialization order of map fields is undefined, and it may change from runtime to runtime.

      • ignoringRepeatedFieldOrderOfFields

        public IterableOfProtosFluentAssertion<M> ignoringRepeatedFieldOrderOfFields​(int firstFieldNumber,
                                                                                     int... rest)
        Specifies that the ordering of repeated fields for these explicitly specified top-level field numbers should be ignored when comparing for equality. Sub-fields must be specified explicitly (via Descriptors.FieldDescriptor) if their orders are to be ignored as well.

        Use ignoringRepeatedFieldOrder() instead to ignore order for all fields.

        See Also:
        for details.
      • ignoringExtraRepeatedFieldElements

        public IterableOfProtosFluentAssertion<M> ignoringExtraRepeatedFieldElements()
        Specifies that, for all repeated and map fields, any elements in the 'actual' proto which are not found in the 'expected' proto are ignored, with the exception of fields in the expected proto which are empty. To ignore empty repeated fields as well, use comparingExpectedFieldsOnly().

        This rule is applied independently from ignoringRepeatedFieldOrder(). If ignoring repeated field order AND extra repeated field elements, all that is tested is that the expected elements comprise a subset of the actual elements. If not ignoring repeated field order, but still ignoring extra repeated field elements, the actual elements must contain a subsequence that matches the expected elements for the test to pass. (The subsequence rule does not apply to Map fields, which are always compared by key.)

      • usingDoubleTolerance

        public IterableOfProtosFluentAssertion<M> usingDoubleTolerance​(double tolerance)
        Compares double fields as equal if they are both finite and their absolute difference is less than or equal to tolerance.
        Parameters:
        tolerance - A finite, non-negative tolerance.
      • usingDoubleToleranceForFields

        public IterableOfProtosFluentAssertion<M> usingDoubleToleranceForFields​(double tolerance,
                                                                                int firstFieldNumber,
                                                                                int... rest)
        Compares double fields with these explicitly specified top-level field numbers using the provided absolute tolerance.
        Parameters:
        tolerance - A finite, non-negative tolerance.
      • usingDoubleToleranceForFields

        public IterableOfProtosFluentAssertion<M> usingDoubleToleranceForFields​(double tolerance,
                                                                                Iterable<Integer> fieldNumbers)
        Compares double fields with these explicitly specified top-level field numbers using the provided absolute tolerance.
        Parameters:
        tolerance - A finite, non-negative tolerance.
      • usingDoubleToleranceForFieldDescriptors

        public IterableOfProtosFluentAssertion<M> usingDoubleToleranceForFieldDescriptors​(double tolerance,
                                                                                          Iterable<Descriptors.FieldDescriptor> fieldDescriptors)
        Compares double fields with these explicitly specified fields using the provided absolute tolerance.
        Parameters:
        tolerance - A finite, non-negative tolerance.
      • usingFloatTolerance

        public IterableOfProtosFluentAssertion<M> usingFloatTolerance​(float tolerance)
        Compares float fields as equal if they are both finite and their absolute difference is less than or equal to tolerance.
        Parameters:
        tolerance - A finite, non-negative tolerance.
      • usingFloatToleranceForFields

        public IterableOfProtosFluentAssertion<M> usingFloatToleranceForFields​(float tolerance,
                                                                               int firstFieldNumber,
                                                                               int... rest)
        Compares float fields with these explicitly specified top-level field numbers using the provided absolute tolerance.
        Parameters:
        tolerance - A finite, non-negative tolerance.
      • usingFloatToleranceForFields

        public IterableOfProtosFluentAssertion<M> usingFloatToleranceForFields​(float tolerance,
                                                                               Iterable<Integer> fieldNumbers)
        Compares float fields with these explicitly specified top-level field numbers using the provided absolute tolerance.
        Parameters:
        tolerance - A finite, non-negative tolerance.
      • usingFloatToleranceForFieldDescriptors

        public IterableOfProtosFluentAssertion<M> usingFloatToleranceForFieldDescriptors​(float tolerance,
                                                                                         Iterable<Descriptors.FieldDescriptor> fieldDescriptors)
        Compares float fields with these explicitly specified top-level field numbers using the provided absolute tolerance.
        Parameters:
        tolerance - A finite, non-negative tolerance.
      • comparingExpectedFieldsOnly

        public IterableOfProtosFluentAssertion<M> comparingExpectedFieldsOnly()
        Limits the comparison of Protocol buffers to the fields set in the expected proto(s). When multiple protos are specified, the comparison is limited to the union of set fields in all the expected protos.

