Contact

INT in Collabora Office: Round Down Correctly, Including Negative Values

Math Intermediate Collabora Office

Round down to an integer in Collabora Office Calc with negative-number, group-boundary and date-time examples.

Verification status: See the evidence and test scope described in this guide. How formula examples are checked

Quick answer: INT rounds down to an integer. =INT(7.8) is expected to return 7, while =INT(-7.8) returns -8. “Down” means toward the smaller number, not toward zero.

Availability and version history

  • Introduced: documented by Collabora Office 26.04; the original introduction release is not established.
  • Availability: INT is documented in the official 26.04 Calc Help.
  • Scope: desktop Calc documentation; separately deployed Online and mobile versions have not been verified.
  • Checked: October 13, 2026, against the official INT reference. Documentation check; not a native application test. Expected examples are independently calculated.

Syntax and meaning

=INT(number)

The single argument is the number to round down. For a value that is already an integer, no rounding is needed. The official documentation specifically describes negative inputs rounding to the integer below. Keep that rule in mind whenever a calculation can cross zero.

Enter formulas in a blank output cell, press Enter, and keep numeric inputs separate from labels. This guide uses semicolons for multi-argument formulas, consistent with the documented Calc notation.

Example 1: a positive and negative comparison

Enter these numeric values in A2:A7, with Input in A1: 7.8, 7, 0.2, 0, -0.2, -7.8. Enter =INT(A2) in B2 and fill down.

Input Expected INT result
7.8 7
7 7
0.2 0
0 0
-0.2 -1
-7.8 -8

The discontinuity around zero is deliberate. A small negative value does not become zero under INT. This matters when tiny negative adjustments are present in data that otherwise looks nonnegative.

Example 2: count complete packs for nonnegative quantities

Suppose D2 contains 53 units and E2 contains a pack size of 12.

=INT(D2/E2)

Expected result: 4. Four complete packs use 48 units, leaving five. A companion formula makes the remainder explicit:

=D2-E2*INT(D2/E2)

Expected result: 5. This construction assumes a positive pack size and nonnegative stock. If stock can be negative, INT’s downward rule may not represent the business meaning of a negative number of packs. Define that policy before reusing the formula.

Example 3: lower boundary of a numeric band

To place measurements in width-10 intervals, enter 27 in D5 and use:

=10*INT(D5/10)

Expected result: 20. The corresponding interval is at least 20 and less than 30. Changing D5 to -3 gives -10, placing it in the interval from -10 inclusive to 0 exclusive.

Test values immediately below a boundary and exactly on it. With this rule, 29.9 belongs to the band starting at 20, while 30 belongs to the band starting at 30. Avoid labeling both interval ends inclusive in a report.

Advanced example: extracting a day from a serial date-time

Calc stores date-times numerically. For a controlled serial-value demonstration, put 45000.75 in G2 and use:

=INT(G2)
=G2-INT(G2)

Expected results: 45000 and 0.75. The second result is three quarters of a day, or 18 hours. This example deliberately does not assign a calendar date to serial 45000 because a workbook’s date base and formatting must be checked separately.

If you use this with actual dates, preserve the workbook’s date settings and format the integer as a date only after checking the source value. Text that resembles a date is a different input problem.

INT versus other rounding questions

INT answers “which integer is no greater than this value?” It is not a nearest-integer rule. For -7.8, rounding toward zero would give -7, while INT gives -8. For 7.8, ordinary nearest-integer rounding would give 8, while INT gives 7.

Do not replace one operation with another simply because they agree for one positive sample. Include a negative example and a boundary value in any migration test.

Troubleshooting and verification

If a result is one below what you expect, first check whether the input is negative or just below an integer because of preceding arithmetic. Display extra decimals and inspect intermediate formulas before adding an arbitrary adjustment.

A pack-size formula can fail because of a zero divisor even though INT itself has valid syntax. Test the inner division separately. Changing the cell’s display to zero decimals does not reproduce INT; formatting can hide a fractional stored value without changing it.

For native verification, record the exact Collabora build and compare all six inputs above, the pack calculation, and both band examples. No such application execution is claimed here.

Official reference