IntToBin2

function IntToBin2(d: LongInt): string;

Example

procedure ScriptEvent(var Value: variant);
begin
  Value := IntToBin2(513);  // '0000001000000001'
end;

Usage

IntToBin2 formats a LongInt as 8, 16, 24 or 32 binary digits by trimming only complete leading zero bytes.

Parameters

Name	Type	Description
d	LongInt	Signed 32-bit value to represent.

Returns

ASCII 0/1 text with one of these lengths:

Input	Result length	Example
Positive value through 255	8	1 -> 00000001
Positive value through 65,535	8 or 16 as required	513 -> 0000001000000001
Larger positive value	24 or 32 as required	Whole leading zero bytes only are removed.
Zero	32	Thirty-two 0 characters.
Negative value	32	Signed two's-complement bit pattern.

Errors

Only normal string allocation/runtime failures are expected for a valid LongInt; the loop and final result size are fixed.

Usage notes

Use this function for display or byte-grouped bit inspection. For an external binary protocol, define signedness and byte order separately and use actual byte operations rather than treating this display string as serialised data.

Additional Technical Info

IntToBin2 creates a 32-character binary representation of a signed LongInt, then removes only complete eight-character groups that precede the first 1. A non-zero positive value therefore returns 8, 16, 24 or 32 bits. Negative values return all 32 two's-complement bits.

The example returns two byte groups because 513 requires nine significant bits. It is source-reviewed and is not executed by the documentation workflow.

Implementation

The Velox-owned function performs exactly 32 iterations. Each iteration prepends 1 if the current value is odd, otherwise 0, then shifts the signed LongInt right by one. It finds the first 1 in the completed string and deletes a multiple of eight leading characters calculated from that position.

The zero case is a quirk of this trim calculation: no 1 is found, so the computed delete count is zero and all 32 zeroes remain.

Edge cases and quirks

• Despite its name, the function does not produce two bits or accept a width argument. The 2 distinguishes it from IntToBin.
• Leading zeroes are removed only in complete byte groups. The returned text is byte-aligned rather than a minimal binary representation.
• Zero returns 32 bits instead of the otherwise plausible 00000000.
• Negative values retain 32 bits and represent the signed LongInt bit pattern. BinToInt reconstructs a valid 32-character result as the same signed bit pattern, but its permissive handling of non-1 characters means external text still requires syntax validation.
• The returned character groups express bit order from most significant on the left to least significant on the right; they do not define byte order for serialising a multi-byte integer.

Side effects

None outside result-string allocation.

Performance and concurrency

Always performs 32 small prepend operations and a short delete. Work and output are bounded, and no shared mutable state is used.

Related entries

• IntToBin produces an exact caller-selected width for non-negative values.
• BinToInt performs a permissive signed 32-bit positional scan and is not a general inverse for every result here.

Created 2026-07-15