Back to all checks

Text quality

Unexpected hidden character — Shopify catalog

Finds hidden characters that are not part of normal catalog content.

Needs your reviewAbout this check

What is a control character in catalog text?

Unicode reserves code points for C0/C1 control characters and explains that their semantics generally depend on the application using them. This check focuses on hidden control characters that are not part of normal field content and can introduce invisible differences or unusual display behavior.

That does not mean every control code is corruption. Some controls can have established uses in a protocol or data format. PROVEN should flag the observed case and preserve context for review rather than automatically assigning a cause or declaring the data compromised.

Examples of hidden characters that need review

These conceptual examples are fictional. They do not publish the exact code list, exceptions, or classification logic used by the PROVEN detector.

Product text + <hidden control> + textShow character position and context

If the control has no legitimate role in the field, it can be a normalization candidate; its cause is not inferred automatically.

An imported value contains a control between wordsCompare import source and rendering

Several workflows could introduce it; the finding does not prove which one did.

A control appears at the edge of a valueCheck that removal preserves useful copy

Remediation must preserve visible characters and field intent.

01

What does PROVEN check?

Checks hidden control characters that are not part of normal field content.

02

Why does it matter?

Control characters can introduce invisible differences or unusual display behavior.

Why does this matter to a merchant?

Control characters can be difficult to see in an editor while making two apparently identical values differ in storage or influencing display and processing. This can complicate comparison, export, or content investigation.

PROVEN does not turn the signal into a verdict about encoding, security, or the integrity of the whole catalog. Normalization can be prepared, but the change requires review to confirm the character has no legitimate field-specific role.

03

What can PROVEN do?

PROVEN can prepare normalization of detected control characters; the change requires your review.

How PROVEN handles a control character

The workflow focuses on position, context, and preservation of visible copy before any normalization.

  1. Locate the characterPROVEN shows where the control character occurs relative to useful text without converting it into a conclusion about the source of the issue.
  2. Check contextual legitimacyThe merchant compares field type, data source, and observable behavior to decide whether the control is unexpected or intentional.
  3. Prepare normalizationWhen the control is confirmed as inappropriate, PROVEN can prepare normalization; the result remains review-gated before application.

What should be checked?

Do not remove a character merely because it is invisible. Unicode describes controls as part of broader protocols and contexts, and some can have meaning established by the application processing the text.

Check related findings such as corruption, whitespace, or line breaks without merging them into a single diagnosis. Each detector has its own identity and evidence.

Interpretation limits

  • The detector does not declare every control character invalid in every context.
  • The finding does not automatically prove encoding corruption, attack, exploit, or compromise.
  • PROVEN does not attribute the cause to an import, editor, or external system without separate evidence.
  • Normalization must preserve visible text and any legitimate control required by context.
  • This page does not promise SEO or commercial improvements from removing the character.

Frequently asked questions

Does Unicode define control characters?

Yes. Unicode 17 describes C0/C1 ranges and notes that their semantics are generally determined by the application or protocol using them.

Does a control character mean the text is corrupted?

Not automatically. It is a context-dependent signal. Encoding corruption is a separate check.

Why can it be hard to find manually?

Some controls have no ordinary visible representation, so the stored value can differ without that difference being obvious in an editor.

Can PROVEN normalize it automatically?

This check allows preparation of normalization but requires review before change so legitimate uses are protected.

Sources and independent context

These sources explain Unicode control-code semantics and Shopify content editing. They do not define the complete set of contexts accepted by the check.