Overflow exception on Numeric field when using up/down arrows to increment value #1275

New Issue

claunia · 2026-01-29T17:51:27Z

claunia commented

2026-01-29 17:51:27 +00:00

Originally created by @phantomkingx on GitHub (Jun 20, 2024).

Describe the bug
Overflow exception on Numeric field when using up/down arrows to increment value and the value overflows specific datatype.

To Reproduce
Steps to reproduce the behavior:

Go to Radzen Blazor examples webpage and modify first example for Numeric field to use byte datatype for value:
=========================================

@code{
byte value;
}

=========================================

Press the 'Run' button to use the modified code in the example. The example will reload with the Numeric field set to '0'.
Press the down arrow on the Numeric field and there will be a Blazor exception:

blazor.web.js:1 crit: Microsoft.AspNetCore.Components.WebAssembly.Rendering.WebAssemblyRenderer[100]
Unhandled exception rendering component: Value was either too large or too small for an unsigned byte.
System.OverflowException: Value was either too large or too small for an unsigned byte.
at System.Number.ThrowOverflowExceptionByte
at System.Decimal.ToByte(Decimal value)
at System.Convert.ToByte(Decimal value)
at System.Decimal.System.IConvertible.ToByte(IFormatProvider provider)
at System.Convert.ChangeType(Object value, Type conversionType, IFormatProvider provider)
at System.Convert.ChangeType(Object value, Type conversionType)
at Radzen.ConvertType.ChangeType(Object value, Type type)
at Radzen.Blazor.RadzenNumeric1[[System.Byte, System.Private.CoreLib, Version=8.0.0.0, Culture=neutral, PublicKeyToken=7cec85d7bea7798e]].ConvertFromDecimal(Nullable1 input)
at Radzen.Blazor.RadzenNumeric`1.d__12.MoveNext()
at Microsoft.AspNetCore.Components.ComponentBase.CallStateHasChangedOnAsyncCompletion(Task task)
at Microsoft.AspNetCore.Components.RenderTree.Renderer.GetErrorHandledTask(Task taskToHandle, ComponentState owningComponentState)

Expected behavior
It would be ideal to avoid generating an exception on the page. It is possible to use Min/Max properties to avoid the exception for this particular example, however Min/Max uses the decimal datatype and an overflow exception will occur if the field is set to decimal MaxValue (79228162514264337593543950335) and the up arrow is pressed:

Additional context
Although this is somewhat of an edge case I'm not clear if there is currently a way to prevent this exception from happening.

A possible fix to this could be for the Radzen field to catch the overflow exception while ensuring no state has changed on the field. A new callback property 'Overflow' would allow for the option to handle the overflow with a user message or some other action. If we don't want this to be a breaking change incase someone expects the Blazor exception then check if the 'Overflow' property is set and rethrow the original exception if it is not.

Originally created by @phantomkingx on GitHub (Jun 20, 2024). **Describe the bug** Overflow exception on Numeric field when using up/down arrows to increment value and the value overflows specific datatype. **To Reproduce** Steps to reproduce the behavior: 1. Go to Radzen Blazor examples webpage and modify first example for Numeric field to use byte datatype for value: ========================================= <div class="rz-p-12 rz-text-align-center"> <RadzenNumeric @bind-Value=@value InputAttributes="@(new Dictionary<string,object>(){ { "aria-label", "enter value" }})" /> </div> @code{ byte value; } ========================================= 2. Press the 'Run' button to use the modified code in the example. The example will reload with the Numeric field set to '0'. 3. Press the down arrow on the Numeric field and there will be a Blazor exception: blazor.web.js:1 crit: Microsoft.AspNetCore.Components.WebAssembly.Rendering.WebAssemblyRenderer[100] Unhandled exception rendering component: Value was either too large or too small for an unsigned byte. System.OverflowException: Value was either too large or too small for an unsigned byte. at System.Number.ThrowOverflowException[Byte]() at System.Decimal.ToByte(Decimal value) at System.Convert.ToByte(Decimal value) at System.Decimal.System.IConvertible.ToByte(IFormatProvider provider) at System.Convert.ChangeType(Object value, Type conversionType, IFormatProvider provider) at System.Convert.ChangeType(Object value, Type conversionType) at Radzen.ConvertType.ChangeType(Object value, Type type) at Radzen.Blazor.RadzenNumeric`1[[System.Byte, System.Private.CoreLib, Version=8.0.0.0, Culture=neutral, PublicKeyToken=7cec85d7bea7798e]].ConvertFromDecimal(Nullable`1 input) at Radzen.Blazor.RadzenNumeric`1.<UpdateValueWithStep>d__12[[System.Byte, System.Private.CoreLib, Version=8.0.0.0, Culture=neutral, PublicKeyToken=7cec85d7bea7798e]].MoveNext() at Microsoft.AspNetCore.Components.ComponentBase.CallStateHasChangedOnAsyncCompletion(Task task) at Microsoft.AspNetCore.Components.RenderTree.Renderer.GetErrorHandledTask(Task taskToHandle, ComponentState owningComponentState) **Expected behavior** It would be ideal to avoid generating an exception on the page. It is possible to use Min/Max properties to avoid the exception for this particular example, however Min/Max uses the decimal datatype and an overflow exception will occur if the field is set to decimal MaxValue (79228162514264337593543950335) and the up arrow is pressed: **Additional context** Although this is somewhat of an edge case I'm not clear if there is currently a way to prevent this exception from happening. A possible fix to this could be for the Radzen field to catch the overflow exception while ensuring no state has changed on the field. A new callback property 'Overflow' would allow for the option to handle the overflow with a user message or some other action. If we don't want this to be a breaking change incase someone expects the Blazor exception then check if the 'Overflow' property is set and rethrow the original exception if it is not.

