incomplete configuration causing some loss of whitespace in speech output

[pkra]

I understand if this a won't-fix but I thought I'd share it in any case.

I was flummoxed over the weekend trying to figure out why something like FU ( \\mathbf{t} ) was giving me upper Fupper Uleft parenthesisboldtright parenthesis (note the dropped whitespace).

Turns out my configuration was missing something:

  await sre.setupEngine({
    domain: 'mathspeak',
    style: 'default',
    locale: 'en',
    speech: 'deep',
    structure: true,
    mode: 'sync',
  });

namely, adding modality: 'speech' fixed the issue.

Since the domain only allows this modality, maybe modality could be set automatically. But again, I understand if this isn't worth the trouble.

[zorkow]

Is it possible that you ran SRE with modality: 'braille' beforehand? Than the standard string separator would be ''.

At the moment there is no automation that maps from locales or domains to modalities.

[pkra]

Is it possible that you ran SRE with modality: 'braille' beforehand?

Effectively, yes.

Than the standard string separator would be ''. At the moment there is no automation that maps from locales or domains to modalities.

Right. Maybe incomplete configurations could cause a warning. But that's a different issue.