ISODD_ADD Function (OpenOffice Calc)

• Introduced: First release not established; documented by OpenOffice.org 3.3.
• Supported in: Apache OpenOffice 4.1.16; no final supported release is established.
• Scope: English desktop Calc, Analysis add-in compatibility identity and semicolon-separated formulas.
• Checked: October 14, 2026. One ordinary English-name formula was directly entered and recalculated in 4.1.16 on Linux x86-64, Build 9816. The table contains independent references; the failed generated-ODF import is separate.

ISODD_ADD returns numeric 1 for an odd integer and numeric 0 for an even integer. It belongs to the Analysis add-in and is separately registered from core ISODD. Its implementation tests the least significant bit of a signed integer. Release implementation.

=ISODD_ADD(Number)

• Number: integer whose parity is required.

The add-in declares exactly one signed 32-bit integer input and an integer return. The examples use small exact integers; they do not establish fractional or out-of-range conversion behavior.

These values follow the stated mathematical formulas and the inspected integer-input contract. They are not presented as eight native executions. Results are rounded to twelve decimal places where needed.

Inputs 1, 7, −1 and −7 return 1 in the independent integer check. Inputs 0, 2, −2 and 100 return 0. An odd negative integer is still odd: −7=2×(−4)+1, so its parity is consistent with the positive case.

The result can act as a numeric indicator in an integer classification worksheet. For example, among the eight inputs listed above there are four odd values, so the independent sum of their indicators is 4. That count assumes these are intentional integer inputs, not unvalidated text identifiers.

The source uses nVal & 1 and chooses integer 1 or 0. For the listed integer set, adding ISODD_ADD and ISEVEN_ADD gives 1 for each value. The eight odd indicators sum to 4, while the eight even indicators also sum to 4.

Checking the saved numeric result type, direct English entry and source-mapped stored identity is important in compatibility work. Core parity functions may have their own argument conversions or logical result formatting. Matching a few ordinary integers does not establish equality across text, blank, error or fractional inputs.

The add-in interface is narrower than an unrestricted mathematical integer domain: it declares a signed 32-bit input. No boundary-size integer, fractional, text, blank or error argument was executed. Validate your intended input before relying on an indicator.

Do not call zero odd because its numeric result is zero; here zero is the false indicator. This function returns 1 when the input is odd. Preserve the _ADD name when documenting an existing formula so a later migration does not silently change which native implementation is being discussed.

The directly entered English formula =ISODD_ADD(-7) produced the saved result 1. The English entry and its serialized identity were reviewed together; an ODF canonical spelling alone cannot distinguish every core function from its compatibility add-in.

An earlier generated ODF used fully qualified Analysis service names. All eight import-path attempts for this function displayed #NAME?, despite passing recalculation controls. Those errors do not establish that the English function is unsupported. They also do not count as successful reference-case executions. Do not copy a fully qualified storage name into a normal formula expecting it to behave like the English name.

ISODD, ISEVEN, mod.

• Function-specific release implementation
• Separate English and ODF add-in identities

Only the single direct-entry case above is successful native function evidence in this guide. The remaining table values and explanatory identities are independently calculated. The failed generated-ODF workbook, including its informational attribute-load warning, is retained separately. No original XLSX execution, other locale, platform or release is implied.