claunia closed this issue

2026-01-29 17:51:27 +00:00

claunia commented

2026-01-29 17:51:28 +00:00

@phantomkingx commented on GitHub (Jun 22, 2024):

Thank you for addressing this so swiftly!

After viewing your solution I came up with this to cover all the native number types for .NET 7 and above. I tested this code with float, byte, int, and decimal types.

To summarize added 3 new private methods to handle step up/down for native number types in .NET >= 7.0 and a modification in UpdateValueWithStep() to uses these methods. For .NET < 7.0 or non number types process step up/down as it does currently.

#if NET7_0_OR_GREATER
        /// <summary>
        /// dynamic type wrapper for UpdateValueWithStepNumeric since TValue is not
        /// constrained to a value type
        /// </summary>
        /// <param name="value"></param>
        /// <param name="stepUp"></param>
        /// <param name="step"></param>
        /// <returns></returns>
        private dynamic UpdateValueWithStepDynamic(dynamic value, bool stepUp, decimal step)
        {
            return UpdateValueWithStepNumeric(value, stepUp, step);
        }

        /// <summary>
        /// Process the step up/down while checking for possible overflow errors
        /// and clamping to Min/Max values
        /// </summary>
        /// <typeparam name="TNum"></typeparam>
        /// <param name="value"></param>
        /// <param name="stepUp"></param>
        /// <param name="step"></param>
        /// <returns></returns>
        private TNum UpdateValueWithStepNumeric<TNum>(TNum value, bool stepUp, decimal step) 
            where TNum : struct, System.Numerics.INumber<TNum>, System.Numerics.IMinMaxValue<TNum>
        {
            var valStep = TNum.CreateSaturating(step);
            var valueToUpdate = TNum.CreateSaturating(value);

            if (stepUp && (TNum.MaxValue - valStep) < valueToUpdate) return valueToUpdate;
            if (!stepUp && (TNum.MinValue + valStep) > valueToUpdate) return valueToUpdate;

            var newValue = valueToUpdate + (stepUp ? valStep : -valStep);

            if (Max.HasValue && newValue > TNum.CreateSaturating(Max.Value) || Min.HasValue 
                && newValue < TNum.CreateSaturating(Min.Value) || object.Equals(Value, newValue))
            {
                return valueToUpdate;
            }

            return newValue;
        }

        /// <summary>
        /// Verify if value is native numeric
        /// </summary>
        /// <param name="value"></param>
        /// <returns></returns>
        private bool IsNumericType(object value) => value switch
        {
            sbyte   => true,
            byte    => true,
            short   => true,
            ushort  => true,
            int     => true,
            uint    => true,
            long    => true,
            ulong   => true,
            float   => true,
            double  => true,
            decimal => true,
            _       => false
        };
#endif

        async System.Threading.Tasks.Task UpdateValueWithStep(bool stepUp)
        {
            if (Disabled || ReadOnly)
            {
                return;
            }

            var step = string.IsNullOrEmpty(Step) || Step == "any" ? 1 : decimal.Parse(Step.Replace(",", "."), System.Globalization.CultureInfo.InvariantCulture);

#if NET7_0_OR_GREATER
            if (IsNumericType(Value))
            {
                Value = UpdateValueWithStepDynamic(Value, stepUp, step);
            }
            else
#endif
            {
                var valueToUpdate = ConvertToDecimal(Value);

                var newValue = valueToUpdate + (stepUp ? step : -step);

                if (Max.HasValue && newValue > Max.Value || Min.HasValue && newValue < Min.Value || object.Equals(Value, newValue))
                {
                    return;
                }

                if ((typeof(TValue) == typeof(byte) || typeof(TValue) == typeof(byte?)) && (newValue < 0 || newValue > 255))
                {
                    return;
                }

