CLEAN in Collabora Office: Remove Documented Control Characters While Preserving Meaningful Spaces

• Introduced: original Collabora release and launch date are not established by the sources reviewed.
• Availability: documented in Collabora Office 26.04 help, scoped to the desktop-help release.
• Platform scope: online and mobile deployments are not inferred from desktop documentation.
• Checked: October 14, 2026. Direct function documentation and independently modeled text controls; no native Collabora test was performed.

=CLEAN(text)

The argument is the source text, either quoted or referenced from a cell. The reference describes removal broadly but its worked evidence explicitly identifies U+0007 and U+0008. The examples below focus on those characters; they do not certify every invisible Unicode character or every import encoding.

CLEAN is not a general whitespace-normalization, case-conversion or data-validation function. A string can be free of these controls and still contain a wrong identifier or unwanted ordinary spaces.

Put this formula in A2:

=CHAR(7)&"Lot 42"&CHAR(8)

In B2 use =CLEAN(A2). Expected text: Lot 42. The control before the label and the control after it are removed; the internal ordinary space remains.

For visible checks, =LEN(A2) is expected to give 8 and =LEN(B2) gives 6. The difference of two counts removed characters in this controlled example. It does not count every kind of business-data defect.

Construct =" Lot 42 " in D2, then enter =CLEAN(D2) in E2. Expected output is unchanged, including two leading spaces, two internal spaces and two trailing spaces. Its length remains 11.

A cell’s display may make those spaces hard to count. Keep the source formula visible and use LEN rather than concluding from appearance that CLEAN trimmed the value. The existing TRIM guide addresses a different cleanup operation; choose it only when changing ordinary spacing is intended.

In H2 construct ="AB"&CHAR(7)&"CD". Direct =CLEAN(H2) produces expected text ABCD, length 4. The control character was the only separator; removing it joins the two parts.

If that specific character is meant to separate fields, first give it a visible replacement:

=CLEAN(SUBSTITUTE(H2;CHAR(7);"-"))

Expected text: AB-CD, length 5. SUBSTITUTE replaces the known delimiter before CLEAN removes any remaining tested controls. Choose a replacement that matches the receiving data format rather than blindly inserting punctuation.

Construct =CHAR(7)&"A"&CHAR(8)&"B"&CHAR(7) in J2. Use:

=LEN(J2)-LEN(CLEAN(J2))

Expected result: 3. The cleaned text is AB. Applying CLEAN again to that cleaned result leaves AB unchanged in this control.

That repeat check helps identify an incorrectly referenced source cell. It is not a proof that the text is safe to publish, unique or valid in an external system. Keep original and cleaned columns separately until the transformation is approved for the actual dataset.

If a suspicious gap survives, determine which character it is before selecting another formula. Ordinary spaces are explicitly retained, and the short reference is not a complete classification table for every Unicode format character. Avoid importing another spreadsheet application’s exact removal list as if it were a verified Collabora guarantee.

Character appearance, stored length and semantic meaning can disagree. The controls here test removal and preservation deliberately, without relying on screenshots, maximum-length benchmarks or native execution.

• Collabora Office 26.04: CLEAN.
• Editorial testing standards.

The dedicated Collabora function page was reviewed. Expected outputs are independently modeled text controls, not captured results from the application. Extended-character, coercion and deployment-specific behavior retain the limits stated above.