What it checks
Some destinations accept a closed set of characters. The BBC, for broadcast subtitles, allows letters without accents, digits, English punctuation and a few symbols. A line containing any other character produces one issue listing those characters.
When a destination forbids only a few characters in an otherwise open set, FORBIDDEN_CHARACTERS says it more directly.
How it is measured
CueNorm compares the text of each line, character by character, with the allowed set. A character here is a grapheme cluster, compared after Unicode normalization: an accented letter counts as one character whether the source encodes it as one code point or as a letter followed by a combining accent. Spaces count as characters, so the set must contain the space; line breaks do not.
Parameters
| Parameter | Type | Meaning |
|---|---|---|
characters |
text | Every allowed character, written one after the other (e.g. 'abc .,') |
Example
With the BBC broadcast set, the line Świat… właśnie is reported as Line 1 contains character(s) outside the allowed set: Ś … ł ś.
How to fix
- Replace accented letters and typographic characters with their plain equivalents allowed by the destination:
…becomes...,’becomes',“ ”become". - If the text is in a language the set cannot write, check that this is the right profile for these subtitles.