TypeTensor - v0.1.0
    Preparing search index...

    Interface ConversionOptions

    Options for controlling type conversion behavior

    interface ConversionOptions {
        allowOverflow?: boolean;
        allowPrecisionLoss?: boolean;
        infinityHandling?: "error" | "clamp";
        nanHandling?: "error" | "zero" | "clamp";
        overflowHandling?: "error" | "clamp" | "wrap";
    }
    Index

    Properties

    allowOverflow? allowPrecisionLoss? infinityHandling? nanHandling? overflowHandling?

    Properties

    allowOverflow?: boolean

    Allow conversions that may overflow/underflow (e.g., int32 to int8)

    false
    allowPrecisionLoss?: boolean

    Allow conversions that may lose precision (e.g., float64 to int32)

    false
    infinityHandling?: "error" | "clamp"

    How to handle Infinity values during conversion

    • 'error': Throw an error (default)
    • 'clamp': Clamp to type bounds
    nanHandling?: "error" | "zero" | "clamp"

    How to handle NaN values during conversion

    • 'error': Throw an error (default)
    • 'zero': Convert to zero
    • 'clamp': Clamp to type bounds
    overflowHandling?: "error" | "clamp" | "wrap"

    How to handle values that exceed the target type's range

    • 'error': Throw an error (default)
    • 'clamp': Clamp to type bounds
    • 'wrap': Use modular arithmetic (JavaScript default)

    Settings

    Member Visibility

    On This Page

    Properties