        The "expected proto(s)" are those passed to the method in IterableOfProtosUsingCorrespondence at the end of the call-chain.

        Fields not set in the expected proto(s) are ignored. In particular, proto3 fields which have their default values are ignored, as these are indistinguishable from unset fields. If you want to assert that a proto3 message has certain fields with default values, you cannot use this method.

      • ignoringFields

        public IterableOfProtosFluentAssertion<M> ignoringFields​(int firstFieldNumber,
                                                                 int... rest)
        Excludes the top-level message fields with the given tag numbers from the comparison.

        This method adds on any previous FieldScope related settings, overriding previous changes to ensure the specified fields are ignored recursively. All sub-fields of these field numbers are ignored, and all sub-messages of type M will also have these field numbers ignored.

        If an invalid field number is supplied, the terminal comparison operation will throw a runtime exception.

      • ignoringFields

        public IterableOfProtosFluentAssertion<M> ignoringFields​(Iterable<Integer> fieldNumbers)
        Excludes the top-level message fields with the given tag numbers from the comparison.

        This method adds on any previous FieldScope related settings, overriding previous changes to ensure the specified fields are ignored recursively. All sub-fields of these field numbers are ignored, and all sub-messages of type M will also have these field numbers ignored.

        If an invalid field number is supplied, the terminal comparison operation will throw a runtime exception.

      • ignoringFieldDescriptors

        public IterableOfProtosFluentAssertion<M> ignoringFieldDescriptors​(Descriptors.FieldDescriptor firstFieldDescriptor,
                                                                           Descriptors.FieldDescriptor... rest)
        Excludes all message fields matching the given Descriptors.FieldDescriptors from the comparison.

        This method adds on any previous FieldScope related settings, overriding previous changes to ensure the specified fields are ignored recursively. All sub-fields of these field descriptors are ignored, no matter where they occur in the tree.

        If a field descriptor which does not, or cannot occur in the proto structure is supplied, it is silently ignored.

      • ignoringFieldDescriptors

        public IterableOfProtosFluentAssertion<M> ignoringFieldDescriptors​(Iterable<Descriptors.FieldDescriptor> fieldDescriptors)
        Excludes all message fields matching the given Descriptors.FieldDescriptors from the comparison.

        This method adds on any previous FieldScope related settings, overriding previous changes to ensure the specified fields are ignored recursively. All sub-fields of these field descriptors are ignored, no matter where they occur in the tree.

        If a field descriptor which does not, or cannot occur in the proto structure is supplied, it is silently ignored.

      • reportingMismatchesOnly

        public IterableOfProtosFluentAssertion<M> reportingMismatchesOnly()
        If set, in the event of a comparison failure, the error message printed will list only those specific fields that did not match between the actual and expected values. Useful for very large protocol buffers.

        This a purely cosmetic setting, and it has no effect on the behavior of the test.

      • unpackingAnyUsing

        public IterableOfProtosFluentAssertion<M> unpackingAnyUsing​(TypeRegistry typeRegistry,
                                                                    ExtensionRegistry extensionRegistry)
        Specifies the TypeRegistry and ExtensionRegistry to use for Any messages.

        To compare the value of an Any message, ProtoTruth looks in the given type registry for a descriptor for the message's type URL:

        • If ProtoTruth finds a descriptor, it unpacks the value and compares it against the expected value, respecting any configuration methods used for the assertion.
        • If ProtoTruth does not find a descriptor (or if the value can't be deserialized with the descriptor), it compares the raw, serialized bytes of the expected and actual values.

        When ProtoTruth unpacks a value, it is parsing a serialized proto. That proto may contain extensions. To look up those extensions, ProtoTruth uses the provided ExtensionRegistry.

        Since:
        1.1