When converting an HTML document that contains non-breaking space characters to a PDF profile that supports tagging, in the tag structure of the resulting document, each individual non-breaking space character ends up wrapped in a Span tag containing only that character.

This can lead to undesirable behaviour in some screen readers, which assume that Span tags are semantically meaningful by default, and consequently may pause for instructions every time they encounter such a Span.

This applies both to breaking space characters produced by HTML entitles ( ) as well as by CSS character codes in generated content (\0a00).

Thanks we will take a look at this.

We have implemented this change in the latest build.

Yep, looks good to me - thanks!