Skip to main content

vxStrToFloatDef

function vxStrToFloatDef(
  const S: string;
  const Default: Extended): Extended;

Example​

procedure ScriptEvent(var Value: variant);
begin
  Value := vxStrToFloatDef('not a number', -1);
end;

Usage​

Use vxStrToFloatDef(Text, Default) to parse a system-locale floating-point value and return Default when ordinary conversion fails. The whole trimmed input must be valid; grouping and currency symbols are not accepted.

Because a successfully parsed value can equal Default, use vxTryStrToFloat when the script must distinguish failure from a legitimate fallback value.

Parameters

NameTypeDescription
Sstring, constLocale-sensitive decimal/scientific text; grouping and currency symbols are invalid.
DefaultExtended, constReturned when syntax, complete-consumption or target-range validation fails.

Returns

The parsed Extended on success; otherwise exactly Default. A successful result equal to Default is indistinguishable from failure by inspecting the return alone.

Behaviour

Accepted ordinary syntax is an optional sign, digits, an optional system decimal separator, and an optional signed decimal exponent with required exponent digits. Leading/trailing ASCII spaces are accepted; other extra characters are not.

Errors

Ordinary syntax and floating-range failures return Default. Locale acquisition, allocation and unexpected runtime exceptions can still propagate.

Usage notes

Use vxTryStrToFloat when a fallback collision matters. Treat a fallback as a deliberate data-quality rule and account explicitly for accepted floating special values.

Additional Technical Info​

vxStrToFloatDef parses complete Windows system-locale decimal/scientific text as Extended. It returns the caller's Default value when ordinary conversion fails. The current script name StrToFloatDef is registered to the same implementation.

The fictional example uses an application-selected sentinel. It is source-reviewed and not executed by the documentation workflow.

Implementation

The formats import registers vxStrToFloatDef and StrToFloatDef against vxFormats.vxStrToFloatDef. The wrapper creates TFormatSettings.Create(LOCALE_SYSTEM_DEFAULT) and calls System.SysUtils.StrToFloatDef(S, Default, AFormatSettings). The installed Delphi terminal runs the same text-to-Extended conversion as StrToFloat and substitutes Default only when that conversion reports False.

Edge cases and quirks

  • A valid parsed value can equal Default, so this function cannot prove whether parsing succeeded.
  • NAN, INF, +INF and -INF are successful Delphi Extended values and do not select Default.
  • The wrong locale decimal character, thousands grouping, a currency symbol, incomplete exponent or overflow selects Default without identifying the reason.
  • Extended precision and range depend on the build target. A source can parse on one target and overflow or round differently on another.
  • Locale is the Windows system default, not a user profile or Delphi global format record. Host migration can change acceptance.
  • StrToFloatDef is an exact Velox scripting alias, not PascalScript's independent built-in converter.

Side effects

No persistent state changes. Local locale and parser records are created for the call.

Performance and concurrency

Parsing is linear in source length. Explicit local settings avoid races on Delphi global formatting state; a later Windows system-locale change can affect later calls.

Related entries

External references

Created 2026-07-15