[#BloombergLP-bdldfp-faux_numeric_limits-02]
= xref:BloombergLP.adoc[BloombergLP]::xref:BloombergLP/bdldfp.adoc[bdldfp]::xref:BloombergLP/bdldfp/faux_numeric_limits-01.adoc[faux_numeric_limits]<xref:BloombergLP/bdldfp/Decimal128.adoc[BloombergLP::bdldfp::Decimal128], DUMMY_TYPE>
:relfileprefix: ../../
:mrdocs:

== Synopsis

Declared in `<bdldfp_decimal.h>`

[source,cpp,subs="verbatim,replacements,macros,-callouts"]
----
template<class DUMMY_TYPE>
class xref:BloombergLP/bdldfp/faux_numeric_limits-01.adoc[faux_numeric_limits]<xref:BloombergLP/bdldfp/Decimal128.adoc[BloombergLP::bdldfp::Decimal128], DUMMY_TYPE>;
----

== Static Data Members

[cols="1,4"]
|===
| Name| Description
| xref:BloombergLP/bdldfp/faux_numeric_limits-02/digits.adoc[`digits`] 
| The maximum number of significant digits, in the native (10) radix of the `BloombergLP::bdldfp::Decimal128` type that the type is able to represent. Defined to be 34 by IEEE‐754.
| xref:BloombergLP/bdldfp/faux_numeric_limits-02/digits10.adoc[`digits10`] 
| The maximum number of significant decimal digits that the `BloombergLP::bdldfp::Decimal128` type is able to represent. Defined to be 34 by IEEE‐754.
| xref:BloombergLP/bdldfp/faux_numeric_limits-02/has_denorm.adoc[`has_denorm`]  [.small]#[deprecated]#
| `BloombergLP::bdldfp::Decimal128` may contain denormal values.
| xref:BloombergLP/bdldfp/faux_numeric_limits-02/has_denorm_loss.adoc[`has_denorm_loss`] 
| `BloombergLP::bdldfp::Decimal128` is able to distinguish loss of precision (floating‐point underflow) due to denormalization from other causes.
| xref:BloombergLP/bdldfp/faux_numeric_limits-02/has_infinity.adoc[`has_infinity`] 
| `BloombergLP::bdldfp::Decimal128` can represent infinity.
| xref:BloombergLP/bdldfp/faux_numeric_limits-02/has_quiet_NaN.adoc[`has_quiet_NaN`] 
| `BloombergLP::bdldfp::Decimal128` can be a non‐signaling Not a Number.
| xref:BloombergLP/bdldfp/faux_numeric_limits-02/has_signaling_NaN.adoc[`has_signaling_NaN`] 
| `BloombergLP::bdldfp::Decimal128` can be a signaling Not a Number.
| xref:BloombergLP/bdldfp/faux_numeric_limits-02/is_bounded.adoc[`is_bounded`] 
| Decimal floating‐point types represent a finite set of values.
| xref:BloombergLP/bdldfp/faux_numeric_limits-02/is_exact.adoc[`is_exact`] 
| `BloombergLP::bdldfp::Decimal128` is not an exact type, i.e.: calculations done on the type are not free of rounding errors. Note that integer and possibly rational types may be exact, floating‐point types are never exact.
| xref:BloombergLP/bdldfp/faux_numeric_limits-02/is_iec559.adoc[`is_iec559`] 
| Decimal floating‐point is not covered by the IEC 559 standard.
| xref:BloombergLP/bdldfp/faux_numeric_limits-02/is_integer.adoc[`is_integer`] 
| `BloombergLP::bdldfp::Decimal128` is not an integer type.
| xref:BloombergLP/bdldfp/faux_numeric_limits-02/is_modulo.adoc[`is_modulo`] 
| Decimal floating‐point types do not have modulo representation.
| xref:BloombergLP/bdldfp/faux_numeric_limits-02/is_signed.adoc[`is_signed`] 
| `BloombergLP::bdldfp::Decimal128` is a signed type.
| xref:BloombergLP/bdldfp/faux_numeric_limits-02/is_specialized.adoc[`is_specialized`] 
| The template instance `std::numeric_limits<BloombergLP::bdldfp::Decimal128>` is meaningfully specialized. Also means that `BloombergLP::bdldfp::Decimal128` is a numeric type.
| xref:BloombergLP/bdldfp/faux_numeric_limits-02/max_digits10.adoc[`max_digits10`] 
| The number of significant decimal digits necessary to uniquely represent the significant digits of any `BloombergLP::bdldfp::Decimal128` value. Note that max_digit10 is the same as digits10 for decimal floating‐point values.
| xref:BloombergLP/bdldfp/faux_numeric_limits-02/max_exponent.adoc[`max_exponent`] 
| The highest possible positive exponent for the native base of the `BloombergLP::bdldfp::Decimal128` type that represents a finite value. Defined to be 385 by IEEE‐754.
| xref:BloombergLP/bdldfp/faux_numeric_limits-02/max_exponent10.adoc[`max_exponent10`] 
| The highest possible positive decimal exponent of the `BloombergLP::bdldfp::Decimal128` type that represents a finite value. Defined to be 6145 by IEEE‐754. Note that `max_exponent10` is the same as `max_exponent` for decimal types.
| xref:BloombergLP/bdldfp/faux_numeric_limits-02/max_precision.adoc[`max_precision`] 
| The highest possible precision in the `BloombergLP::bdldfp::Decimal128` type that is large enough to output the smallest non‐zero denormalized value in fixed notation.
| xref:BloombergLP/bdldfp/faux_numeric_limits-02/min_exponent.adoc[`min_exponent`] 
| The lowest possible negative exponent for the native base of the `BloombergLP::bdldfp::Decimal128` type that does not yet represent a denormal number. Defined to be ‐6143 by IEEE‐754.
| xref:BloombergLP/bdldfp/faux_numeric_limits-02/min_exponent10.adoc[`min_exponent10`] 
| The lowest possible negative decimal exponent in the `BloombergLP::bdldfp::Decimal128` type that does not yet represent a denormal number. Defined to be ‐6142 by IEEE‐754. Note that `min_exponent10` is the same as `min_exponent` for decimal types.
| xref:BloombergLP/bdldfp/faux_numeric_limits-02/radix.adoc[`radix`] 
| The base for `BloombergLP::bdldfp::Decimal128` is decimal or 10.
| xref:BloombergLP/bdldfp/faux_numeric_limits-02/round_style.adoc[`round_style`] 
| Decimal floating‐point rounding style is defined to be indeterminate by the C and C++ Decimal TRs.
| xref:BloombergLP/bdldfp/faux_numeric_limits-02/tinyness_before.adoc[`tinyness_before`] 
| Decimal floating‐point types are able to detect if a value is too small to represent as a normalized value before rounding it.
| xref:BloombergLP/bdldfp/faux_numeric_limits-02/traps.adoc[`traps`] 
| Decimal floating‐point types implement traps to report arithmetic exceptions (required by IEEE‐754).
|===



[.small]#Created with https://www.mrdocs.com[MrDocs]#