Luis Filipe Nassif

After the fix:  One last question: now, in situation above, when a string value comes in a new unknown field previously detected as number or date, the field type...

Side note: for built-in fields with types fixed as number or date, a corrupted value was already put in a `property:string` field.

> Maybe after a string value is added all values can be indexed as string and the detected type. Or maybe always indexing unknown fields as string (like approach 2)...

> This would use a little bit more index space like (2) Probably not that much, because this would apply just to unknown fields.

> Or maybe always indexing unknown fields as string (like approach 2) and the detected value type? This would use a little bit more index space like (2), but the...

Tests also have shown misses of Russian, Arabic, Chinese and Korean strings doing indexed searches in unallocated space, they are found in allocated and recovered files, but this is an...

Actually ligatures and libertà (NFD) misses in unallocated are expected, since RawStringParser was not designed to extract the first and the last was a known limitation of current implementation since...

Hi @aberenguel. > 1. The transcription could be enabled by default when processing audio files attached to chats This was already proposed informally by other users, thanks for opening this...

A simple workaround to recorded audios of some chat apps, like WhatsApp, that I already suggested to 2 users in the past, is to define new audio mimetypes, children of...

Would above be enough? What recorded audio name patterns and extensions are more common out there?