                Value = ConvertFromDecimal(newValue);
            }

            await ValueChanged.InvokeAsync(Value);
            if (FieldIdentifier.FieldName != null) { EditContext?.NotifyFieldChanged(FieldIdentifier); }
            await Change.InvokeAsync(Value);

            StateHasChanged();
        }

@phantomkingx commented on GitHub (Jun 22, 2024): Thank you for addressing this so swiftly! After viewing your solution I came up with this to cover all the native number types for .NET 7 and above. I tested this code with float, byte, int, and decimal types. To summarize added 3 new private methods to handle step up/down for native number types in .NET >= 7.0 and a modification in UpdateValueWithStep() to uses these methods. For .NET < 7.0 or non number types process step up/down as it does currently. ``` #if NET7_0_OR_GREATER /// <summary> /// dynamic type wrapper for UpdateValueWithStepNumeric since TValue is not /// constrained to a value type /// </summary> /// <param name="value"></param> /// <param name="stepUp"></param> /// <param name="step"></param> /// <returns></returns> private dynamic UpdateValueWithStepDynamic(dynamic value, bool stepUp, decimal step) { return UpdateValueWithStepNumeric(value, stepUp, step); } /// <summary> /// Process the step up/down while checking for possible overflow errors /// and clamping to Min/Max values /// </summary> /// <typeparam name="TNum"></typeparam> /// <param name="value"></param> /// <param name="stepUp"></param> /// <param name="step"></param> /// <returns></returns> private TNum UpdateValueWithStepNumeric<TNum>(TNum value, bool stepUp, decimal step) where TNum : struct, System.Numerics.INumber<TNum>, System.Numerics.IMinMaxValue<TNum> { var valStep = TNum.CreateSaturating(step); var valueToUpdate = TNum.CreateSaturating(value); if (stepUp && (TNum.MaxValue - valStep) < valueToUpdate) return valueToUpdate; if (!stepUp && (TNum.MinValue + valStep) > valueToUpdate) return valueToUpdate; var newValue = valueToUpdate + (stepUp ? valStep : -valStep); if (Max.HasValue && newValue > TNum.CreateSaturating(Max.Value) || Min.HasValue && newValue < TNum.CreateSaturating(Min.Value) || object.Equals(Value, newValue)) { return valueToUpdate; } return newValue; } /// <summary> /// Verify if value is native numeric /// </summary> /// <param name="value"></param> /// <returns></returns> private bool IsNumericType(object value) => value switch { sbyte => true, byte => true, short => true, ushort => true, int => true, uint => true, long => true, ulong => true, float => true, double => true, decimal => true, _ => false }; #endif async System.Threading.Tasks.Task UpdateValueWithStep(bool stepUp) { if (Disabled || ReadOnly) { return; } var step = string.IsNullOrEmpty(Step) || Step == "any" ? 1 : decimal.Parse(Step.Replace(",", "."), System.Globalization.CultureInfo.InvariantCulture); #if NET7_0_OR_GREATER if (IsNumericType(Value)) { Value = UpdateValueWithStepDynamic(Value, stepUp, step); } else #endif { var valueToUpdate = ConvertToDecimal(Value); var newValue = valueToUpdate + (stepUp ? step : -step); if (Max.HasValue && newValue > Max.Value || Min.HasValue && newValue < Min.Value || object.Equals(Value, newValue)) { return; } if ((typeof(TValue) == typeof(byte) || typeof(TValue) == typeof(byte?)) && (newValue < 0 || newValue > 255)) { return; } Value = ConvertFromDecimal(newValue); } await ValueChanged.InvokeAsync(Value); if (FieldIdentifier.FieldName != null) { EditContext?.NotifyFieldChanged(FieldIdentifier); } await Change.InvokeAsync(Value); StateHasChanged(); } ```

claunia commented

2026-01-29 17:51:28 +00:00

@enchev commented on GitHub (Jun 23, 2024):

Thanks @phantomkingx! You can submit pull request!

@enchev commented on GitHub (Jun 23, 2024): Thanks @phantomkingx! You can submit pull request!

claunia commented

2026-01-29 17:51:29 +00:00

@phantomkingx commented on GitHub (Jun 23, 2024):

Yes created PR https://github.com/radzenhq/radzen-blazor/pull/1570

@phantomkingx commented on GitHub (Jun 23, 2024): Yes created PR https://github.com/radzenhq/radzen-blazor/pull/1570

Sign in to join this conversation.